A STUDY OF PROTEIN DYNAMICS AND COFACTOR
INTERACTIONS IN PHOTOSYSTEM I
A Dissertation
Presented to
The Academic Faculty
By
Shana L. Bender
In Partial Fulfillment
of the Requirements for the Degree
Doctorate of Philosophy in the
School of Chemistry and Biochemistry
Georgia Institute of Technology
December 2008
A STUDY OF PROTEIN DYNAMICS AND COFACTOR
INTERACTIONS IN PHOTOSYSTEM I
Approved by:
Dr. Bridgette Barry, Advisor
School of Chemistry and Biochemistry
Georgia Institute of Technology
Dr. Ingeborg Schmidt-Krey
School of Biology
Georgia Institute of Technology
Dr. Donald Doyle
School of Chemistry and Biochemistry
Georgia Institute of Technology
Dr. Nael McCarty
Department of Pediatrics
Emory University
Dr. Wendy Kelly
School of Chemistry and Biochemistry
Georgia Institute of Technology
Date Approved: October 15, 2008
ACKNOWLEDGEMENTS
I would like to thank my parents, who have given my strength, moral support and
practical advice throughout my educational endeavors. I would also like to thank my
husband who has always put my goals and education above his own ambitions. I would
like to acknowledge my entire family for their constant support in my education. I would
like to thank my grandparents who have kept me in their prayers, and my sister who has
never doubted my career goals. I would like to acknowledge my advisor, Dr. Bridgette
Barry for her valuable support throughout the years. She has been a great advisor, always
challenging me to think critically. She has helped me become a better scientist. I would
like to thank the past members of the Barry group: Dr. Idelisa Ayala who trained me, Dr.
Colette Sacksteder who taught me to be critical of everything, and Dr. Lori Anderson for
her guidance and support. I would like to thank the current members of the Barry group:
Ian Cooper, Robin Sibert, David Jensen, Jun Chen, Adam Offenbacher, Tina Dreaden,
James Keough, and Brandon Polander.
The members of the Barry group have been
supportive of each other, and have created a great working environment.
iii
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS
iii
LIST OF TABLES
vii
LIST OF FIGURES
viii
LIST OF SYMBOLS
x
LIST OF ABBREVIATIONS
xi
SUMMARY
xiii
CHAPTER
1
INTRODUCTION
1.1 Photosystem I
2
1.2 Primary Donor
3
1.3 Primary Electron Acceptor: A0
4
1.4 Secondary Electron Acceptor: A1
5
1.5 Terminal Electron Donor: P700
6
1.6 Iron Sulfur Clusters: FX, FA and FB
8
1.7 Interactions of PSI with Ferredoxin and Flavodoxin
10
1.8 Fourier-Transform Infrared Spectroscopy
10
1.9 Thesis outline
13
1.10 References
15
2. BIOSYNTHESIS PATHWAYS OF CHLOROPHYLL, PHYLLOQUINONE
AND AROMATIC AMINO ACIDS
2.1 Biosynthetic Pathway of Chlorophyll
29
2.2 Chorismate Biosynthesis
33
iv
2.3 Phylloquinone Biosynthesis
34
2.4 Aromatic Amino Acid Biosynthesis
36
2.5 References
40
3. THE VIBRATIONAL SPECTRUM OF THE SECONDARY ELECTRON
ACCEPTOR, A1 IN PHOTOSYSTEM I
3.1 Abstract
52
3.2 Introduction
54
3.3 Materials and Methods
57
3.4 Results
62
3.5 Discussion
69
3.6 Acknowledgements
75
3.7 References
76
4 LIGHT-INDUCED DYNAMICS IN PHOTOSYSTEM I ELECTRON
TRANSFER
4.1 Abstract
93
4.1 Introduction
95
4.3 Materials and Methods
98
4.4 Results
103
4.5 Discussion
108
4.6 Acknowledgements
114
4.7 References
113
5 PROBING ELECTRON T RANSFER IN PHOTOSYSTEM I
5.1 Abstract
133
5.2 Introduction
134
5.3 Materials and Methods
136
5.4 Results
139
v
5.5 Discussion
145
5.6 Acknowledgements
148
5.7 References
149
6 CONCLUSION
162
vi
LIST OF TABLES
Page
Chapter 3
Table 1
85
Table 2
86
Chapter 4
Table 1
120
Table 2
121
Table 3
122
vii
LIST OF FIGURES
Page
Chapter 1
Figure 1
22
Figure 2
23
Figure 3
24
Figure 4
26
Figure 5
27
Figure 6
28
Chapter 2
Figure 1
45
Figure 2
46
Figure 3
47
Figure 4
48
Figure 5
49
Figure 6
50
Figure 7
51
Chapter 3
Figure 1
87
Figure 2
88
Figure 3
89
Figure 4
90
Figure 5
91
Figure 6
92
viii
Chapter 4
Figure 1
123
Figure 2
124
Figure 3
125
Figure 4
127
Figure 5
129
Figure 6
131
Figure 7
132
Chapter 5
Figure 1
153
Figure 2
154
Figure 3
155
Figure 4
156
Figure 5
157
Figure 6
158
Figure 7
160
ix
LIST OF SYMBOLS
δ
delta
A1
phylloquinone electron acceptor in Photosystem I
A0
chlorophyll electron acceptor in Photosystem I
FA
iron-sulfur cluster electron acceptor in Photosystem I
FB
iron-sulfur cluster electron acceptor in Photosystem I
FX
iron-sulfur cluster electron acceptor in Photosystem I
P700
chlorophyll dimer electron donor in Photosystem I
Q
quinone
x
LIST OF ABBREVIATIONS
δ-aminolevulinic acid
ALA
Asp
aspartic acid
chl
chlorophyll
CoA
coenzyme A
CDRP
carboxyphenylamino-1-deoxyribulose
DAHP
3-deoxy-D-arabino-heptulosonate-7-phosphate
DFT
density functional theory
E. coli
Escherichia coli
EPR
electron paramagnetic resonance
EPSP
enolpyruvyl shikimate-3-phosphate
FT-IR
Fourier-transform infrared
Glu
glutamic acid
IGP
indole-3-glycerol phosphate
Leu
leucine
MALDI
matrix assisted laser desorption ionization
NADPH
nicotinamide adenine dinucleotide phosphate
PCC
pasteurized cell culture
PEP
enolpyruvate
PhQ
phylloquinone
PSI
Photosystem I
PSII
Photosystem II
Thr
threonine
tRNA
transfer ribonucleic acid
Trp
tryptophan
xi
UVRR
ultraviolet resonance Raman
xii
SUMMARY
Previous research has underscored the importance of protein dynamics during
light-induced electron transfer; however, specific interactions have not been well
characterized. It is of particular importance to understand the role of protein dynamics
and cofactor interactions in controlling electron transfer in oxygenic photosynthesis.
These factors include hydrogen bonding, π-stacking and electrostatic interactions.
Reaction-induced FT-IR spectroscopy is sensitive to these interactions as well as isotopic
incorporation, and is useful to probe protein dynamics associated with light-induced
electron transfer in Photosystem I (PSI). Density functional theory (DFT) provides
information concerning the vibrational frequencies of molecules as well as the amplitudes
of the vibrations and sensitivity to isotope incorporation. Combining these approaches,
protein dynamics associated with light-induced electron transfer in PSI were studied.
The work presented here describes specific protein cofactor interactions and specific
protein relaxation events associated with light-induced electron transfer. The results
reported here are consistent with noncovalent protein cofactor interactions that modulate
the redox potential of the secondary electron acceptor of PSI. Furthermore, the studies
presented here describe novel protein dynamics associated with the oxidation of the
terminal electron donor of PSI. These results characterize specific protein dynamics that
may be associated with interactions of the soluble electron donors.
highlight the importance of protein dynamics in oxygenic photosynthesis.
xiii
These studies
CHAPTER 1
INTRODUCTION
Photosynthesis is a biological process, which utilizes light energy to produce a
charge separated state across a membrane. This biological process occurs in plants, green
algae and cyanobacteria. In plants, the organelles known as the chloroplasts house the
photosynthetic machinery responsible for light-induced charge separation. Inner
membranes known as thylakoid membranes are stacked within the chloroplasts.
Embedded within the thylakoid membranes are the proteins involved in light-induced
charge separation and electron transfer. The overall reaction consists of CO2 being
reduced to an aldehyde via the Calvin cycle, and water being oxidized to molecular
oxygen and H+.
Figure 1 shows a schematic diagram of the proteins involved in the electron
transfer reactions of photosynthesis.
There are two protein complexes directly
responsible for light-induced charge separation and electron transfer across the thylakoid
membrane, Photosystem II (PSII) and Photosystem I (PSI). PSII is responsible for the
reduction of quinone to quinol via two sequential electron transfer events, and the
oxidation of water to molecular oxygen and H+ via four sequential photon absorption
events. The cytochrome b6f complex oxidizes the quinol molecule produced by PSII to
quinone. The cytochrome b6f complex transfers electrons between PSII and Photosystem
I (PSI), and participates in the formation of the transmembrane electrochemical proton
gradient by transferring protons from the stromal side of the thylakoid membrane to the
1
luminal side (Figure 1). The hydrogen gradient produced by these reactions is used by
ATP synthase to produce ATP from ADP and inorganic phosphate. The cytochrome b6f
complex is also responsible for the reduction of plastocyanin, a monomeric, coppercontaining protein.
PSI is responsible for the light-induced reduction of ferredoxin, an iron-sulfur
protein, which reduces the enzyme NADP+ reductase.
NADP+ reductase is an
oxidoreductase, and functions to reduce NADP+ to NADPH. NADPH is the reducing
power for the biosynthetic reactions of the Calvin cycle in photosynthesis. Plastocyanin
reduces PSI, resetting the system.
1.1 Photosystem I
Photosystem I (PSI) is responsible for the reduction of ferredoxin and the
oxidation of plastocyanin or cytochrome c6. It is a membrane-bound protein consisting of
twelve subunits, three of which are responsible for binding the electron transfer cofactors.
Figure 2 depicts the polypeptide composition of PSI from the 2.5 Å resolution crystal
structure (Jordan et al. 2001). The crystal structure reveals that subunits PsaA and PsaB
form a heterodimer spanning the membrane and bind chlorophyll (chl), phylloquinone,
and an iron-sulfur cluster as shown in Figure 3A. The subunit PsaC, located on the
stromal side of the thylakoid membrane, is responsible for binding the terminal electron
acceptors, two iron-sulfur clusters as shown in Figure 3B (Jordan et al. 2001).
Charge separation is believed to begin on an accessory chl monomer, which
reduces a second chl monomer designated A0 (Holzwarth et al. 2006). A0 reduces a
phylloquinone known as A1. From here, the iron sulfur cluster known as FX accepts the
2
electron from A1, and reduces FA, a second iron-sulfur cluster. FA reduces FB, a third
iron-sulfur cluster, and the terminal electron acceptor of PSI.
The primary donor is reduced by P700, a chl heterodimer comprised of a chl a
molecule and a chl a’ molecule. P700 is the terminal electron donor of PSI and is reduced
by the soluble electron carrier plastocyanin, or cytochrome c depending on copper
availability. The electron transfer scheme in PSI starting with the radical species P700+
A0- is shown in Figure 4. The arrangement of electron cofactors in PSI is depicted in
Figure 5. The accessory chl along with the electron acceptors A0 and A1 are arranged
with apparent C2 structural symmetry around P700. The arrangement of the cofactors
results in two branches; the A branch representing the cofactors bound by the PsaA
subunit, and the B branch representing the cofactors bound by the PsaB subunit. Recent
studies have suggested both branches support light-induced electron transfer (GuergovaKuras et al. 2001; Hastings et al. 2001b; Muhiuddin et al. 2001; Rigby et al. 2002;
Fairclough et al. 2003; Ramesh et al. 2004; Santabarbara et al. 2005a; Santabarbara et al.
2005b; Li et al. 2006; Santabarbara et al. 2006).
1.2 Primary Donor
As stated previously, recent studies have revealed an accessory chl monomer may
be the primary donor in PSI (Holzwarth et al. 2006). These studies utilized ultrafast
transient absorption spectroscopy in the femtosecond and nanosecond time range to
characterize the electron transfer reactions. These studies reported charge separation
occurring in the picosecond time scale, with the first radical pair being accessory chl+/A0. P700 reduces the accessory chl cation during a second electron transfer event on the
3
picosecond time scale. The ultrafast transient absorption studies indicate that electron
transfer in PSI is similar to PSII, with the initial charge separated state originating on an
accessory chl and the positive charge being transferred to the chl dimer (Diner et al.
2002; Groot et al. 2005; Holzwarth et al. 2006).
1.3 Primary Electron Acceptor: A0
The first electron acceptor in PSI is A0, a chl monomer located between the
accessory chl and the phylloquinone, A1.
