ULTRAFAST DYNAMICS OF ENERGY AND ELECTRON TRANSFER IN DNA-PHOTOLYASE
DISSERTATION
Presented in Partial Fulfillment of the Requirements for The Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By
Chaitanya Saxena, M.Phil.
******
The Ohio State University 2007
Dissertation Committee:
Professor Dongping Zhong, Advisor Approved by Professor Richard P. Swenson
Professor Zucai Suo
Professor Michael G. Poirier
Advisor Biophysics Graduate Studies Program
ABSTRACT
One of the detrimental effects of UV radiation on the biosphere is the formation
of cyclobutane pyrimidine dimers (Pyr<>Pyr) between two adjacent thymine bases in
DNA. Pyr<>Pyr dimers can not be repaired by normal DNA repair machinery and may
result in gene mutation or cell death. Photolyase, a photoenzyme harnesses blue or near-
UV light energy to cleave the cyclobutane ring of the Pyr<>Pyr and thus protects against
the harmful effects of UV radiation. In the proposed hypothesis for the catalysis, the enzyme binds a Pyr<>Pyr in DNA, independent of light. The photoantenna, a photolyase cofactor methenyltetrahydrofolate (MTHF) harvests a UV/blue-light photon, and transfers the excitation energy (dipole-dipole interaction) to another photolyase cofactor, a fully reduced flavin (FADH−). Excited FADH−* then transfers an electron to the
Pyr<>Pyr, which consequently splits the Pyr<>Pyr into two pyrimidine moieties and
hence repairs the damaged DNA. As proposed, the repair cycle ends when the excess
electron from the repaired pyrimidine moieties is transferred back to the nascent-formed
neutral FADH• species and regenerates the active FADH− form. The complex
mechanism of energy and electron transfer in photolyase enzyme involved in performing
its DNA repair function was investigated using femtosecond-resolved fluorescence up-
conversion and transient absorption methods. Under physiological conditions, the
excitation energy transfer from the antenna molecule MTHF to the FADH− occurs in 292 ii ps, but it takes 19 ps to the in vitro oxidized neutral cofactor FADH•. The orientation factors were found to be 0.11 for the MTHF- FADH− pair and 0.28 for MTHF- FADH•, unfavorable for energy transfer, indicating the existing structural constraints probably placed by three functional binding sites. The photoreduction of the neutral FADH• to the
catalytically active cofactor FADH− was revealed to evolve along two electron-transfer
pathways: one is along a tryptophan triad with the initial electron hop in 10 ps; the other
route starts with an initial electron separation in 40 ps through the neighboring
phenylalanine followed by either tunneling along an α-helix or hopping through the
tryptophan triad again. Ultrafast libration/rotation motions of local protein residues and
trapped water molecules at the active site were observed to initially occur in ~2 ps. These
ultrafast ordered-water motions are critical to stabilizing the photoreduction product
FADH− instantaneously to prevent fast charge recombination.
Monitoring the catalytic processes we observed direct electron transfer from the
FADH−* to the Pyr<>Pyr in 170 ps and back electron transfer from the repaired thymines in 560 ps. Both reactions are strongly modulated by active-site solvation to achieve maximum repair efficiency. These results show that the photocycle of DNA repair by photolyase is through a radical mechanism and completed on subnanosecond time scale at the dynamic active site, with no net change in the redox state of the flavin cofactor.
The photophysics of FADH− cofactor was studied in aqueous solution. Dramatic shortening of the excited state lifetime of FADH− in aqueous solution compare to its iii lifetime in protein environment compelled us to propose that enzyme photolyase also modulates photophysical properties of the flavin cofactor to perform the essential biological function of electron transfer to repair damaged DNA.
iv
Dedicated to Family, Friends,
And Gurus
v ACKNOWLEDGMENTS
I would like to thank my advisor, Prof. Dongping Zhong, for his unwavering support and guidance throughout my graduate education. His guidance and insight were instrumental in the success of the research presented in this dissertation.
I am grateful to Prof. Aziz Sancar, University of North Carolina (Chapel Hill) who conceded his contagious motivation for pursuing DNA-repair research and shared more than two decades of research experience in planning the key experiments. I am thankful to Lijuan Wang and Ya-Ting Kao who closely remain involved in the present work and provided help and insight in performing and analyzing the experiments.
I would like to thank Prof. Richard P. Swenson and Prof. Russ Hill for helping me in developing understanding of flavin biochemistry and photochemistry. I would like to extend my thanks to Prof. Dehua Pei, of the Ohio State University who willingly shared his resources for our bacterial growth needs during initial time of this work. I am also thankful to Kari Green-Church and Chunhua Yuan of the Ohio State University Campus
Chemical Instrument Center (CCIC) for the technical support that they provided me for the characterization of photolyase substrates using Mass-spectrometry and Nuclear
vi Magnetic Resonance techniques. Thanks are also due for Prof. Michael G. Poirier who shared his resources for purification of the photolyase substrates.
The Ohio State Biophysics Graduate Studies Program and The Department of
Physics, The Ohio State University deserves special thanks for allowing me to use all the available administrative and technical facilities to perform this work and advance my research career. I would like to thank all the members of Zhong group for their stimulating discussions and for their help throughout the research.
Very special thanks to my wife Dr. Yanping Yan who provided moral support, encouragement and professional help throughout my doctoral work.
This research was supported by the funds available to Prof. Dongping Zhong from
‘Department of Physics’ and ‘National Institute of Health’ and to research collaborator
Prof. Aziz Sancar from ‘National Institute of Health’. Partial support to this research also comes from ‘American Heart Association’ in the form of Predoctoral fellowship.
vii VITA
August 28, 1975, ______Born-Dewas, India
1993-1996 ______B.Sc. Chemistry (honors), Devi Ahilya University, Indore India
1996-1998 ______M.Sc. Pharmaceutical Chemistry, Devi Ahilya University, Indore India
1998-1999 ______‘Production Chemist’, Lupin Laboratories Pvt. Ltd. Bhopal, India
1999-2001 ______M.Phil. Biophysics, NIMHANS, Bangalore , India
2001-present______Graduate Research Associate, The Ohio State Universty, Coumbus, USA
PUBLICATIONS
C. Saxena, Y.-T. Kao, L. Wang, A. Sancar and D. Zhong “Direct observation of DNA repair by photolyase” in Femtochemistry VII: Fundamental Ultrafast Processes in
Chemistry, Physics, and Biology, A. W. Castleman, Jr., M. L. Kimble, eds., Elsevier:
Amsterdam (2006) p407.
Y.-T. Kao*, C. Saxena*, L. Wang, A. Sancar and D. Zhong “Direct observation of thymine dimer repair in DNA by photolyase” Proc. Natl. Acad. Sci. USA 102:16128-
16132 (2005). (*authors contributed equally to this work)
viii C. Saxena, H. Wang, I. H. Kavakli, A. Sancar and D. Zhong “Ultrafast dynamics of
resonance energy transfer in cryptochrome” J. Am. Chem. Soc. 127:7984-7985 (2005).
H. Wang, C. Saxena, D. Quan, A. Sancar and D. Zhong “Femtosecond dynamics of
flavin cofactor in DNA photolyase: Radical reduction, local solvation, and charge
recombination” J. Phys. Chem. B. 109:1329-1333 (2005).
C. Saxena, A. Sancar and D. Zhong “Femtosecond dynamics of DNA photolyase:
Energy transfer of antenna initiation and electron transfer of cofactor reduction” J. Phys.
Chem. B 108:18026-18033 (2004).
FIELDS OF STUDY
Major Field: Biophysics
ix TABLE OF CONTENTS
Page
Abstract ……………………………………………………………………………………………………...ii
Dedications ………………………………………………………………………………………………….v
Acknowledgments …………………………………………………………………………………………..vi
Vita...……………………………………………………………………………………………………….viii
List of Tables ..……………………………………………………………………………………………...xv
List of Figures……………………………………………………………………………………………...xvi
Abbreviation..………………………………………………………………………………………………xix
Chapter 1
Introduction...... 1
1.1 Enzyme dynamics...... 1
1.1.1 Enzyme catalysis and enzyme dynamics ...... 2
1.1.2 Dynamic fluctuations in enzyme ...... 3
1.1.3 Time-resolved approaches to characterize enzyme dynamics ...... 5
1.1.4 Applications ...... 10
1.2 DNA photolyase...... 11
1.2.1 Biological function...... 11
1.2.2 Significance...... 12
x 1.2.3 The photolyase enzyme...... 13
1.2.4 The damaged DNA substrate...... 17
1.2.5 Molecular mechanism of DNA-photolyase ...... 19
1.3 Research project...... 23
1.3.1 Dynamics of resonance energy transfer in photolyase ...... 24
1.3.2 Dynamics of intraprotein electron transfer in photoreduction of
photolyase ...... 25
1.3.3 Dynamics of DNA repair in photolyase...... 26
1.3.4 Dynamics of reduced flavin and flavin model systems involved in repair
...... 26
1.4 The value added in enzyme dynamics...... 27
Chapter 2
Experimental Methodologies ...... 37
2.1 Femtosecond laser spectroscopy...... 37
2.1.1 Fluorescence up-conversion technique...... 38
2.1.2 Transient absorption technique...... 39
2.2 DNA photolyase enzyme, CPD substrates and reduced flavin species .... 41
2.2.1 Purification and characterization of photolyase...... 41
xi 2.2.2 Photoreduction of flavin cofactor & photodecomposition of folate
cofactor ...... 44
2.2.3 Preparation and characterization of CPD substrates...... 46
2.2.4 Photorepair of damaged DNA (CPDs) ...... 46
2.2.5 Preparation and characterization of reduced flavin species...... 48
Chapter 3
Resonance Energy Transfer from the Folate to the Flavin Cofactor...... 56
3.1 Resonance energy transfer in photolyase ...... 56
3.1.1 Introduction...... 56
3.1.2 Dynamics of energy transfer from MTHF to FADH (neutral radical) .. 57
3.1.3 Dynamics of energy transfer from MTHF to FADH– (anionic reduced
radical) ...... 60
3.2 Comparative study of resonance energy transfer in cryptochrome...... 61
3.2.1 Introduction...... 61
3.2.2 Resonance energy transfer from MTHF to flavin cofactor...... 62
3.3 Conclusions...... 65
Chapter 4
Dynamics of the Flavin Cofactor...... 75
4.1 Introduction...... 75 xii 4.2 Dynamics of the flavin cofactor reduction...... 76
4.3 Solvation dynamics at the active site...... 81
4.4 Dynamics of the back electron transfer ...... 83
4.5 Conclusions...... 84
Chapter 5
Dynamics of Damaged DNA Repair...... 93
5.1 Introduction...... 93
5.2 Active-site solvation dynamics...... 94
5.3 Capture of the FADH• intermediate...... 95
5.4 Electron return and catalytic photocycle...... 96
5.5 van der Waals contacts and Adenine mediation ...... 97
5.6 Dynamic control and repair efficiency...... 99
5.6 Conclusions...... 100
Chapter 6
Dynamics of Reduced Flavins...... 108
6.1 Introduction...... 108
6.2 Time-resolved fluorescence studies of the reduced flavins ...... 110
xiii 6.3 Time-resolved transient absorption studies of the reduced flavins...... 114
6.4 Conclusions...... 117
Chapter 7
Epilogue ...... 124
LIST OF REFERENCES...... 131
Appendix A ...... 147
DNA-Photolyase Protein Purification Protocol
Appendix B ...... 153
Photoreduction and Photodecomposition of Cofactors in Photolyase
Appendix C...... 156
Preparation of Photolyase Substrates
Appendix D...... 159
Preparation of Reduced Flavin Species
xiv LIST OF TABLES
Table 1.1 Time scales, amplitudes and type of motions in proteins ...... 29
Table 1.2 Biochemical properties of E. coli. CPD photolyase...... 30
Table 1.3 Photophysical properties of E. coli. CPD photolyase ...... 31
Table 4.1 Time constants (ps) and relative amplitudes (Dynamics of flavin cofactor) ...... 86
Table 5.1 Fitting results of the fluorescence transients shown in figure 5.2B ...... 101
Table 6.1 Time constants (ps) and relative amplitudes (Dynamics of reduced flavins) ...... 119
xv LIST OF FIGURES
Figure 1.1 Absorption spectra of E. coli photolyase in different states ...... 32
Figure 1.2 X-ray crystal structure of DNA photolyase from E. coli...... 33
Figure 1.3 Chemical structure of cis-syn thymine dimer ...... 34
Figure 1.4 Schematic representation of the catalytic mechanism of photolyase ...... 35
Figure 1.5 Absolute absorption and action spectra of folate class photolyases ...... 36
Figure 2.1 Layout of femtosecond pump-probe methods ...... 49
Figure 2.2 Schematic representation of fluorescence up-conversion technique ...... 50
Figure 2.3 Schematic representation of the experimental setup with both the fluorescence up- conversion and the transient absorption configurations...... 51
Figure 2.4 Typical photoreduction profile of flavin neutral radical in photolyase ...... 52
Figure 2.5 Typical photodecomposition profile of folate cofactor in photolyase...... 53
Figure 2.6 Typical preparation of cyclobutane pyrimidine dimers (CPDs)...... 54
Figure 2.7 Typical steady-state DNA repair ...... 55
Figure 3.1 The absorption and emission spectra of various forms of photolyase enzyme ...... 68
Figure 3.2 Fluorescence transient profile of MTHF in photolyase and the cofactor flavin is present in the neutral radical form ...... 69
Figure 3.3 Femtosecond-resolved fluorescence transients of MTHF in photolyase with and without O2 ...... 70
Figure 3.4 Femtosecond-resolved fluorescence transient profile of MTHF in photolyase and the flavin cofactor is present in the anionic reduced form...... 71
xvi Figure 3.5 Absorption and emission spectra of V. cholerae cryptochrome ...... 72
Figure 3.6 Femtosecond-resolved fluorescence transients of the MTHF* decay in VcCRY1 ...... 73
Figure 3.7 Schematic representation of the molecular mechanism and the measured resonance energy transfer dynamics ...... 74
Figure 4.1 Time-resolved dynamics of the flavin neutral radical ...... 87
Figure 4.2 Potential electron transfer pathways for photoreduction of the flavin neutral radical cofactor ...... 88
Figure 4.3 Femtosecond-resolved transient absorption measurements with excitation at 400 nm 89
Figure 4.4 Excited state lifetime of reduced flavin cofactor...... 90
Figure 4.5 Configuration of the catalytic flavin cofactor from the X-ray crystal structure of E. coli photolyase ...... 91
Figure 4.6 Femtosecond-resolved absorption transients probed at 500 nm with and without O2.92
Figure 5.1 Repair of damaged DNA by photolyase...... 102
Figure 5.2 Solvation dynamics at the active site of photolyase ...... 103
Figure 5.3 Ultrafast fluorescence transient of a photolyase in the absence and presence of substrate ...... 104
Figure 5.4 Determination of forward and back electron transfer in photolyase photocycle by ultrafast absorption spectroscopy...... 105
Figure 5.5 Electron transfer dynamics in the presence of different substrates ...... 106
Figure 5.6 Evolution of catalytic reactions of DNA repair by photolyase along the reaction coordinate...... 107
Figure 6.1 The absorption and fluorescence emisssion spectra of reduced flavins ...... 120
xvii Figure 6.2 Femtosecond-resolved fluorescence transients of reduced flavins...... 121
Figure 6.3 Femtosecond-resolved transient absorption measurements of reduced flavins...... 122
Figure 6.4 Schematic representation of the two potential energy surfaces, illustrating non- radiative decay of excited state to ground state through conical intersection...... 123
xviii ABBREVIATIONS
α alpha, used for alpha-helix of proteins
β beta, used for beta-sheets of proteins, also used for degree of
stretch-mode exponential fit
λ Wavelength
Å Angstroms (10-10)
τ tau, used for time constants
∆ delta, used for difference / change in two parameters
Φr quantum efficiency of repair
Φe quantum efficiency of electron transfer
Km Specific constant kcat Catalytic rate constant
˚C Degree centigrade
UV Ultra-violet
NaCl Sodium chloride
DTT Ditheothreitol
IPTG Isopropyl β-D-1-thiogalactopyranoside
EDTA Ethylene diamine tetra-acetic acid
ΒME β-mercaptoethanol
xix KCl Potassium chloride
SDS-PAGE Sodium dodecyl sulphate – polyacrylamide gel electrophoresis
HPLC High pressure liquid chromatography
MW Molecular weight
MTHF 5, 10-methenyltetrahydropteroylpolyglutamate
FAD Flavin adenine dinucleotide
FADH Anionic reduced flavin
FADH• / FADH Flavin semiquinone, neutral radical
EPL – FADH Enzyme photolyase without pterin cofactor and flavin is in anionic
reduced form
• EPL- FADH Enzyme photolyase without pterin cofactor and flavin is in neutral
radical form
EPL-MTHF- FADH Enzyme photolyase with both the cofactor where flavin is in
anionic reduced form
• EPL-MTHF- FADH Enzyme photolyase with both the cofactor where flavin is in
neutral radical form h hour(s) m minute(s) s second(s) ns nanosecond(s) (10-9)
xx ps picosecond(s) (10-12) fs femtosecond(s) (10-15)
kHz kilohertz
KD dissociation constant
NMR Nuclear Magnetic Resonance
xxi CHAPTER 1
INTRODUCTION
1.1 Enzyme dynamics
For more than a century, the catalytic rate augmented by enzymes has engrossed the attention of chemists, biochemists and biologists. Though our understanding of enzyme catalysis evolved slowly, present development in structural biology techniques has enriched the knowledge rather rapidly in recent past. High resolution three dimensional structural data available from X-ray crystallography and NMR techniques have provided an opportunity to understand the orchestra of enzyme catalysis. However, it has still been difficult to fully account for the enzymatic power from these static structures (Schramm 2006). Enzymes are dynamic in nature and today it is increasingly clear that molecular dynamic contributions to catalysis must be incorporated to provide a full understanding of catalysis (Schramm 2006).
1 1.1.1 Enzyme catalysis and enzyme dynamics
Enzymes are fascinating molecules. For biologists, they are important in catalyzing life sustaining physiological processes and for chemists they are the ultimate catalyst which multifolds the reaction rate to achieve the product formation. Enzymatic reactions typically proceed at rates more than millions of times faster than the corresponding uncatalyzed ones.
In attempt to understand the enzyme catalysis, a simple kinetic scheme which holds the key in having an intermediate transition step of enzyme substrate complex was proposed. It was suggested that complex further dissociates in ‘product’ leaving the enzyme unchanged as catalyst for the next cycle. Although the observed kinetics in many cases can be described using this model, the actual chemical pathway involved in converting “substrate” into “product” is rather complicated (Callender et al. 2006). For enzymes, where comprehensive mechanisms have been established, multiple intermediates are almost always detected (Smiley et al. 2006). These multiple intermediates divide a chemical reaction into multiple steps of relatively low activation energy, as contrasted to a single step of high activation energy. Today this appears to be a general feature of the catalytic process (Smiley et al. 2006). In terms of ‘energetics’ in a biological system, these multiple transition steps use minimum energy, favor the reaction in desired direction and thus perform the preferred function with greater efficiency.
To stabilize the multiple intermediate steps, dynamic motions of the enzyme play a critical role. The actual catalytic reaction site involves considerable and specific atomic
2 motions of both the bound substrate and the atoms of the protein along the reaction
coordinate (Callender et al. 2006). In addition, not only the atoms which are close to the active site are important but also the overall macromolecular nature and motion of the whole enzyme play an appreciable role in the overall catalysis (Hammes 2002). Thus studies on the dynamic fluctuations in enzymes are essential in developing a comprehensive understanding of the enzyme catalysis.
1.1.2 Dynamic fluctuations in enzyme
Enzyme structures are dynamic in nature. To understand the molecular basis of
the catalytic efficiency gained by enzymes, it is essential that dynamic fluctuations in
enzyme get appreciable considerations. The polypeptide backbone and specially the side
chains are constantly moving due to either thermal motion or kinetic energy of the atoms.
These thermal motions are essential for high substrate specificity and stabilization of
multiple transition steps. Along with the atomic fluctuations at the active site, which
largely affect the actual catalytic step, macromolecular nature of the enzyme also
significantly affects the catalysis. Synchronized motion of various amino acid residues in
the macromolecular enzyme allows it to adopt multiple conformations which appear to be
a general attribute of enzymatic mechanisms. The role of these conformational changes in
the catalytic process is not well understood, but several possibilities are obvious. For
example, the structure can be optimized for each step over the reaction coordinate by
restrained cooperative conformational changes (Hammes 2002). The binding of the
substrate can be enhanced by a change in the enzyme structure, as explained by the
"induced fit model" (Koshland 1960), and the substrates can be spatially oriented for 3 redox reactions or for electrostatic catalysis. Also, the dielectric properties of the protein nano-space can be markedly altered: water can be pushed in/out of the active site to create a more/less polar environment for the catalytic reaction to carry out. On a larger scale, involving coupled conformational processes, enzyme may also play a more dynamic role in the catalysis. The first explicit model for this role was proposed by
Eyring and Lumry (Lumry 1959). A similar model in which noncovalent bond breaking and making within the macromolecule is coupled to the catalytic event has been postulated as a dynamic mechanism for lowering the activation energy (Hammes 1964).
For any given case, dynamics of conformational flexibility of the macromolecules, atomic fluctuations along the reaction coordinate (of protein and substrate), and reaction
‘‘nanoenvironment’’ are vital for enzyme function.
Enzymes in general exhibit several different motions over a wide range of time scales (Boehr et al. 2006, Wang et al. 2004). Small scale displacements of atoms, necessary for steric placing of substrate and product in enzyme complex, take place at picosecond to nanosecond time scale. Activated motions of molecular groups, involved in initial substrate recognition, occur within nanosecond to microsecond time scale. The motions of protein domain or small scale group motions with high activation energy, involved in enzyme allostery are observed to take place on the millisecond or longer time scale. Table 1.1 provides representative time scales, amplitudes and types of different motions observed in proteins. To study these motions experimental and theoretical methods are just now being developed that can adequately cover the broad time range of
4 atomic motions (Callender et al. 2006). Section 1.1.3 will summarize a few time-resolved approaches currently in use for characterizing enzyme dynamics.
1.1.3 Time-resolved approaches to characterize enzyme dynamics
Any of these motions as tabulated (table 1.1) may be functionally significant and directly related to the ligand exchange and/or catalysis (Boehr et al. 2006). The time scales of these dynamic motions vary over a wide range and hence many spectroscopic and computational techniques have been developed to monitor and analyze these motions. In this section a brief outline of the current available time-resolved methods to monitor these dynamic processes will be presented. By no means can one experimental method of investigation be classified as superior over other; however at the end of the section I will briefly summarize why the use of ultrafast optical spectroscopy was necessary to investigate the dynamic processes of energy and electron transfer in photolyase enzyme, which was the focus of this doctoral research.
1.1.3.1 Nuclear Magnetic Resonance
NMR is potentially a very powerful experimental technique for studying enzyme dynamics. It provides a complete and thorough description of enzyme dynamics at a wide range of time scales at atomic resolution without the introduction of bulky and potentially interfering probes (Boehr et al. 2006).
With the advancements in newer labeling and pulse sequence techniques (Goto et al. 2000, Riek et al. 2002, Fernandez et al. 2003), the present time scale available to
5 NMR technique ranges from picoseconds to hours, i.e., over 17 order of magnitude
(Boehr et al. 2006). This wide range covers all of the relevant dynamic motions in enzymes. Newer isotope labeling strategies can probe nearly every significant atomic site. These studies not only provide a qualitative picture of enzyme dynamics but in many cases also detail the complete kinetic and thermodynamic profile of the dynamic processes (Boehr et al. 2006). In early 1990s, the analysis of high molecular weight enzymes using NMR was a tougher task. However recently transverse relaxation optimized spectroscopy (TROSY) technique is thriving to monitor signals from these proteins. Recently investigators have published dynamic studies on high molecular weight enzyme complexes such as malate synthase G (82kDa) (Tugarinov et al. 2004).
Although NMR is a powerful tool in analyzing dynamic motions of proteins however, it is an ensemble technique and the results from NMR studies may not reveal superior complexities associated with the data, which were gathered from ensemble averaging. The observed higher energy conformation from such studies may themselves give a time-averaged view of the energy landscape and hence may not decipher the true and broader dynamic processes involved in enzyme dynamics. The research on photo- enzymes involving ultrafast changes (at the order of 10-12) in redox, hydride and / or proton and electron transfer reactions largely can not be characterized using NMR techniques.
6 1.1.3.2 X-ray crystallography
X-ray crystallography provides atomic-level spatial details of the enzyme and its
interaction with the substrate/product. To gain a mechanistic understanding into how
enzymes execute their designed function, it is crucial to know how their structure evolves
over time (Schotte et al. 2003). Though largely X-ray crystallography has been used to
get the static snapshot of enzyme structure, recently there are reports where time-resolved diffraction patterns were obtained to visualize the enzyme in action (Srajer et al. 2001,
Schotte et al. 2003, Bourgeois et al. 2003, Brunori et al. 2004).
The time-resolved crystallography gets advantage over all other available techniques as it provides the structural details, including the direction, amplitude, and time scale of functionally important enzyme motions (Schotte et al. 2003). Time-resolved
X-ray crystallography has been developed by pioneering efforts of Moffat and co- workers and led to the first nanosecond time-resolved structure of myoglobin (Moffat
2001, Srajer et al. 1996). Subsequent studies improved the spatial resolution attained
(Bourgeois et al. 2003; Brunori et al. 2004, Srajer et al. 2001). Recently Schotte et al.
(2003) developed the method of picosecond X-ray crystallography and used this technique to structurally characterize the L29F mutant of myoglobin as it evolves from the carboxy to the deoxy state. They obtained the data in picosecond to microsecond time regime. Constraint dihedral, side chain and loop motions were investigated in the study to understand the ligand binding and dissociation.
7 Today time-resolved X-ray crystallography seems to be emerging as a very
powerful technique to understand the enzyme dynamics. However, the technique is in its
infancy and experiments are technically very challenging. Also results obtained from
crystalline form of enzyme may not represent true and physiologically relevant dynamics.
As of now the time resolution obtained from such technique is limited and ultrafast
processes especially electron transfer and small atom/nuclei hopping can not be
investigated. Also on the other side measurements of slower time scales involving helix
to coil transition and/or domain movements are very difficult to study using this
technique.
1.1.3.3 Optical spectroscopy
The essential feature of time-resolved studies with optical spectroscopy is the use
of a trigger method to initiate chemical or structural changes in a system combined with a
structural probe, which follows the evolution of the system in time (Callender et al.
2006). This technique has been applied in investigating ultrafast dynamics in picosecond
to nanosecond regime in multiple proteins (Denisov et al. 2005, Nienhaus et al. 2005,
Nibbering et al. 2005, McMohan et al. 2004, Gennis et al. 2004, Wang et al. 2004,
Zhong et al. 2001). These studies have focused on proteins with an absorbing prosthetic group such as heme, retinal or chlorophyll because the chromophore provides a convenient means to both initiate and probe the dynamics (Callender et al. 2006).
Recently other rapid initiation techniques such as T-jump, pH-jump, photolytic release of caged reagents have been developed to provide a more generalized strategy to study the enzyme dynamics. 8 Time resolution obtained using optical spectroscopy offers unique advantage over
other techniques and is a very useful in studying ultrafast, nanosecond and sub-
nanosecond processes in proteins. For examples, during catalysis bond breaking and
making occur at picosecond and femtosecond time scale (Mcfarlane et al. 2003); other
important enzymatic processes such as redox reactions, electron transfer and proton
transfer, also occur on picosecond time scales (Zhong et al. 2001, Mataga et al. 2002);
ligand assisted side-chain motions occur in nano-second and sub-nanosecond time scale.