From the crystal structure, A0 is most likely
the chlorophyll designated eC2 or eC3 (Figure 5) (Fromme et al. 2001; Jordan et al.
2001; Grotjohann et al. 2005). Measurements of the first electron transfer step by optical
spectroscopy is difficult; due to the common time scale of excitation energy transfer from
antenna chl to the primary donor and electron transfer from the primary electron donor to
acceptor. Furthermore, both processes give rise to absorbance changes in the same
spectral region (Brettel, K. et al. 2001). However, spectroscopic studies have suggested
A0 is reduced in 7-10 ps ((Brettel, Klaus 1997; Brettel, K. et al. 2001; Ramesh et al.
2007) and references therein). The time constant of electron transfer from A0- to the
phylloquinone electron acceptor A1 has been estimated from kinetic absorption
spectroscopy by monitoring the reoxidation of A0-, and has been reported to be 30 ps
(reviewed in (Brettel, K. et al. 2001)). Other spectroscopic techniques used to determine
the time constant for the oxidation of A0- include photo-voltage measurements, and
monitoring the reduction of A1 by ultra-fast spectroscopy in the near-UV (Hecks et al.
1994; Brettel, K. et al. 1999). All the mentioned spectroscopic studies have assigned the
time constant for oxidation of A0- to be in the range of 21-50 ps.
4
1.4 Secondary Electron Acceptor: A1
The secondary electron acceptor of PSI is a phylloquinone molecule (vitamin K)
designated A1. It is reduced by A0, a chl monomer, and is oxidized by the iron-sulfur
[4Fe4S] cluster FX. There are two A1 molecules per PSI reaction center designated A1A
and A1B (Figure 5). Previous studies have revealed biphasic oxidation kinetics of A1,
with a fast decaying component of 20 ns and a longer decaying component of 150 ns
(Setif et al. 1993; Brettel, K. et al. 1999). These two phases have been proposed to
represent electron transfer from A1B and A1A to FX respectively. Both molecules have
been suggested to be active during electron transfer reactions. However, quantitative
treatments reveal slightly different kinetics in eukaryotes and prokaryotes (GuergovaKuras et al. 2001; Hastings et al. 2001b; Muhiuddin et al. 2001; Purton et al. 2001; Rigby
et al. 2002; Fairclough et al. 2003; Xu et al. 2003a; Xu et al. 2003b; Cohen et al. 2004;
Ramesh et al. 2004; Dashdorj et al. 2005; Santabarbara et al. 2005a; Santabarbara et al.
2005b; Ali et al. 2006; Li et al. 2006; Santabarbara et al. 2006). A1 has the lowest known
redox potential of any quinone found in nature. Substitution of the A1 phylloquinone
with other quinones such as ubiquinone give a reduction in midpoint potential of up to
300 mV (Iwaki et al. 1994; Itoh et al. 2001). The redox potential measured for
phylloquinone in DMF is -581 mV (Prince et al. 1983). These results suggest that protein
interactions at the A1 site lower the midpoint potential of the quinone bound. The protein
binding pocket consists of a conserved tryptophan within π-π stacking distance, with the
ring planes being almost parallel, and an asymmetric hydrogen bond to the C4O carbonyl
group from the amide backbone of Leu (Jordan et al. 2001; Grotjohann et al. 2005).
Recent studies have also revealed an aspartate residue which may interact
5
electrostatically with A1 (Feldman et al. 2007).
Figure 6 depicts the noncovalent
interactions of the A1 binding site in PSI.
The protein environment of A1 in PSI has been compared to the QA binding site in
the bacterial reaction center. In this system, a tryptophan residue is within π-π stacking
distance to the quinone. The bacterial reaction center also contains a hydrogen bond to
the C4O group of QA, however a second hydrogen bond is observed at the C1O position
as well. One method applied to the study of the secondary electron acceptor in PSI and
the bacterial reaction center is EPR spectroscopy.
This technique has determined
hyperfine coupling (hfc) tensors of the protons of the two quinones. Drastically different
hfc tensors have been measured for the two molecules, illustrating the difference in
protein interactions during reduction, and protein dynamics during electron transfer in
PSI and the bacterial reaction center (Van der Est et al. 1997).
1.5 Terminal Electron Donor: P700
P700, the terminal electron donor of PSI, is comprised of two chl molecules
designated PB and PA. The difference between the chl molecules comprising P700 arises
from the stereochemistry at the 132 carbon position. An examination of the protein
environment of the two halves of the dimer reveals an asymmetric hydrogen bonding
pattern. PA’s keto group is known to be involved in a hydrogen bonding network with a
conserved threonine (Thr 743). A water molecule and tyrosine (Tyr 603) complete the
network, which also involves the ester group of PA (Fromme et al. 2001; Jordan et al.
2001; Pantelidou et al. 2004). The corresponding protein environment around PB lacks
hydrogen bond donors to the carbonyl or ester group of PB. Comparing the midpoint
6
potential of P700 to chl in solution reveals a 400mV decrease in redox potential of the
terminal electron donor (Wasielewski et al. 1981). The hydrogen bonding network of PA
has been implicated to modulate the redox properties.
EPR studies of P700+ have revealed the spin density is mainly localized on the PB
half of P700, and that PA carries little to no charge density (Kaess et al. 1995; Kaess et al.
2001). The lack of spin density delocalization of the dimer might be the result of small
coupling between the two chl molecules, or because the energetic difference between the
two chls is rather large. In contrast, FTIR studies have concluded the spin density is
equally shared between PB and PA (Breton et al. 1999; Breton et al. 2002).
Reduction of P700+ to P700 is catalyzed by plastocyanin or cytochrome c in
cyanobacteria. Unlike eukaryotes, which form a transient complex with plastocyanin, in
cyanobacteria, PSI interacts with plastocyanin or cytochrome c by following a collisional
reaction mechanism, with no formation of any kinetically detectable transient complex in
vitro (Bottin et al. 1985; Drepper et al. 1996; Hervas et al. 1996; Hippler et al. 1996).
However, recent in vivo studies of whole cells have indicated a transient complex of PSI
with cytochrome c. The interaction between plastocyanin and PSI is believed to proceed
through long-range electrostatic interactions and hydrophobic contacts during random
collisions. In both plastocyanin and cytochrome c, there is a large flat hydrophobic
surface around the redox centers. Mutations to residues residing in this hydrophobic
patch have revealed this area is essential for electron transfer to PSI (Sigfridsson et al.
1997; Young et al. 1997; Sigfridsson et al. 1998; Xue et al. 1998).
7
The subunits PsaA and PsaB contain extra-membrane luminal loops that have
been reported to be important for interactions with plastocyanin (Navarro et al. 2000;
Fromme et al. 2003). In eukaryotes, the subunit PsaF has an amphipatic helix consisting
of basic residues facing the binding site of plastocyanin (Hippler et al. 1998; Sommer et
al. 2006). In contrast to eukaryotic organisms, the cyanobacterium Synechocystis sp PCC
6803 does not require the PsaF subunit for efficient electron transfer from plastocyanin or
cytochrome c to PSI (Van Der Est et al. 2004).
Therefore, PsaA and PsaB are solely
responsible for interacting with plastocyanin and cytochrome c in cyanobacteria.
Mutation studies of the luminal J loop of the PsaB protein from cyanobacteria
demonstrated the importance of the loop in interacting with soluble electron donor
proteins (Fromme et al. 2003).
Two conserved tryptophan residues have been reported to be important in electron
transfer from plastocyanin to P700+. Mutation studies in Chlamydomonas reinhardtii
show both tryptophan molecules are crucial for high affinity binding of plastocyanin and
cytochrome c to PSI (Sommer et al. 2004). These two Trp residues are within π-π
stacking distance and are located in the proposed binding pocket of the soluble electron
carriers to PSI (Fromme et al. 2003).
1.6 Iron Sulfur Clusters: FX, FA and FB
PSI contains three iron sulfur clusters: FX, FA and FB. Before the crystal structure
was resolved to 2.5 Å, and the arrangement of cofactors was resolved, it was difficult to
distinguish spectroscopically among the three iron sulfur clusters (Brettel, K. et al. 2001;
Vassiliev et al. 2001). FX is bound by the subunits PsaA and PsaB, (Figure 3A) while FA
8
and FB are bound to the PsaC subunit located on the stromal side of the membrane
(Figure 3B). FX is reduced by the phylloquinone molecule A1-, and is oxidized by the
iron-sulfur cluster FA. The terminal electron acceptor of PSI FB, which mediates electron
transfer to ferredoxin, by oxidizing FA-.
The reduction potential for FX measured by reductive titration of PSI by low
temperature EPR spectroscopy was found to be -705 ± 15 mV ((Vassiliev et al. 2001) and
references therein).
Higher values of -670 mV using room temperature optical
spectroscopy were reported in PSI samples that did not contain the iron-sulfur clusters FA
and FB (Parrett et al. 1989). The protein interactions that make FX one of the most
electronegative iron-sulfur clusters is not well understood.
Electron transfer from FX- to FA occurs in about 200 ns ((Brettel, K. et al. 2001;
Vassiliev et al. 2001) and references therein). The midpoint potential of FA, measured by
low temperature EPR studies, was found to be -540 mV and -590 mV for FB. However, a
different technique, employing optical detection of charge recombination between the FA,
FB and P700, measured a midpoint potential of -440 mV for FB and -465 mV for FA
((Vassiliev et al. 2001) and references therein). The range of redox potentials comes
from the different techniques used in the studies. Both techniques required the prereduction of FA to measure the redox potential of FB. Furthermore, neither study used
PSI samples containing ferredoxin, which may influence the redox potential of FB.
Binding of ferredoxin may decrease the reduction potential of FB. This mechanism may
avoid reducing oxygen, which is potentially harmful to PSI (Setif 2001).
9
1.7 Interaction of PSI with Ferredoxin and Flavodoxin
Electrons from PSI are transferred to the iron sulfur protein ferredoxin in the
microsecond time regime ((Setif 2001) and references therein).
Flavodoxin may
substitute for ferredoxin in cyanobacteria and some algae under iron limiting conditions.
Ferredoxin reduces ferredoxin-NADP+ reductase, which acts in reducing NADP+ to form
NADPH.
The subunits PsaC, PsaD, and PsaE, located on the stromal side of the
membrane, participate in the docking of ferredoxin/flavodoxin. A patch of basic residues
on the PsaD, PsaA and PsaE subunits has been suggested to be the docking site for
ferredoxin (Fromme et al. 2001; Grotjohann et al. 2005). Mutagenesis studies along with
chemical crosslinking experiments have shown direct evidence for the interaction of
ferredoxin to this docking site (Zanetti et al. 1987; Zilber et al. 1988; Wynn et al. 1989;
Andersen et al. 1992; Lelong et al. 1994; Fischer et al. 1998). Upon binding to PSI, it has
been postulated that ferredoxin modulates the redox potential of FB for efficient electron
transfer between the proteins.
1.8 Fourier-Transform Infrared Spectroscopy
Vibrational spectroscopy such as Fourier transform-infrared (FT-IR) spectroscopy
is a highly sensitive technique used to obtain information about enzymatic reactions,
dynamics, and structure. For membrane-bound proteins, such as PSI, it is one of the few
techniques sensitive enough to identify changes accompanied with charge separation. An
FT-IR spectrum reflects important information concerning amino acid side chains such as
charge, hydrogen bonding, conformation, and the protein environment of cofactors. For a
diatomic molecule, the vibrational frequency is proportional to the reduced masses of the
10
atoms that are involved in the chemical bond, as well as the bond strength. Assignment
of the stretching frequencies of the molecule can be approximated with the application of
Hooke’s law in which the bond connecting the two atoms together is regarded as a spring.
The system is treated as a simple harmonic oscillator (Wilson et al. 1980).
However, larger molecules consisting of more than two atoms have 3N-6
vibrations, where N is the number of atoms comprising the molecule (Wilson et al. 1980).
Normal mode of vibrations of molecules consists of each atom of the molecule oscillating
about its equilibrium position in space by simple harmonic motion (outlined above).
When the frequency and phase of the motion of the atoms are the same, each atom will
reach its maximum displacement position at the same time. Furthermore, each atom will
pass through its equilibrium at the same time. The position and momentum of the atoms
can be represented by vectors and theoretical calculations of the vibrational frequencies
of molecule can be calculated. Density functional theory (DFT) is a quantum mechanical
theory used to describe the electronic structure of a molecule. DFT calculations describe
normal mode vibrations of a molecule and are able to predict frequencies, amplitudes,
and isotope shifts of the molecule. In DFT calculations, the electronic state of the
molecule can be described as a wavefunction, satisfying the many-electron Shrödinger
equation. Gaussian basis sets are functions used as atomic orbitals for the computation of
electron orbitals in molecules.
The Gaussian basis DFT method breaks down the
Hamiltonian described in the wavefunction into basic one-electron and two electron
components, which can be solved by computational chemistry (Lewer 2003).
Vibrational transitions are narrow in line-shape; however in non-homogeneous
mixtures, such as proteins, the lineshape will broaden due to the multiple vibrations
11
contributing to the absorption spectrum.
Difference FT-IR spectroscopy is technique
that simplifies a rather complex spectrum.
To obtain a difference spectrum, an
absorption spectrum is obtained before and after the reaction has occurred. These spectra
are subtracting from one another, the resulting spectrum consists of frequencies that were
perturbed due to the reaction.
Those vibrational frequencies not affected will be
subtracted from the difference spectrum. Light-induced difference spectroscopy is a
technique able to monitor the oxidized and reduced states of PSI electron-transfer
cofactors. Detection of changes in the electronic structure is possible because of the high
signal to noise and internal frequency calibration of FT-IR spectrophotometers (Kim et
al. 1998; Breton et al. 1999)
Although the FT-IR spectrum is characteristic to an entire molecule, certain
groups of atoms give rise to characteristic frequencies regardless of the structure of the
rest of the molecule. Therefore, these characteristic bands can be used as molecular
probes of the system.
One application that has been widely used in the study of
photosynthetic proteins is isotope incorporation into amino acids or cofactors involved in
electron transfer reactions. IR frequencies are dependent on the reduced masses of the
atoms involved; therefore FT-IR spectroscopy is a technique that is sensitive to isotope
incorporation. Previous studies have incorporated various isotopes into the electron
donor and acceptor molecules of PSI (Kim et al. 2000; Hastings et al. 2001a; Hastings et
al. 2001b). Isotope-edited spectra have been invaluable in the study of electron transfer in
various photosystems and bacterial reaction centers (Rothschild et al. 1981; Rothschild et
al. 1982; MacDonald et al. 1993; Breton et al. 1994; Breton 1997).
12
1.9 Thesis Outline
To gain greater knowledge about protein dynamics and electron cofactor interactions,
isotope labeling of amino acid side chains, electron acceptors, and electron donors of PSI
was performed to generate isotope-edited spectra. The spectra presented here will have
contributions for only those groups perturbed by the isotope incorporation, and lightinduced electron transfer. The results presented here have shown profound effects on
protein environment of the vibration spectra of the electron cofactors of PSI, and have
given novel insight into protein dynamic effects on electron transfer.
In this thesis, light-induced Fourier-transform infrared (FT-IR) spectroscopy is
utilized to probe the spectroscopic properties of the secondary electron acceptor, A1, the
terminal electron donor, P700, and protein dynamics during light-induced electron
transfer. This work utilizes isotope incorporation into particular functional groups of the
cofactors and amino acids.
Chapter 2 reviews the biosynthetic pathway of chl,
phylloquinone and the aromatic amino acids. The particular steps in the biosynthetic
pathways that were utilized to incorporate the isotopes are also discussed. In Chapter 3,
the isotope-edited vibrational spectrum of A1--minus-A1 is presented. Bands arising from
the reduced and oxidized forms of A1 are assigned, and the effects of protein interactions
with A1 are discussed.
Chapter 4 described the spectroscopic studies performed on the terminal electron
donor P700. Assignments for the reduced and oxidized forms of P700 are discussed, as
well as protein dynamics associated with oxidation of P700. These studies are very
important because they suggest novel protein dynamics that may be associated with
13
charge transfer from plastocyanin to PSI. An examination of the protein dynamics
associated with the PsaA and PsaB subunits is given in Chapter 5. Isotope-edited spectra
of amino acid side chains perturbed doing electron transfer are analyzed and a discussion
of the importance of the protein dynamics is presented.
14
1.10 References
Ali, K., S. Santabarbara, P. Heathcote, M. C. W. Evans and S. Purton (2006).
"Bidirectional electron transfer in photosystem I: replacement of the symmetry-breaking
tryptophan close to the PsaB-bound phylloquinone A1B with a glycine residue alters the
redox properties of A1B and blocks forward electron transfer at cryogenic temperatures."
Biochim Biophys Acta 1757: 1623-1633.
Andersen, B., B. Koch and H. V. Scheller (1992). "Structural and functional analysis of
the reducing side of Photosystem I." Plant Cell Physiology 84: 154-161.
Bottin, H. and P. Mathis (1985). "Interaction of plastocyanin with the photosystem I
reaction center: a kinetic study by flash absorption spectroscopy " Biochemistry 24: 64536460.
Breton, J. (1997). "Efficient exchange of the primary quinone acceptor QA in isolated
reaction centers of Rhodopsuedomonas viridis." Proc. Natl. Acad. Sci. USA 94: 1131811323.
Breton, J., C. Boullais, J. R. Burie, E. Nabedryk and C. Mioskowski (1994). "Binding
sites of quinones in photosynthetic bacterial reaction centers investigated by lightinduced FTIR difference spectroscopy: assignment of the interactions of each carbonyl of
QA in Rhodobacter sphaeroides using site-specific 13C-labeled ubiquinone." Biochemistry
33: 14378-14386.
Breton, J., E. Nabedryk and W. Leibl (1999). "FTIR study of the primary electron donor
of photosystem I (P700) revealing delocalization of the charge in P700+ and localization of
the triplet character in 3P700." Biochemistry 38: 11585-11592.
Breton, J., W. Xu, B. A. Diner and P. R. Chitnis (2002). "The two histidine axial ligands
of the primary electron donor chlorophylls (P700) in photosystem I are similarly perturbed
upon P700+ formation." Biochemistry 41: 1200-1210.
Brettel, K. (1997). "Electron transfer and arrangement of the redox cofactors in
Photosystem I." Biochim. Biophys. Acta 1318: 322-373.
Brettel, K. and W. Leibl (2001). "Electron transfer in photosystem I." Biochim. Biophys.
Acta 1507: 100-114.
Brettel, K. and M. H. Vos (1999). "Spectroscopic resolution of the picosecond reduction
kinetics of the secondary electron acceptor A1 in photosystem I." FEBS Letters 447: 315317.
Cohen, R. O., G. Shen, J. H. Golbeck, W. Xu, P. R. Chitnis, A. I. Valieva, A. van der Est,
Y. Pushkar and D. Stehlik (2004). "Evidence for asymmetric electron transfer in
15
cyanobacterial photosystem I: analysis of a methionine-to-luecine mutation of the ligand
to the primary electron acceptor A0." Biochemistry 43: 4741-4754.
Dashdorj, N., W. Xu, R. O. Cohen, J. H. Golbeck and S. Savikhin (2005). "Asymmetric
electron transfer in cyanobacterial Photosystem I: charge separation and secondary
electron transfer dynamics of mutations near the primary electron acceptor A0." Biophys.
J. 88: 1238-1249.
Diner, B. A. and F. Rappaport (2002). "Structure, dynamics, and energetics of the
primary photochemistry of Photosystem II of oxygenic photosynthesis." Annual Review
of Plant Biology 23.
Drepper, F., M. Hippler, W. Nitschke and W. Haehnel (1996). "Binding dynamics and
electron transfer between plastocyanin and photosystem I." Biochemistry 35: 1282-1295.
Fairclough, W. V., A. Forsyth, M. C. W. Evans, M. C. Evans, S. E. J. Rigby, S. Purton
and P. Heathcote (2003). "Bidirection electron transfer in photosystem I: electron transfer
on the PsaA side is not essential for phototrophic growth in Chlamydomonas." Biochim.
Biophys. Acta 1606: 43-55.
Feldman, K. S., D. K. Hester 2nd and J. H. Golbeck (2007). "A relationship between
amide hydrogen bond strength and quinone reduction potential: Implications for
photosystem I and bacterial reaction center quinone function." Bioorg. Med. Chem. Lett.
17: 4891-4894.
Fischer, N., M. Hippler, P. Setif, J.-P. Jacqout and J. D. Rochaix (1998). "The PsaC
subunit of Photosystem I provides an essential lysine residue for fast electron transfer to
ferredoxin." EMBO Journal 17: 849-858.
Fromme, P., P. Jordan and N. Krauss (2001). "Structure of photosystem I." Biochim
Biophys Acta 1507: 5-31.
Fromme, P., A. Melkozernv, P. Jordan and N. Krauss (2003). "Structure and function in
Photosystem I: interaction with its soluble electron carriers and external antenna
systems." FEBS Letters 555: 40-44.
Groot, M. L., N. P. Pawlowicz, L. J. van Wildren, J. Breton, I. H. van Stokkum and R.
van Grondelle (2005). "Initial electron donor and acceptor in isolated Photosystem II
reaction centers identified with femtosecond mid-IR spectroscopy." Proc. Natl. Acad. Sci.
USA 102: 13087-13092.
Grotjohann, I. and P. Fromme (2005). "Structure of cyanobacterial photosystem I."
Photosynth. Res. 85: 51-72.
Guergova-Kuras, M., B. Boudreaux, A. Joliot, P. Joliot and K. Redding (2001).
"Evidence for two active branches for electron transfer in photosystem I." Proc. Natl.
Acad. Sci. USA 98: 4437-4442.
16
Hastings, G., V. M. Ramesh, R. Wang, V. Sivakumar and A. Webber (2001a). "Primary
donor photo-oxidation in photosystem I: a re-evaluation of (P700+ - P700) Fourier
transform infrared difference spectra." Biochemistry 40: 12943-12949.
Hastings, G. and V. Sivakumar (2001b). "A Fourier transform infrared absorption
difference spectrum associated with the reduction of A1 in Photosystem I: are both
phylloquinones involved in electron transfer?" Biochemistry 40: 3681-3689.
Hecks, B., K. Wulf, J. Breton, W. Leibl and H.-W. Trissl (1994). "Primary charge
separation in Photosystem I: a two-step electrogenic charge separation connected with
P700+A0- and P700+A1- formation." Biochemistry 447: 8619-8624.
Hervas, M., J. A. Navarro, A. Diaz and M. A. De la Rosa (1996). "A comparative
thermodynamic analysis by laser-flash absorption spectroscopy of Photosystem I
reduction by plastocyanin and cytochrome c6 in Anabaena PCC 7119, Synechocystis PCC
6803, and spinach." Biochemistry 35: 2693-2698.
Hippler, M., F. Drepper, W. Haehnel and J. D. Rochaix (1998). "The N-terminal domain
of PsaF: Precise recognition site for binding and fast electron transfer from cytochrome
c(6) and plastocyanin to photosystem I of Chlamydomonas reinhardtii." Proc. Natl.
Acad. Sci. USA 95: 7339-7344.
Hippler, M., J. Reichert, M. Sutter, E. Zak, L. Altschmied, U. Schroer, R. G. Herrmann
and W. Haehnel (1996). "The plastocyanin binding domain of Photosystem I." EMBO
Journal 15: 6374-6384.
Holzwarth, A. R., M. G. Muller, J. Niklas and W. Lubitz (2006). " Ultrafast transient
absorption studies on photosystem I reaction centers from Chlamydomonas reinhardtii. 2:
Mutations near the P700 reaction center chlorophylls provide new insight into the nature
of the primary electron donor." Biophys. J. 90: 552-565.
Itoh, S., M. Iwaki and I. Ikegami (2001). "Modification of photosystem I reaction center
by the extraction and exchange of chlorophylls and quinones." Biochim. Biophys. Acta
1507: 115-138.
Iwaki, M. and S. Itoh (1994). "Reaction of reconstituted acceptor quinone and dynamic
equilibration of electron transfer in the Photosystem I reaction center." Plant Cell
Physiol. 35: 983-993.
Jordan, P., P. Fromme, H. T. Witt, O. Klukas, W. Saenger and N. Krauss (2001). "Threedimensional structure of cyanobacterial photosystem I at 2.5A resolution." Nature 411:
909-917.
Kaess, H., E. Biitersmann-Weidlich, L.-E. Andreasson, B. Boenigk and W. Lubitz
(1995). "ENDOR and ESEEM of the 15N labeled radical cations of chlorophyll and the
primary donor P700 in photosystem I." Chem. Phys. 194: 419-432.
17
Kaess, H., P. Fromme, H. T. Witt and W. Lubitz (2001). "Orientation and electronic
structure of the primary donor radical cation P700+ in photosystem I: a single crystal EPR
and ENDOR study." J. Phys. Chem. B 105: 1225-1239.
Kim, S. and B. A. Barry (1998). "The protein environment surrounding tyrosyl radicals
D⋅ and Z⋅ in photosystem II: a difference Fourier-transform infrared spectroscopic study."
Biophys. J. 74: 2588-2600.
Kim, S. and B. A. Barry (2000). "Identification of carbonyl modes of P700 and P700+ by in
situ chlorophyll labeling in photosystem I." J. Am. Chem. Soc. 122: 4980-4981.