All these events can be characterized precisely using the optical spectroscopy (Wang et
al. 2004).
In present studies we used optical spectroscopy to understand enzyme dynamics
in photolyase because of following reasons. In past various techniques including NMR,
EPR and also optical spectroscopy were applied to understand the key catalytic processes
(Sancar 2003). However, only recently has optical spectroscopy been successful in
developing a comprehensive understanding of photolyase catalyzed key reactions of
damaged DNA repair (Langenbacher et al. 1997, Mcfarlane et al. 2003, Kao et al. 2005).
It is mainly because that optical spectroscopy offers a unique advantage of ultrafast time resolution. Damaged DNA bond breaking and repaired DNA bond making occur at ultrafast time scale and hence in past it was almost impossible to probe the dynamics of these events using NMR techniques. In addition, photolyase catalyzed reactions are photon triggered and it is advantageous to adopt a ‘trigger- based’ approach which works similarly as in nature.
9 During the time of this doctoral work, studies on isolated flavin chromophore of photolyase were pointing on a totally new concept in enzyme dynamics. Excited state lifetime studies performed on reduced flavin were suggesting that the dynamic enzyme and its spatial configuration place the flavin cofactor in special orientation. This special orientation extends the lifetime of excited state, which only in fact can perform the required electron transfer to repair the DNA with a highest catalytic efficiency. No other techniques but ultrafast optical spectroscopy could have provided the details of the photolyase catalyzed DNA repair as presented in this thesis.
1.1.4 Applications
The correlation between structure, dynamics and function has practical importance. If dynamics determines the function, then understanding dynamics is vital for both protein and ligand design. Thus for de novo enzyme design, not only correct positioning of catalytic group needs to be engineered but also it is necessary to incorporate appropriate dynamic modes into the protein of interest for full efficiency
(Boehr et al. 2006, Bolon et al. 2002, Hilvert et al. 2000). Many enzymes are key pharmacological targets and in past their structural information has provided the basis for the discovery of many drugs. Today it has become clear that flexibility of protein and/or substrate is very important to drug-receptor interaction and incorporating dynamic information in design is necessary in developing potential new drugs (Teague 2003).
10 1.2 DNA photolyase
Enzyme catalysis is an inherently dynamic process. As discussed earlier, studies of dynamic processes involved in enzyme catalysis are essential in developing a complete understanding of catalysis. To advance the present knowledge of enzyme dynamics we choose DNA photolyase as our model. Photolyase is a photoenzyme which repairs the ultraviolet (UV) radiation induced DNA damages and is present in all the three branches of life (Sancar 2003).
1.2.1 Biological function
Ultraviolet radiation, especially UV-B (290nm-320nm), endangers all forms of sun-exposed life by the formation of various genotoxic photoproducts in DNA which consequentially cause mutation, growth delay and even cell death (Essen et al. 2006). To maintain the genetic integrity of the organism, in presence of strong sunlight, nature invented several ways to repair difference types of UV lesions (Essen et al. 2006). The two major lesions in DNA induced by UV are the cyclobutane pyrimidine dimers
(Pyr<>Pyr) and the pyrimidine-pyrimidone (6 – 4) photoproduct (Pyr [6 – 4] Pyr)
(Patrick et al. 1976). Both of these lesions are repaired by DNA photolyase enzyme of similar sequences and most likely similar structures and reaction mechanisms.
Interestingly, a photolyase that repairs one cannot repair the other, and hence the enzymes are often referred to as cyclopyrimidine dimer (CPD) photolyase and (6 – 4) photolyase respectively (Sancar 2003).
11 1.2.2 Significance
DNA damage and repair are two of the most important fields in molecular biology
(Schärer 2003, Sancar 1996). Although UV-induced damage in human cells is usually repaired by nucleotide/base excision repair, severe damage can be toxic and mutagenic, and will trigger cell apoptosis. Under strong UV exposure, severe DNA damage blocks normal replication and transcription, finally leading to cell death. The main cause of skin cancer is UV light that induces the formation of dimers and (6-4) photoproducts (Sancar
2003 and references there in). Recent stratospheric ozone depletion increases skin cancer risk drastically (Diffey 2004). Studies suggest that a 1% decrease in the ozone layer will cause an estimated 2% increase in UV-B irradiation; it is estimated that this will lead to a
4% increase in basal carcinomas and 6% increase in squamous-cell carcinomas [NASA edu.]. 90% of the skin carcinomas are attributed to UV-B exposure [NASA edu.]. More than 1.3 million Americans are diagnosed each year with the disease (Chigancas et al.
2000). In addition, increased formation of UV-photoproducts in other organisms in the biosphere is likely to have severe consequences as well (NASA edu.). Although photolyase does not exist in humans, two types of research related to human recently been reported. One used a recombinant adenovirus vector containing the gene encoding a marsupial CPD-photolyase to investigate CPD repair in situ and consequent UV-induced apoptosis in primary human cells (Chigancas et al. 2000, 2004 and Lo et al. 2005). The other experiment generated transgenic mice that ubiquitously express photolyase to allow rapid light-dependent repair of photoproducts in the skin (Garinis et al. 2005 and reference there in). The results are very encouraging as they showed that mice with
12 photolyase exhibit superior resistance to sunlight-induced tumorigenesis. These studies provide direct evidence that photolyase enzyme can be employed as effective tools, such as through gene therapy, to control skin cancer, especially in individuals with the DNA- repair defect disease called Xeroderma Pigmentosum (Masutani et al. 1999, Johnson et al. 1999).
1.2.3 The photolyase enzyme
CPD photolyases have been widely found in all three kingdom of life.
Photolyases are monomeric proteins of 450-550 amino acids and two noncovalently bound chromophore cofactors. One of the cofactors is always flavin adenine dinucleotide
(FAD), and the second is either methenyltetrahydrofolate (MTHF) or 8-hydroxy-7, 8- didemethyl-5-deazariboflavin (8-HDF). Accordingly, the enzymes have been classified into folate class and deazaflavin class photolyases (Sancar et al. 1988). The FAD is the essential cofactor for both specific binding to damaged DNA and catalysis (Jorns et al.
1987). The second chromophore (MTHF or 8-HDF) is not necessary for catalysis and has no effect on specific enzyme-substrate binding. However, under limited light the second chromophore may increase the rate of repair 10-100-fold depending on the wavelength used to affect catalysis. This is because the second chromophore has a higher extinction coefficient than FADH and an absorption maximum at longer wavelength relative to that of the two-electron-reduced flavin that is the active form of the flavin in the enzyme
(Jorns et al. 1990). The overall DNA repair quantum yield is as high as 0.7-0.9.
13 1.2.3.1 Primary Structure
The amino acid sequences of about 50 photolyases are known. The sequences of
these proteins reveal varying degree of homology ranging from 15% to 70 % (Todo et al.
1999). Sequence alignment results provide several points of interest. A) The C-terminal
150 amino acids exhibit the highest degree of homology among all photolyases classes
(Ramsey 1992). B) Photolyases show no obvious sequence homology to flavoproteins
oxidoreductases (Sancar 1994). C) Recent molecular phylogenetic analysis suggests that
photolyases are distantly related with other nucleotide binding proteins (Aravind et al.
2002).
1.2.3.2 Chromophore Cofactors
As mentioned earlier, all photolyases contain FAD and MTHF/8-HDF as
cofactors. These cofactors have different biochemical properties and they perform
different biological functions. A brief description of their properties is given below.
1.2.3.2.1 Flavin
FAD is bound noncovalently but very tightly to photolyase and can be released
only after mild denaturation of the enzyme. FAD can be found in three redox states:
oxidized, one-electron-reduced (neutral blue radical or anionic red radical), and two-
electron-reduced (neutral or anionic) forms. The active form of flavin in photolyase is the two-electron-reduced form (Payne 1987). In vitro, with the exception of S. cervisiae photolyase the catalytically active cofactor FADH is always oxidized into flavin neutral
14 radical form FADH• during purification and consequently photolyase loses repair
activity. However, this inactive radical form FADH• can be restored into the active form
FADH by absorption of visible light and intraprotein electron transfer (photoreduction) or simply by chemical reducing agents, such as dithionite (Heelis et al. 1986). FADH• has characteristic absorption bands at 480, 580 and 625 nm compared with its reduced form FADH which has a single characteristic absorption band at 366 nm in UV-visible absorption spectrum (figure 1.1).
1.2.3.2.2 Pterin
The majority of the photolyases contains a pterin as photoantenna. In E. coli
photolyases the pterin is in the form of 5, 10-methenyltetrahydropteroylpolyglutamate
(methenyltetrahydrofolate, MTHF). Unlike flavin, folate cofactor dissociates from some
of the photolyases readily. Thus, during protein purification1:1 stochoimetric of MTHF
and FAD can be rarely obtained. Folate can also be removed using photodecomposition
method by irradiating the photolyase sample with 366 nm light. The 5, 10-methenyl
bridge of the folate is responsible for the near-UV absorption at 360 nm. However, upon
binding to the apoenzyme polar interactions with the positive charge on the methenyl
group combined with the hydrophobic interactions of the pterin ring within the binding
pocket cause a red shift. And hence folate in CPD photolyase exhibits characteristic
absorption maxima at 384 nm (figure 1.1).
15 1.2.3.3. Crystal Structure
Crystal structures of three photolyases from E. coli (Park et al. 1995), A. nidulans
(Tamada et al. 1997), and T. thermophilus (Komori et al. 2001) have been determined and all three are remarkably similar although there is only 25% sequence identity among these three enzymes. The complex structure of A. nidulans photolyase with a repaired decamer substrate product has recently been reported (Mees et al. 2004). Figure 1.2 shows the X-ray structure of E. coli photolyase at 2.3Å. It is composed of two well
defined domains: an N-terminus α/β domain (residues 1-131) and a C-terminal α-helical
domain (residues 204-471), which are connected to each other with a long interdomain
loop (residues 132-203) that wraps around the α/β domain.
The MTHF photoantenna is located in a shallow cleft between the two domains
and partially sticks out from the enzyme surface. Positive charge on the five membered
ring of MTHF makes contact with the carbonyl side chain of Cys292. One glutamate
moiety of MTHF establishes a salt bridge with Lys293 that increases affinity of MTHF
with apoenzyme. Higher binding affinity of MTHF containing multiple glutamate
residues with apoenzyme can be rationalized with the probability of increased numbers of
salt bridges (Park et al. 1995).
The FADH cofactor is deeply buried within the α-helical domain and has an
unusual U-shaped conformation with the isoalloxazine and adenine rings in close
proximity. The FADH is held tightly in place by contact with 14 amino acids and is accessible to the flat surface of the α-helical domain through a hole in the middle of this
16 domain. The hole is too small to allow the diffusion of FADH in and out of the enzyme
but is easily accessible to oxygen. This explains the relative ease with which FADH is
converted to FADH• in most photolyases. Interestingly, this hole has the right dimensions
and polarity to allow the entry of a thymine dimer within van der Waals contact distance to the FADH, as also shown in the recent complex structure (Mees et al. 2004).
1.2.4 The damaged DNA substrate
Ultraviolet light damage produces several types of mutagenic DNA
photoproducts. The substrate for CPD photolyases is cyclopyrimidine dimers. The CPDs
are formed via a cyclobutane ring connecting the 5, 6 positions of two adjacent
pyrimidine bases. Formation of CPD requires an extensive rotation of the neighboring
pyrimidines from their usual B-form DNA (Pfeifer 1997). The two 5, 6 double bonds
must overlap in order for dimerization to occur efficiently. The torsional flexibility of the
DNA helix at the site of the photoproduct thus plays a significant role in determining the ability of a given dipyrimidine sequence to form a CPD (Pfeifer 1997). The thymine cyclobutane pyrimidine photodimer is the highest quantum yield photolesion formed in
DNA (Φ = 0.019). This dimerization reaction occurs from the thymine triplet state and hence during their synthesis in vitro it can be sensitized by triplet sensitizers such as acetone and acetophenone (section 2.2.3). The regiochemistry and stereochemistry are controlled by the geometry of the DNA double helix. Out of six isomers of the thymine<
>thymine (T<>T) dimers, the major isomer formed in DNA has the cis-syn stereochemistry (figure 1.3).
17 Photolyase is a structure-specific DNA binding protein and not sequence specific
(Myles et al. 1985, Svoboda et al. 1993). Photolyase binds to CPDs with essentially the
same affinity in single and double stranded DNA (Sancar 1985, Husain 1987). The
-9 specific binding constant to a T<>T in DNA is KS = 10 M. Nonspecific binding of
-4 undamaged DNA with photolyase is KNS = 10 M, i.e. photolyase has a selectivity factor
of 105. Also because the enzyme bind to a T<>T dinucleotide with an equilibrium
-4 constant KD = 10 M. In other words half of the binding energy is contributed by the
interaction of dimer with flavin binding pocket (Jorns et al. 1987, Jordan et al. 1989, Kim et al. 1991).
Site directed mutagenesis studies have shown that positively charged groove running across the surface of helical domain plays an important role in binding the substrate. The cavity/hole hosting flavin cofactor has the hydrophobic lining and right dimensions to accommodate the CPD. Interestingly, replacement of key amino acid residue Trp277 even eliminates the specific binding of the substrate (Li et al. 1990). The
presence or absence of the ‘photoantenna’ MTHF does not affect the binding of
photolyase with CPDs.
The enzyme contacts the three phosphates that are 3' and the phosphate that is 5'
to the T<>T, as well as the phosphate that is opposite the dimer across the minor groove
on the complementary strand. Majority of these phosphates make ionic bonds with the
positively charged residues in the binding groove of the enzyme. It also obstructs the
major groove for about half of a turn 3' to the dimer (Husain et al. 1987). Significantly, it
does not contact the intradimer phosphate. Thus, nearly all of the contacts are with the 18 damaged strand. In addition to the phosphate backbone, type of cyclobutane dipyrimidine
structure, the nature of the pentose and the stereochemistry of the dimer also affect
binding. For example, Uridine<>Uridine (U<>U) in RNA binds with <105 –fold lower affinity than that of U<>U in DNA (Kim et al. 1991).
Recently solved crystal structure of a DNA photolyase bound to duplex DNA suggests that DNA is bent by 50° (Mees et al. 2004). DNA in the structure comprises a synthetic CPD lesion. This CPD lesion was found flipped into the active site and split there into two thymines.
1.2.5 Molecular mechanism of DNA-photolyase
Photolyases catalyze cycloreversion of pyrimidine dimers (Pyr<>Pyr) generated
in DNA by ultraviolet light. In the proposed hypothesis for the catalysis, the enzyme
binds a Pyr<>Pyr in DNA, independent of light, and flips the dimer out of the double
helix into the active site cavity. The antenna chromophore MTHF harvests UV/blue-light
photon from the ambience and transfers the excitation energy (dipole-dipole interaction)
to FADH−. Excited state of FADH− then transfers an electron to the Pyr<>Pyr. With an excess electron, 5 – 5 and 6 – 6 bonds of the cyclobutane rings are now in violation of
Hückel rule and therefore splits the Pyr<>Pyr into two pyrimidines and hence repairs the damaged DNA. As proposed, the repair cycle ends when an electron is transferred back to the nascent-formed FADH• and regenerates the active FADH−. Figure 1.4 outlines the
proposed catalytic mechanism.
19 The behavior of catalysis follows Michaelis-Menten kinetics. Damaged DNA (S)
binds to photolyase (E) to form enzyme-substrate complex (ES). The enzyme performs
catalysis to yield enzyme-product complex (EP), and then repaired DNA (P) dissociates.
The only difference is that catalytic step of actual repairing DNA is absolutely light
dependent (Rupert 1960, 1962, Harm et al. 1968, 1970, 1976).
Using a variety of techniques, multiple lines of investigation have been pursued to
understand the mechanistic details of the DNA repair carried out by photolyase. The key
areas of investigation include characterizing elementary photophysical processes during
energy transfer from MTHF to FADH− and electron transfer from excited state of FADH−
to damaged DNA.
1.2.5.1 Photophysics of photoreactivation
This section will outline the photophysical properties and dynamics of the energy
and electron transfer available in the scientific literature before the start of this doctoral
research. Strict focus will be on E. coli CPD photolyase studies unless otherwise mentioned. For comprehensive understanding, biochemical and photophysical properties of photolyase are tabulated in Table 1.2 and 1.3 respectively.
1.2.5.1.1 Action spectrum
Quantitative analysis of the CPD repair with photolyases in presence and absence of either or both cofactor has established that the flavin but not the folate is essential cofactor for the repair. Also, only reduced form of flavin FADH−, but not the oxidized
20 FAD form, can repair the CPD. Quantitatively, the photolytic cross section (ε X φ) in vitro of reduced enzyme closely matched the values obtained in vivo, thus validating the use of reduced enzymes for photochemical characterization (Malhotra et al. 1994).
Absolute action spectra of the E-FADH− and E-MTHF-FADH− are shown in figure 1.5
(adopted from Sancar 2003). The quantum yield of repair was calculated to be 0.85. This
value clearly suggests that the quantum yield of electron transfer from FADH− to
Pyr<>Pyr should be equal to or higher than the quantum yield of repair.
1.2.5.1.2 Energy transfer
Emission spectrum of MTHF overlaps the absorption spectra of FAD in all the three oxidation states. Steady-state fluorescence analysis has also shown that MTHF fluorescence in E. coli gets quenched in presence of any of three oxidation states of the
FAD, indicating singlet-singlet energy transfer (Jordan et al. 1988). Time resolved spectroscopy was used to calculate the energy transfer efficiency either by monitoring fluorescence decay of MTHF in presence of flavin species or by transient absorption techniques. Table 1.3 outlines the photophysical properties of photolyases.
1.2.5.1.3 Electron transfer
Excited state of FADH− hypothetically ejects electron to Pyr<>Pyr dimer and
splits the cyclobutane ring but there was no direct experimental evidence. However free
energy calculations for single electron transfer (SET) supported the idea that SET from
FADH− to Pyr<>Pyr is the only possible mechanism for photolyase. In early 1990’s time-
21 resolved fluorescence and transient absorption spectroscopy were used to calculate
electron transfer rate, repair efficiency and direction of electron transfer. Table 1.3
outlines the major findings. Time-resolved EPR studies were also performed to capture
the intermediate FADH• state; however a clear picture could not emerge.
1.2.5.1.4 Knowledge from model systems
Most of our knowledge about the cleavage of pyrimidine photo dimers has come
in the past from model studies (Carell et al. 1998). Studies with model systems using
simple organic compounds capable of photoexcited redox chemistry to mimic kinetics,
thermodynamics, and mechanistic aspects of the dimer splitting by photolyase have been
instrumental in designing and analyzing experiments conducted with photolyase in
developing a mechanistic model for the enzyme (Sancar 2003). In late 1980s and early
1990s, very simple organic compounds such as aniline, tryptophan were used to
understand the cyclobutane ring opening. The experiments were diffusion limited, where
mixing of such electron donating compounds with Pyr<>Pyr would bring the splitting of
cyclobutane pyrimidine ring after the photoirradiation of the samples. In order to gain a
deeper understanding of the dependencies of the flavin-mediated dimer cleavage,
covalently linked model compounds were developed. They a) enabled a systematic
investigation of cleavage processes, b) developed the experimental proof to what extent
photolyases interfere with the repair processes to maximize the repair-efficiency and c)
helped deducing the function of specific structural and electronic properties of the photolyase cofactor and active-site (Carell et al. 1998). In general, studies with model systems have shown that Pyr<>Pyr can be split by either electron donors or electron 22 acceptors. These studies also advanced the understanding of the redox environment around the flavin binding pocket by quantifying the dielectric properties and pH of the medium. Suggestions were made that the redox environment can have significant effect on the efficiency of splitting.
1.2.5.2 Photoreduction of photolyase
Understanding electron transfer processes involved in the photoreduction of flavin is crucial in developing the catalytic model for photolyases. During the purification of photolyase, flavin cofactor gets semi-oxidized to FADH• and enzyme loses its repair activity. Exposure of the enzyme with FADH• form to light, in the presence of a reducing agent, converts it to the FADH− form (Heelis et al. 1986). Inspection of the crystal structure revealed two potential pathways for electron transfer (Park et al. 1995): one pathway passes through Trp306 → Trp358 → Trp382 and involved three electron hops, and the other consists of the α-helix (α-15) between residues 382 and 366 and the side chain of Phe366.
1.3 Research project
To add to the present developing knowledge of enzyme dynamics we focused our studies in understanding aspects of energy and electron transfer mechanism involved in enzyme catalysis. From the photolyase research point of view there were several unaddressed issues. Specifically, 1) the resonance energy transfer dynamics from folate to flavin was not precisely determined, 2) multiple electron routes that might be involved
23 in photoreduction of photolyase were not established, 3) a direct interpretation of repair mechanism in model systems in terms of enzyme catalysis was obscure and finally and most importantly 4) the proposed catalytic mechanism of electron transfer from excited flavin to damaged DNA and back to flavin, was lacking experimental proof.
1.3.1 Dynamics of resonance energy transfer in photolyase
During catalysis, the first step is the absorption of a photon by the photoantenna
MTHF, which then transfers the excitation energy to the catalytic cofactor FADH− to enhance the DNA-repair efficiency (Payne et al. 1990). Earlier studies reported that the energy transfer from MTHF to FADH− takes 140-180 ps (lifetime) but occurs in less than
30 ps from MTHF to FADH• (Kim et al. 1993). The actual time scales have not yet been determined precisely. In this study, we combined femtosecond-resolved fluorescence up- conversion and transient absorption methods to determine the actual time scales of the initial resonance energy transfer from the antenna to the fully reduced and semioxidized cofactors. Chapter 3 will discuss the studies performed to understand the dynamics of resonance energy transfer processes in photolyase. For comparative analysis we also performed resonance energy transfer studies on cryptochrome protein, one another member protein of photolyase family.
24 1.3.2 Dynamics of intraprotein electron transfer in photoreduction of
photolyase
The flavin cofactor of the enzyme in vitro is usually oxidized to the inactive
neutral radical form FADH• after purification. Under illumination with visible light, the
cofactor can be converted to the active form through electron transfer from neighboring
aromatic residues. The residue tryptophan 306 (W306) was identified as the ultimate donor by site-specific mutagenesis, fast kinetics, and EPR analyses (Li et al. 1991, Kim et al. 1993). Two recent studies (Aubert et al. 2000, Byrdin et al. 2003) have reported the dynamics occurring in ~25 ps through one electron transfer from a nearby tryptophan
(W382) favoring one of the two pathways proposed on the basis of the crystal structure
(Park et al. 1995). However, a steady-state photoreduction study indicated that the
W382F mutant was photoreduced with 2-fold higher quantum yield relative to that of the wild type (Sancar 2003). Clearly, more detailed work was necessary to understand the intraprotein electron transfer in the photoreduction of photolyase. Intraprotein electron transfer in protein itself is a wide research area; photoreduction processes in photolyase provide us a unique opportunity to develop understanding of dynamics of intraprotein electron transfer. Chapter 4 discusses the dynamics of flavin cofactor in photolyase, involving the understanding of radical reduction, local solvation and charge recombination processes.
25 1.3.3 Dynamics of DNA repair in photolyase
The model for the catalytic reaction (Sancar 2003, Park et al. 1995) proposes that the excited flavin cofactor ejects an electron to the Pyr<>Pyr to generate a charge- separated radical pair (FADH• + Pyr<>Pyr•−). The anionic ring of the dimer is split by a
[2 + 2] cycloreversion, and the excess electron returns to the flavin radical to restore the catalytically competent FADH− form and closes the catalytic photocycle. This hypothetical radical mechanism has not been directly proven and the radical intermediates have not yet been captured, although it was proposed ~20 years ago
(Sancar et al. 1987) and supported by extensive biochemical data (Payne et al. 1990,
Ramsey et al. 1992), spectroscopic studies (Kim et al. 1990, Langenbacher et al. 1997,
MacFarlane et al. 2003), and computer modeling (Sanders et al. 1999, Antony et al.
2000), as well as recent structural determination (Komori et al. 2001, Mees et al. 2004).
Chapter 5 will discuss our experimental results on direct observation of DNA repair processes. An extensive understanding of active-site solvation dynamics will be presented with a brief note on how solvation might modulate the electron transfer processes in enzyme catalysis.
1.3.4 Dynamics of reduced flavin and flavin model systems involved in repair
The quantum yield of DNA repair by CPD photolyase is in the order of ~0.9-1.0.
However, model systems developed to understand the repair mechanism typically were found to repair the CPDs with quantum efficiency of about ~0.1-0.2 (Carell et al. 1998).
Molecular mechanism of this low efficient repair process in the existing model systems
26 has been unknown. Several hypotheses were developed to understand this low efficiency, however a definitive answer was not available (Carell et al. 1998). After searching the literature we realized that even photophysical properties of reduced flavin as such had not been characterized very well. A mechanism which can explain lower quantum yield of repair in these reduced flavin model systems was surely a far reachable goal. Chapter 6 will elaborate our present efforts in charactering photophysical properties of reduced flavin species in aqueous solution. Also studies on covalently linked dimers with flavin species were undertaken. Preliminary discussions of repair processes in a flavin-model system are also presented in chapter 6.
1.4 The value added in enzyme dynamics
Chapter 7 will summarize the understanding of enzyme dynamics as gained from the present studies. We intend to significantly add to the body of knowledge concerning the effect of dynamics on enzyme catalysis. We have demonstrated that the understanding of ultrafast processes is very essential in investigation of enzyme catalysis.
Using modern molecular biology techniques coupled with femtosecond time-resolved studies we have unambiguously confirmed that use of these techniques can add in developing comprehensive knowledge of enzyme dynamics and catalysis. We have also established the dynamics of reduced flavin chromophore in aqueous solution. We have provided significant development in the present understanding that how protein
‘nanoenvironment’ can tune the photophysical properties of cofactors to perform essential biological functions. We could for the first time directly observe the DNA repair processes and captured the long-envisioned intermediate species involved in catalysis. 27 The investigation of active-site solvation dynamics, which could have a profound effect in molecular recognition studies, also added insight in enzyme dynamics by providing a direct proof that local ‘nanoenvironment’, specially the dielectric properties in proteins/enzymes, holds the key for any chemistry which can happen around it.
28
Time Scale (s) Amplitude Protein Motion
Ultrafast • Bond stretching, angle bending 0.001 – 0.1 Å Femto, pico (10-15, 10-12) • Constraint dihedral motions
Fast • Unhindered surface side chain motion 0.1 – 10 Å Pico, nano (10-12, 10-9) • Loop motion, collective motion
Medium • Folding in small peptides 1 – 100 Å nano, micro (10-9, 10-6) • Helix-coil transition
Long • Domain motion 10-100 Å Micro, second (10-6, 1) • Protein folding
Table 1.1 Time scales, amplitudes and type of motions in proteins
29 Structural Parameters
Protein Size (amino acids) 471 Molecular Weight 53994 Subunit Monomer Cofactor FADH−+ MTHF
Absorption / Emission Properties
Absorption Maxima E-FADH−-MTHF 384 nm E-MTHF 384 nm E-FADH− 366 nm Emission Maxima E-FADH•-MTHF (weak) 465 nm, 505 nm E-FADH−-MTHF 465 nm E-MTHF 465 nm E-FADH− 505 nm
Substrate Binding / Catalytic Parameters
-8 -9 Binding Constant (KD) 10 to 10 M -1 Catalytic Constant (kcat) 1.0 s
Quantum Yield of Repair (Φr) E-FADH−-MTHF 0.7 – 0.75 E-FADH− 0.8 – 0.9
Table 1.2 Biochemical properties of E. coli. CPD photolyase
E-Enzyme, FADH•-neutral radical Form, FADH-reduced form (adopted from Sancar 2003).