Lelong, C., P. Setif, B. Lagoutte and H. Bottin (1994). "Identification of the amino acids
involved in the functional interaction between photosystem I and ferredoxin from
Synechocystis sp. PCC 6803 by chemical crosslinking." J. Biol. Chem. 269: 1003410039.
Lewer, E. (2003) Computational Chemistry:Introduction to the Theory and Applications
of Molecular and Quantum Mechanics. Norwell, Kluwer Academic Publisher.
Li, Y., A. van der Est, M. G. Lucas, V. M. Ramesh, F. Gu, A. Petrenko, S. Lin, A. N.
Webber, F. Rappaport and K. Redding (2006). "Directing electron transfer within
Photosystem I by breaking H-bonds in the cofactor branches." Proc. Natl. Acad. Sci. USA
103: 2144-2149.
MacDonald, G. M., K. A. Bixby and B. A. Barry (1993). "A difference Fourier-transform
infrared study of two redox-active tyrosine residues in photosystem II." Proc. Natl. Acad.
Sci. USA 90: 11024-11028.
Muhiuddin, I. P., P. Heathcote, S. Carter, S. Purton, S. E. J. Rigby and M. C. W. Evans
(2001). "Evidence from time resolved studies of the P700+/A1- radical pair for
photosynthetic electron transfer on both the PsaA and PsaB branches of the photosystem
I reaction centre." 503: 56-60.
Navarro, J. A., M. Hervas, J. Sun, B. De la Cerda, P. R. Chitnis and M. A. De la Rosa
(2000). "Negatively charged residues in the H loop of PsaB subunit in Photosystem I
from Synechocystis sp PCC 6803 appear to be responsible for electrostatic repulsions
with plastocyanin." Photosynth. Res. 65: 63-68.
Pantelidou, M., P. R. Chitnis and J. Breton (2004). "FTIR spectroscopy of Synechocystis
6803 mutants effected on the hydrogen bonds to the carbonyl groups of the PsaA
chlorophyll of P700 supports an extensive delocalization of the charge in P700+."
Biochemistry 43: 8380-8390.
Parrett, K. G., T. Mehari, P. G. Warren and J. H. Golbeck (1989). "Purification and
properties of the intact P700 and FX containing Photosystem I core protein." Biochim
Biophys Acta 973: 324-332.
18
Prince, R.C., P. Dutton, L. Bruce, J. Malcolm (1983). "Electrochemistry of ubiquinones.
Menoquinones and plastoquinones in aprotic solvents." FEBS Letters 160: 273-276.
Purton, S., D. R. Stevens, I. P. Muhiuddin, M. C. W. Evans, S. Carter, S. E. J. Rigby and
P. Heathcote (2001). "Site-directed mutagenesis of PsaA residue W693 affects
phylloquinone binding and function in the photosystem I reaction center of
Chlamydomonas reinhardtii." Biochemistry 40: 2167-2175.
Ramesh, V. M., K. Gibasiewicz, S. Lin, S. E. Bingham and A. N. Webber (2004).
"Bidirectional electron transfer in photosystem I: accumulation of A0- in A-side or B-side
mutants of the axial ligand to chlorophyll A0." Biochemistry 43: 1369-1375.
Ramesh, V. M., K. Gibasiewicz, S. Lin, S. E. Bingham and A. N. Webber (2007).
"Replacement of the methionine axial ligand to the primary electron acceptor A0 slows
the A0- reoxidation dynamics in photosystem I." Biochim Biophys Acta 1767: 151-160.
Rigby, S. E. J., I. P. Muhiuddin, M. C. W. Evans, S. Purton and P. Heathcote (2002).
"Photoaccumulation of the PsaB phyllosemiquinone in photosystem I of Chlamydomonas
reinhardtii." Biochim. Biophys. Acta 1556: 13-20.
Rothschild, K. J. and H. Marrero (1982). "Infrared evidence that the Schiff base of
bacteriorhodopsin in protonated: bR570 and K intermediates." Proc. Natl. Acad. Sci. USA
79: 4045-4049.
Rothschild, K. J., M. Zagaeski and W. A. Cantore (1981). "Conformational changes of
bacteriorhodopsin detected by Fourier transform infrared difference spectroscopy."
Biochem. Biophys. Res. Commun. 103.
Santabarbara, S., P. Heathcote and M. C. Evans (2005a). "Modeling of the electron
transfer reactions in Photosystem I by electron tunneling theory: the phylloquinones
bound to the PsaA and the PsaB reaction centre subunits of PS I are almost isoenergetic
to the iron-sulfur cluster FX." Biochim Biophys Acta 1708: 283-310.
Santabarbara, S., I. Kuprov, W. V. Fairclough, S. Purton, P. J. Hore, P. Heathcote and M.
C. W. Evans (2005b). "Bidirectional electron transfer in photosystem I: determination of
two distances between P700+ and A1- in spin-correlated radical pairs." Biochemistry 44:
2119-2128.
Santabarbara, S., I. Kuprov, P. J. Hore, A. Casal, P. Heathcote and M. C. Evans (2006).
"Analysis of the spin-polarized electron spin echo of the P700+ A1- radical pair of
photosystem I indicates that both reaction center subunits are competent in electron
transfer in cyanobacteria, green algae, and higher plants." Biochemistry 45: 7389-7403.
Setif, P. (2001). "Ferredoxin and flavodoxin reduction in Photosystem I." Biochimica
Biophysica Acta 1507: 161-179.
Setif, P. and K. Brettel (1993). "Forward electron transfer from phylloquinone A1 to ironsulfur centers in spinach photosystem I." Biochemistry 32.
19
Sigfridsson, K., S. Young and O. Hansson (1997). "Electron transfer between spinach
plastocyanin mutants and photosystem 1." Eur. J. Biochem. 245: 805-812.
Sigfridsson, K., S. Young and O. Hansson (1998). "Electron transfer between spinach
plastocyanin mutants and photosystem 1 (vol 245, pg 805, 1998)." Eur. J. Biochem. 255:
324-324.
Sommer, F., F. Drepper, W. Haehnel and M. Hippler (2004). "The hydrophobic
recognition site formed by residues PsaA-Trp651 and PsaB-Trp627 of photosystem I in
Chlamydomonas reinhardtii confers distinct selectivity for binding of plastocyanin and
cytochrome c6." J. Biol. Chem. 279: 20009-20017.
Sommer, F., F. Drepper, W. Haehnel and M. Hippler (2006). "Identification of precise
electrostatic recognition sites between cytochrome c(6) and the photosystem I subunit
PsaF using mass spectrometry." J. Biol. Chem. 281: 35097-35103.
Van der Est, A., T. Prisner, R. Bittl, P. Fromme, W. Lubitz, K. Mobius and D. Stehlik
(1997). "Time-resolved X-, K-, and W-Band EPR of the radical pair State P700+A1- of
Photosystem I in comparison with P865+Q- in bacterial reaction centers." J. Phys. Chem. B
101: 1437-1443.
Van Der Est, A., A. I. Valieva, Y. E. Kandrashkin, G. Shen, D. A. Bryant and J. H.
Golbeck (2004). "Removal of PsaF alters forward electron transfer in photosystem I:
evidence for fast reoxidation of QK-A in subunit deletion mutants of Synechococcus sp.
PCC 7002." Biochemistry 43: 1264-1275.
Vassiliev, I. R., M. L. Antonkine and J. H. Golbeck (2001). "Iron-sulfur clusters in type I
reaction centers." Biochimica Biophysica Acta 1507: 139-160.
Wasielewski, M., J. Norris, L. Shipman, C. Lin and W. Svec (1981). "Monomeric
chlorophyll a enol: evidence for its possible role as the primary electron donor in
photosystem I of plant photosynthesis." Proc. Natl. Acad. Sci. USA 78: 2957-2961.
Wilson, E. B., J. C. Decium and P. C. Cross (1980). Molecular Vibrations: The theory of
infrared and Raman vibrational spectra. New York, Dover Publications Inc.
Wynn, R. M., J. Omaha and R. Malkin (1989). "Structural and functional properties of
the cyanobacterial Photosystem I complex." Biochemistry 28: 5554-5560.
Xu, W., P. Chitnis, A. Valieva, A. van der Est, Y. N. Pushkar, M. Krzystyniak, C.
Teutloff, S. G. Zech, R. Bittl, D. Stehlik, B. Zybailov, G. Shen and J. H. Golbeck
(2003a). "Electron transfer in cyanobacterial photosystem I: I. Physiological and
spectroscopic characterization of site-directed mutants in a putative electron transfer
pathway from A0 through A1 to Fx." J. Biol. Chem. 278: 27864-27875.
Xu, W., P. R. Chitnis, A. Valieva, A. van der Est, K. Brettel, M. Guergova-Kuras, Y. N.
Pushkar, S. G. Zech, D. Stehlik, G. Shen, B. Zybailov and J. H. Golbeck (2003b).
"Electron transfer in cyanobacterial photosystem I: II. Determination of forward electron
20
transfer rates of site-directed mutants in a putative electron transfer pathway from A0
through A1 to Fx." J. Biol. Chem. 278: 27876-27887.
Xue, Y. F., M. Okvist, O. Hansson and S. Young (1998). "Crystal structure of spinach
plastocyanin at 1.7 angstrom resolution." Protein Sci. 7: 2099-2105.
Young, S., K. Sigfridsson, K. Olesen and O. Hansson (1997). "The involvement of the
two acidic patches of spinach plastocyanin in the reaction with photosystem I." Biochim.
Biophys. Acta 1322: 106-114.
Zanetti, G. and G. Merati (1987). "Identification by chemical crosslinking of the
polypeptide which binds ferredoxin." Eur. J. Biochem. 169: 143-146.
Zilber, A. L. and R. Malkin (1988). "Ferredoxin cross-links to a 22 kD subunit of
Photosystem I." Plant Physiol. 88: 810-814.
21
Figure 1. The protein complexes embedded in the thylakoid membrane involved in the
light-induced electron transfer reactions in photosynthesis.
oxidation
of
water
is
not
implied.
Original
figure
Stoichiometry of the
can
be
found
at
http://en.wikipedia.org/wiki/Image:Thylakoid_membrane.png.
22
Figure 2. The arrangement of subunits in PSI: PsaA subunit, yellow; PsaB, blue; PsaC,
green; PsaD, red; PsaE, grey; PsaF, purple; PsaI, cyan; PsaJ, pink; PsaK, dark grey; PsaL,
light brown; and PsaM, dark brown. The figure is reproduced from the 2.5 Å resolution
crystal structure of PSI from S. elongatus (PDB file accession number 1JB0 (Jordan et al.
2001)) using Swiss-Pdb viewer (v3.7; www.expasy.ch/spdbc) and rendered using POVRay (v3.5; www.povray.org).
23
Figure 3. The arrangement of electron cofactors bound to PsaA, PsaB, or PsaC. (A). The
PsaA and PsaB subunits that form the heterodimeric core of PSI and the electron transfer
molecules bound to each subunit. PsaA is depicted in yellow and PsaB is depicted in
blue. The electron transfer cofactors from the luminal side toward the stromal side of the
membrane include P700, eCA2, eCB2, eCB3, eCA3, A1A, A1B and FX.
The subscripts
designate the subunit which binds the cofactor. FX is bound to both the PsaA and PsaB
subunits. (B). The terminal electron acceptors bound to the PsaC subunit located on the
24
Figure 3 continued
stromal side of PSI. FA and FB are depicted as spherical balls in yellow and the PsaC
subunit is colored green. The figure is reproduced from the 2.5 Å resolution crystal
structure of PSI from S. elongatus (PDB file accession number 1JB0 (Jordan et al. 2001))
using Swiss-Pdb viewer (v3.7; www.expasy.ch/spdbc) and rendered using POV-Ray
(v3.5; www.povray.org).
25
Figure 4.
Electron transfer scheme in PSI starting with the radical pair P700+ A0-
(Brettel, K. et al. 2001; Vassiliev et al. 2001 and references therein).
26
Figure 5. Arrangement of the electron cofactors of PSI. The figure is reproduced from
the 2.5 Å resolution crystal structure of PSI from S. elongatus (PDB file accession
number 1JB0 (Jordan et al. 2001)) using Swiss-Pdb viewer (v3.7; www.expasy.ch/spdbc)
and rendered using POV-Ray (v3.5; www.povray.org).
The nomenclature of the
accessory chl comes from ref (Grotjohann et al. 2005). The subscripts indicate the
subunit to which the cofactor is bound.
27
Figure 6. The prominent features of the A1 binding pocket. Both A1 molecules are in
hydrogen bonding distance to the backbone of a leucine molecule as well as π-stacking
distance to a tryptophan residue. The aspartate B575 residue has been implicated in the
modulation of the redox potential of A1.
28
CHAPTER 2
BIOSYNTHETIC PATHWAYS OF CHLOROPHYLL,
PHYLLOQUINONE AND AROMATIC AMINO ACIDS
The research presented in the following chapters applies isotope labeling
techniques to characterize protein dynamics during electron transfer in PSI.
These
labeling techniques utilize the biosynthetic pathways of chlorophyll, phylloquinone, and
the aromatic amino acids to specifically isotope label hydrogen, carbon or oxygen atoms.
In order to incorporate isotopes into specific positions of the molecules, the biosynthetic
pathways were manipulated at precise steps. Here I present a brief overview of the
biosynthetic pathways for chlorophyll, phylloquinone and the aromatic amino acids and
outline the enzymatic steps that were utilized to incorporate the desired isotope.
2.1 Biosynthetic Pathway of Chlorophyll
Chlorophyll contains a tetrapyrrole ring structure known as a prophyrin, common
to that found in the heme prosthetic group of hemoglobin and cytochrome. There are
several biosynthetic steps shared by chlorophyll and heme, however chlorophyll contains
a magnesium atom coordinated by four nitrogen atoms whereas heme binds an iron atom.
The biosynthetic pathway of chlorophyll has been elucidated through the use of various
biochemical techniques such as isotope labeling, enzyme biochemistry, and mutation
studies.((Beale 1999) and references therein).
29
In all organisms that synthesize tetrapyrroles the universal precursor is
aminolevulinic acid (ALA).
In animals, the enzyme known as aminolevulinic acid
synthase performs three enzymatic steps which convert glycine and succinyl-CoA to
ALA (Gibson et al. 1958; Kikuchi et al. 1958). In plants, green algae, and cyanobacteria,
there are three enzymes responsible for the production of ALA. The chl biosynthetic
pathway begins with glutamic acid, ligated to a tRNA molecule (Beale et al. 1974; Beale
et al. 1975; Kannangara et al. 1988). Glutamyl-tRNA synthatase (EC 6.1.1.17) activates
glutamate by ligating it to a tRNAglu molecule. Glutamyl tRNA reductase reduces the
activated carboxyl group of the tRNAglu to a formyl group; producing glutamate-1semialdehyde.
This step is followed by an amino-exchange reaction, to form δ-
aminolevulinic
acid
(ALA),
and
is
catalyzed
by
glutamate-1-semialdehyde
aminotransferase (EC 5.4.3.8) (Beale 1999). These three enzymatic steps are outlined in
Figure 1.
Two molecules of ALA are condensed to form a pyrrole molecule known as
porphobilinogen. This step in the chlorophyll biosynthetic pathway is catalyzed by the
enzyme ALA dehydratase (Schmidt et al. 1955; Calisson et al. 1966; Nandi et al. 1967).
Four
molecules
of
porphobilinogen
are
combined
in
the
synthesis
of
1-
hydroxymethylbilane by the enzyme porphobilinogen deaminase; with four molecules of
NH3 being released (Frydman et al. 1970). The next step in the synthesis pathway is the
production of the macrocycle ring. The macrocycle ring formation is carried out by the
enzyme referred to as uroporphyrinogen III synthase (EC 4.2.1.75) (Hart et al. 1985).
This enzyme has undergone intensive studies due to the proposed mechanism of
cyclization (Crockett et al. 1991; Beale 1999) and references therein). Uroporphyrinogen
30
III synthase catalyzes a ring inversion, which interchanges of the carbons that constitute
part of the tetrapyrrole ring From Figure 2, the acetate group and propionate group do
not simply exchange; instead the enzyme catalyzes a ring flip and subsequent
interchange. Uroporhyrinogen III decarboxylase decarboxylates the macrocycle ring,
resulting in four molecules of CO2 to be released. The formation of the macrocycle ring
is shown in Figure 2.
Three enzymes are involved in the conversion of uroporphyrinogen III to
protoporphyrin IX and are responsible for transforming the molecule from a hydrophilic
and photochemically unreactive molecule to a hydrophobic, photochemically active
molecule. These reactions are shown in Figure 3. The first two steps in the conversion
require oxygen to oxidize multiple single bonds to double bonds on the macrocycle ring,
resulting in the production of water and CO2.
These reactions are carried out by
coporphyrinogen III oxidase (EC 4.1.1.37) and protoporphyrinogen IX oxidase (EC
1.3.34) to produce protoporphyrin IX (Jacobs et al. 1979; Hart et al. 1985; Robinson et al.
1985). The reactions outlined above to form protoporphyrin IX are common to both
plants and animals. Although the synthesis of ALA differs, the reactions that convert
ALA to protoporphyrin are common to both groups of organisms. To form a heme
molecule, ferrochelatase (EC 4.99.1) inserts an iron atom into the center of
protoporphyrin IX. In plants and cyanobacteria, magnesium chelatase is responsible for
the addition of a magnesium atom into the center of protoporphyrin IX, as shown in
Figure 3 (Gorchein 1972; Walker et al. 1991).
Once the magnesium has been incorporated into the macrocycle ring, a methyl
group is added to the 132 carboxyl group to form an ester by the enzyme Mg31
protporphyrin IX methyltransferase (EC 2.1.1.11) (Figure 4) (Ellswort et al. 1972). This
enzyme uses the substrate S-adenosylmethionine as the methyl donor. Previous studies
have used this step to incorporate C2H3 at this carbon position (Kim et al. 2000). This
was accomplished by growing methionine tolerant Synechocystis supplemented with LC2H3-methionine. After the methyl group has been added to the macrocycle, the fifth
ring is formed.
Figure 4 depicts the last steps of the chl biosynthesis pathway. The fifth ring is
formed by an oxygen dependent cyclase; the mechanism of which is not well understood.
The cyclase catalyzes carbon bond formation as well as the oxidation of the 131 carbon to
a carbonyl, using O2 as a substrate (Bollivar et al. 1995, 1996). Further processing of the
macrocycle is accomplished by 8-vinyl reductase which reduces a double bond on the 82
carbon to a single bond to produce monovinyl protochlorophyllide a, and uses the
oxidation of NADPH to NADP+ to drive the reaction to completion. The reduction of
ring IV by protochlorophyllide oxidoreductase (EC 1.3.1.33) requires light energy and
NADPH to produce chlorophyllide a (Oliver et al. 1980). The final step in the formation
of chlorophyll a is the attachment of the phytol tail, which is accomplished with the use
of chlorophyll synthatase (EC 3.1.1.14) (Rudiger et al. 1980).
For the isotope labeling studies presented here, I was able to utilize the first three
enzymatic steps in the production of ALA, which utilizes glutamate as the sole carbon
source. Eight molecules of glutamate are used in the production of one molecule of chl.
In order to specifically label the 131 keto group of chl, a glutamate-tolerant strain of
Synechocystis PCC 6803 was cultured in glutamic -13C-3-acid. Figures 1-4 illustrate the
incorporation glutamic-13C-3- acid with the green carbons showing the 13C isotope.
32
A second set of studies were conducted to incorporate 18O into chl. As outlined
above, the chl synthesis pathway contains an oxygen-dependent cyclase that catalyzes the
formation of ring V and the oxidation of the 131C to a carbonyl carbon. Synechocystis
cultures were grown in the presence of 18O2, bubbled through the media. Figure 4 depicts
the cyclase reaction and the incorporation of molecular oxygen (colored blue) into chl.
2.2 Chorismate Biosynthesis
Synthesis of chorismate constitutes a common step in aromatic amino acids and
quinone biosynthetic pathways in plants and microorganisms (Herrmann 1995, 1995).
The seven enzymes involved in the production of chorismate have been studied
extensively in plants and bacteria (Bentley 1990; Pittard 1996).
The biosynthetic
pathway for chorismate production, also referred to as the shikimate pathay, is depicted
in Figure 5. The biosynthetic pathway begins with phosphoenolpyruvate, an intermediate
in glycolysis, and erythrose-4-phosphate derived from the pentose phosphate pathway. 3deoxy-D-arabino-heptulosonate-7-phosphate
(DAHP)
synthase
catalyzes
the
condensation reaction to produce DAHP. DAHP synthase has similar kinetic properties
in bacteria and plants, however the bacterial DAHP synthase is feedback inhibited by any
of the three aromatic amino acids, whereas plant DAHP synthase is activated by
tryptophan (Srinivasan et al. 1959; Smith 1962).
The second step in the biosynthetic pathway involves the formation of a six
carbon ring, catalyzed by 3-dehydroquinate synthase to produce 3-dehydroquinate. This
molecule is converted to 3-dehydroshikimate by a dehydration step catalyzed by 3dehydroquinnate dehydratase. A reduction reaction utilizing NADPH and shikimate
33
dehydrogenase reduces 3-dehydroshikimate to shikimate (Balinsky et al. 1971).
In
plants, both these steps are catalyzed on the bifunctional enzyme shikimate:NADP+
oxidoreductase (Bentley 1990).
Shikimate is phosphorlated by shikimate kinase to produce shikimate-3-phosphate
(Gollube 1966, 1966; Koshiba 1979).
enolpyruvylshikimate-3-phosphate
(EPSP)
Once this molecule is formed, 5synthase
catalyzes
the
transfer
of
enolpyruvate (PEP) to shikimate-3-phosphate to produce EPSP (Bondinell et al. 1971).
This enzyme is the most studied of all the enzymes involved in chorismate biosynthesis
and has been the target of many herbicides including glyphosphate (Roundup ™) and for
novel antimicrobials (Marques et al. 2007). The final step of the pathway, the conversion
of EPSP to chorismate, involves a 1,4-trans elimination of phosphate from EPSP. This
reaction is catalyzed by chorismate synthase to produce chorismate (Gaertner et al. 1973).
The reaction is considered unusual due to its absolute requirement for a reduced flavin
mononucleotide, which is not consumed during the reaction (Macheroux et al. 1999).
2.3 Phylloquinone Biosynthesis
The biosynthetic pathway of the phylloquinone A1 which functions as a secondary
electron acceptor in PSI has not been fully investigated. The bacterial menoquinone
biosynthesis pathway has been elucidated, and it believed that formation of the
napthoquinone ring follows the same enzymatic steps for phylloquinone.
The best
studied system for menoquinone biosynthesis is E. coli, and the genes encoding the
enzymes are designated as men genes. Genetic comparisons of E. coli genome to the
Synechocystis sp. PCC 6803 genome have shown that all the expected E. coli men genes
are present in Synechocystis. Furthermore, knockout mutants of the five genes for PhQ
34
biosynthesis have been created and disruption of four of them, menD, menE, menB, and
menA, resulted in a complete absence of PhQ accumulation (Johnson et al. 2000; Johnson
et al. 2001). The addition of exogenous phylloquinone and other naphthoquinones can
partially restore the activity of PSI (Johnson, T.W. et al. 2000; Johnson, T. W. et al.
2001) confirming the biosynthetic pathway of phylloquinone in cyanobacteria. The
proposed biosynthetic scheme for phylloquinone is shown in Figure 6.
The phylloquinone biosynthetic pathway begins with chorismate, a six membered
ring synthesized through the shikimate biosynthetic pathway (as described previously).
Isochorismate synthetase catalyzes the isomerization reaction of chorismate to
isochorismate (Simantiras et al. 1989). Little research has been done on the mechanism
of the enzyme in cyanobacteria and higher plants. The next enzymatic reaction involves
the decarboxylation of isochorismate and the addition of 2-ketoglutarate by the enzyme
known as 2-succinyl-6-hydroxy-2,4-cyclohexadiene-1-carboxylate synthase to form 2succinyl-6-hydroxy-2,4-cyclohexadiene-1-carboxylate (Palaniappan et al. 1992).
The
following step in the phylloquinone biosynthetic pathway involves dehydration of the
molecule to form o-succinyl benzoate by o-succinyl benzoate synthase (Sharma et al.
1993).
Formation of the quinone ring is catalyzed by 1,4-dihydroxy-2-naphthoyl-CoA
synthase. In this enzymatic step, the napthoquinone ring structure is formed, and a CoA
is attached via a sulfur bond to the ring structure. The CoA is released by thioesterase
resulting in production of a 1,4-dihydroxy-2-napthoate molecule (Sharma et al. 1992;
Meganathan 2001). Once CoA is removed from the ring structure, an isoprenyl tail is
added, catalyzed by 1,4-dihydroxy-2-napthoate phytyltransferase, resulting in the
35
formation of demethylphylloquinone (Johnson et al. 2000). The last enzymatic step is the
transfer of a methyl group from methionine to produce phylloquinone, which is catalyzed
by the enzyme demethylphylloquinone methyl transferase (Suvarna et al. 1992; KoikiTakeshita et al. 1997).