30
Excited-singlet-state Lifetime (ns)
E-MTHF* 0.35 (F) – 0.48 (A) E- FADH−-MTHF* 0.14 (F) – 0.18 (A) E- FADH•-MTHF* <0.03
Energy Transfer (SC*- FADH−)
rate (s-1) 4.6 X 109 efficiency (%) 62 interchromophore distance (Å) 16.8
Excited-singlet-state Lifetime (ns)
E-FADH−* 1.5 (F) – 1.7 (A) E-FADH−* + T<>T (U<>U) 0.16 (F) – 0.2 (0.05) (A)
Electron Transfer (FADH−*-T<>T)
rate (s-1) ~ (1-2) E 1010 efficiency (%) 89
Table 1.3 Photophysical properties of E. coli. CPD photolyase
(F), from time-resolved fluorescence measurements; (A), from time-resolved absorbance measurements. Electron transfer to U<>U is about 2-fold faster than to T<>T (adapted from Sancar 2003 and references cited there in).
31
Figure 1.1 Absorption spectra of E. coli photolyase in different states
(Upper) Absorption spectra of E. coli photolyase in semioxidized (thick dark line) and fully reduced (thin dark line) flavin forms. (Lower) Absorption spectra of the folate- depleted enzyme with neutral radical and reduced forms of flavin. MTHF emission spectrum (dashed line) is shown to overlap with the absorption spectra of flavin, indicating resonance energy transfer from MTHF to flavin. Semi-oxidized FADH• is represented as FADH.
32
Figure 1.2 X-ray crystal structure of DNA photolyase from E. coli
X-ray crystal structure (Park et al. 1995) of DNA photolyase from E. coli with the antenna molecule (MTHF), the catalytic cofactor (FADH), the active site, and the tryptophan triad for the photoreduction of the oxidized neutral FADH•.
33
Figure 1.3 Chemical structure of cis-syn thymine dimer
34
Figure 1.4 Schematic representation of the catalytic mechanism of photolyase
Schematic representation of the overall molecular mechanism involved in the DNA repair by the photolyase enzyme. RET, resonance energy transfer. CPD, cyclobutane pyrimidine dimer (Pyr<>Pyr).
35
Figure 1.5 Absolute absorption and action spectra of folate class photolyases
Solid and dashed lines are the absorption spectra of the E-FADH--MTHF and E-FADH- forms of the enzyme. The triangles and squares are the values for the quantum yield (Φ) times the molar extinction coefficient (ε) of the two forms (adapted from Sancar 2003).
36 CHAPTER 2
EXPERIMENTAL METHODOLOGIES
2.1 Femtosecond laser spectroscopy
All experimental measurements in current studies were carried out by using the femtosecond-resolved fluorescence up-conversion and transient absorption techniques.
The brief description of the theory and experiment set up are given in following sections.
The ultrafast spectroscopic technique used is of the type pump-probe. The photo-process was synchronized by a relatively weak (70-150nJ) and necessarily short pump-pulse (110 fs). The spectroscopic response of the sample was probed by another short and weak probe (gate) pulse. The dynamic response was reconstructed by repeating the experiment at various values for the delay time between pump and probe pulses. This is achieved by varying the relative optical path length of the pump and probe pulse pair (in air a path length difference of 10 µm corresponds to 33 fs). Principal of femtosecond pump-probe spectroscopy is more clearly illustrated in figure 2.1.
37 2.1.1 Fluorescence up-conversion technique
The basic principles of fluorescence up-conversion technique are described in
figure 2.2. Typically a femtosecond pump pulse excites the molecules and the
fluorescence is emitted by the sample. In present set-up we used a pair of parabolic
mirrors and the emitted fluorescence was focused on the same spot of the non-linear
crystal as the probe pulse. When the photons, from both fluorescence and gate pulse are
present in the crystal, through a non-linear process called sum (or difference) frequency generation, the sum (or difference) frequency photons are generated. The upconverted signals were sent to a calibrated computer controlled photomultiplier tube (PMT). By changing the delay of gate pulse (τ, figure 2.2) that usually has the duration smaller then
100-200 fs, we “cut” the fluorescence in time (observe experimental points on figure 2.2) and hence obtain a time-resolved fluorescence intensity.
For fluorescence up-conversion experiments, we used the pump wavelength at
400 nm by direct doubling of the 800-nm fundamental from the first OPA through a 0.2- mm-thick BBO crystal. Pump wavelength of 325 nm was generated by doubling the output (650 nm) of an optical parametric amplifier (OPA) which was generated through a mixing of the idler (2109 nm) and the fundamental (800 nm). The pulse energy was attenuated to 70-150 nJ before entering the sample cell. The fluorescence emission was collected by a pair of parabolic focus mirrors and mixed with another fundamental pulse in a 0.2-mm BBO crystal through a noncollinear configuration. The pump beam polarization for fluorescence up-conversion experiments was set at a magic angle (54.7°) with respect to the acceptance axis of the up-conversion crystal (vertical), and the gating 38 beam polarization was set parallel to this axis using a half-wave plate. Up-converted signals from 283 to 330 nm were detected by a photomultiplier (PMT) after passing through a double-grating monochromator. The response time in this noncollinear geometry is between 350 and 450 fs as determined from the up-conversion signal of
Raman scattering by water in the range of 450-460 nm.
2.1.2 Transient absorption technique
Femtosecond resolved transient absorption measurements were obtained using pump and probe technique. The pump/excitation pulse initiates the reaction in the sample cell and the probe pulse captures transient absorption of molecular species as a function of the delay time.
The Beers’ law of the absorption describes that absorption of the incident light I0 passing through a unit cell is directly proportional to the concentration of the light absorbing species in the cell. Light transmitted out It through the cell obeys the following relationship with the incident light I0
Absorption = − log( I t / I 0 ) = ε cl ……. (2. 1)
Where, ‘ε’ is the molecular extinction coefficient (a constant for a particular molecular species at a given wavelength), ‘c’ is the concentration of the molecular species and ‘l’ is the path-length of the sample cell.
The transient absorption A (τ, λ) at a given wavelength λ is defined as the ratio
– log It (t, λ) / I0 (t, λ), where It and I0 are the probe intensity measured with and without 39 the pump pulse respectively. A (t, λ) is obtained by two subsequent acquisitions. It, pump off is the transmitted light through the sample with no pump, i.e., there are no photoexcited states present and It, pump on is the transmitted light right after excitation. Comparing the ‘It,
pump off’ and ‘It, pump on’ with equation 2.1, Beer’s law of the absorption for transient absorption can be represented as
Absorption = −[log( I t pump on / I 0 ) − log( I t pump off / I 0 )] ……….. (2. 2)
During the data acquisition the difference in absorption in the absence or presence
of pump beam was measured with respect to the delay time between pump and probe
For transient absorption measurements, we used pump wavelengths of 325, 400,
580 and 620 nm. The pump pulse at 580 and 620 nm was generated from a mixing of the
idler (2109 nm) and the fundamental (800 nm) from the output of the first OPA through the same BBO crystal. Pump wavelengths of 400 and 325 nm were generated as described in section 2.1.1. By translating the movable mirror (MM in figure 2.3) out of the laser path, the half-intensity fundamental beam (1 mJ) pumps the second OPA, and various probe wavelengths from 400 to 700 nm were generated by a mixing of the idler or signal with the fundamental. Both the pump and probe pulses were compressed through a pair of prisms with double paths to reach a temporal resolution of 60 fs. For the transient absorption measurements, the pump beam polarization was set at a magic angle directly with respect to the probe beam, which was vertical. By translating the movable
parabolic mirror (MP in figure 2.3) out of the probe beam path, we quickly switched from
fluorescence detection to transient absorption measurements. The transient absorption
40 method can reach sensitivity of 10-4- 10-5 of the absorbance change. All signals were digitized and processed by a computer (PC).
The experimental setup is schematically shown in figure 2.3. Specifically, the femtosecond pulse after the two-stage amplifier (Spitfire, Spectra-Physics) has a temporal width of 110 fs with energy of more than 2 mJ and a repetition rate of 1 kHz. The laser beam is then split into two equal parts to pump two optical parametric amplifiers (OPA-
800C, Spectra-Physics).
2.2 DNA photolyase enzyme, CPD substrates and reduced flavin species
2.2.1 Purification and characterization of photolyase
2.2.1.1 Purification of DNA photolyase
Bacterial strains and plasmids. Bacterial strains (UNC523) and plasmids containing phr gene with IPTG induction site and Ampicilin resistance gene were obtained from the
Aziz Sancar’s lab at University of North Carolina at Chapel Hill.
Protein purification. All column materials for protein purification were obtained from the
BioRad®. Basic chemical needs as outlined in buffer composition were fulfilled either by
Fisher-scientific® or by Sigma-Aldrich® . The cell lysis buffer contained 50 mM Tris-
HCI, pH 8.0, 100 mM NaCl, 1 mM EDTA, and 10 mM β-mercaptoethanol. Buffer A contained 50 mM Tris-HCI, pH 7.4, 1 mM EDTA, 0.1 M KCl, 10 mM β- mercaptoethanol and 10% glycerol (v/v). Buffer B is Buffer A containing 2.0 M of KCl.
Buffer C contained 67 mM potassium phosphate, pH 6.8, 1 mM EDTA, 10 mM β- 41 mercaptoethanol and 10% glycerol (v/v). Buffer D is buffer C with 330.0 mM of
potassium phosphate. Photolyase storage buffer contained 50 mM Tris, pH 7.4, 50 mM
NaCl, 1 mM EDTA, 10 mM dithiothreitol, 50% glycerol (v/v). A brief discussion of the
protein purification protocol is provided below and the detailed information is attached in
Appendix A.
Cell-free extract. Cultures were routinely grown in Luria broth which contained 100ug /
ml ampicilin. To induce photolyase expression, 500 uM IPTG (isopropylthio-β-
galactoside) was added after cells were grown to A600 = 0.6 – 0.8 and incubation was then
continued for 4 more hours. Cells were harvested by centrifuging at 4000g. Cell pellets
were resuspended and washed with PBS and stored in PBS at -80°C until further use.
For further use, frozen cells were thawed in ice-water bath at 0°C. Once thawed,
lysis buffer was added to lyse the cells. The cells with lysis buffer were stirred in an ice-
water bath for 6-8 hrs for efficient lysis. In addition, lysis cells were sonicated with a
Branson Model 450 sonifier, set at 40% duty cycle for the 1 mm sonicator tip. Cell debris
was removed by centrifugation at 21,000 x g for 1 h to obtain a clear cell-free extract.
Ammonium sulfate precipitation. Ammonium sulfate with 43gm c/o was added to cell- fraction extract over a period of 45 minutes with gentle magnetic stirring. The stirring was then continued for another 45 minutes before collecting the precipitate by centrifugation. Precipitated proteins were dissolved in buffer A. To further remove the ammonium sulfate, the dissolved protein mass was dialyzed against the buffer A.
42 Affinity chromatography. The protein mass obtained as described above was slowly loaded onto a 60-ml affinity-gel Blue Gel (Bio-Rad) column equilibrated with Buffer A.
After the loading, the column was washed with 120 ml of the same buffer. Buffer B with high salt concentration was used to elute the protein. 5 ml fraction were collected and analyzed using UV-visible spectrophotometer by monitoring 580 and 625 nm absorption peaks of flavin neutral radical in photolyase. The fractions were also analyzed using
SDS-PAGE. Fractions containing observable flavin neutral radical and having SDS-
PAGE band at around 55,000 were pooled together.
Size-exclusion/Gel-Filtration chromatography. The photolyase fractions as obtained from
affinity chromatography were concentrated and loaded over BioGel P-100 which was
previously equilibrated with Buffer C. Bio Gel resin is hydrophilic and essentially free of
charge, and provides efficient, gentle gel filtration of the compounds. Gravity run with
buffer C was performed; fractions of 5 ml were collected and analyzed using UV-visible
spectrophotometer and SDS-PAGE. Fractions containing observable flavin neutral
radical and having SDS-PAGE band at around 55,000 were pooled together.
Hydroxyapatite chromatography. The photolyase fractions pooled from size
chromatography were loaded onto hydroxyapatite column (Bio-Rad HT gel) which was
previously equilibrated with Buffer C. A linear gradient elution with buffer C and D was
used to elute the purified fractions of photolyase. Purity of the protein was judged based
on the SDS-PAGE analysis and UV-visible spectrophotometer. At this stage protein was
dialyzed against the storage buffer E. Dialyzed protein was stored at -80°C for further use
in experiments.
43 2.2.1.2 Characterization of DNA photolyase
Purified photolyase was characterized using the SDS-PAGE analysis and UV-
visible absorption spectroscopy. After three column purification a single, isolated band at
around 55 KD was obtained and hence photolyase was considered pure to homogeneity.
During the purification, the flavin cofactor in photolyase get oxidized to flavin neutral radical and has characteristic absorption bands at 480, 580 and 625 nm (Sancar et al.
2003). The other cofactor MTHF also shows characteristic absorption at 384 nm.
Presence of characteristic absorption bands in purified proteins confirmed the presence of
photolyase (figure 1.1). Functional assay of enzyme was performed by monitoring the
CPD repair by photolyase at 264 nm in the steady-state experiment (section 2.2.4).
2.2.2 Photoreduction of flavin cofactor & photodecomposition of folate
cofactor
2.2.2.1 Photoreduction of flavin cofactor
The photoreduction of oxidized flavin neutral radical to fully reduced flavin was
achieved essentially using the protocol as described by Heelis et al. (1986). The protocol
was further modified depending on the needs of the experiment. For a typical steady-state
experiment micromolar concentration of photolyase enzyme in photoreactivation buffer
(50 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, 10 mM dithiothreitol, 5% glycerol
(v/v)) was illuminated with a high-intensity lamp with a cutoff filter to ensure that the
sample was exposed at wavelengths longer than 550 nm. The sample was kept over ice to
avoid thermal degradation. Absorption spectra were measured from time to time to 44 monitor the decrease in flavin neutral radical absorption (figure 2.4). For the time- resolved studies a concentration of 300 uM was used in a buffer solution at pH 7.4, containing 50 mM Tris, 50 mM NaCl, 1mM EDTA, 10 mM DTT, and 50% (v/v) glycerol. The sample was extensively purged with nitrogen to remove oxygen and was sealed in the air tight reaction cell. Then the reaction cell containing photolyase sample was illuminated with a high-intensity lamp with a cutoff filter to ensure that the sample was exposed at wavelengths longer than 550 nm. A detailed protocol for flavin reduction is attached as Annexure B.
2.2.2.2 Photodecomposition of folate cofactor
The photodecomposition of folate cofactor from photolyase enzyme was achieved using the protocol as described by Heelis et al. (1987). For a typical steady-state experiment micromolar concentration of photolyase enzyme in a buffer solution at pH
7.4, containing 50 mM Tris, 50 mM NaCl, 1mM EDTA, 10 mM DTT, and 50% (v/v) glycerol was irradiated with high-intensity lamp (Black–Ray, Model 100B-SP UVP lamps) with peak intensity at 366 nm. The sample was kept over ice to avoid thermal degradation. Absorption spectra were measured from time to time. Decrease in absorption at 384 nm and fluorescence emission at 468 nm was monitored to ensure the complete decomposition of folate cofactor (figure 2.5). In the present experimental conditions, the anaerobic environment was found to be essential for 100% removal of the folate cofactor. A detailed protocol for flavin reduction is attached as Annexure B.
45 2.2.3 Preparation and characterization of CPD substrates
Cyclobutane dimers may be formed from the excited triplet state of pyrimidines
following singlet-triplet intersystem crossing (Lamola 1968). Hence, a triplet sensitizer
such as acetone, which populates the thymine π,π* triplet state, was used for the
preparation of thymine dimers as described by Meistrich et al. (1972). In a typical
preparation, 100 ODs of thymine dinucleotide (Sigma T-8880) or oligo dT15 (IDT) were
dissolved in 1 ml of double distilled water. 15% (v/v) acetone was added in the solution
and the sample was placed in UV-transparent reaction cells. Then the sample was
irradiated using a 15W, 302 nm lamp (UVP 34-0039-01). During irradiation the reaction
cell containing dinucleotide sample was kept over ice to avoid thermal degradation. A
very slow flow of Argon gas was maintained over the dinucleotide solution to avoid air
oxygen which might quench the effect of acetone as sensitizer. From time to time during
irradiation 1ul aliquots of sample were taken out from the reaction mixture to monitor the
decrease in dinucleotide absorption at 260 nm. Under the experimental condition as described above, 2-3 hr irradiation was found sufficient to convert more than 95% of dinucleotide to thymine dimers (figure 2.6). Conversion efficiency was calculated using the decrease in absorption values at 260 nm. A detailed protocol of CPD preparation is attached in Appendix C.
2.2.4 Photorepair of damaged DNA (CPDs)
Photolyase enzyme and substrates were characterized by following the CPD
repair in the presence of blue-light as described by Sancar et al. (2006). Assay buffer
46 containing 50 mM Tris, pH 7.4, 50 mM NaCl, 1 mM EDTA, 10 mM dithiothreitol, 5%
glycerol (v/v) was prepared. 400 nM photolyase in assay buffer was incubated with the
750 uM of thymine dimer (CPD) for 5 minutes in a 5 mm quartz cell at 4°C in aerobic
conditions. The solution in the quartz cell was exposed to 366 nm light and absorption at
264 nm was monitored every five minutes. Repaired thymine dimer by photolyase
becomes thymine monomer which has a characteristic absorption at 264 nm (Sancar et al.
2003 and reference cited therein). The thymine dimer has no absorption at 264 nm and
the background absorption of enzyme at 264 nm was subtracted every time. The time
dependent build-up of thymine absorption (figure 2.7) with 366 nm exposure confirmed
that purified photolyase was in its functional form and the prepared substrates were cis-
syn cyclopyrimidine dimers. For time-resolved studies a 400 µM folate degraded
photolyase was used in the reaction buffer containing 50 mM Tris (pH 7.4), 50 mM
NaCl, 1 mM EDTA, 10 mM DTT, and 50% (v/v) glycerol. The sample was extensively
purged with nitrogen and was photoreduced as described in section 2.2.2.1. Thymine
dimer substrates was added 30-fold (final concentration 12 mM) in excess with the
enzyme concentration to ensure the saturation of the enzyme with substrate. The
equilibrium binding constants of E. coli photolyase for T<>T dinucleotide dimer
substrates are typically in order of 103 to 104 M–1 (Sancar 2003). This mixture of the fully reduced enzyme with substrates was prepared under yellow light and anaerobic conditions, and it had no measurable absorption at >500 nm.
47 2.2.5 Preparation and characterization of reduced flavin species
1, 5-dihydroflavin adenine dinucleotide (FAD) and mononucleotide (FMN) were
purchased from the Sigma®. 300 ul of 325-350 uM flavin solution in 12.5 mM phosphate buffer either at pH 8.5 or pH 5.0 was extensively purged with oxygen free, ultra-purified nitrogen gas for more than 120 minutes in the specially designed reaction cell. Flavins were chemically reduced either by adding final concentration of 2-4 mM sodium dithionite (Fox 1974) or 30-250 mM of sodium borohydride (Muller et al. 1969).
Reduced flavins were also generated photochemically (Ghisla et al. 1975) by adding final
concentration of 10-20 mM of sodium-oxalate in nitrogen purged samples and later by
irradiating the samples using a UV-lamp at 360 nm for more than 10 hrs. The reaction
cell was designed to maintain anaerobic conditions. In most cases no reoxidation of flavin
was observed up to six days. Recovery of fully oxidized flavin immediately upon
admission of air was up to 95%, as measured by absorption spectroscopy. No detectable
changes in absorption and fluorescence spectra were observed after time resolved studies.
Monitored at specific wavelengths, excited-state dynamics of reduced flavins were found
to be independent of the flavin reduction method in use. To note, in presence of oxygen,
flavin reduced by sodium borohydride yields a 3, 4-hydroxyl product (Muller et al.
1969). Because of different absorption spectral properties (Muller et al. 1969), the 3, 4-
hydroxyl product worked as oxygen sensor for the reduced samples. The excited-state
dynamics of this 3, 4-hydroxyl product was found to be significantly different (Chapter
6). A detailed protocol of reduced flavin preparation is attached in Appendix D.
48
Figure 2.1 Layout of femtosecond pump-probe methods
The time delay between pump and probe pulses is determined by the optical path length difference between the two pulses. (A) Set-up for fluorescence emission measurements using the up-conversion technique. The pump/excitation pulse induces the fluorescence in the sample cell. Using a parabolic mirror pair emitted fluorescence were focused on the non-linear crystal at the same spot as the gate/probe pulse. After sum-frequency generation, as described in the text, upconverted signals were detected using a computer controlled photomultiplier tube (PMT). (B) Set-up for transient absorption measurements. The pump/excitation pulse initiates the reaction in the sample cell and the probe pulse captures transient absorption of molecular species as a function of the delay time. (Figure adopted from Vos et al. 1999 and modified).
49
Figure 2.2 Schematic representation of fluorescence up-conversion technique
This representation explains that the fluorescence signals appearing at the same time delay τ (shaded area) as the gate pulse are up-converted (squares). (This is a modified figure adapted from Zgrablic 2006).
50
Figure 2.3 Schematic representation of the experimental setup with both the fluorescence up- conversion and the transient absorption configurations
The dashed line represent the transient-absorption probe pathway. F, filter. MM, movable mirror. MP, movable parabolic mirror. PD, photodiode. Millenium, Tusnami, Spitfire, Evolution 30, and SSA are the pump laser, femtosecond oscillator, two-stage amplifier, amplifier’s pump laser, and single-shot autocorrelator, respectively.
51
Figure 2.4 Typical photoreduction profile of flavin neutral radical in photolyase
Photoreduction of flavin neutral radical was carried out as described in text. Characteristic flavin neutral radical absorption bands at 485, 580 and 625 nm gradually disappears upon photoirradiation. Curve 1, 2, 3, 4 and 5 are at time of 0 min, 5 min, 10 min, 20 min and 30 min, respectively, during irradiation.
52
Figure 2.5 Typical photodecomposition profile of folate cofactor in photolyase
Photodecomposition of folate cofactor MTHF was carried out as described in text. (Upper) Characteristic MTHF absorption band at 384 nm gradually disappears upon photoirradiation. During photodecomposition flavin neutral radical also gets reduced and looses its characteristic absorption bands at longer wavelength. (Lower) Characteristic MTHF emission at 468 nm gradually disappears upon photoirradiation. Curve 1, 2, 3 and 4 are at time of 0 min, 60 min, 120 min and 180 min respectively during irradiation.
53
Figure 2.6 Typical preparation of cyclobutane pyrimidine dimers (CPDs)
Cyclobutane pyrimidine dimers (CPDs) were prepared as described in the text. Decrease in monomer absorption at 264 nm was monitored to characterize and quantify the dimer preparation.
54
Figure 2.7 Typical steady-state DNA repair
Enzyme assays were developed and substrates were characterized as given in the text. In a typical steady-state DNA repair experiment, photolyase converts CPDs into pyrimidine monomers, which strongly absorb at 264 nm. Increase in absorption at 264 nm indicates that with multiple catalytic cycles more and more monomer get accumulated. CPDs were incubated in the presence of photolyase and the absorption spectrum was obtained every 5 minutes under exposure of 366 nm light.
55 CHAPTER 3
RESONANCE ENERGY TRANSFER FROM THE FOLATE
TO THE FLAVIN COFACTOR
3.1 Resonance energy transfer in photolyase
3.1.1 Introduction
Photolyase is one of a few photoenzymes existing in nature and performs an
important biological function of repairing damaged DNA (Begley 1994, Sancar 2003).
The function is trigged by absorption of one photon and thus the dynamics can be well
synchronized with the functional process by a femtosecond laser pulse initiation.
DNA photolyase from E. coli has been extensively studied (section 2.2). During
catalysis, the first step is the absorption of a photon by the antenna MTHF which then
transfers the excitation energy to the catalytic cofactor FADH– to enhance the DNA- repair efficiency (Payne et al. 1990). Earlier studies (Kim et al. 1991) reported that the
energy transfer from MTHF to FADH– takes 140-180 ps (lifetime) while it occurs in less
than 30 ps from MTHF to FADH•. The actual time scales have not yet been determined
precisely. We used femtosecond resolved fluorescence up-conversion technique to 56 determine the actual time scales of the initial resonance energy transfer from the antenna
(MTHF) to the fully-reduced (FADH–) and semi-oxidized cofactors (FADH•).
3.1.2 Dynamics of energy transfer from MTHF to FADH (neutral radical)
The absorption and emission spectra of different forms of photolyase enzyme is
given in figure 3.1. The femtosecond-resolved fluorescence transients of the enzyme
• complex EPL- FADH -MTHF at 400-nm excitation under aerobic conditions are shown in
figure 3.2 for three typical wavelengths from the blue side to the red side. All transients
were well fitted by a single-exponential decay with a time constant of 18 ps. At 400-nm
excitation, the MTHF absorption is dominant and the absorption coefficient is ~25000 M-
1cm-1 while it is ~3000 M-1cm-1 for FADH• (Sancar 2003). The excitation of FADH• in the enzyme results in ultrafast electron-transfer reactions with neighboring aromatic residues in less than 50 ps (chapter 4) and the excited state (FADH•*) is basically
nonfluorescent; even if it were emitting, the fluorescence profile should be at longer than
650 nm because its absorption extends to 700 nm (figure 3.1). Thus, the observed
fluorescence dynamics is purely from the MTHF* state. The fluorescence lifetime of the
* EPL-MTHF complex without the flavin cofactor was reported to be 354 ps at 355-nm excitation (Kim et al. 1991). Therefore, the observed dynamics of 18 ps represents the excitation energy transfer from MTHF* to FADH•. After subtracting the population-decay
contribution (354 ps), the time constant of resonance energy transfer is 19 ps and the
energy-transfer efficiency is 95%.
57 According to Förster energy-transfer theory, the rate of resonance energy transfer
(RET) from MTHF* to FADH depends on the relative position (r) and orientations of donor (MTHF) and acceptor (FADH•), and can be expressed as follows:
1 R0 6 kRET = ( ) ……(3.1) τ D r
2 2 −4 1/ 6 R0 = 9.78x10 (κ n QD J ) …….(3.2)
R0 (in nm), the critical transfer distance, is defined as the donor-acceptor distance at which the transfer efficiency is 50%. κ2 is the orientation factor, n is the refractive index of the medium (≈1.4), τD and QD are the donor excited state lifetime and quantum yield in the absence of the acceptor, respectively, and J is the spectral overlap integral (in unit of cm3⋅M-1) between donor-emission and acceptor-absorption. The x-ray structure reported a
• distance (r) of 16.8 Å between MTHF and FADH (Park et al. 1995). With τD=354 ps, the
-14 3 -1 derived R0 is 27.35 Å. Given that QD=0.32 and J=2.06x10 cm ⋅M (Kim et al. 1991), we obtained a value of 0.28 for the orientation factor κ2. This low value is consistent with the observed unfavorable orientations of two chromophores in the x-ray structure. The poor alignment of the two molecules may be due to the structural constrains from retaining three functional binding sites (MTHF, FADH• and the substrate DNA). The favorable orientation has been observed in another class of photolyases (Kim et al. 1992).
The observed single 18-ps exponential decay is independent of fluorescence wavelength detection. Thus, in buffer solution with 50% (v/v) glycerol the solvation process of MTHF* in the enzyme makes negligible contributions to the observed energy- transfer dynamics. The hydration dynamics on protein surfaces has been recently 58 observed on the time scale of tens of picoseconds using intrinsic tryptophan as a local optical probe (Zhong et al. 2002, Pal et al. 2002, Lu et al. 2004). In rubredoxin, the resonance energy-transfer process from the excited W3/W36 to the Fe-S cluster is convoluted with the hydration dynamics of tryptophan and both occur in ~15 ps (Zhong et al. 2002). In human thioredoxin, the electron transfer dynamics from W31* to the disulfide bond (S-S) mixes with the tryptophan hydration process and both also occur in
~20 ps (Qiu et al. 2006). In those systems, we typically observed fluorescence wavelength-dependent transients and the dynamics systematically slow down from the blue side to the red side. Here, we observed the similar transients for all fluorescence wavelengths, indicating that the chromophore MTHF is buried in the shallow cleft between the two domains and the solvation dynamics, if any, would be much longer than the observed energy-transfer process of 18 ps.