In order to incoporate C2H3 into the methyl group of phylloquinone, I utilzed the
last step in the biosynthetic pathway. Methionine-tolerant Synechocystis PCC 6803 were
grown in the presence of L- C2H3 -methionine. As outlined above, methionine is used as
a methyl donor in the last enzymatic step of phylloquinone biosynthesis. The chlorophyll
biosynthesis pathway also utlizes methionine as a methyl donor. Previous experiments
have shown 42-68% incorporation of C2H3 into chl and plastoquinone using this
experimental method (Barry 1995; Razeghifard et al. 1999; Bender et al. 2008)
2.4 Aromatic Amino Acid Biosynthesis
Tyrosine and phenylalanine are produced by the same pathway, branching at the
last enzymatic step as shown in Figure 7. First, chorismate is converted to prephenate by
the enzyme chorismate mutase.
This enzymatic step catalyzes the intramolecular
rearrangement of the enolpyruvyl side chain of chorismate, reducing the double bond,
and moves the enolpyruval group in a Claisen rearrangement (Poulson et al. 1991). This
is the committed step for phenylalanine and tyrosine synthesis. Chorismate mutase is
feedback inhibited by each of the end products, phenylalanine and tyrosine, and is
activated by tryptophan (Poulson et al. 1991).
In plants and cyanobacteria, prephenate aminotransferase is responsible for the
production of arogenate, which contains the amide group common to both phenylalanine
36
and tyrosine. The enzyme uses glutamate to transfer the amino group to prephenate,
producing alpha-ketoglutarate (Bonner et al. 1987).
precursor of phenylalanine and tyrosine.
Arogenate is the last common
Arogenate dehydratase catalyzes the
decarboxylation of arogenate to produce phenylalanine (Hall et al. 1982). Arogenate
dehydrogenase uses NADP+ as an oxidizing agent to oxidize arogenate to tyrosine
through a two step process. First the enzyme oxidizes the alcohol group on the ring to a
carbonyl, and then catalyzes a decarboxylation to form tyrosine and CO2(Hall et al.
1982).
Figure 7 outlines the biosynthetic pathway and the structure of the intermediates
involved in the synthesis of tryprophan from chorismate. Anthranilate synthase catalyzes
the committing step in tryptophan biosynthesis. This enzyme catalyzes the amination of
chorismate and the removal of the enolpyruvyl side chain to produce anthranilate.
Anthranilate synthase utilizes either glutamine or ammonium as the amine donor, a Mg2+
as a cofactor and is feedback inhibited by the end product of the pathway, tryptophan
(Braus 1991). The formation of anthranilate and the following biosynthetic pathway to
produce tryptophan are identical in all the prokaryotes and eukaryotes that have been
examined (Gilchrist et al. 1980).
The second step in the tryptophan biosynthesis pathway involves a transfer of a 5phosphoribosyl pyrophosphate to anthranilate by the enzyme phosphoribosylanthranilate
transferase. N-phosphoribosyl-anthranilate is converted to 1-(o-carboxyphenylamino)-1deoxyribulose-5-phosphate by the enzyme known as phosphoribosylanthranilate
isomerase. Relatively little is known about the previous two enzymes in plants; however,
37
both enzymes have been the focus of genetic and molecular testing and analysis (Furter et
al. 1986; Hutter et al. 1986; Hommel et al. 1989).
The
enzyme,
indole-3-glycerol-phosphate
(IGP)
synthase
catalyzes
the
decarboxylation and ring closure of carboxyphenylamino-l-deoxyribulose 5-phosphate
(CDRP). This is the least well-studied step in the plant tryptophan synthesis pathway.
The crystal structures of indole-3-glycerol-phosphate (IGP) synthase have been reported
from E. coli, and research comparing the indole-3-glycerol-phosphate (IGP) synthase
genes from various organisms has been performed (Priestle et al. 1987; Braus 1991).
From these studies it is known that IGP synthase is bifunctional, and it is the second
domain that catalyzes the second ring closure of IGP. Tryptophan synthase catalyzes the
final step in tryptophan synthesis by converting indole-3-glycerol phosphate and serine to
tryptophan. Tryptophan synthase from E. coli is the best studied enzyme of the
biosynthetic pathway (Pittard 1996).
In order to incorporate ring-labeled 2H5-tryptophan into PSI, I utilized feedback
inhibition in Synechocystis PCC 6803 cells. Cultures were grown in the presence of
millimolar concentrations of phenylalanine, tyrosine, with tryptophan or 2H5-trpytophan.
This method will inhibit DAHP synthase, an enzyme in the chorismate biosynthesis
pathway (explained above). Chorismate is a common precursor to the aromatic amino
acids. Previous studies using ring-labeled, 2H4-tyrosine have shown little scrambling of
the isotope into chlorophyll and plastoquinone (Barry et al. 1987).
The previous sections have outlined the biosynthetic approaches for incorporation
of isotopes into cofactors and amino acids in PSI. These in vivo biochemical techniques
allow precise assignments of the light-induced FT-IR difference spectrum of PSI. The
38
following chapters illustrate the usefulness of these techniques in identifying protein
dynamics and protein-cofactor interactions during light-induced electron transfer.
39
2.5 References
Balinsky, D., A. W. Dennis and W. W. Cleland (1971). "Kinetic and isotope-exchange
studies on shikimate dehydrogenase from Pisumsativum." Biochemistry 10: 19471952.
Barry, B. A. (1995). "Tyrosyl radicals in photosystem II." Methods Enzymol. 258: 303319.
Barry, B. A. and G. T. Babcock (1987). "Tyrosine radicals are involved in the
photosynthetic oxygen evolving system." Proc. Natl. Acad. Sci. USA 84: 70997103.
Beale, S. I. (1999). "Enzymes of chlorophyll biosynthesis." Photosynth. Res. 60: 43-73.
Beale, S. I. and P. A. Castefranco (1974). "The biosynthesis of delta-aminolevulinic acid
in higher plants." Plant Physiol. 53: 297-303.
Beale, S. I., S. P. Gough and S. Granik (1975). "Biosynthesis of delta-aminolevulinic acid
from the intact carbon skeleton of glutamic acid in greening barley." Proc. Natl.
Acad. Sci. USA 72: 2719-2723.
Bender, S. L., J. Keough, S. E. Boesch, R. A. Wheeler and B. A. Barry (2008). "The
vibrational spectrum of the secondary electron acceptor, A1, in photosystem I." J.
Phys. Chem. B 112: 3844-3852.
Bentley, R. (1990). "The shikimate pathway - a metabolic tree with many branches."
CRC Critical Reviews of Biochemistry and Molecular Biology 25: 307-384.
Bollivar, D. W. and S. I. Beale (1995). "Formation of the isocyclic ring of chlorophyll by
isolated Chlamydomonas reinhardtii chloroplasts." Photosynth. Res. 43: 113-124.
Bollivar, D. W. and S. I. Beale (1996). "The chlorophyll biosynthetic enzyme Mgprotoporphyrin IX monomethyl ester (oxidative) cyclase. Characterization and
partial purification from Chlamydomonas reinhardtii and Synechocystis sp. PCC
6803." Plant Physiol. 112: 105-114.
Bondinell, W. E., J. Vnek, P. F. Knowles, M. Sprecher and D. B. Sprinson (1971). "On
the mechanism of 5-enolpyruvylshikimate 3-phosphate synthetase." J. Biol.
Chem. 246: 6191-6196.
Bonner, C. and R. Jensen (1987). "Prephenate aminotransferase." Methods Enzymol. 142:
479-487.
40
Braus, G. H. (1991). "Aromatic amino acid biosynthesis in the Yeast Saccharomyces
cerevisiae: a model system for the regulation of eukaryotic biosynthetic pathway."
Microbiol. Rev. 55: 349-370.
Crockett, N., P. Alefounder, A. Battersby and C. Abell (1991). "Uroporphyrinogen III
synthase: studies on its mechanism of action, molecular biology and
biochemistry." Tetrahedron 47: 6003-6014.
Ellswort, R. K. and J. P. Dullagha (1972). "Activity and properites of (-)-S-adenosyl-Lmethionine: magnesium-protoporphyrin IX methyl transferase in crude
homogenates from wheat seedlings." Biochim. Biophys. Acta 268: 327-333.
Frydman, R. B. and B. Frydman (1970). "Purification and properites of porphobilinogen
deaminase from wheat germ." Arch. Biochem. Biophys. 136: 193-202.
Furter, R., G. Paravicini, M. Aebi, G. Braus, F. Prantl, P. Niederberger and R. Hutter
(1986). "The TRP4 gene of Saccharomyces cerevisiae: isolation and structural
analysis." Nucleic Acids Res. 14: 6357-6373.
Gaertner, F. H. and K. W. Cole (1973). "Properties of chorismate synthase in Neurospora
crassa." J. Biol. Chem. 248: 4602-4609.
Gibson, K., W. Laver and A. Neuberger (1958). "Formation of delta-aminolevulinic acid
in vitro from succinyl-conenzyme A and glycine." Biochem. J. 70: 71-81.
Gilchrist, D. G. and T. Kosuge (1980). The Biochemistry of Plants. New York, Academic
Press.
Gollube, E. (1966a). "Correlation of genes and enzymes and studies on regulation of
aromatic pathway in Salmonella." J. Biol. Chem. 242: 5323.
Gollube, E. (1966b). "Enzymes associated with unlinked genes of aromatic pathway in
Salmonella." Federation Proceedings 25: 337.
Gorchein, A. (1972). "Magnesium protoporphyrin chelatase activity in
Rhodopseudomonas spheroides. Studies with whole cells." Biochem. J. 127: 97106.
Hall, G. C., M. B. Flick, R. L. Gherna and R. Jensen (1982). "Biochemical Diversity for
biosynthesis of aromatic amino acids among the cyanobacteria." J. Bacteriol. 149:
65-78.
Hart, G. and A. Battersby (1985). "Purification and properties of uroporphyrinogen II
synthase (co-synthase) from Euglena gracilis." Biochem. J. 232: 151-160.
Herrmann, K. M. (1995a). "The shikimate pathway as an entry to aromatic secondary
metabolism." Plant Physiol. 107.
41
Herrmann, K. M. (1995b). "The shikimate pathway; early steps in the biosynthesis of
aromatic compounds." Plant Cell 7: 909-917.
Hommel, U., A. Lustig and K. Kirschner (1989). "Purification and characterization of
yeast anthranilate phosphoribosyltransferase." Eur. J. Biochem. 180: 33-40.
Hutter, R., P. Niederberger and J. A. DeMoss (1986). "Tryptophan biosynthetic genes in
eurkaryotic microorganisms." Annu. Rev. Microbiol. 40: 55-77.
Johnson, T. W., G. Shen, B. Zybailov, D. Kolling, R. Reategui, S. Beauparlant, I. R.
Vassiliev, D. A. Bryant, A. D. Jonesi, J. H. Golbeck and C. P.R. (2000).
"Recruitment of a foreign quinone into the A1 site of photosystem I. I. Genetic
and physiological characterization of phylloquinone biosynthetic pathway
mutants in Synechocystis sp. pcc 6803." J. Biol. Chem. 275: 8523-8530.
Johnson, T. W., B. Zybailov, A. D. Jones, R. Bittl, S. Zech, D. Stehlik, J. H. Golbeck and
P. R. Chitnis (2001). "Recruitment of a foreign quinone into the A1 site of
Photosystem I. In vivo replacement of plastoquinone-9 by media-supplemented
naphthoquinones in phylloquinone biosynthetic pathway mutants of Synechocystis
sp. PCC 6803." J. Biol. Chem. 276: 39512-39521.
Kannangara, C. G., S. P. Gough, P. Bruyant, J. K. Hoober, A. Kahn and D. von Wettstein
(1988). "tRNAGlu as a cofactor in delta-aminolevulinate biosyntheisis. Steps that
regulate chlorophyl synthesis." Trends Biochem. Sci. 13: 139-143.
Kikuchi, G., A. Kumar, P. Talmage and D. Shemin (1958). "The enzymatic synthesis of
delta-aminolevulinic acid." J. Biol. Chem. 233: 1214-1219.
Kim, S. and B. A. Barry (2000). "Identification of carbonyl modes of P700 and P700+ by in
situ chlorophyll labeling in photosystem I." J. Am. Chem. Soc. 122: 4980-4981.
Koiki-Takeshita, A., T. Koyama and K. Ogura (1997). "Identification of a novel gene
cluster participating in menaquinone (vitamin K2) biosynthesis." J. Biol. Chem.
272: 12380-12383.
Koshiba, T. (1979). "Alicyclic acid metabolism in plants. 12. Partial purification and
some properties of shikimate kinase from Phaseolus-mungo seedlings." Plant Cell
Physiol. 20: 803-809.
Macheroux, P., R. Schmid, N. Amrhein and A. Schaller (1999). "A unique reaction in
common pathway: mechanism and function of chorismate synthase in the
shikimate pathway." Planta 207: 325-334.
Marques, M. R., J. H. Pereira, J. S. Oliveira, L. A. Basso, W. F. de Azevedo, D. S. Santos
and M. S. Palma (2007). "The inhibition of 5-enolpyruvylshikimate-3-phosphate
synthase as a model for development of novel antimicrobials." Curr. Drug
Targets 8: 445-457.
42
Meganathan, R. (2001). "Biosynthesis of menoquinone (vitamin K2) and ubiquinone
(coenzyme Q): a perspective on enzymatic mechanisms." Vitamin and Hormones
61: 173-218.
Oliver, R. P. and W. T. Griffiths (1980). "Identification of the polypeptides of NADPHprotochlorophyllide oxidoreductase." Biochem. J. 191: 277-280.
Palaniappan, C., V. Sharma, M. E. S. Hudspeth and R. Meganathan (1992).
"Menoquinone (vitamin K2) biosynthesis: evidence that the Escherichia coli
menD gene encodes both 2-succinyl-6-hydroxy-2,4-cyclohexadiene-1-carboxylic
acid synthase and alpha-ketoglutarate decarboxylase activities." J. Bacteriol. 174:
8111-8118.
Pittard, A. J. (1996). Biosynthesis of the aromatic amino acids, in Escherichia coli and
Salmonella. Cell. Mol. Biol. F. Neidardt, American Society of Bacteriology: 458484.
Poulson, C. and R. Verpoorte (1991). "Roles of chorismate mutase, isochorismate
synthase and anthranilate synthase in plants." Phytochemistry 30: 377-386.
Priestle, J. P., M. G. Gruetter, J. White, M. G. Vincent, M. Kania, E. Wilson, T. S.
Jardetzkey, K. Kirschner and J. N. Jansonius (1987). "3-dimensional structure of
the bifunctional enzyme phosphoribosylanthranilate isomerase: indoleglycerol
phosphate synthase from Escherichia coli." Proc. Natl. Acad. Sci. USA 84: 56905694.
Razeghifard, M. R., S. Kim, J. Patzlaff, R. S. Hutchison, T. Krick, I. Ayala, J. Steenhuis,
S. E. Boesch, R. A. Wheeler and B. A. Barry (1999). "In vivo, in vitro and
calculated vibrational spectra of plastoquinone and the plastosemiquinone anion
radical." J. Phys. Chem. B 103: 9790-9800.
Rudiger, W., J. Benz and Guthoff, C. (1980). "Detection and partial characterization of
activity of chlorophyll synthetase in etioplast membranes." Eur. J. Biochem. 109:
193-200.
Sharma, V., R. Meganathan and M. E. S. Hudspeth (1993). "Menaquinone (vitamin K2)
biosynthesis: nucleotide sequence and expression of the menC gene from
Escherichia coli." J. Bacteriol. 175: 4917-4921.
Sharma, V., K. Suvarna, R. Meganathan and M. E. S. Hudspeth (1992). "Menaquinone
(vitamin K2) biosynthesis: nucleotide sequence and expression of the menB gene
from Eschericia coli." J. Bacteriol. 174: 5057-5062.
Simantiras, M. and E. Leistner (1989). "Formation of o-succinylbenzoic acid from isochorismic acid in protein extracs from anthraquinone-producing plant cell
suspensions." Phytochemistry 28: 1381-1382.
43
Smith, L. C. (1962). "Control of 3-deoxy-D-arabino-heptulosonic acid 7-phosphate
synthesis by phenylalanine and tyrosine." J. Biol. Chem. 237: 3566-3570.
Srinivasan, P. R. and D. B. Sprinson (1959). "2-keto-3-deoxy-D-arabo-heptonic acid 7phosphate synthetase." J. Biol. Chem. 234: 716-722.
Suvarna, K., D. Stevenson, R. Meganathan and M. E. S. Hudspeth (1992). "Menoquinone
(vitamin K2) biosynthesis: localization and characterization of the menA gene
from Escherichia coli." J. Bacteriol. 180: 2782-2787.
Walker, C. J. and J. Wienstein (1991). "In vitro assay of the chlorophyll biosynthetic
enzyme Mg-chelatase: resolution of the activity into soluble and membranebound fractions." Proc. Natl. Acad. Sci. USA 88: 5789-5793.
44
Figure 1. Formation of delta-aminolevulinic acid from L-glutamic acid. Carbon 3
(green) will be incorporated into eight positions of the tetrapyrrole ring of chlorophyll,
including the 131 keto position.
45
Figure 2. Formation of uroporphyrinogen III from porphobilinogen. The carbons in
green are from L-glutamic-3-13C-acid. Uroporphyrinogen is the branching point for heme
synthesis and chlorophyll synthesis.
46
Figure 3.
Enzymatic steps involved in synthesis of Mg-protophorphyrin IX.
The
carbons in green are from L-glutamic-3-13C-acid.
47
Figure 4. Final enzymatic steps in the formation of chlorophyll a including formation of
ring V. The oxygen (blue) will be incorporated from O2 by the cyclase enzymatic
reaction. The carbons in green are from L-glutamic-3-13C-acid.
48
Figure 5.
Biosynthetic steps involved in the production of chorismate from
phosphoenolpyruvate and erythrose 4-phosphate.
49
Figure 6.
Enzymes and cofactors involved in the production of phylloquinone from
chorismate. The methyl group (blue) will be incorporated from S-adenosylmethionine.
50
Figure 7. Biosynthetic pathway of the aromatic amino acids: phenylalanine, tryptophan, and
tyrosine, starting with the common precursor, chorismate.
51
CHAPTER 3
THE VIBRATIONAL SPECTRUM OF THE SECONDARY ELECTRON
ACCEPTOR,
A1, IN PHOTOSYSTEM I
Reproduced with permission from “The Vibrational Spectrum of the Secondary Electron
Acceptor, A1, in Photosystem I.” Bender, S. L.; J. M. Keough; , S. E. Boesch; R. A. Wheeler;
B. A. Barry, Journal of Physical Chemistry B, 112; 3844-3852. Copyright 2008 American
Chemical Society
3.1 Abstract
Photosystem I (PSI) is a multisubunit protein complex, which carries out lightinduced, transmembrane charge separation in oxygenic photosynthesis. In PSI, the electron
transfer pathway consists of chlorophyll and phylloquinone molecules, as well as iron-sulfur
clusters.
There are two phylloquinone molecules, which are located in structurally
symmetric positions in the reaction center. It has been proposed that both phylloquinone
molecules are active as the A1 secondary electron acceptor in bi-directional electron transfer
reactions. The PSI A1 acceptors are of interest because they have the lowest reduction
potential of any quinone found in nature.
In this work using light-induced FTIR
52
spectroscopy, isotope-edited spectra are presented, which attribute vibrational bands to the
carbonyl stretching vibrations of A1 and A1- and the quinoid ring stretching vibration of A1.
Bands are assigned by comparison with hybrid Hartee-Fock density functional calculations,
which predict vibrational frequencies, amplitudes, and isotope shifts for the phylloquinone
singlet and radical anion states. The results are consistent with an environmental interaction
increasing the frequency of the singlet CO vibration and decreasing the frequency of the
anion radical CO vibration, relative to model compounds. This environmental interaction
may be the asymmetric hydrogen bond to A1/A1-, electrostatic interactions with charged
amino acid side chains, or a pi-pi interaction with the indole ring of a nearby tryptophan.
Such differential effects on the structure of A1 and A1- may be associated with a
destabilization of the anion radical. These studies give novel information concerning the
effect of the protein matrix on the PSI electron transfer cofactor.
53
3.2 Introduction
In oxygenic photosynthesis, photosystem I carries out the light-driven oxidation of
plastocyanin and the reduction of ferredoxin. These reactions lead to a charge separation
across the thylakoid membrane. Electron transfer is initiated by photoexcitation of the
photosystem I primary chlorophyll donor, which reduces the primary acceptor, A0, a chl a
monomer (Holzwarth et al. 2006).
The resulting positive charge resides on P700, a
heterodimeric chlorophyll, which is composed of chl a and its epimer, chl a' (Fromme et al.
2001; Jordan et al. 2001). A0 reduces the secondary electron acceptor, A1, which is a
phylloquinone (Rustandi et al. 1990). Sequential electron transfer reactions then lead to the
reduction of FX, FA and FB, which are iron-sulfur clusters. FA and FB are bound to the
extrinsic PsaC subunit, while the other electron transfer cofactors are bound to the intrinsic,
membrane-associated PsaA and PsaB subunits (Fromme et al. 2001; Jordan et al. 2001).
The structure of PSI from Synechococcus elongates has been solved at 2.5 Å
resolution by X-ray diffraction (Jordan et al. 2001). There are two phylloquinone acceptor
molecules, which are located in C2 structurally symmetric positions. The two phylloquinone
cofactors will be designated here as A1A and A1B, where the subscripts designate the subunit
to which the phylloquinone molecule is bound. It has been proposed that both sets of
phylloquinone cofactors are reduced in PSI electron transfer reactions, but with different time
constants (Rigby, S. E. et al. 1996; Joliot et al. 1999; Boudreaux et al. 2001; Guergova-Kuras
et al. 2001; Purton et al. 2001; Rigby, S. E. J. et al. 2002; Ramesh et al. 2004; Bautista et al.
2005; Poluektov et al. 2005; Santabarbara et al. 2005a; Santabarbara et al. 2005b; Ali et al.
2006; Li et al. 2006; Santabarbara et al. 2006; Ramesh et al. 2007). In Chlamydomonas
54
reinhardtii, electron transfer has been deduced to be bidirectional with a 1:1 ratio of the
kinetic phases (Joliot et al. 1999; Guergova-Kuras et al. 2001). In cyanobacteria, electron
transfer is more unidirectional along the PsaA branch with ratios of 1:2-3 (Klughammer et al.
1999; Guergova-Kuras et al. 2001; Xu et al. 2003a; Xu et al. 2003b; Cohen et al. 2004;
Dashdorj et al. 2005). In particular, PSI from Synechocystis PCC 6803 shows a fast phase
(time constant 11 ns) and slow phase, which have been assigned to electron transfer down the
B and A branch, respectively. The slow phase was observed to have a 340 ns time constant
at 295 K and to slow upon cooling, with an activation energy of 110 meV (Aglarov et al.
2003). The minor fast phase was observed to be temperature independent.
The structure of phylloquinone is shown in Figure 1, along with its IUPAC
numbering scheme. Figure 2 displays the binding pockets of A1A (left) and A1B (right) from
the 2.5 Å PSI structure (Jordan et al. 2001). A predominant feature of both binding pockets
is a tryptophan residue, which is within π-stacking distance of each A1 acceptor. These two
conserved tryptophans are located on the stromal side of A1A and A1B. The distances between
the indole ring and the quinone are approximately 3 Å for A1A and 3.5 Å for A1B. Another
striking feature of the A1 protein environment is an asymmetric hydrogen bond from the
amide NH groups of LeuA722 and LeuB706 respectively, to the C4=O groups of each A1
molecule (Jordan et al. 2001; Pushkar, Y. et al. 2004; Teutloff et al. 2004).
The lack of a
second hydrogen bond to C1=O distinguishes the A1 binding site, when compared to the QA
binding site in bacterial reaction centers, in which two hydrogen bonds are formed to QA-. A
tryptophan-QA pi-pi interaction also occurs in the bacterial reaction center (Allen et al.;
Deisenhofer et al. 1989; Kamlowski et al. 1998; Fromme et al. 2001; Pushkar, Y. et al.
2004).
55
A1 has the lowest known redox potential of any quinone found in nature.
Measurements for the A1 redox potential range from -750 to -810 mV (Brettel, Klaus 1997).
Previous studies in which A1 has been exchanged for various quinone molecules, such as
benzoquinone, have shown a -0.3 V change in the redox potential compared to the measured
reduction potential in solution (Itoh et al. 2001). These results indicated that specific proteincofactor interactions, such as the NH...CO asymmetric hydrogen bond (Pushkar, Y. et al.
2004; Li et al. 2006; Feldman et al. 2007) and possibly the pi-pi interaction, modulate the
redox potential of quinones bound to the A1 binding site (van der Est 2001). Recent work
has also implicated an electrostatic interaction with an aspartate as a modulator of A1 redox
potential (Karyagina et al. 2007).
To understand the structure of A1/A1- and their interaction with the protein
environment, we have used FTIR spectroscopy. Vibrational spectroscopy supplies dynamic
structural information concerning the electron transfer cofactors in vivo. Previously, A1-/A1
spectra have been constructed, and multiple assignments have been suggested for the A1 and
A1- vibrational bands (Hastings et al. 2001; Sivakumar et al. 2005; Bandaranayake, K. M.; et
al. 2006; Bandaranayake, K. M. et al. 2006a; Bandaranayake, K. M. et al. 2006b). However,
there are discrepancies in these assignments, and different spectroscopic methods have
yielded contradictory results.