As shown in chapter 4, the excited FADH• proceeds through ultrafast electron transfer with neighboring aromatic residues to form the fully-reduced FADH–. However, the signal in figure 3.2 showed no changes for relatively long irradiation by laser pulses in our experiments, indicating the efficient reaction cycling of FADH• during the pulse interval (2 ms). Thus, in the presence of oxygen the FADH– could be oxidized back to
FADH• in less than 2 ms. In figure 3.3, we show the fluorescence transients gated at 530- nm emission in the presence and absence of O2. With O2 (figure 3.3a), the transient gave the same energy-transfer dynamics of 18 ps. Without O2 (figure 3.3b), a new component of 160 ps (20%) started to appear. As the irradiation continued, the percentage of the 160- ps component (65%) increased while the 18-ps component (35%) decreased, as shown in
– figure 3.3c. These observations show that without O2 the FADH could not be oxidized to 59 FADH• in 2 ms. Thus, the observed 160 ps results from resonance energy transfer from
MTHF* to the fully-reduced FADH– in agreement with previous measurements (Kim et al. 1991). Recent studies (Aubert et al. 2000, Byrdin et al. 2003), also showed that it takes more than 10 ms to finish reoxidation of FADH– under anaerobic conditions.
3.1.3 Dynamics of energy transfer from MTHF to FADH– (anionic reduced radical)
To confirm the observed energy transfer between MTHF and FADH–, we fully reduced FADH to FADH– by irradiation of the sample at visible light (λ≥550nm) under anaerobic conditions (Heelis et al. 1986), see figure 3.1. Figure 3.4 shows the femtosecond-resolved fluorescence transient gated at 440-nm emission. The signal is best fitted by a single-exponential decay of 160 ps, consistent with the result obtained at 530- nm emission. The excitation of FADH– (~4000 M-1cm-1) gives a weak emission peaked at
505 nm but no emission at 440 nm. Thus, the observed transient at 440 nm is purely from
MTHF* state and the observed 160 ps does represent the energy-transfer process from
MTHF* to FADH–. After considering the population-decay contribution (354 ps), the resulting time constant of resonance energy transfer is 292 ps and the energy-transfer efficiency is only 55%.
– According to eq. (3.1), the resulting R0 for MTHF-FADH energy transfer is
17.35 Å. Using eq. (3.2) and J=0.34x10-14 cm3⋅M-1 (Kim et al. 1991), we derived a value of 0.11 for the orientation factor κ2, which is smaller than the one (0.28) obtained for the relative MTHF- FADH• orientations. After the reduction of the neutral FADH•, the negatively charged flavin cofactor FADH– may optimize electrostatic and stacking 60 interactions with neighboring charged residues such as R344, D372 and W277 resulting in the orientation change as we observed here. Otherwise, if we assume the same orientation factor of 0.28 as for FADH•, we would obtain a spectral overlap integral of
0.13x10-14 cm3⋅M-1 for MTHF-FADH– energy transfer.
We also observed that over long irradiation of the sample by laser pulses the signal in figure 3.4 gradually decreased but the overall shape remained the same. This is due to the photodecomposition of MTHF caused by one electron transfer from the excited
FADH–, as reported before (Heelis et al. 1987).
3.2 Comparative study of resonance energy transfer in cryptochrome
3.2.1 Introduction
Cryptochrome, a recently discovered blue-light photoreceptor, synchronizes the circadian rhythm in animals and regulates growth and development in plants (Cashmore et al. 1999, Sancar 2000, Sancar, 2003). It is highly homologous to photolyase in sequence, a well-studied photoenzyme for repairing damaged DNA (Lin et al. 2003).
Both are flavoproteins containing two noncovalently-bound chromophores. One chromophore is a flavin-adenine dinucleotide (FAD) as a key cofactor to carry out initial biological function upon photo-excitation and the other is a pterin in the form of methenyltetrahydrofolate (MTHF) as a light-harvesting antenna to enhance biological efficiency. In vivo, it is believed that the flavin is in the fully-reduced FADH– form but during purification it is oxidized to FADH• or FAD. The photophysics and photochemistry of the two chromophores in cryptochrome have not been studied so far for lack of purified cryptochrome containing both cofactors in significant quantities and 61 thus there is no data indicating interchromophore energy transfer in cryptochrome as has been found in photolyase. Recently, it was found that a cryptochrome purified from
Vibrio cholerae, called VcCry1, contained both cofactors in near stoichiometric amount that makes ultrafast kinetic studies feasible. In this communication, with femtosecond resolution we report our first measurement of resonance energy-transfer dynamics from the photoantenna MTHF to the fully-reduced active cofactor FADH– in V. cholerae cryptochrome, VcCry1.
VcCry1 is a monomeric protein with 466 amino acid residues and its absorption and emission spectra are shown in figure 3.5. In contrast to all cryptochromes purified so far that contain FAD in the two-electron oxidized form, in VcCry1 the flavin is mostly in
FADH– form with minor amount of oxidized FAD, which is unavoidable in the protein preparation (Worthington et al. 2003). The neutral radical FADH• has absorption at longer than 500 nm thus no FADH• cofactor exists in our protein preparation. The fluorescence spectrum is dominated by the MTHF* emission with a λmax=480 nm; at longer wavelength overlapping with the emission from both FAD* and FADH–*.
3.2.2 Resonance energy transfer from MTHF to flavin cofactor
In figure 3.6, four typical femtosecond-resolved fluorescence transients are shown, covering the blue-side to the red-side emission. At 400-nm excitation, the fluorescence emission at 440 nm is purely from MTHF*; both FAD* and FADH–* emit at longer wavelengths with a peak at about 530 nm. The 440-nm transient can be best fitted by three exponential components: 10 ps decay with 14% of the total amplitude, 60 ps
(58%) and 845 ps (28%). Protracted handling of the sample under aerobic conditions
62 results in gradual oxidation of FADH– to FAD as revealed by absorption increase in the range of 450-500 nm (not shown). We repeated femtosecond-resolved fluorescence studies for the oxidized sample and the obtained transient decayed with the same three time constants but with different amplitudes, as shown in the inset of figure 3.6 with removal of the lifetime components. The 10-ps component increased to 65% and the 60- ps component decreased to ~10%. Thus, the 60-ps decay represents the resonance energy-transfer process from MTHF* to FADH– and the 10-ps component reflects the transfer dynamics from MTHF* to FAD. The 845 ps is therefore the lifetime of MTHF* in the protein because free MTHF that might be released from the protein is not stable at neutral pH and is converted to 10-formyltetrahydrolfolate that does not absorb at λ>300 nm (Kim et al. 1991). Using the obtained percentages for each chromophore and the known molar extinction coefficients of MTHF, FADH– and FAD, (Sancar 2003, Massey
2000) the calculated absorption spectrum exactly reproduced the measured one in figure
3.5.
The fluorescence transients of VcCry1 gated at other three wavelengths of 480 nm, 520 nm and 560 nm show a similar temporal behavior. All transients can be represented by the similar three decay components with 10 ps (14-16%), 60 ps (58-47%) and 845-1045 ps (28-37%). At the longer wavelengths, the FADH* (and minor FAD*) fluorescence emissions make some contributions. Overall, the observed ultrafast decay
(10 and 60 ps) is independent of fluorescence wavelength detection. Thus, in a buffer with 50% (v/v) glycerol, the solvation process of MTHF* in the protein makes negligible contributions to the observed energy-transfer dynamics, as observed in E. coli photolyase.
63 Thus, the chromophore MTHF is buried in the protein and the solvation process, if any, would be much longer than the observed energy-transfer dynamics (10 ps and 60 ps).
We also measured the anisotropy dynamics of MTHF* at 440-nm emission and the obtained anisotropy was constant with a value of ~0.32 within 1.5 ns (not shown).
The result revealed a rigid local structure around the MTHF, indicating that the chromophore is tightly bound to the protein with minimal local motions, also consistent with the absence of ultrafast solvation processes. The recently solved X-ray structure of the photolyase homology (PHR) domain of A. thaliana cryptochrome (Brautigam et al.
2004) and of the Synechocystis cryptochrome (Brudler et al. 2003) revealed flavin binding sites very similar to that of photolyase (Park et al. 1995). Thus the local motion of the FADH– is highly restricted in this family of proteins. Therefore, the relative orientations of the energy-transfer pairs, MTHF with FAD/FADH–, are well aligned and the orientation factor (κ2) is well defined, as we observed in E. coli photolyase. The observed long lifetime component in figure.3.6 can be ascribed to MTHF in the absence of flavin as in a fraction of VcCry1 protein the flavin is lost during purification
(Worthington et al. 2003). According to Förster resonance energy transfer formulation, the ratio of two energy-transfer rates from MTHF* to FADH– and to FAD is
2 2 k − / k = (J − κ − ) /(J κ ). FADH FAD FADH FADH FAD FAD …….(3.3) where, J is the spectral overlap integral and was calculated to be 0.34 X 10-14 cm3•M-1 for the MTHF*-FADH– pair and 3.48 X 10-14 cm3•M-1 for MTHF*-FAD. From our measurements, the ratio of the energy-transfer rates of MTHF* to FADH– and FAD is
1/6. Thus, we obtained a ratio of 1.7 for the orientation factors for the two energy transfer
64 2 2 pairs (κFADH − to κFAD ). This result indicates a local structural change from the charged
(FADH–) to the polar (FAD) flavin redox state due to the strong electrostatic interactions with the surrounding residues at the binding site. Similar results were observed in photolyase containing FADH– and FADH•, respectively.
3.3 Conclusions
We reported our first studies of the enzymatic dynamics on DNA photolyase with femtosecond resolution. We used femtosecond fluorescence up-conversion method to follow the resonance energy transfer by photoantenna initiation. The molecular mechanism and the measured dynamics are summarized in figure 3.9. Specifically, we elucidated the following important processes:
(1) The dynamics of excitation energy transfer from the antenna molecule to the
cofactor were determined from both the fluorescence detection and transient-
absorption measurements consistently: 292 ps for the physiologically relevant
cofactor FADH– and 19 ps for the in-vitro oxidized cofactor FADH•. The derived
orientation factors are 0.11 for the MTHF-FADH– pair and 0.28 for MTHF-
FADH•.
(2) Both cryptochrome and photolyase have similar flavin binding sites with a unique
hole configuration (Brautigan et al. 2004, Brudler et al. 2003). Structure of
photolyase containing both cofactors is available. However, the available
cryptochrome structures were obtained with proteins devoid of MTHF and hence
the relative configuration of the two cofactors in cryptochromes is not known.
The results reported here suggest different binding interactions and local 65 structures of MTHF in the two proteins, at least with respect to VcCry1. The
MTHF* lifetime of 845 ps in VcCry1 cryptochrome is more than two times
longer than that in E. coli photolyase (354 ps), indicating a more hydrophobic and
rigid environment. The energy-transfer rate is more than four times faster than
that in photolyase, suggesting a shorter distance or a highly favorable orientation
of the two chromophores. The ultrafast energy-transfer dynamics in
cryptochrome reported here (60 ps) must enhance its functional efficiency.
Importantly, our results for the first time show energy transfer from MTHF to
flavin in cryptochrome suggesting some mechanistic similarities between
photolyase that repairs damaged DNA and cryptochrome that mediates blue-light
signaling.
New biochemical (Selby et al. 2006) studies suggest that many cryptochromes, including VcCRY1, actually are photolyases with high degree of specificity for cyclobutane pyrimidine dimers in ssDNA. Crystal structure of cryptochrome3, a sequence homologue of CPD photolyase and VcCRY1, obtained from Arabidopsis thaliana has been solved recently (Huang et al. 2006). Structure of cryptochrome3 shows significant similarity with structure of CPD photolyase. In view of these new reports, results presented in this chapter become even more relevant. After comparing the excited state lifetime of MTHF cofactor in cryptochrome and photolyse we predicted that in cryptochrome, MTHF might be buried in a rigid and hydrophobic protein pocket. Crystal structure of cryptochrome3 supports our predication. Crystal structure of the cryptochrome shows that MTHF makes eight hydrogen bonds to the side-chain atoms of various amino acids. These hydrogen bonds, oriented in different directions, both 66 enhance the affinity between MTHF and the protein and stabilize the conformation of
MTHF (Huang et al. 2006). VcCRY1 has been shown to have CPD repair capabilities, where CPDs are in single stranded DNA (Selby et al. 2006). It is possible that VcCRY1 might follow the similar mechanism of CPD repair as CPD photolyase does. In such case present studies will provide a good starting point to follow the catalytic mechanism of
CPD repair by VcCRY1.
67
Figure 3.1 The absorption and emission spectra of various forms of photolyase enzyme
Upper: The absorption spectra of E. coli photolyase in semi-oxidized (thick-gray line) and fully-reduced (thin-dark line) flavin forms. Lower: The absorption spectra of the folate-depleted enzyme with neutral radical and reduced forms of flavin. MTHF emission spectrum (dashed line) is shown to overlap with the absorption spectra of flavin, indicating resonance energy transfer from MTHF to flavin. Semi-oxidized FADH• is represented as FADH. (This figure is same as figure 1.1 but reproduced here to maintain the flow of the text).
68
Figure 3.2 Fluorescence transient profile of MTHF in photolyase and the cofactor flavin is present in the neutral radical form
Femtosecond-resolved fluorescence transients of MTHF* probed at several typical wavelengths. All three transients show a single-exponential decay of 18 ps. Note that the cofactor flavin is in the neutral radical form FADH•.
69
Figure 3.3 Femtosecond-resolved fluorescence transients of MTHF in photolyase with and without O2
Femtosecond-resolved fluorescence transients probed at 530 nm in: (a) aerobic conditions (with O2), (b) anaerobic conditions (no O2), and (c) anaerobic conditions with longer-time data collection; see text. The 18-ps and 160-ps dynamics represent the MTHF*- FADH• and MTHF*-FADH– energy transfer, respectively.
70
Figure 3.4 Femtosecond-resolved fluorescence transient profile of MTHF in photolyase and the flavin cofactor is present in the anionic reduced form
Femtosecond-resolved fluorescence transient of the MTHF* decay probed at 440 nm. The transient follows a single-exponential decay of 160 ps. Note that the cofactor flavin is in the fully-reduced form FADH–.
71
Figure 3.5 Absorption and emission spectra of V. cholerae cryptochrome
Absorption and emission spectra of V. cholerae cryptochrome: The absorption is dominated by the MTHF cofactor. The upward arrows indicate the excitation wavelength (400 nm) and the four gated fluorescence emissions. Note that a small fraction of FAD (15%) is also present as indicated by an arrow in the 450-500 nm region.
72
Figure 3.6 Femtosecond-resolved fluorescence transients of the MTHF* decay in VcCRY1
Femtosecond-resolved fluorescence transients of the MTHF* decay probed from the blue- side to red-side emission at 400-nm excitation. All four transients show a three- exponential decay of 10 ps, 60 ps and ~845 ps. In the inset, the signal with the faster decay (square symbol) represents the protein sample with FAD as the dominant flavin cofactor; see text.
73
Figure 3.7 Schematic representation of the molecular mechanism and the measured resonance energy transfer dynamics
Schematic representation of the molecular mechanism and the measured dynamics of energy transfer between two cofactors in photolyase and in cryptochrome proteins.
74 CHAPTER 4
DYNAMICS OF THE FLAVIN COFACTOR
4.1 Introduction
The flavin cofactor of the enzyme in vitro is usually oxidized into the inactive neutral radical form FADH• after purification. Under illumination with the visible light, the cofactor can be converted to the active form through electron transfer from neighboring aromatic residues. The residue tryptophan 306 (W306) was identified as the ultimate donor by site-specific mutagenesis and fast kinetics and EPR analyses (Kim et al. 1993, Li et al. 1991). Two recent studies (Aubert et al. 2000, Byrdin et al. 2003) reported the dynamics occurring in ~25 ps through one electron transfer from a nearby tryptophan (W382) favoring one of the two pathways proposed based on the crystal structure (Park et al. 1995). However, a steady-state photoreduction study indicated that the W382F mutant was photoreduced with 2-fold higher quantum yield relative to the wild type (Sancar 2003). Clearly, more detailed work was necessary to understand the intraprotein electron transfer. We carefully studied the photoreduction of oxidized neutral radical cofactor FADH•. We observed two reduction pathways and the observed
75 dynamics are different from those reported in the previous studies. The dynamic of the
FADH– reoxidation after photoreduction under aerobic conditions was also examined.
4.2 Dynamics of the flavin cofactor reduction
To understand the electron-transfer (ET) mechanism of FADH photoreduction,
we performed a series of transient absorption measurements to observe the ET state and
the ground-state recovery of FADH•. Figure 4.1 gives three typical transients probed in
the range of 400-690 nm. At excitation of 580 nm, only FADH• was excited and MTHF has no absorption. Specifically, at 690 nm we only observed a decay signal and thus the
FADH•* dynamics was dominant. The ET products of FADH– and W+, if the electron
donor is tryptophan, have no absorption at 690 nm. The transient was best fitted by a
double-exponential decay with two time constants of 11 ps (36%) and 42 ps (64%),
respectively. At 500 nm, we observed an initial bleaching and a formation/recovery
signal. Thus the absorption of the ET states and the ground-state FADH• were dominant.
The transient can be best fitted with two time constants of 42 ps (60%) and ~3.3 ns
(40%), if the signal returns to zero. We did not observe any contributions from the 11-ps
component as observed at 690 nm. When the probe wavelength was tuned to 400 nm, we
observed a positive decay signal and thus the FADH•* dynamics dominated again. The
transient has the similar dynamics as observed at 690 nm with the two time constants 10
ps (42%) and 42 ps (58%).
Previous site-specific mutagenesis (Li et al. 1991) identified W306 as the ultimate
electron donor for the flavin photoreduction. Based on the crystal structure, two potential
76 electron-transfer pathways of the FADH photoreduction were proposed (Park et al.
1995): One passes through a tryptophan triad (W382←W358←W306) with three
electron hops and the other route to W306 consists of an α-helix (α-15) between residues
F366 and D358 through electron tunneling; see figure 4.2. It was recently reported
(Aubert et al. 2000, Byrdin et al. 2003) that photoreduction exhibits a single-exponential
decay of 24 ps for the FADH•* dynamics and this was attributed to electron transfer from
W382 (figure.4.2) to FADH* to form W382+ and FADH–. This process was then
followed by two subsequent electron hops from W358 and W306 in less than 10 ns for
each. These observations favored the electron-hopping pathway. However, steady-state
photoreduction measurements with the W382F mutant gave results inconsistent with this
model (Sancar 2003). In contrast to the time-resolved transient absorption measurements
(Byrdin et al. 2003) that indicated a nearly complete turn-off of the photoreduction, the
quantum yield measurements by steady-state irradiation revealed 2-fold enhancement of
photoreduction (Sancar 2003), which favored an electron tunneling mechanism, also
supported by quantum calculations (Cheung et al. 1999).
The double-exponential decay of the electron-transfer dynamics from aromatic
residues to flavin have also been observed in other flavoproteins such as riboflavin- binding protein (Zhong et al. 2001), glucose oxidase (Zhong et al. 2001, Mataga et al.
2000), and flavodoxin (Mataga et al. 2002). These double decays were attributed to the
multiple neighboring electron donors of aromatic residues (W, Y, or F) accompanying
with local dynamic heterogeneous configurations. Here, the observed double-exponential
decay of FADH•* in photolyase also reflects the dynamic heterogeneity of local
77 structures, resulting in two electron-transfer pathways: Electron hopping through the
tryptophan triad with the initial electron jump from W382 in ~10 ps, and electron
tunneling through the α-15 helix with the initial electron separation through F366 in ~40
ps. The distance from edge to edge of the rings between F366 and FADH• is 4 Å. The
electron hop from W382 is faster because of the favored redox potentials (W/W+ vs.
F/F+) although the separation is larger (4.2 Å). The observed initial two electron-transfer
pathways are also strongly supported by the transient probed at 500 nm (figure 4.1); we
only observed the 42-ps component and did not observe the 11-ps contribution. The
absence of the 11-ps dynamics is due to the weak absorption of FADH•* and W382+
(~1000 M-1cm-1 at 500 nm) while for the tunneling pathway we could detect the F366+ formation because it has an absorption maximum (~3000-4000 M-1cm-1) at ~500 nm
(Shida et al. 1988, Shida et al. 1966). The observed 3.3-ns dynamics at 500 nm reflects
the recovery process of FADH• either by back electron transfer or through oxygen
reduction.
Similar dynamics as described above was obtained after exciting the samples at
620 nm. The oxidation potential of tryptophan (W/W+) (DeFelippis et al. 1991, Harriman
1987) is ~ +1.0 V vs normal hydrogen electrode (NHE), and the reduction potential of the
flavin cofactor (FADH/ FADH–) (Heelis et al. 1992) was estimated between -0.33 and -
0.5 V vs NHE. Under excitation of 620 nm (2 eV, close to 0-0 transition), the net free
energy change (∆G°) of the electron-transfer reaction from W382 to FADH is negative,
between -0.67 and -0.5 V, and the reaction is feasible. The oxidation potential of
phenylalanine is not known but is estimated to be less than +2 V vs NHE (Vorsa et al.
78 1999, Kim et al. 2000). The net free energy change (∆G°) for electron transfer from F366
to FADH is around zero, and the reactivity highly depends on the local environment.
Most of the surrounding residues around F366 are polar and charged, and four oxygen
atoms of peptide bonds are in proximity of the F366 ring within 4.5 Å. All of these favor
charge separation and stabilize the F366+ cation. Another possibility is that the reaction
of FADH•* with phenylalanine (F366) follows Mulliken’s charge-transfer mechanism
(Mulliken 1950) as observed in the benzeneiodine complex (Cheng et al. 1996, Zhong et
al. 1999). Because of the van der Waals contact and the strong orbital overlap between
the donor and acceptor, there might exist a charge-transfer excited state. The charge-
transfer potential surface can cross with the excited local-state one (Cheng et al. 1996,
Zhong et al. 1999); thus, excitation of a local state (FADH•*) finally results in a charge
separation. The distance between FADH and F366 is 4Å, also favoring a charge-transfer
reaction. Thus, we believe that the electron-transfer reaction between FADH•* with F366 is feasible in E. coli photolyase, which is also supported by recent theoretical calculations
(Cheung et al. 1999).
To further confirm the observed two initial electron-transfer pathways, we tuned the pump wavelength to 400 nm to dominantly excite MTHF. Figure 4.3 shows two transients probed at 690 nm and 500 nm. As we observed above (chapter 3), the excitation energy transfer from MTHF* to FADH• takes 18 ps and the electron transfer of
the flavin FADH• reduction proceeds along two pathways with 11 ps and 42 ps. Thus, for
the electron-hopping route, since the FADH•* decay (11 ps) is faster than its formation
(18 ps), the FADH•* molecules could not be easily accumulated and we should also
79 observe an apparent formation signal of 11 ps. At 690 nm, we did observe a formation
signal of 11 ps (-18%) and a dominant decay component of 42 ps (100%). Thus, the
overall signal in the range of 0-10 ps appeared very flat due to the signal superposition of
the 11-ps formation and the 42-ps decay; see the insert in figure 4.3. We did not observe
any 18-ps dynamics at this wavelength, indicating the overall cancellation of the signal
decay and formation (MTHF* and FADH•*). These observations are striking and
convincing, and elucidate the two initial parallel photoreduction pathways through W382
and F366, respectively, sampled by the local dynamic configurations. When the probe
wavelength was tuned to 500 nm, we observed a strong positive signal and the transient
was best represented by a single-exponential decay of 18 ps, equal to the observed
energy-transfer dynamics and consistent with the fluorescence up-conversion
measurement. This result shows the dominant absorption of MTHF* at 500 nm.
It should be pointed out here that site-specific mutagenesis studies (Sancar 2003) of W382A showed complete elimination of photoreduction pathways. Those studies indicated that W382 is a structurally and chemically important residue. The mutation not only eliminates the electron-hopping pathway through the tryptophan triad but probably also changes the local structure and shuts off the initial step of the tunneling route.
Another scenario is that after the initial electron transfer from F366 to FADH•*, the
electron tunneling (or hole migration) does not follow the α-15 helix instead evolves
along the tryptophan triad again. The x-ray structure shows a close distance of 3.8 Å
between F366 to W382. Clearly, more femtosecond-resolved experiments of mutated
80 enzymes are needed to fully resolve these electron-transfer processes, especially mutation
of F366.
We also performed measurements of the DNA photolyase with the fully-reduced
state under anaerobic conditions. The femtosecond-resolved transient absorption is shown
in figure 4.4 with excitation at 400 nm and a probe at 500 nm. The dynamics is best
represented by a double-exponential decay with two time constants of 160 ps (52%) and
775 ps (48%). The 160-ps component, the decay of MTHF*, represents the excitation
energy transfer from MTHF* to FADH–, consistent again with the fluorescence up- conversion result. The observed 775 ps is the lifetime of the excited FADH– in excellent
agreement with the fluorescence up-conversion measurement of the FADH–* decay in the
MTHF-depleted enzyme, which is also shown in figure 4.4.
4.3 Solvation dynamics at the active site
The flavin cofactor is accessible to the flat enzyme surface through a hole in the
middle of the α-helical domain, and water molecules can easily diffuse in and out of the hole. The X-ray structure (Park et al. 1995) does show several water molecules involved in the H-bonding network around the cofactor in the active site. Fourteen amino acid residues are also bound to the cofactor, and most of them are polar and charged. The excited state flavin FADH•* can be easily solvated by such an environment through
strong electrostatic interactions. To understand the solvation dynamics we excited the
flavin neutral radical with 620 nm pump pulse and measured the excited state dynamics
at various wavelengths. Table 4.1 summarizes the measured dynamics. The ~ 2-ps decay
81 components appear in most transients (see Table 4.1) and must reflect solvation processes
of the excited state FADH•*. Ultrafast hydration dynamics on protein surfaces or in
shallow clefts have recently been observed to follow a double-exponential decay, for
most cases, in several and tens of picoseconds using an intrinsic tryptophan as a local
optical probe (Zhong et al. 2002, Pal et al. 2002, Lu et al. 2004) observed electron
transfer dynamics in ~10 and ~40 ps are longer than the initial solvation process
occurring in several picoseconds, resulting in the observation of the 2-ps solvation
component in most transients. The second longer solvation process, especially in the
active site, is probably much slower than the observed electron-transfer reactions (11 and
42 ps). Thus, the excited state population is mainly quenched by the reactions before
reaching the final relaxed state. The observation of ultrafast solvation dynamics at the
active site in the enzyme is significant because solvation processes, especially water
motions, are believed to play an important functional role in enzymatic reactions (Mattos
2002). Ultrafast water motions are crucial to protein conformational flexibility, which is
essential for enzymatic activity. The catalytic process is mediated by ordered water
molecules located at the molecular distance scale (figure 4.5). To maintain the order and
selectivity of water molecules in function, the observed picosecond time scale is ideal for
ordered water motions, for the case studied here, to stabilize the biologically active
cofactor FADH– as well as the reaction intermediates during the DNA-repair process.
These processes are faster than the local protein motions to allow the protein to optimize an ideal configuration immediately for efficient biological function.
82 4.4 Dynamics of the back electron transfer
The charge recombination is critical to photoreduction and the overall quantum
efficiency of photoreduction in E. coli photolyase is only about 0.05-0.1 (Heelis et al.
1987) All experiments shown in figure 4.6 were done in the presence of oxygen and oxygen has been shown to oxidize FADH– to the neutral radical FADH• in less than 2 ms
– through reduction of O2 to O2 . Figure 4.6 shows two transients which were taken in the
presence and absence of oxygen. Both transients are identical and have the same
dynamics. The electron transfer processes can be written as
* k − + + k FADH + W 382 / F366 →1 FADH + W 382 / F366 →2 FADH + W 382 / F366 k − →3 FADH + W 382 / F366
Thus, the observed ~3 ns is the time of the first charge-recombination process (k2) in a series of electron-transfer reactions in photoreduction. The second step of photoreduction (k3) must also occur on the similar time scale of several nanoseconds
because if k3 were much larger than k2 we would not observe considerable recovery of
FADH and if k3 were much smaller than k2 we would not have detectable flavin
photoreduction. The oxidation of FADH– by oxygen takes longer than ~3 ns and it is
probably in the microsecond range.
In the insert of figure 4.6, we also show the transients taken under anaerobic
– conditions (without O2) for three time intervals. After a period of time, the FADH molecules accumulated in the sample but they have no absorption at excitation of 620
83 nm. Thus, the total signal probed at 500 nm decreased. As shown in figure.4.6, after 4 hours of laser irradiation the signal only dropped 38% in total. Therefore, the overall charge-recombination reactions are severe.
4.5 Conclusions
Here, we used femtosecond transient absorption methods to dissect the elementary steps of electron transfer involved in photoreduction of the flavin neutral cofactor. Specifically, we elucidated the following important processes:
(1) The physiological form of the enzyme contains the flavin in two-electron reduced
FADH– form and the non-photoreducible mutants are as active as the wild-type
enzyme indicating that the photoreduction of FADH•, which is generated in vitro
during purification, may not be directly relevant to the action mechanism.
Nevertheless, photolyase does provide an excellent system to study mechanisms
of intraprotein electron transfer. Our results show that the dynamics of the FADH•
photoreduction, either through excitation energy transfer from MTHF* or by
direct excitation, evolve along two electron-transfer pathways to reach the final
electron donor W306: One involves with a tryptophan triad by the initial electron
hop from W382 in ~10 ps while the other route starts with the initial electron
separation through F366 in 40 ps followed by either the tunneling along the α-15
helix or the hopping through the tryptophan triad again.
(2) The ultrafast solvation dynamics was observed to occur in about 2 ps at the active
site through several local order-water motions. The electron-trasnfer reactions (11 84 and 42 ps) quenched the excited state and thus the longer solvation processes,
which typically occur in tens to hundreds of picoseconds, were not observed.
(3) The dynamics of the FADH– reoxidation after photoreduction under aerobic
conditions occurs in several nanoseconds (~3.3 ns) through back electron transfer.
In the absence of oxygen, the fully-reduced cofactor takes milliseconds to be
reoxidized.
In summary, the results reported in this chapter and in chapter 3 reveal significant
details on the nature of the ultrafast dynamics in DNA photolyase and lay a solid
foundation for further studies of mapping out the entire DNA-repair process. The
structure and dynamics of photolyase are beautifully intertwined to funnel energy along
the functional coordinate. During these processes the ultrafast dynamics are necessary to
reach high biological efficiency (Wang et al. 1994, Chosrowjan et al. 2004).
Femtosecond methods are ideal and powerful to reveal the actual dynamics and molecular mechanism on the local atomic scale as shown here in DNA photolyase.
85
λpr(nm) τ1 a1 τ2 a2 τ3 a3 A3/a2 τ4 a4 690 - - 11 0.35 42 0.65 1.86 - - 580 2 0.15 11 -0.25 42 -0.39 1.56 3400 -0.36 550 2 0.15 11 0.34 42 0.51 1.5 3200 -1.11 500 2 0.16 11 -0.16 42 -0.48 3.00 3100 -0.36 400 2 0.06 11 0.36 42 0.58 1.61 3000 -0.02
Table 4.1 Time constants (ps) and relative amplitudes
4 a −t /τi All transients were fitted by ∑ aie where i = 1 – 4 and the positive amplitude i=1 represents decay dynamics while the negative one means species formation (rise).
86
Figure 4.1 Time resolved dynamics of the flavin neutral radical
Femtosecond-resolved transient absorption measurements probed at 690 nm (top), 500 nm (middle) and 400 nm (bottom). The excitation wavelength is at 580 nm and only the FADH molecules were excited. The signal in the long-time range (more than 1 ns) probed at 500 nm is shown in the insert.
87
Figure 4.2 Potential electron transfer pathways for photoreduction of the flavin neutral radical cofactor
Potential electron-transfer pathways for neutral cofactor photoreduction: Excited FADH can be reduced by one electron hop from two amino acid residues W382 or F366 as shown by arrows. One route passes along the tryptophan triad (W382←W359←W306) while the other one evolves after the initial electron hop along the α-15 helix through the electron tunneling (F366←α-helix←W306) or by the electron hopping through the tryptophan triad again (F366←W382←W359←W306).
88
Figure 4.3 Femtosecond-resolved transient absorption measurements with excitation at 400 nm
Lower: At the probe wavelength of 690 nm, the transient is best fitted by an 11-ps formation component and a 42-ps decay contribution. Note that the flat signal in the 0-10 ps range in the insert is due to the superposition of these two components. The dashed line is the pure 42-ps decay shown for comparison. Upper: At the 500-nm probe, the MTHF* dynamics dominates with a signal-exponential decay of 18 ps, consistent with the fluorescence up-conversion result shown in figure 3.2.
89
Figure 4.4 Excited state lifetime of reduced flavin cofactor
Upper: Femtosecond-resolved transient absorption detection of the fully-reduced enzymes under anaerobic conditions at excitation of 400 nm and probe of 500 nm. The first decay component (160 ps) represents the energy transfer from MTHF* to FADH–, consistent with the fluorescence up-conversion result as shown in figure 3.4. The other component of 775 ps is the lifetime of FADH–* in the enzyme. Lower: Femtosecond- resolved fluorescence up-conversion signal of FADH–* in the MTHF-depleted enzyme at 520-nm emission with 400-nm excitation. The transient gives a ~775 lifetime of FADH–* in agreement with the transient-absorption result in the upper panel.
90
Figure 4.5 Configuration of the catalytic flavin cofactor from the X-ray crystal structure of E. coli photolyase
Configuration of the catalytic flavin cofactor from the X-ray crystal structure (Park et al. 1994) of E. coli photolyase with six water molecules and four surrounding amino acid residues in less than 6 Å to the isoalloxazine ring of the cofactor: Two aromatic residues (W382 and F366) are involved in initial electron transfer in photoreduction. For clarity, only one polar (N378) and one charged (R344) residues out of fourteen surrounding ones are shown. Water molecules and highly polar/charged residues at the active site stabilize the photoreduction product, the active catalytic form FADH–, as well as reaction intermediates during the repair process.
91
Figure 4.6 Femtosecond-resolved absorption transients probed at 500 nm with and without O2
Two transients are identical and the shift of the transient in absorbance change without O2 is just for clarity. In the insert, the transients with three different data-collecting time intervals (1 h, 2 h and 3 h) under anaerobic conditions are shown. After a 3-h period, the signal only dropped 38 % in total, see text. The long-time recovery of the signal takes ~3 ns.
92 CHAPTER 5
DYNAMICS OF DAMAGED DNA REPAIR
5.1 Introduction
The model for the catalytic reaction (Sancar 2003, Park et al. 1995) proposes that
the excited flavin cofactor transfers an electon to the Pyr<>Pyr to generate a charge-
separated radical pair FADH• + Pyr<>Pyr˚–). The anionic ring of the dimer is split by a [2
+ 2] cycloreversion, and the excess electron returns to the flavin radical to restore the catalytically competent FADH– form and close the catalytic photocycle (figure 5.1A).
This hypothetical radical mechanism has not been directly proven [although it was
proposed 20 years ago (Sancar et al. 1987) and supported by extensive biochemical data
(Payne et al. 1990, Ramsey et al. 1992), spectroscopic studies (Kim et al. 1991,
Langenbacher et al. 1997, Macfarlane et al. 2003), and computer modeling (Sanders et
al. 1999, Antony et al. 2000), as well as recent structural determination (Komori et al.
93 2001, Mees et al. 2004], and the radical intermediates have not yet been captured. In this chapter we summarize results obtained by direct mapping of the repair processes by following the temporal evolution of reactants and intermediate states, to uncover the complete dynamics of the catalytic photocycle.
5.2 Active-site solvation dynamics
We first characterized the photophysics of FADH– in the active site without substrate. By gating a series of fluorescence wavelengths of the weak FADH–* emission from the blue side to the red side (figure 5.2A), we obtained the wavelength-resolved fluorescence dynamics shown in figure 5.2B. The observed transients, corresponding to systematic decays from 475 to 620 nm, show significant solvation dynamics at the active site, after the sudden change of the cofactor dipole upon excitation (Pal et al. 2004, Lu et al. 2004). All blue-side transients (fl <510 nm) have an initial decay within 2 ps (figure
5.2B Inset), showing ultrafast liberation of local protein residues and trapped water molecules (Wang et al. 2005, and chapter 4). Subsequently, slow solvation occurs from
66 ps until merging with the cofactor lifetime of 1.3 ns in the entire emission, reflecting long-time relaxation of the local environment. Table 5.1 shows the time constants and relative amplitudes of the transients with a series of exponential decays. The observation of continuous solvation processes, consistent with the highly polar active site from the x- ray structure (Park et al. 1995), reveals a dynamic active site upon function initiation.
Although the substrate recognition may push certain water out of the binding pocket, we observed similar solvation dynamics with repaired thymine monomers in the active site.
94 The observed active-site solvation directly controls the catalytic reactions of DNA repair,
as described below.
5.3 Capture of the FADH• intermediate
In the presence of substrate, the fluorescence transients drastically changed and
became much faster (figure 5.3). At the blue side, the ultrafast solvation is still clear
(figure 5.3 Inset). For example, the transient gated at 480-nm emission shows initial
solvation in 2.7 ps. At the red side (fl >510 nm), all fluorescence transients become identical. At 550-nm emission, the transient can be represented either by a stretched- single-exponential decay, A e – (t/τ)β, with = 170 ps and = 0.71, or a double-exponential
decay of 60- and 335-ps lifetimes, with 45% and 55% of the total amplitude, respectively.
The dynamics does not follow a single-exponential decay because of its strong coupling
with the slow active-site solvation. Thus, the observed 170 ps (β = 0.71) represents the
dynamics (k1) of excited state quenching directly by the substrate in the highly solvated
active site.
To determine whether the quenching occurs by an electron-transfer process, we
searched for the proposed intermediate species, FADH•. At >500 nm, only FADH• and
the excited state FADH–* have absorption (figure 5.4A). Thus, the dynamics of FADH• intermediate can be followed by probing at appropriate wavelengths. At 690 nm, FADH–* absorption dominates, and, at this wavelength, we observed drastically different dynamics in the absence and presence of substrate (figure 5.4B Inset), consistent with the fluorescence decay dynamics at 550-nm emission (figure 5.3). In the presence of
95 substrate, the transient is dominantly represented by FADH–* with the same
stretched-single-exponential or double-exponential decay (figure 5.4B). When the probe
was tuned to 625 nm, we observed very different dynamics (figure 5.4C), and the
transient decayed much slower. This result is significant because it shows the capture of
the hypothesized FADH• intermediate, proving that the excited state quenching of
FADH–* by the substrate is through an electron-transfer process and the catalytic reaction
follows a radical mechanism, which has been a longstanding unresolved issue. The
transient at 625 nm is the sum of two dynamic processes of FADH–* and FADH•. With
the same stretched value of 0.71, we obtain a time constant of 560 ps (1/k2) for the decay
of FADH•, which represents the electron-return process from the repaired thymine to
FADH• to restore the catalytically active FADH–. By using double-exponential decay for
the charge-separation process, we also obtained a double-exponential decay of FADH• in
460 and 1,240 ps. When the probe is tuned at wavelengths of 625–500 nm, all resulting
transients gave the same decay dynamics of FADH•, as shown also with the probe of 510
nm (figure 5.4C Inset). The observed non-single-exponential decay of FADH• presumably reflects the strong coupling with the slow solvation at the active site. After charge separation, the local environment relaxes again, and the active site is in a continuous dynamic motion.
5.4 Electron return and catalytic photocycle
The restoration of FADH– by electron return from the repaired thymine
monomers in 560 ps ( = 0.71) is supported by a number of observations. With 170 ps for
96 the charge separation and 1.3 ns of the FADH–* lifetime in the active site, we obtained a
quantum yield of 0.87 for the k1 process, consistent with the thymine-formation quantum
yield of 0.89 given in (Sancar et al. 1984), indicating that the efficiency of the ring
cleavage is nearly 100% and the charge recombination before dimer splitting (which
would result in a futile cycle and, hence, lower quantum yield) must occur in a much
longer time than 560 ps. The observed complete repair process in 560 ps is also
consistent with the recent observation (MacFarlane et al. 2003) of thymine formation in
589 ps probed at 260 nm for A. nidulans photolyase. Also, the FADH• signal decays to
zero (figure 5.4C), and no detectable steady-state FADH• absorption was observed after the repair reaction (figure 5.4A), excluding the possibility that FADH• stays in the neutral
radical form after repair and is subsequently reduced photochemically by an electron
from a neighboring tryptophan residue in the enzyme (Chapter 4 and Saxena et al. 2004,
Wang et al. 2005, Li et al. 1991) or by a chemical donor in the active site. Thus, the excess electron in the repaired thymine bases does return to FADH• to close the catalytic
photocycle and lead to net-zero electron changes. The entire DNA repair is completed in
560 ps through a ‘radical mechanism’.
5.5 van der Waals contacts and Adenine mediation
We further examined the repair dynamics by using substrates of oligo(dT)12–18 and poly(dT) containing T<>Ts. The results of these experiments with the 690-nm probe are shown in figure 5.5. It was conceivable that the complex binding might be different for various substrates resulting in different reaction dynamics, but our repair results
97 surprisingly showed nearly identical temporal behaviors, indicating that the distances
between FADH–* and T<>T in complexes with dinucleotide, oligonucleotide, and polynucleotide substrates are very similar. This observation is consistent with the recent
X-ray photolyase-product complex structure of A. nidulans photolyase with repaired
DNA, showing a direct van der Waals contact in 3–4 Å between the cofactor and the dimer thymines inside the active-site hole (Mees et al. 2004).
Crystal structures of photolyases from three different species (Park et al. 1995,
Komori et al. 2001, Tamada et al. 1997) show a bent configuration of the flavin cofactor in the active site. The adenine moiety was proposed to mediate the electron transfer between the cofactor and the substrate (Antony et al. 2000). In A. nidulans photolyase- product structure (Mees et al. 2004), the adenine has two hydrogen bonds with the thymine residues of the dimer at 3.1 and 3.2 Å, and the isoalloxazine ring is 4.3 Å away from the thymine residues. However, our results suggest a direct electron jump from the cofactor to the substrate, not following the hopping mechanism bridged by the adenine moiety. Otherwise, without substrate, we would observe intramolecular electron transfer from the isoalloxazine ring to the adenine moiety in both fluorescence and absorption measurements. In the absence of substrate, we observed only solvation dynamics and lifetime emission of FADH–* (figures. 5.2B and 5.3B Inset), and we did not observe evidence for alternate decay pathways. Also, the reduction potentials [– (1.9–1.4) V vs. normal hydrogen electrode (NHE) for T<>T (Heelis et al. 1992) and –2.52 V vs. NHE for adenine (Seidel et al. 1996)] strongly favor electron transfer to the dimer. However, it is possible that the adenine moiety of FADH–, because of its proximity to both the
98 isoalloxazine ring and the T<>T, may mediate the repair reaction by anchoring the dimer
through hydrogen bonding and modulating the electron jump between the cofactor and
the substrate through a superexchange mechanism.
5.6 Dynamic control and repair efficiency
Figure 5.6 shows a reaction scheme based on data from this study and previous
studies on DNA repair by photolyase. Upon function initiation by a blue-light photon, the
active site starts a continuous dynamical motion, which is strongly coupled with catalytic
electron transfer reactions. To achieve high repair efficiency, the quantum yields of both
charge-separation and ring-splitting reactions must be optimized to minimize the
nonproductive pathway of charge recombination before ring cleavage. Unlike other
sensitizer systems that have fast charge recombination leading to a low-repair quantum
yield (Song et al. 2005), the active-site solvation and proximate adenine in photolyase appear to be critical to strategically slowing down the charge separation (170 ps) and recombination by dynamically tuning the redox potentials of reaction species and stabilizing the charge-separated radical intermediates, leaving enough time to cleave the cyclobutane ring (560 ps) to reach a maximum-repair quantum yield (0.87).
One subtle issue that remains unresolved is whether the splitting of the cyclobutane ring in 560 ps is asynchronously concerted or sequential. Recent quantum chemical studies (Voityuk et al. 1997, Durbeej et al. 2000) predicted stepwise splitting in which the first-step cleavage is a downhill reaction and is expected to be ultrafast. Thus, for effective repair, the charge recombination needs to take longer time than the complete
99 ring cleavage, as described here, to prevent cyclobutane ring reclosure. With the fully resolved dynamic behavior of the flavin cofactor reported here, it should become possible to probe the dimer radical and the thymine product in the UV region and understand their dynamics.
5.6 Conclusions
With femtosecond resolution, we followed the photocycle and mapped out the temporal evolution of catalytic reactions. We captured the catalytic intermediate of flavin radical (FADH•) and directly proved the electron-transfer radical mechanism of the photocycle that was proposed approximately two decades ago (Sancar, G. B. et al. 1987).
Active-site solvation was observed to occur on picosecond-to-nanosecond time scales and to have a critical role in the continuous modulation of catalytic reactions. These synergistic motions in the active site of the damaged DNA–enzyme complex, optimized by evolution, reveal its perfect correlation of structural integrity and dynamical locality to ensure maximum repair efficiency (0.87) on the ultrafast time scale of 560 ps.
100
Wavelength 475 nm 480 nm 490 nm 550 nm
a1 0.38 0.29 0.21 -0.99
a2 0.36 0.34 0.21 0.20 Amplitudes a3 0.22 0.28 0.36 -
a4 0.04 0.09 0.22 0.80
τ1 0.98 1.10 1.21 0.31
Time τ 2 66 77 87 212
constants τ 3 277 308 690 -
τ 4 1,300 1,300 1,300 1,300
Table 5.1 Fitting results of the fluorescence transients shown in figure 5.2B
4 −t /τi All femtosecond-resolved transients are fitted by the following equation∑ aie where i i=1 = 1 - 4. The time constants are given in picoseconds. The lifetime of FADH–* is 1.3 ns in the enzyme, and all other time constants represent solvation dynamics. The negative coefficient corresponds to formation dynamics at the red-side emission.
101
Figure 5.1 Repair of damaged DNA by photolyase
(A) Schematic representation of DNA-repair processes by photolyase through an electron-transfer radical mechanism. The key catalytic reactions, charge separation (k1) and ring splitting (k2), are given at the bottom. Our study follows the evolution of the flavin cofactor. (B) The steady-state repair of dinucleotide thymine dimer (200 µM) by photolyase (10 µM) by illumination of the reaction mixture under 360-nm light. The thymine monomer formation was detected by increase in absorption at 260 nm.
102
Figure 5.2 Solvation dynamics at the active site of photolyase
– (A) Absorption and emission spectra of photolyase containing FADH cofactor (EPL- FADH–) and no second chromophore in the absence of substrate. The pump wavelength was fixed at 400 nm for all experiments. Nine fluorescence wavelengths were gated from the blue side to the red side. (B) Four typical gated fluorescence transients, with systematic decays from 475 to 550 nm, reflecting significant solvation at the active site. Inset shows fluorescent transients at early time points.
103
Figure 5.3 Ultrafast fluorescence transient of a photolyase in the absence and presence of substrate
The fluorescence transients at 550 nm with and without the substrate thymine dimer. The enzyme concentration was 0.4 mM, and the substrate concentration was 8 mM. Inset shows the fluorescence transients at 480 and 550 nm in the presence of substrate. The initial ultrafast solvation at 480 nm is still present.
104
Figure 5.4 Determination of forward and back electron transfer in photolyase photocycle by ultrafast absorption spectroscopy
• –* (A) Absorption spectra of EPL-FADH (red) and EPL-FADH (blue), as well as – EPL-FADH before (black) and after (green) the repair experiment. The absorption profile –* of EPL-FADH was obtained from ref. 8 and calibrated by our four transient-absorption data at 510, 580, 620, and 690 nm, relative to FADH•. (B) The absorption transient probed at 690 nm, showing a dominant contribution of FADH–* decay (95%) with a minor signal from FADH•. (B) The absorption transient probed at 690 nm, showing a dominant contribution of FADH–* decay (95%) with a minor signal from FADH•. (C) The absorption transient probed at 625 nm, and 510 nm (inset) showing a contribution of FADH–* decay with a major signal from formation and decay of FADH•. The enzyme concentration is 0.4 mM, and the substrate concentration is 8 mM. Inset shows the drastically different dynamics with and without the substrate.
105
Figure 5.5 Electron transfer dynamics in the presence of different substrates
Absorption transients probed at 690 nm for a series of substrates, dinucleotide thymine dimer, and oligo(dT)12–18 and poly(dT) with dimer. The oligomer concentration is 1 mM, and the total dimer concentration in poly(dT) is 5 mM. Inset shows the early time points of the reaction.
106
Figure 5.6 Evolution of catalytic reactions of DNA repair by photolyase along the reaction coordinate
Active-site solvation strongly modulates the charge-separation, ring-splitting, and electron-return processes, resulting in slow charge separation (170 ps) and a stretched- single-exponential-decay dynamics (β = 0.71). The charge recombination must be slower than the complete ring splitting (560 ps) to eliminate possible ring re-closure and achieve a maximum-repair quantum yield (0.87).
107 CHAPTER 6
DYNAMICS OF REDUCED FLAVINS
6.1 Introduction
Rich redox chemistry of flavin chromophore makes flavoproteins versatile
biological molecules. The electronic properties of the different redox states of the flavin
chromophore in flavoproteins and in aqueous solutions have been the object of theoretical
and experimental investigations (Weber 1948, Müller et al. 1969, Ghisla et al. 1974,
Visser et al. 1990, Heelis et al. 1993, Enescu et al. 1998). Though the redox chemistry of oxidized flavin has been understood at a great extent, explicit steady-state and time- resolved absorption and emission studies of reduced flavin are meager in literature.
Earlier spectroscopic studies of reduced flavins showed that there was no well- resolved structure in the near-ultraviolet absorption spectrum or fluorescence at ambient temperature in any solvent. Extensive efforts were undertaken to characterize the fluorescence and optical properties of reduced flavins and the fluorescence spectra of several substituted reduced flavins were obtained (Ghisla et al. 1974).
108 Time-resolved transient absorption and emission studies with sub-picosecond time resolution were performed on several substituted and non-substituted reduced flavins
(Visser et al. 1979, Visser et al. 1990, Heelis et al. 1993, Enescu et al. 1998). Both transient singlet and triplet excited states after excitation of fully reduced carboxymethyllumiflavin have been studied (Stanley 2001). The lifetime of singlet excited state was found to be 100 + 15 ps (Heelis et al. 1993). In another study, time- resolved transient absorption measurements on reduced, neutral and anionic flavins showed that the excited state of flavins decays biexponentially (Enescu et al. 1998).
Depending on the flavin species, the lifetime of 4 to 130 ps was obtained for the faster decaying component. The slower decay component with a lifetime above 1 ns was also observed. The two lifetimes were attributed to the changes in the degree of nonplanarity of flavin molecule (Visser et al. 1990). It was also proposed that stacking interactions may be responsible for the ultrafast 4 ps component (Enescu et al. 1998). Earlier a nanosecond time-resolved polarized fluorescence study on reduced flavins showed that fluorescence decay kinetics was multimodal (Visser et al. 1979), and suggested that the fluorescence quantum yield of reduced flavins is very low and a biexponential decay model could not explain the excited state quenching of reduced flavins.
All the previous studies arrive at the general agreement that excited state decay kinetics of reduced flavins is at least bimodal. However because of the limited time- resolution of fluorescence studies (Visser et al. 1979) and different experimental conditions with time-resolved transient absorption studies (Heelis et al. 1993, Enescu et al. 1998), an explicit understanding of the excited state dynamics of the reduced flavins
109 can not be established. A mechanistic perspective sketching the ultrafast processes
involved in low fluorescence quantum yield is also not available. The flavin molecule is
a redox system and catalyzes one and two electron transfer reactions in proteins. In such
reactions, modified forms of the reduced coenzyme are likely to occur as intermediates
(Visser et al. 1979). Thus the knowledge of their excited state properties might yield
significant information for a complete understanding of flavin catalysis (Visser et al.
1979). Studies of reduced flavins become even more important as reduced flavin cofactor
in flavoproteins performs a variety of essential biological functions, including the repair of UV-induced DNA damages as in the case of photolyase enzyme (Sancar 2003).
The present study was undertaken to understand the excited state dynamics of reduced flavins. Steady-state and femtosecond-resolved absorption and emission spectroscopy of excited reduced flavin were performed. For the first time, at least to our knowledge, we observed the steady-state fluorescence spectrum of unsubstituted reduced
1, 5-dihydroflavin adenine dinucleotide and mononucleotide in aqueous solution at room temperature (figure 6.1). In view of present knowledge of protein chromophores in aqueous solution, we are also proposing a model to understand the low fluorescence quantum yield of reduced flavins.
6.2 Time-resolved fluorescence studies of the reduced flavins
– – The anionic forms FADH and FMNH and the neutral species FADH2 were
studied at pH 8.5 and pH 5.0 respectively (ionization at N1 is reported to have pKa =
6.7). The femtosecond-resolved fluorescence transients of the reduced flavin species were
110 obtained after exciting the samples at 325 nm under anaerobic conditions. Fluorescence
decay for anionic and neutral forms was monitored at wavelengths 450 nm and 470 nm
respectively. Figure 6.2 shows fluorescence decay profile of the reduced flavins. FADH– transients were well fit by a bi-exponential decay with the time constants of 5 ps (89%) and 120 ps (11%). Similarly, FMNH– transients also follow a bi-exponential decay
kinetics and can be well fit with the time constants of 8 ps (89%) and 118 ps (11%).
FADH2 fluorescence transients monitored at 470 nm showed a significantly different
behavior. A three exponential decay fit was necessary to well fit the FADH2 transients.
The transient was fit with the ultrafast components of 3 ps (58%), 113 ps (25%) and a long component of 2 ns (17%).
The ultrashort fluorescence lifetime of anionic and neutral reduced 1,5 dihydroflavin in solution is quite striking and is in general agreement with the previously reported fluorescence lifetime of 1,3,5 trimethylated 1,5 dihydrolumiflavins (Visser et al.
1990). Photoinduced electron (or hydrogen) transfer to and from the isoalloxazine ring is a well known property of flavins (van den Berg et al.1998, Sancar 2003). However, in the case of reduced anionic flavins, the possibility of involvement of electron transfer, especially from isoalloxazine ring to adenine moiety, in shortening the fluorescence lifetime can be ruled out by two reasons. First, in the case of FMNH–, where adenine
moiety is not present, the fluorescence decay profile still shows the ultrafast component
of 8 ps. Second, in the case of FADH–, the free energy calculations using the Rehm-
Weller equation (Rehm et al. 1970) with the redox potential values of 1.5 V for adenine
(Seidel et al. 1996) and -0.17 V for second-electron reduction of FAD at excitation
111 energy of 325 nm does not favor an electron transfer from isoalloxazine ring to adenine
moiety. Thus the present study suggests that the stacking of adenine moiety does not play
a direct role in manipulating the excited state lifetime. The possibility of the protolytic reactions in the excited state is also meager. Such reactions are not expected to be fast enough to occur in the time range observed and also no remarkable difference in the behavior of ultrafast component for FADH2 transient was observed (Figure 6.2). A similar conclusion that the protolytic reaction was not involved in excited state quenching was drawn in the earlier study (Enescu et al. 1998).
For a long time, reduced flavins in aqueous solution were considered as non- fluorescing molecules. Fluorescence quantum yields of anionic reduced flavin cofactors in proteins are much higher (0.1-0.8) and normally reduced flavin exhibits a much longer lifetime, at the scale of hundreds of picoseconds as in photolyase (Chapter, 3, 4 and 5). In contrast, their fluorescence quantum yield decreases multifold in solution. Our time resolved fluorescence study suggests that the absence of fluorescence or ultrafast decay of fluorescence intensity is due to the presence of nonradiative channels in reduced flavins in aqueous solution, which are possibly curtailed in the highly structured protein environment.
The presence of nonradiative channels due to the crossing or near-crossing of potential energy surfaces is a well-known problem in polyatomic molecules (Zhong et al.
1998). The relevance of conical intersections (CIs) to nonradiative relaxation and to
photochemical and thermal reaction has gained significant attention in understanding the
mechanistic details of excited state dynamics. Recently, experimental and theoretical
112 studies done on isolated model chromophore of the green fluorescence protein suggest
the possibility that rotational flexibility of the phenyl paddle allows potential energy
surfaces of the excited and ground state to form CIs. Such CIs provides a channel for
nonradiative excited state decay and explains the poor fluorescence quantum yield of the
chromophore (Yarkony 1998, Weber et al. 1999, Gepshtein et al. 2006, Voityuk et al.
1998). In case of reduced flavins NMR spectroscopy suggests that reduced flavins can
exist in multimeric forms (Moonen et al. 1984). They exhibit ring inversion and the
rotation of the N (5) acetyl side chain was commonly observed (Moonen et al. 1984). X- ray crystal structure analysis (Werner et al. 1971, Porter et al. 1978) and recent
theoretical studies on model reduced flavins compounds imply the possibility of a bent
geometry of isoalloxazine ring compared to a planar geometry in oxidized flavins
(Rodriguez-Otero et al. 2000). Given the flexibility in its structure, ultrafast time scale of
the fluorescence decay and absence of other fluorescence quenching chemistry such as
electron or proton transfer, we propose that the excited and ground state potential energy surfaces of reduced flavins in aqueous solution may conically intersect to provide a nonradiative decay channel for excited state relaxation. We assign the dominant ultrashort component of 8-12 ps in anionic reduced flavins to the deactivation dynamics of excited state to ground state through the conical intersection (figure 6.4).
Non-planarity of the isoalloxazine ring along with the possibility of the presence of rapidly interconverting mesostructures of reduced flavin in aqueous solution appears to
be critical in determining the lifetime of reduced flavins. Difference in fluorescence
decay profiles of anionic and neutral flavins can be rationalized with the fact that the
113 bending properties of isoalloxazine ring for the anionic and neutral forms are different
(Werner et al. 1971, Porter et al. 1978, Rizzo 2001). Because the bending influences the fluorescence lifetime, it is possible that in case of FADH2, where studies were done at pH
5.0, some of the mesostructures were dominated to give a longer fluorescence lifetime,
which were apparently either absent or probably minor at pH 8.5. However, in absence
of unambiguous comparative data of reduced flavin structures at different pH, this
conclusion may be tentative. The hypothesis that different tautomeric structures are present in solution was proposed earlier to explain the multimodal fluorescence decay of reduced flavins in aqueous solution (Visser et al. 1990). Later NMR studies on reduced flavins confirmed the fact that more than one mesomeric structure of reduced flavin may be present in solution (Voityuk 1998).
6.3 Time-resolved transient absorption studies of the reduced flavins
To further understand the ultrafast excited state decay of reduced flavins, we
performed a series of transient absorption measurements. After exciting reduced flavins
at 325 nm the transient absorption of the excited state was probed at 585, 625 and 690
nm. Although anionic flavin transients were well fit with three exponential-decay time
constants of 11, 65 and 1500 ps, the decay dynamics was dominated by the ultrafast
components. Similarly, neutral reduced flavin transients were well fit with the three
exponential-decay time constants of 10, 60 and 2000 ps. Table 6.1 provides the different
time constants and their amplitudes obtained by the three exponential-decay fit at
different wavelengths. Figure 6.3 (upper and lower panels) represents the typical excited
– state decay dynamics of the FADH and FADH2, respectively. For simplicity, long 114 nanosecond components were subtracted from the fit and are not shown in the figure. A
rise component was necessary to fit the initial part of the transient, this rise component
time constant increases as the probe was tuned to longer wavelength (insert figure 6.3 for
FADH–). Similar behavior for FMNH– transients was observed (data not shown). The
transients of neutral reduced flavins with longer components are shown in the insert of
figure 6.3.
The time constants obtained in the present study are in general agreement with the
earlier published studies. The ultrafast decay component of 4 ps was observed for FADH– earlier (Enescu et al. 1998) and as well as in current studies. However, in the earlier study the intermediate decay of 60 ps was not observed. Another earlier reported study determine that the excited state of reduced flavins decays with 100 + 15 ps (Heelis et al.
1993) but interestingly, the ultrafast decay component was not observed. Enescu et al.
(1998) found that FMNH– transients decay slower than the FADH– transients. They
implicitly suggested that the stacking interaction might be the origin of the faster 4 ps
component in case of FADH– . But they also agreed that there might be other factors
involved in shortening of the lifetime. In the present study, we found that both the
FADH– and FMNH– transient absorption dynamics can be represented by the similar decay mode. Time-resolved fluorescence data also showed the similar behavior for
FADH– and FMNH–. Hence we rule out the possibility that the stacking interaction plays
a major role in manipulating the excited state dynamics of reduced flavins. However
changes in isoalloxazine bending properties due to the attachment of heavier chain at N10
as in case of FADH– may influence the excited state decay dynamics. We assign the
115 ultrafast component of 11 ps to the initial decay of excited state where excited state
population passes through the conical intersection to ground state (figure 6.4). The 60 ps
component represents a bifurcation of such dynamics.
As described earlier, all the transients were fit with a long nanosecond
component. The observation is consistent with the earlier report (Enescu et al. 1998),
however the magnitude of the long component in present study was found to be only 6%,
in contrast to 40% as reported earlier. The amplitude of the long component increases,
though not significantly, in all the cases as the probe wavelength was tuned longer
wavelength. A similar behavior was also observed earlier (Enescu et al. 1998). The amplitude of the long component was also found to be dependent of excitation pump energy. With increase in pump energy from 200 nJ to 1 µJ the amplitude of the long component increased by 50% (data not shown). This also explains the higher amplitude of the long component observed in the earlier work (Enescu et al. 1998), where pump
energy of ~15 µJ was used in the experiment. Pump power and wavelength-dependent
increase of the longer component suggests that this component may belong to low energy
excited state absorption band. Excited flavins have been shown to undergo intersystem
crossing to generate triplet state (Visser et al. 1980), which shows absorption in 450-850 nm region (Visser et al. 1974). The long components in present study may represent the triplet state but with present data it can not be unambiguously concluded.
116 6.4 Conclusions
In this chapter, we reported our studies of the excited state dynamics of reduced
flavins in aqueous solution with femtosecond time resolution. Time-resolved
fluorescence and transient absorption studies were performed. Ultrafast decay of the
excited state of reduced flavins was observed. In absence of a quenching chemistry such
as electron or proton transfer, we propose the presence of a nonradiative channel in form
of conical intersection, which funnels the excited state population to the ground state.
The present study provides obvious evidence that the presence of a nonradiative
channel is responsible for the lower fluorescence quantum yield of reduced flavins in
aqueous solution. Reduced flavins in protein environment exhibit higher fluorescence
quantum yield. The present study supports the hypothesis proposed earlier that the
protein environment controls the structure of flavins (Massey et al. 1969, Simondsen et
al. 1980). In addition, we proposed that the protein curtails the flexibility of reduced
flavins by providing a confined environment to close the nonradiative channels, which is
manifested in form of increased fluorescence quantum yield and long nanosecond lifetime. Thus, the proteins elegantly use the increased excited state lifetime of its chromophore to perform relevant biological functions, such as electron transfer in case of photolyase to repair damaged DNA. This also explains why the quantum yield of dimer repair for flavin-model compounds is very low compared to photolyase enzyme. The probability of electron transfer from the excited state of the donor compounds, i.e., flavin, reduces significantly owing to very short lifetime of the excited state. Less populated excited state decreases the chance of electron transfer to dimers and hence the repair 117 efficiency of these compounds reduces. In contrast, in photolyase long nanosecond excited state significantly increases the chance of electron transfer to dimers and hence repairs dimers efficiently.
118
λpr Compound τ1 a1 τ2 a2 τ3 a3 τ4 a4 (nm)
585 11 0.47 62 0.43 1500 0.09 0.4 (rise)
FADH– 625 11 0.60 65 0.28 1500 0.11 0.55 (rise)
690 11 0.64 65 0.17 1500 0.19 0.98 (rise)
585 8 0.57 60 0.35 2000 0.8 0.40 (rise)
FADH2 625 10 0.53 60 0.33 2000 0.14 0.50 (rise)
690 10 0.68 60 0.13 2000 0.19 0.95 (rise)
Table 6.1 Time constants (ps) and relative amplitudes
4 −t /τi All transients were fitted by ∑ aie function, where i = 1- 4. i=1
119
Figure 6.1 The absorption and fluorescence emisssion spectra of reduced flavins
The absorption and fluorescence emission spectra of reduced flavins, in 12.5mM phosphate buffer at room temperature. The thin and thick lines represent the absorption – – and emissions spectra of FMNH (green line), FADH (blue) and FADH2 (red line), respectively. The anionic and neutral reduced species were obtained at pH 8.5 and pH 5.0, respectively (see experimental section).
120
Figure 6.2 Femtosecond-resolved fluorescence transients of reduced flavins
– – Femtosecond-resolved fluorescence transients of FADH (□) , FMNH (∆) and FADH2 (○). The anionic and neutral reduced flavin emissions were probed at 450 and 470 nm respectively.
121
Figure 6.3 Femtosecond-resolved transient absorption measurements of reduced flavins
– Femtosecond-resolved transient absorption measurements of FADH (upper) and FADH2 (lower ). The transient absorption was probed at 585 nm (∆), 625 nm (□) and 690 nm (○). The transients are shown after subtracting the long nanosecond time component (see text). Upper panel insert shows the increasing amplitude of the rise component as probe wavelength was tuned toward the longer wavelength. Lower panel insert shows the FADH2 transient without subtracting the long nanosecond components.
122
Figure 6.4 Schematic representation of the two potential energy surfaces, illustrating non-radiative decay of excited state to ground state through conical intersection
Schematic representation of the two potential energy surfaces, illustrating non-radiative decay of excited state to ground-state through conical intersection. Note that the bent geometry of the reduced isoalloxazine ring is crucial to reaching the conical intersection. The protein can gate different bent structures, i.e., different excited state dynamics to manipulate flavin activities.
123 CHAPTER 7
EPILOGUE
Biological processes are very complex in nature and gargantuan factors control
one ‘happening’ event, which looks really gloomy to even start working with. However, the development of cutting edge technologies and knowledge of macromolecular structures allow adopting a plausible approach to unravel the mysteries of complex biological systems. To develop the understanding in enzyme dynamics we took advantage of state-of-the-art femtosecond pulsed laser system with well-established pump-probe technique. Using fluorescence up-conversion and transient absorption methods, we could observe and measure the key dynamic events during the DNA repair by photolyase enzyme. The available knowledge of photolyase structure, biochemistry and function allowed us to prepare appropriate biological samples to perform the key experiments and also permitted us to interpret the ultrafast events in consistent with the relevant biological processes.
124 The understandings gained through this doctoral work demonstrate that enzymes indeed are dynamic molecules and almost all the key reactions involved in performing their functions are dynamic in nature. This chapter attempts to draw a broader picture of the
importance of dynamics in enzyme function in the light of presented results in current
studies.
7.1 Energy transfer studies and enzyme dynamics
Förster resonance energy transfer describes an energy transfer mechanism
between two fluorescent molecules. These studies are now common in developing
concepts in enzyme dynamics and in answering many biologically important questions.
Investigators make use of endogenous fluorescence (in case of native and modified GFP)
or directly attach chromophores to biomolecules to quantify important dynamic
biological processes, such as protein-protein interactions, protein-DNA interactions, and
protein conformational changes. In the present research we found a system where nature
inherently uses this approach to enhance the functional efficiency of photolyase enzyme.
Absorption coefficient of photon-acceptor antenna cofactor MTHF is four times higher
than the other cofactor FADH− which actually performs the catalytic function. Higher absorption coefficient allows the enzyme to carry out its function even in presence of low ambience light, where the antenna cofactor captures the light and transfers energy to catalytic flavin cofactor by dipole-dipole interaction mechanism. Experiments performed to understand this dynamics of resonance energy transfer are not new in terms of concepts, however, with the use of advanced femto-second resolved techniques, we could precisely determine the actual time-scales of energy transfer. The measured dynamics of 125 energy transfer assisted in developing a simple but comprehensive experiment plan for studying the important dynamic processes of DNA repair. Learning gained though this work also helped us in elucidating the energy transfer dynamics in related ‘cryptochrome’ protein. We were able to predict hydrophobic and rigid protein environment around the folate cofactor in ‘cryptochrome’ based entirely on anisotropy and wave-length resolved fluorescence measurements (Saxena et al. 2005). The study also suggested that ultrafast energy transfer processes might facilitate formation of initial signaling state in cryptochrome protein (Saxena et al. 2005). Present study undoubtedly concludes that dynamic energy transfer processes play a significant role in enzyme catalysis.
7.2 Electron transfer and enzyme dynamics
Intraprotein electron transfer is ubiquitous and is a key step in functions of many proteins including various enzymes. In many cases such as in mitochondrial membrane proteins, an impaired electron transfer may even lead to the death of an organism. Studies of intraprotein electron transfer is challenging because a) electron transfer processes are ultrafast in nature, b) electron transfer can follow multiple paths in a dynamic protein environment. As a general feature, most of the electron transfer proteins have two catalytic sites connected by redox chains (Page et al. 2003). Catalytic sites are defined as multi-electron redox centers or clusters of single-electron redox centers that interact with substrates and act as sources or sinks of pairs of electrons. Redox chains are defined as groups of single electron redox centers that are more widely separated, with fewer near neighbors (Page et al. 2003). Because photolyase incorporates all the characteristics of an electron transfer protein it is a good model for the studies of intraprotein electron transfer. 126 Understanding gained through electron transfer studies in photolyase suggests several possible mechanisms of electron transfer in proteins such as electron tunneling and/or electron hoping. Electron tunneling mechanism has been widely used in explaining long-range electron transfer especially in mitochondrial and photosynthetic systems. On the other side the electron hopping mechanism is critical for a shorter distance and is an efficient way for proteins to transfer electrons. From present studies it can be inferred that most of the short-range electron transfer reaction in proteins take place on ultrafast time scale to maintain the efficiency and integrity of the proteins. It is known that the electron transfer reaction depends on multiple factors such as redox pair, distance between donor and acceptor, and reorganization energy of the system (Marcus
1993). In addition, from present studies we found that even the physical properties of protein nanoenvironment, particularly altering electron donor capabilities of aromatic amino acids residues in proteins, can also influence the electron transfer rate. This study concludes that dynamic protein environment manipulates intraprotein electron transfer processes which are essential for enzyme catalysis.
7.3 Enzyme active-site solvation dynamics
Theoretically and experimentally, it has been shown that dynamic motion of water around protein surfaces might play an important role in molecular recognition, protein stability and even in catalytic chemistry (Finney 1996, Robinson et al. 1999, Mattos
2002, Pal et al. 2004). My present research group and also other investigators around the world are putting a lot of efforts in developing a comprehensive understanding of protein- water interactions. To characterize the water dynamics around protein surfaces, 127 investigators use tryptophan, dyes and other chromophore labeling methods to label a
specific site in proteins (Riter et al. 1996, Jordanides et al. 1999, Zhong et al. 2001,
Zhang et al. 2006). In the experiment, either tryptophan or attached chromophore can be excited at an appropriate wavelength and the time resolved fluorescence decay dynamics from the blue to the red side of emission of tryptophane/chromophore can be interpreted in terms of local ultrafast solvation and electronic quenching (Zhang et al. 2006).
Interestingly, photolyases, which host flavin at their active sites, gave us an opportunity to understand the hydration dynamics at the active-site of an enzyme without artificially manipulating the amino acid composition of the enzyme. Experimentally it was for the first time that we could observe a direct involvement of hydration dynamics at the active site of photolyase in modulating the electron transfer (the key catalytic reaction) rate.
Though the characterization widely remained qualitatively in nature and as of now we could not establish a quantitative description of this modulation, it is a significant progress in developing a molecular pictures that how dynamic motions of water and amino acid residues can play a significant role in enzyme catalysis.
7.4 Dynamics of enzyme cofactor
The release of the final product in enzymatic reactions involves the formation and
decay of multiple intermediates (Smiley et al. 2006). The emerging trends suggest that
dynamic motions in protein control the formation and dissociation of these intermediates
and allow the enzyme to function efficiently by providing multiple activation barriers of
relatively lower energy instead of a single but larger activation barrier. To add in present
understanding, our studies on enzyme cofactor (reduced flavin) in aqueous solution 128 suggested that the dynamic protein conformation stabilizes a particular configuration of the cofactor in the protein to perform biologically important functions.
Usually, reduced flavins in aqueous solutions are known for their non-fluorescent behavior. However it has been demonstrated earlier (Ghisla et al. 1974) and again in this study that reduced flavins in aqueous solutions do exhibit fluorescence. By monitoring time resolved fluorescence decay profile we found that more than 90% of the excited state population decays to ground state in less than 10 ps, which explains why in earlier studies reduced flavins were considered non-fluorescent. Interestingly, dynamic photolyase stabilizes a different configuration of reduced flavin in its nanoenvironment and the excited state of this flavin decays in nanosecond time span. Hence photolyase with reduced flavin exhibits measurable fluorescence. Long-lived excited state of reduced flavin in photolyase enzyme provides an opportunity for the excited flavin to eject an electron to CPDs for DNA repair. However for the reduced flavin in aqueous solution the excited state is stable only for the fraction of the time that an electron might take to travel to CPDs. Hence, reduced flavin itself can not repair CPDs with higher efficiency as the reduced flavin cofactor in enzyme does. This observation is direct evidence that the dynamic protein can control the photophysical properties of its cofactor to perform the required biological functions, such as electron transfer in case of photolyase.
7.5 The final words
The correlation of structure, dynamics and function has practical importance.
Today it has become clear that the information of dynamic modes of enzymes and
129 substrates is essential for the design of potential new enzymes and drugs. The knowledge of enzyme dynamics gained through current studies can be directly applied to the
enhancement of the existing skin-care lotion therapies used for the repair of UV-induced
DNA lesions. In the past peptide-like model compounds containing tryptophan as
electron donor have been used to repair the UV-induced DNA lesions (Helene et al.
1977), but the quantum yield of DNA repair is very low. As photolyases host flavins as
catalytic cofactors, it was suggested that introducing flavin as electron donor in the
peptide mimics may increase the repair quantum yield (Carell et al. 1997). But it is
known that the quantum yield of repair even for the flavin based model compound is very
low as compared with photolyase enzyme itself. In the present work we have clearly
demonstrated that one of the key factors for higher repair quantum yield is the excited
state lifetime of the flavin cofactor. Photolyases place flavin cofactor in a special
configuration to extend its excited state lifetime, which is necessary to perform the
electron transfer to DNA lesions. Thus all the future research on peptide mimics hosting
flavins as electron donor should consider the correct positioning of the flavins in order to
enhance the excited state lifetime and consequently repair quantum yield. No doubt other
factors such as correct binding configuration of peptide mimic with damaged DNA,
residues stabilizing the charge separated species and favored redox potential for electron
transfer are also important in designing a potential drug to repair the UV-induced DNA lesions. The achievements made through present work make us optimistic that the objective of developing artificial small-molecule DNA photolyase can be reached.
130 LIST OF REFERENCES
Antony, J., D. M. Medvedev, et al. (2000). "Theoretical study of electron transfer between the photolyase catalytic cofactor FADH(-) and DNA thymine dimer." Journal of the American Chemical Society 122(6): 1057-1065.
Aravind, L., V. Anantharaman, et al. (2002). "Monophyly of class I aminoacyl tRNA synthetase, USPA, ETFP, photolyase, and PP-ATPase nucleotide-binding domains: Implications for protein evolution in the RNA world." Proteins-Structure Function and Genetics 48(1): 1-14.
Aubert, C., M. H. Vos, et al. (2000). "Intraprotein radical transfer during photoactivation of DNA photolyase." Nature 405(6786): 586-590.
Ballew, R. M., J. Sabelko, et al. (1996). "Observation of distinct nanosecond and microsecond protein folding events." Nature Structural Biology 3(11): 923-926.
Begley, T. P. (1994). "Photoenzymes - a novel class of biological catalysts." Accounts of Chemical Research 27(12): 394-401.
Boehr, D. D., H. J. Dyson, et al. (2006). "An NMR perspective on enzyme dynamics." Chemical Reviews 106(8): 3055-3079.
Bolon, D. N., C. A. Voigt, et al. (2002). "De novo design of biocatalysts." Current Opinion in Chemical Biology 6(2): 125-129.
Bourgeois, D., B. Vallone, et al. (2003). "Complex landscape of protein structural dynamics unveiled by nanosecond Laue crystallography." Proceedings of the National Academy of Sciences of the United States of America 100(15): 8704-8709.
Brash, D. E. (1997). "Sunlight and the onset of skin cancer." Trends in Genetics 13(10): 410-414.
131 Brautigam, C. A., B. S. Smith, et al. (2004). "Structure of the photolyase-like domain of cryptochrome 1 from Arabidopsis thaliana." Proceedings of the National Academy of Sciences of the United States of America 101(33): 12142-12147.
Brudler, R., K. Hitomi, et al. (2003). "Identification of a new cryptochrome class: Structure, function, and evolution." Molecular Cell 11(1): 59-67.
Brunori, M., D. Bourgeois, et al. (2004). "The structural dynamics of myoglobin." Journal of Structural Biology 147(3): 223-234.
Byrdin, M., A. P. M. Eker, et al. (2003). "Dissection of the triple tryptophan electron transfer chain in Escherichial coli DNA photolyase: Trp382 is the primary donor in photoactivation." Proceedings of the National Academy of Sciences of the United States of America 100(15): 8676-8681.
Byrdin, M., V. Sartor, et al. (2004). "Intraprotein electron transfer and proton dynamics during photoactivation of DNA photolyase from E-coli: review and new insights from an "inverse" deuterium isotope effect." Biochimica Et Biophysica Acta-Bioenergetics 1655(1-3): 64-70.
Callender, R. and R. B. Dyer (2006). "Advances in time-resolved approaches to characterize the dynamical nature of enzymatic catalysis." Chemical Reviews 106(8): 3031-3042.
Carell, T. and R. Epple (1998). "Repair of UV light induced DNA lesions: A comparative study with model compounds." European Journal of Organic Chemistry 7: 1245-1258.
Carell, T. and J. Butenandt (1997). "Towards artificial DNA-repair enzymes: Incorporation of a flavin amino acid into DNA-binding oligopeptides." Angewandte Chemie-International Edition in English 36(13-14): 1461-1464.
Cashmore, A. R., J. A. Jarillo, et al. (1999). "Cryptochromes: Blue light receptors for plants and animals." Science 284(5415): 760-765.
Cheng, P. Y., D. Zhong, et al. (1996). "Femtosecond real-time probing of reactions .21. Direct observation of transition-state dynamics and structure in charge-transfer reactions." Journal of Chemical Physics 105(15): 6216-6248.
Cheung, M. S., I. Daizadeh, et al. (1999). "Pathways of electron transfer in Escherichia coli DNA photolyase: Trp306 to FADH." Biophysics Journal 76(3): 1241-1249.
Chigancas, V., A. Sarasin, et al. (2004). "CPD-photolyase adenovirus-mediated gene transfer in normal and DNA-repair-deficient human cells." Journal of Cell Sciences 117: 3579-92.
132 Chigancas, V., E. N. Miyaji, et al. (2000). "Photorepair prevents ultraviolet-induced apoptosis in human cells expressing the marsupial photolyase gene." Cancer Research 60(9): 2458-2463.
Chosrowjan, H., S. Taniguchi, et al. (2004). “Low-Frequency Vibrations and Their Role in Ultrafast Photoisomerization Reaction Dynamics of Photoactive Yellow Protein.” Journal of Physical Chemistry B 108(8): 2686-2698.
Defelippis, M. R., C. P. Murthy, et al. (1991). "Electrochemical Properties of Tyrosine Phenoxy and Tryptophan Indolyl Radicals in Peptides and Amino-Acid-Analogs." Journal of Physical Chemistry 95(8): 3416-3419.
Deisenhofer, J. (2000). "DNA photolyases and cryptochromes." Mutation Ressearch 460(3-4): 143-149.
Denisov, I. G., T. M. Makris, et al. (2005). "Structure and chemistry of cytochrome P450." Chemical Reviews 105(6): 2253-2277.
Diffey, B. (2004). "Climate change, ozone depletion and the impact on ultraviolet exposure of human skin." Physics in Medicine and Biology 49(1): R1-R11.
Durbeej, B. and L. A. Eriksson (2000). "Thermodynamics of the photoenzymic repair mechanism studied by density functional theory." Journal of the American Chemical Society 122(41): 10126-10132.
EllisDavies, G. C. R., J. H. Kaplan, et al. (1996). "Laser photolysis of caged calcium: Rates of calcium release by nitrophenyl-EGTA and DM-nitrophen." Biophysical Journal 70(2): 1006-1016.
Enescu, M., L. Lindqvist, et al. (1998). "Excited-state dynamics of fully reduced flavins and flavoenzymes studied at subpicosecond time resolution." Photochemistry and Photobiology 68(2): 150-156.
Essen, L. O. and T. Klar (2006). "Light-driven DNA repair by photolyases." Cellular and Molecular Life Sciences 63(11): 1266-1277.
Fernandez, C. and G. Wider (2003). "TROSY in NMR studies of the structure and function of large biological macromolecules." Current Opinion in Structural Biology 13(5): 570-580.
Finney, J. L. (1996). "Overview lecture. Hydration processes in biological and macromolecular systems." Faraday Discussions 103: 1-18.
Fox, J. L. (1974). "Sodium dithionite reduction of flavin." FEBS Letters 39(1): 53-55.
133 Garinis, G. A., J. R. Mitchell, et al. (2005). "Transcriptome analysis reveals cyclobutane pyrimidine dimers as a major source of UV-induced DNA breaks." The EMBO Journal 24(22): 3952-3962.
Gennis, R. B. (2004). "Coupled proton and electron transfer reactions in cytochrome oxidase." Frontiers in Biosciences 9: 581-591.
Gepshtein, R.; D. Huppert , et al. (2006). “Deactivation Mechanism of the Green Fluorescent Chromophore.” 110: 4434-4442.
Ghisla, S. and V. Massey (1975). "Mechanism of inactivation of the flavoenzyme lactate oxidase by oxalate." Journal of Biological Chemistry 250(2): 577-84.
Ghisla, S., V. Massey, et al. (1974). "Fluorescence and optical characteristics of reduced flavines and flavoproteins." Biochemistry 13(3): 589-597.
Gindt, Y. M., E. Vollenbroek, et al. (1999). "Origin of the transient electron paramagnetic resonance signals in DNA photolyase." Biochemistry 38(13): 3857-3866.
Goto, N. K. and L. E. Kay (2000). "New developments in isotope labeling strategies for protein solution NMR spectroscopy." Current Opinion in Structural Biology 10(5): 585- 592.
Hammes, G. G. (1964). "Mechanism of Enzyme Catalysis." Nature 204: 342-343.
Hammes, G. G. (2002). "Multiple conformational changes in enzyme catalysis." Biochemistry 41(26): 8221-8228.
Harm, H. and C. S. Rupert (1968). "Analysis of photoenzymic repair of UV lesions in DNA by single light flashes. I. In vitro studies with Haemophilus influenzae transforming DNA and yeast photoreactivating enzyme." Mutation Research 6(3): 355-370.
Harm, H. and C. S. Rupert (1970). "Analysis of photoenzymic repair of UV lesions in DNA by single light flashes. VII. Photolysis of enzyme-substrate complexes in vitro." Mutation Research 10(4): 307-318.
Harm, H. and C. S. Rupert (1976). "Analysis of photoenzymatic repair of UV lesions in DNA by single light flashes. XI. Light-induced activation of the yeast photoreactivating enzyme." Mutation Research 34(1): 75-92.
Harriman, A. (1987). "Further Comments on the Redox Potentials of Tryptophan and Tyrosine." Journal of Physical Chemistry 91(24): 6102-6104.
134 Heelis, P. F. and A. Sancar (1986). "Photochemical properties of Escherichia coli DNA photolyase: a flash photolysis study." Biochemistry 25(25): 8163-8166.
Heelis, P. F., D. J. Deeble, et al. (1992). "Splitting of Cis-Syn Cyclobutane Thymine- Thymine Dimers by Radiolysis and Its Relevance to Enzymatic Photoreactivation." International Journal of Radiation Biology 62(2): 137-143.
Heelis, P. F., G. Payne, et al. (1987). "Photochemical Properties of Escherichia-Coli DNA Photolyase - Selective Photodecomposition of the 2nd Chromophore." Biochemistry 26(15): 4634-4640.
Heelis, P. F., R. F. Hartman, et al. (1993). "Detection of the excited singlet state of a deprotonated, reduced flavin." Photochemistry and Photobiology 57(6): 1053-1055.
Heelis, P. F., R. F. Hartman, et al. (1995). "Photoenzymic Repair of Uv-Damaged DNA - a Chemists Perspective." Chemical Society Reviews 24(4): 289-297.
Helene, C. and M. Charlier (1977). “Photosensitized splitting of pyrimidine dimers by indole derivatives and by tryptophan containing oligopeptides and proteins.” Photochemistry and Photobiology 25(5): 429-434.
Hilvert, D. (2000). "Critical analysis of antibody catalysis." Annual Review of Biochemistry 69: 751-793.
Huang, Y. H., R. Baxter, B. S. Smith, C. L. Partch, C. L. Colbert and J. Deisenhofer. (2006). “Crystal structure of cryptochrome 3 from Arabidopsis thaliana and its implications for photolyase activity.” Proceedings of the National Academy of Sciences of the United States of America 103, 17701-17706.
Husain, I. and A. Sancar (1987). "Binding of Escherichia-Coli DNA Photolyase to a Defined Substrate Containing a Single T Reversible T Dimer." Nucleic Acids Research 15(3): 1109-1120.
Johnson, R. E. et al. (1999). “hRAD30 mutations in the variant form of xeroderma pigmentosum.” Science 285: 263-265.
Jordan, S. P. and M. S. Jorns (1988). "Evidence for a singlet intermediate in catalysis by Escherichia coli DNA photolyase and evaluation of substrate binding determinants." Biochemistry 27(25): 8915-23.
Jordan, S.P., J. L. Alderfer, L. P. Chanderkar, M. S. Jorns (1989). " Reaction of Escherichia coli and yeast photolyases with homogeneous short-chain oligonucleotide substrates." Biochemistry 28(20): 8149-8153.
135 Jordanides, X. J., M. J. Lang, et al. (1999). "Solvation dynamics in protein environments studied by photon echo spectroscopy." Journal of Physical Chemistry B 103(37): 7995- 8005.
Jorns, M. S., B. Y. Wang, et al. (1987). " DNA repair catalyzed by Escherichia coli DNA photolyase containing only reduced flavin: elimination of the enzyme's second chromophore by reduction with sodium borohydride." Biochemistry 26(21): 6810-6816.
Jorns, M. S., B. Y. Wang, et al. (1990). "Chromophore function and interaction in Escherichia-Coli DNA Photolyase - reconstitution of the apoenzyme with Pterin and or Flavin derivatives." Biochemistry 29(2): 552-561.
Jorns, M. S., E. T. Baldwin, et al. (1987). "Action mechanism of Escherichia-Coli DNA Photolyase .2. Role of the chromophores in catalysis." Journal of Biological Chemistry 262(1): 486-491.
Kao, Y. T., C. Saxena, et al. (2005). "Direct observation of thymine dimer repair in DNA by photolyase." Proceedings of the National Academy of Sciences of the United States of America 102(45): 16128-16132.
Kim, K. W., M. Kuppuswamy, et al. (2000). "Electrochemical oxidation of benzene at a glassy carbon electrode." Journal of Applied Electrochemistry 30(5): 543-549.
Kim, S. T. and A. Sancar (1991). "Effect of base, pentose, and phosphodiester backbone structures on binding and repair of pyrimidine dimers by Escherichia coli DNA photolyase." Biochemistry 30(35): 8623-30.
Kim, S. T., A. Sancar (1993). “Photochemistry, photophysics, and mechanism of pyrimidine dimer repair by DNA photolyase.” Photochemistry and Photobiology 57(5), 895-904.
Kim, S. T., A. Sancar, et al. (1993). "Time-resolved EPR studies with DNA photolyase: excited-state FADH(˚) abstracts an electron from Trp-306 to generate FADH(-), the catalytically active form of the cofactor." Proceedings of the National Academy of Sciences of the United States of America 90(17): 8023-7.
Kim, S. T., P. F. Heelis, et al. (1991). "Determination of rates and yields of interchromophore (Folate- Flavin) energy-transfer and intermolecular (Flavin- DNA) electron-transfer in Escherichia-Coli Photolyase by time-resolved fluorescence and absorption-spectroscopy." Biochemistry 30(47): 11262-11270.
Kim, S. T., P. F. Heelis, et al. (1992). "Energy transfer (deazaflavin-->FADH2) and electron transfer (FADH2-->T <> T) kinetics in Anacystis nidulans photolyase." Biochemistry 31(45): 11244-11248.
136 Komori, H., R. Masui, et al. (2001). "Crystal structure of thermostable DNA photolyase: Pyrimidine-dimer recognition mechanism." Proceedings of the National Academy of Sciences of the United States of America 98(24): 13560-13565.
Koshland, D. E., Jr. (1960). “The active site and enzyme action.” Advances in Enzymology 22: 45-97.
Kotting, C. and K. Gerwert (2005). "Proteins in action monitored by time-resolved FTIR spectroscopy." Chemphyschem 6(5): 881-888.
Lamola, A. A. (1968). "Applications of electronic energy transfer in solution." Photochemistry and Photobiology 8(6): 601-16.
Langenbacher, T., X. D. Zhao, et al. (1997). "Substrate and temperature dependence of DNA photolyase repair activity examined with ultrafast spectroscopy." Journal of the American Chemical Society 119(43): 10532-10536.
Li, Y. F. and A. Sancar (1990). "Active site of Escherichia coli DNA photolyase: mutations at Trp277 alter the selectivity of the enzyme without affecting the quantum yield of photorepair." Biochemistry 29(24): 5698-5706.
Li, Y. F., P. F. Heelis, et al. (1991). "Active site of DNA photolyase: tryptophan-306 is the intrinsic hydrogen atom donor essential for flavin radical photoreduction and DNA repair in vitro." Biochemistry 30(25): 6322-6329.
Lin, C. T. and D. Shalitin (2003). "Cryptochrome structure and signal transduction." Annual Review of Plant Biology 54: 469-496.
Lo, H. L., S. Nakajima, et al. (2005). "Differential biologic effects of CPD and 6-4PP UV-induced DNA damage on the induction of apoptosis and cell-cycle arrest." BMC Cancer 5: 135-143.
Lu, W. Y., J. Kim, et al. (2004). "Femtosecond studies of tryptophan solvation: correlation function and water dynamics at lipid surfaces." Chemical Physics Letters 388(1-3): 120-126.
Lu, W. Y., W. H. Qu, et al. (2004). "Feratosecond studies of crown ethers: supramolecular solvation, local solvent structure and cation-pi interaction." Chemical Physics Letters 394(4-6): 415-422.
Lumry, R. (1959) in The Enzymes (Boyer, P., Ed.) 2nd ed., Vol. 1, p 157, Academic Press, New York.
137 MacFarlane, A. W. and R. J. Stanley (2003). "Cis-syn thymidine dimer repair by DNA photolyase in real time." Biochemistry 42(28): 8558-8568.
Malhotra, K., Kim, S. T. and A. Sancar (1994). "Characterization of a Medium Wavelength Type DNA Photolyase: Purification and Properties of Photolyase from Bacillus firmus." Biochemistry 33(29): 8712-8718.
Marcus, R. A. (1993). “Electron transfer reactions in chemistry. Theory and experiment.” Reviews of Modern Physics 65: 599-610.
Massey, V. (2000). "The chemical and biological versatility of riboflavin." Biochemical Society Transactions 28: 283-296.
Massey, V., F. Muller, et al. (1969). "The reactivity of flavoproteins with sulfite. Possible relevance to the problem of oxygen reactivity." Journal of Biological Chemistry 244(15): 3999-4006.
Masutani, C. et al. (1999). “The XPV (xeroderma pigmentosum variant) gene encodes human DNA polymerase. Nature 399: 700-704.
Mataga, N., H. Chosrowjan, et al. (2000). “Dynamics and mechanisms of ultrafast fluorescence quenching reactions of Flavin chromophores in protein nanospace.” Journal of Physical Chemistry B 104(45): 10667-10677.
Mataga, N., H. Chosrowjan, et al. (2002). "Femtosecond fluorescence dynamics of flavoproteins: Comparative studies on flavodoxin, its site-directed mutants, and riboflavin binding protein regarding ultrafast electron transfer in protein nanospaces." Journal of Physical Chemistry B 106(35): 8917-8920.
Mattos, C. (2002). "Protein-water interactions in a dynamic world." Trends in Biochemical Sciences 27(4): 203-208.
McMahon, B. H., M. Fabian , et al. (2004). “FTIR studies of internal proton transfer reactions linked to inter-heme electron transfer in bovine cytochrome c oxidase.” Biochimica Et Biophysica Acta 1655: 321-331.
Mees, A., T. Klar, et al. (2004). "Crystal structure of a photolyase bound to a CPD-like DNA lesion after in situ repair." Science 306(5702): 1789-1793.
Meistrich, M. L. and A. A. Lamola (1972). "Triplet-state sensitization of thymine photodimerization in bacteriophage T4." Journal of Molecular Biology 66(1): 83-95.
Moffat, K. (2001). "Time-resolved biochemical crystallography: A mechanistic perspective." Chemical Reviews 101(6): 1569-1581.
138 Moonen, C. T. W., J. Vervoort, et al. (1984). "C-13 Nuclear Magnetic-Resonance Study on the Dynamics of the Conformation of Reduced Flavin." Biochemistry 23(21): 4868- 4872.
Moonen, C. T. W., J. Vervoort, et al. (1984). "Reinvestigation of the Structure of Oxidized and Reduced Flavin - C-13 and N-15 Nuclear Magnetic-Resonance Study." Biochemistry 23(21): 4859-4867.
Muller, F., V. Massey, et al. (1969). "The reduction of flavins by borohydride: 3,4- dihydroflavin. Struction, absorption and luminescence." European Journal of Biochemistry 9(3): 392-401.
Myles, G.M., B. Van Houten and A. Sancar (1987). " Utilization of DNA photolyase, pyrimidine dimer endonucleases, and alkali hydrolysis in the analysis of aberrant ABC excinuclease incisions adjacent to UV-induced DNA photoproducts." Nucleic Acids Research 15: 1227-1243.
NASA education: http://www.nas.nasa.gov/About/Education/Ozone/radiation.html
Nibbering, E. T. J., H. Fidder, et al. (2005). "Ultrafast chemistry: Using time-resolved vibrational spectroscopy for interrogation of structural dynamics." Annual Review of Physical Chemistry 56: 337-367.
Nienhaus, K., Nienhaus, G. U. (2005). “In Protein-Ligand Interactions: methods and applications”. Nienhaua, G.U., Ed.; Humana Press: Totowa, NJ, vol. 305, pp 215-242.
Okamura, T., A. Sancar, et al. (1991). "Picosecond laser photolysis studies on the photorepair of pyrimidine dimers by DNA Photolyase .1. Laser photolysis of Photolyase- 2-Deoxyuridine Dinucleotide photodimer complex." Journal of the American Chemical Society 113(8): 3143-3145.
Page, C.C., C. C., Moser, et al. (2003). “Mechanism for electron transfer within and between proteins.” Current Opinion in Chemical Biology 7(5): 551-556.
Pal, S. K. and A. H. Zewail (2004). "Dynamics of water in biological recognition." Chemical Reviews 104(4): 2099-2123.
Pal, S. K., J. Peon, et al. (2002). "Biological water: Femtosecond dynamics of macromolecular hydration." Journal of Physical Chemistry B 106(48): 12376-12395.
Pal, S. K., J. Peon, et al. (2002). “Biological water at the protein surface: dynamical solvation probed directly with femtosecond resolution.” Proceedings of the National Academy of Sciences of the United States of America 99(4): 1763-1768.
139 Pal, S. K., J. Peon, et al. (2002). “Biological water: Femtosecond dynamics of macromolecular hydration.” Proceedings of the National Academy of Sciences of the United States of America 106(48): 12376-12395.
Pal, S. K., J. Peon, et al. (2002). “Ultrafast surface hydration dynamics and expression of protein functionality: alpha -Chymotrypsin.” Proceedings of the National Academy of Sciences of the United States of America 99(24): 15297-15302.
Pan, J., M. Byrdin, et al. (2004). "Excited-state properties of flavin radicals in flavoproteins: Femtosecond spectroscopy of DNA photolyase, glucose oxidase, and flavodoxin." Journal of Physical Chemistry B 108(28): 10160-10167.
Park, C. H. and R. S. Givens (1997). "New photoactivated protecting groups .6. p- hydroxyphenacyl: A phototrigger for chemical and biochemical probes." Journal of the American Chemical Society 119(10): 2453-2463.
Park, H. W., S. T. Kim, et al. (1995). "Crystal-Structure of DNA Photolyase from Escherichia-Coli." Science 268(5219): 1866-1872.
Pascher, T., J. P. Chesick, et al. (1996). "Protein folding triggered by electron transfer." Science 271(5255): 1558-1560.
Patrick, M. H. and R. O. Rahn (1976). "Photochemistry of DNA and polynucleotides: Photoproducts, in Photochemistry and Photobiology of Nucleic Acids, Vol. 2 Wang, S. Y., Ed., Academic Press, New York, 1.
Payne, G. and A. Sancar (1990). "Absolute Action Spectrum of E-Fadh2 and E-Fadh2- Mthf Forms of Escherichia-Coli DNA Photolyase." Biochemistry 29(33): 7715-7727.
Payne, G., P. F. Heelis, et al. (1987). "The Active Form of Escherichia-Coli DNA Photolyase Contains a Fully Reduced Flavin and Not a Flavin Radical, Both Invivo and Invitro." Biochemistry 26(22): 7121-7127.
Peon, J., S. K. Pal, et al. (2002). “Hydration at the surface of the protein monellin: Dynamics with femtosecond resolution.” Proceedings of the National Academy of Sciences of the United States of America 99(17): 10964-10969.
Pfeifer, G. P. (1997). "Formation and processing of UV photoproducts: Effects of DNA sequence and chromatin environment." Photochemistry and Photobiology 65(2): 270- 283.
Porter D. J. T. and D. Voet (1978). “The crystal structure of a flavin-nicotinamide biscoenzyme in two oxidation states: Models for flavin-nicotinamide interactions” Acta Crystallography B 34: 598–610.
140 Qiu, W. H., L. Y. Zhang, et al. (2005). "Ultrafast hydration dynamics in melittin folding and aggregation: Helix formation and tetramer self-assembly." Journal of Physical Chemistry B 109(35): 16901-16910.
Qiu, W. H., Y. T. Kao, et al. (2006). " Protein surface hydration mapped by site-specific mutations." Proceedings of the National Academy of Sciences of the United States of America 103: 13979-13984.
Ramsey, A. J. and M. S. Jorns (1992). "Effect of 5-Deazaflavin on energy transduction during catalysis by Escherichia-Coli DNA Photolyase." Biochemistry 31(36): 8437-8441.
Ramsey, A. J., J. L. Alderfer, et al. (1992). "Energy transduction during catalysis by Escherichia-Coli DNA Photolyase." Biochemistry 31(31): 7134-7142.
Rehm, D., and A. Weller. (1970). “Kinetics of fluorescence quenching by electron and H- atom transfer.” Israel Journal of Chemistry 8:259–271.
Riek, R., J. Fiaux, et al. (2002). "Solution NMR techniques for large molecular and supramolecular structures." Journal of the American Chemical Society 124(41): 12144- 12153.
Riter, R. R., M. D. Edington, et al. (1996). "Protein-matrix solvation dynamics in the alpha subunit of C-phycocyanin." Journal of Physical Chemistry 100(33): 14198-14205.
Rizzo, C. J. (2001). "Further computational studies on the conformation of 1,5- dihydrolumiflavin." Antioxidants & Redox Signaling 3(5): 737-746.
Robinson, G. W. and C. H. Cho (1999). "Role of hydration water in protein unfolding." Biophysical Journal 77(6): 3311-3318.
Rodriguez-Otero, J., E. Martinez-Nunez , et al. (2000). “The Role of Aromaticity in the Planarity of Lumiflavin.” Journal of Organic Chemistry 67:6347-6352.
Rupert, C.S. (1958). “Photoreactivation in-vitro of ultraviolet inactivated hemophilus influenzae transforming factor.” Journal of General Physiology 41: 451-452.
Rupert, C.S. (1960). “Photoreactivation of transforming deoxyribonucleic acid by an enzyme from bakers' yeast.” Journal of General Physiology 43: 573-595.
Rupert, C.S. (1962). “Photoenzymic repair of ultraviolet damage in deoxyribonucleic acid (DNA). I. Kinetics of the reaction.” Journal of General Physiology 45: 703-724.
141 Rupert, C.S. (1962). “Photoenzymic repair of ultraviolet damage in deoxyribonucleic acid (DNA). II. Formation of an enzyme-substrate complex.” Journal of General Physiology 45: 725-741.
Sancar, A. (1988). "DNA repair enzymes." Annual Review of Biochemistry 57: 29-67.
Sancar, A. (1994). "Structure and function of DNA Photolyase." Biochemistry 33(1): 2-9.
Sancar, A. (2000). "Cryptochrome: the second photoactive pigment in the eye and its role in circadian photoreception." Annual Review of Biochemistry 69: 31-67.
Sancar, A. (2003). "Structure and function of DNA photolyase and cryptochrome blue- light photoreceptors." Chemical Reviews 103(6): 2203-2237.
Sancar, A., F. W. Smith, et al. (1984). "Purification of Escherichia coli DNA photolyase." Journal of Biological Chemistry 259(9): 6028-6032.
Sancar, G. B. (1990). "DNA photolyases: physical properties, action mechanism, and roles in dark repair." Mutation Research 236(2-3): 147-160.
Sancar, G. B. and A. Sancar (1987). "Structure and function of DNA Photolyases." Trends in Biochemical Sciences 12(7): 259-261.
Sancar, G. B. and A. Sancar (2006). "Purification and characterization of DNA photolyases." Methods in Enzymology 408: 121-56.
Sancar, G. B., F. W. Smith, et al. (1985). "Binding of Escherichia coli DNA photolyase to UV-irradiated DNA." Biochemistry 24(8): 1849-55.
Sancar, G. B., F. W. Smith, et al. (1987). "Action mechanism of Escherichia-Coli DNA Photolyase .1. formation of the enzyme-substrate complex." Journal of Biological Chemistry 262(1): 478-485.
Sancar, G. B., M. S. Jorns, et al. (1987). "Action mechanism of Escherichia-Coli DNA Photolyase .3. photolysis of the enzyme-substrate complex and the absolute action spectrum." Journal of Biological Chemistry 262(1): 492-498.
Sanders, D. B. and O. Wiest (1999). "A model for the enzyme-substrate complex of DNA photolyase and photodamaged DNA." Journal of the American Chemical Society 121(22): 5127-5134.
Saxena, C., A. Sancar, et al. (2004). "Femtosecond dynamics of DNA photolyase: Energy transfer of antenna initiation and electron transfer of cofactor reduction." Journal of Physical Chemistry B 108(46): 18026-18033.
142 Saxena, C., H. Y. Wang, et al. (2005). "Ultrafast dynamics of resonance energy transfer in cryptochrome." Journal of the American Chemical Society 127(22): 7984-7985.
Schärer, O. D. (2003). "Chemistry and biology of DNA repair." Angewandte Chemie- International Edition 42(26): 2946-2974.
Schotte, F. et al. (2003). “Watching a Protein as it Functions with 150-ps Time-Resolved X-ray Crystallography." Science 300: 1944-1947.
Schelvis, J. P. M., M. Ramsey, et al. (2003). "Resonance Raman and UV-Vis spectroscopic characterization of FADH(center dot) in the complex of photolyase with UV-damaged DNA." Journal of Physical Chemistry B 107(44): 12352-12362.
Schramm, V. L. (2006). "Introduction: Principles of enzymatic catalysis." Chemical Reviews 106(8): 3029-3030.
Seidel, C. A. M., A. Schulz, et al. (1996). "Nucleobase-specific quenching of fluorescent dyes .1. Nucleobase one-electron redox potentials and their correlation with static and dynamic quenching efficiencies." Journal of Physical Chemistry 100(13): 5541-5553.
Selby, C. P. and A. Sancar. (2006) “A cryptochrome/photolyase class of enzymes with single-stranded DNA-specific photolyase activity.” Proceedings of the National Academy of Sciences of the United States of America 103: 17696-17700.
Seidel, C.A.M., A. Schulz , et al. (1996). “Nucleobase-specific quenching of fluorescent dyes. 1. nucleobase one-electron redox potentials and their correlation with static and dynamic quenching efficiencies.” Journal of Physical Chemistry B 106: 8858-8869.
Shida, T. Electronic Absorption Spectra of Radical Ions; Elsevier: Amsterdam, 1988.
Shida, T., W. H. Hamill (1966). “Molecular ions in radiation chemistry. II. Aromatic- hydrocarbon cations in CCl4 at 77°K.” Journal of Chemical Physics 44(6): 2375-2377.
Simondsen, R. P. and G. Tollin (1980). "Structure-Function Relations in Flavodoxins." Molecular and Cellular Biochemistry 33(1-2): 13-24.
Smiley, R. D. and G. G. Hammes. (2006). "Single molecule studies of enzyme mechanisms." Chemical Reviews 106(8): 3080-3094.
Srajer, V., T. Y. Teng, et al. (1996). "Photolysis of the carbon monoxide complex of myoglobin: Nanosecond time-resolved crystallography." Science 274(5293): 1726-1729.
143 Srajer, V., Z. Ren, et al. (2001). "Protein conformational relaxation and ligand migration in myoglobin: A nanosecond to millisecond molecular movie from time-resolved Laue X-ray diffraction." Biochemistry 40(46): 13802-13815.
Stanley, R. (2001). “Advances in flavin and flavoproteins optical spectroscopy.” Antioxidant & Redox Signaling 3:847-866.
Svoboda, D.L., C. A. Smith, J. S. Taylor and A. Sancar. (1993). “Effect of sequence, adduct type, and opposing lesions on the binding and repair of ultraviolet photodamage by DNA photolyase and (A)BC excinuclease.” Journal of Biological Chemistry 268:10694-10700.
Tamada, T., K. Kitadokoro, et al. (1997). "Crystal structure of DNA photolyase from Anacystis nidulans." Nature Structure Biology 4(11): 887-91.
Taylor J. S. (1994). “Unraveling the Molecular Pathway from Sunlight to Skin Cancer.” Accounts of Chemical Research 27(3): 76-82.
Teague, S. J. (2003). "Implications of protein flexibility for drug discovery." Nature Reviews Drug Discovery 2(7): 527-541.
Todo, T. (1999). "Functional diversity of the DNA photolyase/blue light receptor family." Mutation Research 434(2): 89-97.
Tugarinov, V., P. M. Hwang, L. E. Kay. (2004). “Nuclear magnetic resonance spectroscopy of high-molecular-weight proteins.” Annual Review of Biochemistry 73, 107-146.
van den Berg, P. A., A. van Hoek, et al. (1998). "Flavin fluorescence dynamics and photoinduced electron transfer in Escherichia coli glutathione reductase." Biophysical Journal 74(4): 2046-2058. van den Berg, P.A.W.; Feenstra, K.A. (2002). “Dynamic conformations of flavin adenine dinucleotide: simulated molecular dynamics of the flavin cofactor related to the time- resolved fluorescence characteristics.” Journal of Physical Chemistry B106: 8858-8869.
Visser, A. J., G. J. Van Ommen, et al. (1974). "Laser photolysis of 3-methyllumiflavin." Photochemistry and Photobiology 20(3): 227-32.
Visser, A. J., S. Ghisla, et al. (1979). "Fluorescence properties of reduced flavins and flavoproteins." European Journal of Biochemistry 101(1): 13-21.
Visser, A., A. Vanhoek, et al. (1980). "Laser-Induced Luminescence of Flavinium Salts." Helvetica Chimica Acta 63(5): 1296-1303.
144 Voityuk, A. A. and N. Rosch (1997). "Ab initio study on the structure and splitting of the uracil dimer anion radical." Journal of Physical Chemistry A 101(44): 8335-8338
Voityuk, A. A., M. E. Michel-Beyerle, et al. (1998). "Quantum chemical modeling of structure and absorption spectra of the chromophore in green fluorescent proteins." Chemical Physics 231(1): 13-25.
Voityuk, A. A., M. E. Michel-Beyerle, et al. (1998). "Structure and rotation barriers for ground and excited states of the isolated chromophore of the green fluorescent protein." Chemical Physics Letters 296(3-4): 269-276.
Vorsa, V., T. Kono, et al. (1999). "Femtosecond photoionization of ion beam desorbed aliphatic and aromatic amino acids: Fragmentation via alpha-cleavage reactions." Journal of Physical Chemistry B 103(37): 7889-7895.
Vos, M. H. and J. L. Martin (1999). “Femtosecond processes in proteins.” Biochemica Et Biophysica Acta 1411: 1-20
Wang, Q., R. W. Schoenlein , et al. (1994). “Vibrationally coherent photochemistry in the femtosecond primary event of vision.” Science 266(5184):422-424.
Wang, Y., J. S. Baskin, et al. (2004). "Human myoglobin recognition of oxygen: dynamics of the energy landscape." Proceedings of the National Academy of Sciences of the United States of America 101(52): 18000-18005.
Weber, G. (1948). “The quenching of fluorescence in liquids by complex formation. Determination of the mean life of the complex.” Transactions of the Faraday Society 44: 185-189.
Weber, W., Helms V. , et al. (1999). “Shedding light on the dark and weakly fluorescent states of green fluorescent proteins.” Proceedings of the National Academy of Sciences of the United States of America 96: 6177-6182.
Werner, P. E, B. Linnros , et al. (1971). “Studies on flavin derivatives. The crystal structure of 5-diethyl-3,7,8,10-tetramethyl-1,5-dihydroisoalloxazine trihydrate.” Acta Chemica Scandinavica 25: 1297–1312.
Williams, S., T. P. Causgrove, et al. (1996). "Fast events in protein folding: Helix melting and formation in a small peptide." Biochemistry 35(3): 691-697.
Worthington, E. N., I. H. Kavakh, et al. (2003). "Purification and characterization of three members of the photolyase/cryptochrome family blue-light photoreceptors from Vibrio cholerae." Journal of Biological Chemistry 278(40): 39143-39154.
145 Wouters, J., F. Durant , et al. (1998). “Electronic properties of flavins: Implications on the reactivity and absorption properties of flavoproteins.” International Journal of Quantum Chemistry 64: 721-733.
Wulff, D. L. (1961). “On nature of thymine photoproduct.” Biochimica Et Biophysica Acta 51: 332 1961.
Yarkony, D.R. (1998). “Conical Intersections: diabolical and often misunderstood.” Accounts of Chemical Research 31: 511-518.
Zgrablic, G. (2006). “Solvent effects on the ultrafast dynamics of the retinal chromophore of bacteriorhodopsin.” Ph.D. thesis Ecole polytechnique fédérale de Lausanne (EPFL), Institut des sciences et ingénierie chimiques, no 3493.
Zhang, L. Y., Y. T. Kao, et al. (2006). "Femtosecond studies of tryptophan fluorescence dynamics in proteins: Local solvation and electronic quenching." Journal of Physical Chemistry B 110(37): 18097-18103.
Zhao, L., S. K. Pal, et al. (2003). “Dynamics of Ordered Water in Interfacial Enzyme Recognition: Bovine Pancreatic Phospholipase A2”. Angewandte Chemie International Edition 43(1): 60-63.
Zhong D., E. W. G. Diau , et al. (1998). “Femtosecond dynamics of valence-bond isomers of azines: transition states and conical intersections.” Chemical Physics Letters 298: 129-140.
Zhong, D. And A. H. Zewail (2001). “Femtosecond dynamics of flavoproteins: charge separation and recombination in riboflavine (vitamin B2)-binding protein and in glucose oxidase enzyme.” Proceedings of the National Academy of Sciences of the United States of America 98(21): 11867-11872.
Zhong, D. P., S. K. Pal, et al. (2001). "Femtosecond studies of protein - DNA binding and dynamics: Histone I." Chemphyschem 2(4): 219-227.
Zhong, D., S. K. Pal, et al. (2002). “Femtosecond dynamics of rubredoxin: tryptophan solvation and resonance energy transfer in the protein.” Proceedings of the National Academy of Sciences of the United States of America 99(1): 13-18.
146
APPENDIX A
DNA-Photolyase Protein Purification Protocol
147
A.1
DNA-photolyase purification protocol
DAY 1-Transformation
1) Take 100 ul of UNC 523 cells in an ependorf over ice. 2) Add 1ul of plasmid DNA in 100 ul of UNC 523 cells. 3) Incubate in ice for 30 minutes. (Meantime switch-on heat block and put little water in the wells and set the temperature to 45 ˚C.) 4) Heat-shock the UNC 523 + Plasmid DNA mixture by placing the cell + DNA mixture in heat block at 42 ˚C for 45 seconds. 5) Take out the mix and incubate over ice for 2 more minutes. 6) Spread the UNC 523 + plasmid DNA mix over the LB+AMP plates. 7) Put the LB +AMP plate upright at 37 ˚C for 30 minutes. 8) Turn the plate upside down and keep it overnight at 37 ˚C.
DAY 2- Preparing the Overnight Culture
1) Take 5 ml autoclaved LB in a sterile test-tube. Add 10 ul of AMP (50 mg/ml) over flame. 2) Pick single colony from yesterday’s culture and dilute in this 5 ml of LB+AMP. 3) Keep the test-tube for 6-8 hours at 37 ˚C. 4) Take 100 ml autoclaved LB in a sterile flask and Add 200 ul of AMP over flame. 5) Dilute one ml of the test-tube culture to this 100 ml LB+AMP. 6) Grow the 100 ml culture overnight. (This is the overnight culture)
DAY 3- Larger cell culture and cell harvesting
1) Take 1 lt autoclaved LB in a sterile flask. Add 2 ml of AMP over flame. 2) Take 10 ml of overnight culture and dilute to 1 lt LB + AMP. 3) Grow at 37 ˚C till the O.D. of the culture reaches to 0.6 (0.6-0.75). (It usually takes 2 hr 30 minutes in the available incubator in the lab. ) 4) Add 2ml of IPTG (0.5M) to 1 lt culture. 5) Let it grow for next 4 hr. 6) Transfer the culture to centrifuge flasks. 7) Centrifuge the culture for 10 minutes at 4 ˚C with 4.5K RPM. 8) Discard supernatant. 9) Dissolve the pallet in PBS (10 ml / lt). 10) Centrifuge the dissolved pallet for 20 minutes at 4˚C with 4.5K RPM in 50 ml plastic tubes. 11) Discard the supernatant. 12) Quick-freeze the pallet in dry ice for 10 minutes. 13) Transfer the 50 ml plastic tubes containing frozen pallet to –80 ˚C for further use.
148
A.2
DAY 4 -Cell Lysis and Protein Purification
1) Thaw the frozen pallet into ice-water bath at 4 ˚C. Add lysis buffer (10 ml / lt of cells) + lysozyme 1.0 mg / ml of buffer). Mix well and leave overnight for mixing on rocker at 4 ˚C, or in an ice-water bath. 2) Next day chill the T-10 rotor in centrifuge machine at 4 ˚C. 3) Take the dissolved pallet tubes in an ice bucket. Sonicate the cells over ice for 15 seconds and 6 times with the available sonicator in the lab. Make sure that the dissolved pallets forms nice slurry and no more clogs / precipitates are there in the tubes. Also don’t over sonicate. Use of ice is essential to avoid any oxidized flavin cofactor in photolyase enzyme. 4) Centrifuge the sonicate at 4 ˚C for 60 minutes with 14K RPM. 5) Take clean supernatant; combine (if multiple centrifuge tubes were used). 6) Measure the supernatant and weigh out the 0.43 gm / ml ammonium sulfate. 7) Stir the supernatant in an ice-bath at 4 ˚C and add the ammonium sulfate slowly in the period of 30-45 minutes. 8) Stir for extra 30-45 minutes. 9) Centrifuge the mix at 12K for 30 min. Make sure that tubes you are using can sustain 12K RPM and every apparatus in use is cold. 10) Resuspend the pallet GENTLY with 5 ml of low-salt buffer per pallet. (Buffer A). 11) Dialyze the resuspend pallets against the low-salt buffer A for (4+4) hours. Change the dialyzing buffer after 4 hrs during the total 8 hr dialysis. 12) After dialysis dilute the resuspend pallet to double the volume (if resuspend pallet volume is 40 ml then add 40 ml extra buffer A) and load (this suspension now called protein) onto the blue sepharose column that has been equilibrated with low-salt buffer A. Usually load the protein overnight with the flow-rate of 0.15 ml/ml or depending on the total volume of loading.
DAY 5- Protein Purification continues
1) Wash the column with low-salt buffer A. Collect fractions while washing. Set fraction collector to 160 drop-not time for 5 ml fraction. Maintain the flow rate to 0.4-0.5 ml / minute. Continue collecting until there is no more yellow color in any of the fraction tubes. 2) Remove the low-salt buffer head and replace with high-salt buffer B. Elute the protein with this buffer. Blue fractions are photolyase. 3) If MTHF degraded photolyase enzyme is required then follow the step 4 and 5, if not then directly got to step 6. 4) Take out the photolyase fraction tubes add DTT (final concentration 10 mM) and keep them tilted at an angle of 45 in an ice bucket. 5) Expose the fractions to 365 nm light for 2-3 hrs (with Black-ray 100 SP) to photodegrade the MTHF cofactor of the photolyase. Keep rotating the fraction
149 A.3
tubes manually to avoid the heating of the tube. Absorption spectra before and after exposure can be measured for surety of MTHF degradation.
6) Combine the fraction and concentrate in Amicon-tubes to about 1-3ml. This might take 2-4 hr. 7) Load the concentrated protein on to gel-filtration (sizing column) that has been equilibrated with buffer C. Once the protein gets in to the resins, connect the column with the buffer C reservoir. Set fraction collector to 160 drop-not time for 5 ml fraction. Maintain the flow rate to 0.4-0.5 ml / minute.
Day 6- Protein Purification continues
1) Watch for the blue color in the column or in the collected fractions. 2) Collect and combine all the blue fractions. 3) Load the fractions on to Hydroxyapatite column that has been equilibrated with buffer C. Allow the protein to go below the resin then add a small amount approx 0.5ml of buffer C. Elute the column with the gradient maker using a gradient of buffer C and buffer D. Set fraction collector to 80 drop-not time for ~3 ml fraction. Maintain the flow rate to 0.4-0.5 ml / minute. Watch for blue color to confirm the elution of protein. Always repack the Hydroxyapatite column before each new use. 4) Depending on the fraction size and numbers, concentrate the protein and dialyze 8 hr-overnight against the 1 L of autoclaved pre-cooled storage buffer, buffer E. Aliquot into cold ependorf tubes and quick-freeze in a dry ice ethanol bath. Store at –80 ˚C.
Buffers
Lysis buffer 4L 50mM Tris pH 8 17.76g Tris-HCl 10.56g Tris-Base 100mM NaCl 23.36g NaCl 1mM EDTA 16.00ml of 250mM stock
Bring upto 4L with DIW and sterilize 10mM BME 2.8ml BME (add the BME after sterilization)
Buffer A (1X) 1L 2L 3L TRIS-HCl (50mM) 6.6g 13.2g 19.8g EDTA (1.0mM) 0.96g 1.92g 2.88g
150 A.4 KCl (0.1M) 7.43g 14.9g 22.29g DIW 900ml 1800ml 2700ml AUTOCLAVE After sterilization add:
10% Glycerol 100ml 200ml 300ml (Sterile) BME (10mM) 0.72ml 1.43ml 2.16ml
Buffer B (1X) 1L 2L 3L TRIS-HCl (50mM) 6.6g 13.2g 19.8g EDTA (1.0mM) 0.96g 1.92g 2.88g KCl (0.1M) 149.12g 298.24 447.36g DIW 900ml 1800ml 2700ml AUTOCLAVE After sterilization add: 10% Glycerol 100ml 200ml 300ml (Sterile) BME (10mM) 0.72ml 1.43ml 2.16ml
1M Potassium Phosphate pH 6.8 (for Buffer C and D) 2L
KH2PO4 (monobasic) 136.09g K2HPO4 (daibasic) 174.18g Add DIW to 2L for stock AUTOCLAVE
Buffer C (1X) 1L
1M potassium phosphate pH 6.8 67ml 0.25 M EDTA (stock) 4ml Glycerol (sterile) 100ml DIW 829ml Add 0.70ml BME
Buffer D (1X) 330mM Potassium Phosphate (KPi)
500ml 1M KPi pH 6.8 165ml 0.25m EDTA 2ml
151 A.5 Glycerol 100ml BME 0.14ml DIW 233ml
10 X Photolyase storage buffer Salts for Buffer E
4L 50mM TRIS 264.4g TRIS-HCl 38.8g TRIS BASE 1mM EDTA 14.8g 50mM NaCl 116.8g Bring up to 4L and autoclave
Buffer E (1X) Storage buffer
500ml 1L DIW (sterile) 200ml 400ml 10X SALTS 50ml 100ml Glycerol (Sterile) 250ml 500ml DTT 0.77g 1.54g
Column Regeneration
Blue-Sepharose – Rinse with high salt buffer B up to 5-column volume then rinse with 2- columns volume 2M guanidine chloride and then rinse with low salt buffer A for at least 5-column volume.
Gel-Filtration – Rinse with Buffer C two-three column volumes.
Hydroxylapatite – After elution Rinse the resins in a beaker two-three times with buffer D and then three times with buffer C. Load the Resin on to the column and then rinse the resins at least two times in column with buffer C. (Avoid using pump while packing the column; use of pump very tightly packs the resins at the bottom and tightness goes down up in with the column height. Column packed using pump makes loading and elution of protein very –very slow).
152
APPENDIX B
Photoreduction and Photodecomposition of Cofactors in Photolyase
153 B.1
Photoreduction of flavin cofactor in photolyase for steady-state and time resolved spectroscopic measurements
1) Take 200-300 ul of appropriate concentration of photolyase enzyme, still in storage buffer. 2) Add 10 mM final concentration of DTT. 3) Measure the absorption spectrum after pipetting the enzyme solution in 5 mm quartz cell. Note flavin neutral radical absorption peaks at 485, 580 and 625 nm. 4) Cap the quartz cell and place horizontally on ice. Irradiate the sample with available high-intensity table lamp in the lab which has a cut-off filter >550 nm. Maintain a constant difference between lamp position and quartz cell, I usually keep them 5-7 cm apart. 5) Measure the absorption spectrum every after 5 minutes of irradiation and monitor decrease in flavin neutral radical absorption peaks. 6) Irradiate the sample as long as flavin neutral radical absorption peaks are visible. 7) Complete reduction of flavin neutral radical, in a protein concentration of 400 uM in 200 ul volume, takes about 30-45 minute.
Notes: Reduction rate is highly dependent on a) concentration of reducing agent (DTT), b) distance of light source from the sample and c) presence of oxygen in the quartz cell. Technically it is impossible to reduce 100% of flavin in photolyase in aerobic conditions. If time-resolved spectroscopic experiments are planned then purge the nitrogen in the sample and make conditions anaerobic.
Photodecomposition of folate cofactor in photolyase for steady-state and time resolved spectroscopic measurements
1) Take 200-300 ul of appropriate concentration of photolyase enzyme, still in storage buffer. 2) Add 10 mM final concentration of DTT. 3) Measure the absorption spectrum after pipetting the enzyme solution in 5 mm quartz cell. Note folate absorption peak at 384 nm and flavin neutral radical absorption peaks at 485, 580 and 625 nm. Also measure fluorescence emission spectra after exciting the sample at 380 nm and monitor emission from 400-700 nm. 4) First photoreduce the flavin cofactor in photolyase as described above. 5) After flavin cofactor reduction is achieved, irradiate the sample on ice with available high-intensity Black-Ray 366 nm lamp. Maintain a constant difference between lamp position and quartz cell, I usually keep them 10-15 cm apart. Black-ray is a high intensity lamp and generates lot of heat and can denature the protein very easily.
154 B.2 6) Measure the absorption and emission spectrum every after 30 minutes of irradiation and monitor decrease in folate absorption and emission peak at 384 and 468 nm respectively. 7) Irradiate the sample as long as folate emission peaks are visible. 8) Complete reduction of folate, in a protein concentration of 400 uM in 200 ul volume, takes about 2-4 hours.
Notes: Reduction rate is highly dependent on a) concentration of reducing agent (DTT), b) distance of light source from the sample and c) presence of oxygen in the quartz cell. Technically it is impossible to reduce 100% folate in photolyase in aerobic conditions. If time-resolved spectroscopic experiments are planned then purge the nitrogen in the sample and make conditions anaerobic.
155
APPENDIX C
Preparation of Photolyase Substrates
156 C.1
Preparation of cyclopyrimidine dimers
1) Dissolve pyrimidine monomer in double distilled water with maximum concentration of 100 ODUs per ml. 2) Measure the UV-Vis absorption spectra of a further diluted sample to confirm the starting concentration, also measure fluorescence emission from the sample after exciting the samples at 360 nm and monitor emission from 380-700 nm. 3) Add 15% acetone by volume in the 1 ml solution. 4) Have an anaerobic quartz cell ready; transfer the solution to the quartz cell. (I used especially designed reaction cell, available in the lab). Maintain a slow flow of argon over the sample and keep the sample over ice. 5) Purge Argon very slowly in the cell and continue purging it during irradiation, which is the next step in Dimer preparation. (Argon purging should be very-very slow, the idea is just to decrease the oxygen concentration in the cell which might work as quencher to acetone sensitization. Higher rate of Argon purging can push the solution out of the quartz reaction cell, if not at least it will evaporate the acetone and monomer sensitization by acetone will be decreased which will eventually lead to the formation of 6-4 photoproduct) (I used the 5 ml injection needle and the rate of Argon purging was tested separately in a beaker filled with water and I kept a rate of one bubble per two seconds) 6) Irradiate the monomer sample over ice with the 302 nm white fluorescence light. (I used UVP Lamp product 34-0039-01 this light tube’s emission ranges from 280–320 nm with the peak at 302 nm). I kept the light source at 5 cm over the reaction cell. 7) Irradiate the sample for 2-3 hr. Take a little sample out after every one hour, dry out the solvent (by vacuum drying or other methods) redissolve in water and check the UV-Vis absorption spectrum. Decrease in 260 nm peak is indication of dimer formation. Irradiate the sample as long as monomer absorption is visible. With the conditions given above usually 2 to 3 hours is enough for 100 % reduction in 260 nm absorption peak. Avoid unnecessary irradiation which may lead to formation of 6-4 photoproducts. 8) Dry out the solvent (by vacuum drying or other methods) and store the sample in desired buffer/ water /dried at –20 ˚C till further use.
Reactivation Experiment
1) Take approximately 1:200 (photolyase enzyme : dimer substrate) concentration in TE buffer (50mM Tris, 100mM NaCl, 1mM EDTA, 1mM DTT, 5% glycerol v/v). Typically I used approximately 400 nM enzyme and 80 uM dimer) 2) Acquire UV-Vis absorption spectrum of enzyme + dimer. (Avoid exposure of light to enzyme dimer mix). Define this as background spectra and subtract this spectrum after every irradiation cycle. 157 C.2 3) Incubate the enzyme + dimer mixture over ice for 5 minutes. 4) Irradiate the enzyme + dimer mixture over ice with 366 nm light for 5 minute; acquire the UV-Vis absorption spectrum. Once the dimer is repaired, the monomer will show characteristic absorption at 260 nm. 5) Irradiate the enzyme + dimer mixture until no more increase at 260 nm is visible in absorption spectra. 6) Compare the change in 260 nm absorption before and after irradiation, and calculate the monomer concentration formed. Assuming enzyme repairs 100% of the dimer present in the solution, calculated concentration of monomer can provide starting concentration of dimer and can be used to cross-check the dimer concentration.
158
APPENDIX D
Preparation of Reduced Flavin Species
159 D.1
Guidelines / protocol for flavin-reduction
Reduced flavin can be obtained either by chemical or photochemical reduction of flavin as described below.
Chemical Reduction Method
Reducing Agent: Sodium BoroHydride (Fisher-Scientific S678-10) CAS No.: 16940-66-2 Molecular Weight: 37.83 Chemical Formula: NaBH4
Precautions: Read the MSDS sheet of NaBH4. It is a very strong reducing agent and corrosive in nature. Avoid contact with oxidizing agents. NaBH4 is hygroscopic in nature and can readily absorb moisture. Aqueous solution of NaBH4 stored in ependorf tube will develop hydrogen pressure and one can observe that after a while ependorf-tube will pop- open because of hydrogen pressure build-up. Similar hydrogen pressure build-up can be severe, if one seal the NaBH4 containing bottle, which already came in contact with moist air. Avoid closing a bottle or ependorf, having high concentration of aqueous NaBH4.
1) Prepare 12.5 mM phosphate buffer at pH 8.0- 8.5. (Notes: The pH value is very critical however, the concentration of phosphate buffer is not very critical and I observed a similar Flavin reduction in 5, 10, 15 and 20 mM phosphate buffer). 2) Dissolve desired concentration of Riboflavin (RF), FMN or FAD in the phosphate buffer. (Notes: For complete reduction of Flavin at least 800 times concentration of reducing agent is needed, so concentration calculations should be done in advance) I normally used ~ 150 uM of flavin in 400 ul volume. To reduce this I need 150 uM X 800 = 120000 uM OR 120 mM of reducing agent. And I prefer addition of not more than 50 ul volume of reducing agent; hence I prepare 1.08 M solution of reducing agent and then take 50 ul of it to be dissolved in 400 ul of 150 uM flavin solution New total volume 400 ul + 50 ul = 450 ul New flavin concentration= (150 uM X 400 ul) / 450 ul = 133.33 uM Final reducing agent concentration = (1.080 M X 50 ul) / 450 ul = 0.12 M OR 120 mM) 3) Deeply purge the flavin solution with nitrogen. (Notes: I normally used 400 uL of solution and purged the nitrogen into the solution through slow bubbling for three hours. I purge the nitrogen in water (attached at the end of the purging system) for at least two hours before purging
160 D.2 the flavin solution. Also I stir the solution, using a magnetic stirrer while purging for efficient oxygen removal). (Though not very critical but I cover the solution with aluminum foil so that it is not exposed to the light. This helps in generating reduced flavin, reduced only by reducing agent, which I will be adding in next step and flavin reduction in anaerobic condition due to light can be minimized)) 4) Freshly prepare the required concentration (800 times of flavin concentration) of NaBH4 in similar phosphate buffer as used in preparation of flavin solution. (Notes: I tried reducing concentration of 200, 500, 800, 1000 and 2000 times to that of flavin concentration but found out that 700-800 times is just sufficient for complete reduction of flavin) (Notes: As soon as I dissolve NaBH4 in buffer I see that hydrogen gas is making bubbles in the solution. This step is very tricky if I have to add this bubbly solution of reducing agent in the flavin solution, than all concentration calculations will go off because it is very difficult to measure the volume occupied by the bubbles, so I remain extra careful in tapping the reducing agent solution again and again, before I aliquot the reducing agent solution to add into the flavin solution. Practically, adding reducing agent little more than 800 times is not harmful but I try to avoid needless excess of reducing agent in the solution). 5) After two hours of nitrogen purging, add required reducing agent concentration to flavin solution with the nitrogen feeding tube. Purge the solution for extra 20 minute with nitrogen. Slowly move the nitrogen tube up and purge the nitrogen in reaction cell 20-20 minute at the mid and neck region of the tube. (Notes: Still keep the sample covered by the aluminum file) 6) Tight the reaction cell knobs and take the reaction cell out of the purging unit. 7) Keep the cell and sample covered by aluminum foil or in dark for extra 30-45 minute. (Notes: This provides sufficient time for complete reduction of flavin; higher concentration of flavin takes little longer for complete reduction.) 8) Measure the absorption and emission spectra; reduced flavin should be ready in the reaction cell. Reduced flavin solution is faint yellow to colorless depending on the higher to lower initial concentration of flavin used in the preparation.
Checking the quality of the preparation:
Stir the sample for one hour and measure the fluorescence spectrum again • If emission peak start moving towards red side, then conditions are not completely anaerobic and oxygen is present in the reaction cell. • If emission peak start moving towards red and also intensity of the fluorescence increased, then along with presence of oxygen, reducing agent is insufficient in the solution.
If no further experiment is planned and to make sure protocol is working, open the cell and shake it for just a minute, solution will immediately becomes yellow.
161 D.3 Measure the absorption; at least 80-90% of the initial oxidized flavin absorption recovers. Further, leave the cell open to oxygen and in couple of hours (3-8 hours depending on initial flavin concentration) solution will turn colorless. Measure the emission; a dominant emission at 470 nm but not around 440 nm will be observed. (However note, this is not the reduced flavin we want but a 4-hydroxy flavin product formed in the presence of oxygen and NaBH4).
Reference: Methods in Enzymology Vol. 18, (1971), pp 471-479
Photo-Chemical Reduction Method
Reducing Agent: Sodium-Oxalate (Fisher-Scientific S487-500) Synonyms: Ethandioic acid disodium salt; disodium salt oxalic acid CAS No.: 62-76-0 Molecular Weight: 134.00 Chemical Formula: NaOCOCOONa
1) Prepare 12.5 mM phosphate buffer at pH 8.0 to 8.5 if you are seeking a deprotonated reduced flavin, and/or use pH 4.5 to 5.0 if you are seeking a protonated reduced flavin species. (generally for pH 4.5 to 5.0 phosphate buffer is not very advisable but for the purpose we can use it) (Notes: The pH value is very critical however, the concentration of phosphate buffer is not very critical and I observed a similar Flavin reduction in 5, 10, 15 and 20 mM phosphate buffer). 2) Dissolve desired concentration of FAD in the phosphate buffer. Measure the absorption. (Notes: For complete reduction of Flavin you need at least 30 times concentration of reducing agent, so calculate concentrations in advance) I normally used ~ 250 uM of flavin in 400 ul volume. To reduce this I need 250 uM X 30 = 7500 uM OR 7.5 mM of reducing agent. And I prefer addition of not more than 50 ul volume, so I prepare 67.5 mM solution of reducing agent and than take 50 ul of it to be dissolved in 400 ul of 250 uM flavin. New total volume 400 ul + 50 ul = 450 ul New flavin concentration= (250 uM X 400 ul) / 450 ul = 222.22 uM Final reducing agent concentration = (67.5 mM X 50 ul) / 450 ul = 7.5 mM 3) Prepare the required reducing agent concentration in phosphate buffer add the reducing agent in the flavin solution and measure the absorption. 4) Deeply purge the flavin + reducing agent solution with nitrogen. (Notes: I normally used 450 uL of solution and purge the nitrogen into the solution through slow bubbling for three hour. I purge the nitrogen in water (attached at the end of the purging system) for at least two hours before purging the flavin solution. Also I stir the solution, using a magnetic stirrer while purging for efficient oxygen removal)
162 D.4
5) Slowly move the nitrogen tube up and purge the nitrogen in reaction cell 20-20 minute at the mid and neck region of the tube. 6) Tight the reaction cell knobs and take the reaction cell out of the purging unit 7) Lock the sample cell on vertical stands and irradiate the sample with 366 nm light perpendicular to the sample. Keep the lamp 8-10 cm away from the sample cell. Measure the absorption and emission time to time. Depending on the starting concentration of flavin (100-500 uM) and reducing agent (50 to 20 times that of flavin), preparation of reduced flavin may take anywhere from 4 to 30 hrs. Irradiation can be stopped when a dominant single fluorescence peak around 440 nm is observed. During the process one will observe that from single FAD emission peak, emission changes to two peaks with varying location depending on the irradiation time, and finally a single fluorescence irradiation peak can be observed after sufficient 366 nm irradiation. (I have observed that if flavin concentration is very high say 1 mM and reducing agent concentration in only 20 times i.e. 20 mM than it takes more than 48hrs to reduce the flavin) 8) Measure the absorption and emission spectra; reduced flavin should be ready in the reaction cell. Reduced flavin solution is faint yellow to colorless depending on the higher to lower initial concentration of flavin used in the preparation.
Checking the quality of the preparation:
Stir the sample for one hour and measure the fluorescence spectrum again • If emission peak start moving towards red side, then conditions are not completely anaerobic and oxygen is present in the reaction cell. • If emission peak start moving towards red and also intensity of the fluorescence increased, then along with presence of oxygen, reducing agent is insufficient in the solution.
If no further experiment is planned and to make sure protocol is working, open the cell and shake it for just a minute, solution immediately becomes yellow. Measure the absorption; at least 80-90% of the initial oxidized flavin absorption recovers.
Method developed from the following references:
1) Barry Halliwell Biochem. J. (Britain) 1972, vol 129, pp 497-498. 2) Sandro Ghisla and Vincent Massey J. Biol. Chem. (USA) 1975, vol 250 (2), pp 577-584. 3) Vincent Massey, marian Stankovich and Peter Hemmerich Biochemistry (USA) 1978, vol 17 (1), pp 1-8. 4) Vincent Massey and Peter Hemmerich Biochemistry (USA) 1978, vol 17 (1), pp 9-17.
163 5) Paul F. Heelis, Rosemarie F Hartman and Seth D. Rose Photochemistry and Photobiology (USA) 1993, vol 57 (6), pp 1053-1055.
164