In this study, we have used isotopic labeling of phylloquinone in order to identify the
vibrational modes of A1 and A1-, under conditions in which A1 is the terminal electron
acceptor in PSI. When the iron-sulfur clusters, Fx, FA, and FB are removed, the P700+A1- state
can be photo-generated (Warren et al. 1990; Van der Est et al. 1994; Shen et al. 2002a). We
have incorporated C2H3 into the methyl group of A1 using a methionine-tolerant strain of the
56
cyanobacterium, Synechocystis sp. PCC 6803. This approach has been used previously to
assign bands to the plastoquinone acceptors, QA and QA-, in cyanobacterial photosystem II
(Razeghifard et al. 1999).
In our current work, light-minus-dark FT-IR spectroscopy
identifies C2H3-sensitive bands of A1 and A1-.
The data presented here support the
interpretation that an asymmetric hydrogen bond, electrostatic interactions, and/or a pi-pi
interaction shifts the CO stretching and quinone ring stretching frequencies of A1 and A1-,
relative to model compounds.
3.3 Materials and Methods
3.3.1 Isotopic labeling and PSI purification.
A methionine-tolerant strain of
Synechocystis PCC 6803 was grown on solid media, containing 90 μM methionine BG-11, 5
mM TES-NaOH, pH 8.0, and 6 mM NaS2O3 (Rippka et al. 1979; Barry, B.A. 1995). Liquid
cultures (15 L) were supplemented either with 200 μM natural abundance C1H3-methionine
or C2H3-methionine (98% enrichment, Cambridge Isotope Laboratories, Andover, MA)
(Razeghifard et al. 1999). Trimeric PSI was purified as described previously (Kim et al.
2001). Purified trimeric PSI samples were dialyzed overnight in 5 mM HEPES-NaOH pH
7.5, 0.04% dodecyl-β-D-maltoside (LM) and were concentrated using an Amicon (Bedford,
MA) Ultra 100,000 MWCO centrifugal filter device to a final concentration of 4.6 mg
chl/ml.
3.3.2 Apo-Fx preparations.
PSI (0.25 mg chl/ml) was incubated in a buffer
containing 6.8 M urea, 50 mM Tris, 10 mM glycine-NaOH, pH 10 for 70 minutes as
previously described (Parrett, K. G. et al. 1989). The PSI sample was then dialyzed overnight
against 50 mM Tris-HCl, pH 8.3. The sample was incubated in 3 M urea, 5 mM Ke3FeCN6
57
50 mM Tris-HCl, pH 8.3 for 2 hours to remove the Fx cluster (Warren et al. 1990). The
sample was dialyzed overnight against 50 mM Tris-HCl, pH 8.3, and then again overnight
against a buffer containing 50 mM Tris-HCl, pH 8.3, 5 mM 4,5-dihydroxy-1,3-benzenedisulfonic acid, which was supplied as the disodium salt. The sample was dialyzed a third
night against 50 mM Tris, pH 8.3, 0.035% LM (Parrett, K. G.; et al. 1990; Warren et al.
1990; Warren et al. 1993). The sample was buffer exchanged with 5 mM HEPES-NaOH, pH
7.5 and concentrated using an Amicon Ultra 100,000 MWCO centrifugal filter device. The
final chl concentration was 3 mg/ml.
Transient absorption spectroscopy, monitoring the decay of the P700+ signal at 820
nm, was performed to verify iron-sulfur cluster removal, as previously described (Warren et
al. 1990). The PSI concentration was 41-58 micrograms chl/ml, and the samples were in 50
mM Tris pH 8.3, and 0.035% LM. Sodium ascorbate (5 mM) and dichloroindophenol
(DCPIP, 4 μM) were added as exogenous electron donors. The 532 nm photolysis laser had
an energy of 1.5-1.8 mJ/cm2. SDS gel electrophoresis was also used to verify PsaC removal
(Piccioni et al. 1982; Kruip et al. 1997).
3.3.3 Assessment of isotope enrichment through mass spectrometry.
Previous
quantitative work has shown that both the 134-methyl of chl (Kim et al. 2000a) and the
phylloquinone ring C2 methyl group (Barry, B. A. et al. 1987; Kim et al. 2001) are labeled
from C2H3-methionine, which is a methyl group donor to both cofactors. Therefore,
phylloquinone isotopic enrichment can be monitored indirectly by measuring the amount of
chl isotopic labeling in cyanobacterial cultures grown on C2H3-methionine (Kim et al.
2000b). The control sample was isolated from cyanobacteria grown on natural abundance
methionine.
Chlorophyll was extracted from PSI with a solution of 80% acetone/20%
58
methanol. The extracted pigments were vortexed, sonicated, centrifuged, and filtered as
previously described (Patzlaff et al. 1996). Chlorophyll was concentrated and dried using a
Thermo-savant SpeedVac concentrator with VLP120 pump and RVT 400 refrigerated vapor
trap (Thermo Electron Corporation, Waltham, MA). Chlorophyll samples were stored at
-70°C until use. Matrix assisted laser desorption ionization mass spectrometry on a 4700
Proteomics Analyzer (Applied Biosystems Foster City, CA) was performed to determine the
amount of C2H3 incorporation.
Immediately before the measurement, chlorophyll was
dissolved in a mixture of 80% methanol, 19% water, and 1% acetic acid; this treatment
removes the central Mg+2 ion of chlorophyll and generates pheophytin. The matrix used was
alpha-cyano-4-hydroxy cinnamonic acid. The 871.5 m/z peak of pheophytin [M+H] was
used for quantitation. Data were analyzed using IGOR (Wavemetrics, Lake Oswego, OR)
software. The data were normalized to the [M+H] peak, and the isotope distribution was
calculated. This approach showed 40 + 1% labeling of the chlorophyll methyl group from
C2H3-methionine and implies a similar amount of phylloquinone labeling (Barry, B. A. et al.
1987; Kim et al. 2000b). Previous work has shown that there is no significant scrambling
into other amino acids besides methionine, suggesting that there will be no detectable
labeling of the peptide backbone (Razeghifard et al. 1999).
3.3.4. FTIR Spectroscopy: FT-IR spectra were collected at -10°C (Kim et al. 1998)
or at room temperature (20oC). Samples contained 3 mM potassium ferricyanide and 3 mM
potassium ferrocyanide and were concentrated at room temperature under a steady flow of
nitrogen gas. Concentration times were 20-30 minutes. Spectral conditions were as follows:
resolution, 4 cm-1; zero filling, 1; data acquisition time, 4.0 min. Difference spectra (lightminus-dark) were generated by taking the ratio of single beam spectra collected before and
59
during illumination and converting to absorbance.
A1-/A1 spectra were generated by
subtraction of data from intact and apo-Fx PSI. The A1-/A1 isotope edited spectra were
generated by subtracting difference FT-IR spectra, acquired from C2H3-PSI samples, from
difference FT-IR spectra, acquired from natural abundance, C1H3-PSI samples (Kim et al.
2000b; Kim et al. 2001; Sacksteder et al. 2005). FT-IR spectra were obtained in the dark or
under continuous illumination with red- and heat-filtered light, as previously described (Kim
et al. 2001; Sacksteder et al. 2005). A 90 min dark relaxation time was used between
successive illuminations. The final averaged difference FT-IR spectra were an average of 45
spectra and were normalized for small differences in sample concentration and path length,
using the amplitude of the amide II band in the infrared absorption spectrum. The infrared
absorption spectrum was created using a dark and a background scan, which were acquired
with infrared windows, but no PSI sample. Previous studies have found that changes in
hydration levels alter the details of PSI charge separation (Sacksteder et al. 2005). Therefore,
water content in all the FTIR samples was held constant, as assessed by the ratio (1.1-1.3) of
the 3300 cm-1 band to the amide I band at 1656 cm-1 (Kim et al. 2001).
3.3.5 Density Functional Calculations. Density functional methods were used to
predict the vibrational frequencies, IR intensities, and isotopic frequency shifts for the normal
modes of models for phylloquinone and the phylloquinone anion radical. Figure 1 displays
the structure of phylloquinone, and for the model the isoprenyl chain of phylloquinone was
truncated to a single isoprenyl unit for ease in computation. The methyl group on the
quinone ring was completely deuterated in order to estimate the maximum isotopic frequency
shifts.
60
The B3LYP hybrid Hartree-Fock density functional (HF/DF) method (Becke 1993)
was used to calculate the vibrational frequencies for a model of phylloquinone and its radical
anion. The B3LYP method was chosen because it is known to give harmonic vibrational
frequencies within approximately 4% of experiment, for approximately the same computer
time investment as a Hartree-Fock calculation, in which estimated frequencies vary from
experiment by more than 10% (Scott et al. 1996). The three-parameter B3LYP method uses
a weighted sum of Hartree-Fock, local DF, and gradient-corrected DF expressions for the
exchange and correlation energies where (EXSlater) is Slater’s local spin density functional for
exchange, (Slater 1974) EXBecke is Becke’s gradient correct exchange functional, (Becke
1988) ECVWN is the local density correlation functional of Vosko, Wilk, and Nusair, (Vosko et
al. 1980) and ECLYP is the gradient-corrected correlation functional of Lee, Yang, and Parr
(Lee et al. 1988). Coefficients giving the relative weights of various approximations for the
exchange and correlation energies in this method were optimized to reproduce
thermochemical data for a variety of small molecules (Becke 1993).
All calculations
reported here were performed using the 6-31G(d) split-valence plus polarization basis set
(Hehre et al. 1986; Helgaker et al. 1995). This basis set was chosen because it reproduces the
properties of medium-sized organic molecules, including a variety of p-benzoquinones and
their radical anions and is small enough for rapid calculations (Boesch et al. 1995, 1997a,
1997b; Wise et al. 1997; Grafton et al. 1998a, 1998b).
The quantum chemistry program Gaussian 03 (Frisch et al. 2004) was used for all
calculations.
Berny’s optimization algorithm was used to perform full geometry
optimizations in C1 symmetry using internal coordinates (Schlegel 1986).
Frequency
calculations were performed for the optimized geometries. It has been customary to scale
61
calculated vibrational frequencies, and we adopt the scaling factor of 0.9614 suggested by
Scott and Radom (Scott et al. 1996).
Vibrational mode assignments were performed by animating each mode using the
program XMOL (Wasikowski et al. 1993) and comparing the modes of one molecule to
another using the program ViPA, an acronym for Vibrational Projection Analysis (Grafton et
al. 1998a, 1998b). The ViPA program (Grafton et al. 1998a, 1998b) exploits the vector
properties of vibrational normal modes to assess the similarity between modes of an object
molecule and a structurally similar basis molecule.
The program first aligns the two
molecules and calculates each molecule’s normal modes and vibrational frequencies. For
each molecule, each of the normal vibrational modes is a column vector, which is
orthonormal to all other normal modes of the same molecule.
The vector projection
operation is done by sequentially projecting each normal mode of the object molecule to the
modes of the basis molecule. The similarity of any mode of the object molecule to any mode
of the basis molecule can then be expressed as a percentage by calculating the sum of the
squares of the matrix elements and multiplying by 100. Vibrational projection analysis has
been
exploited
to
compare
normal
modes
modified
by
isotopic
substitution,
oxidation/reduction, as well as covalent (Grafton et al. 1998a; Wise et al. 1999) and noncovalent (Razeghifard et al. 1999) modifications.
3.4 Results
Transient absorption experiments were performed to verify the removal of the ironsulfur clusters in the apo-Fx samples. In the absence of FX, FA, and FB, P700+ recombines
with A1-, and an acceleration in the decay rate of P700+, is expected (Parrett, K. G.; et al.
62
1990; Warren et al. 1990; Golbeck 1994; Brettel, Klaus 1997; Brettel, K. et al. 2001; Shen et
al. 2002a; Shen et al. 2002b). P700+ decay can be detected at 820 nm (Figure 3) after a 532
nm flash. At the low 532 nm excitation energies employed here, previous work has shown
that chl triplets do not make a significant contribution to 820 nm decay kinetics either in
intact or apo-FX PSI (Vassiliev et al. 1997).
Therefore, the decay of 820 nm absorbance is
attributed to P700+ recombination with PSI electron acceptors (Vassiliev et al. 1997).
Note
that the effects of changes in 532 nm excitation energy on intact and apo-Fx PSI have been
previously described (Vassiliev et al. 1997).
Figures 3A and 3B show P700+ decay kinetics in intact PSI samples, which contain FX,
FA, and FB. Figure 3A was obtained from cells cultured on C1H3 methionine; Figure 3B was
obtained from cells cultured on C2H3 methionine. These intact samples show the expected
P700+ FA-/FB- recombination kinetics (Parrett, K. G.; et al. 1990; Warren et al. 1990) with
lifetimes of 15-17 ms and 84-95 ms. A third slower phase (lifetime 3-6 s) is assigned to the
reduction of P700+ by the exogenous electron donor, DCPIP. Figure 3C and 3D exhibit the
P700+ decay kinetics, which were measured in apo-FX PSI. Figure 3C was obtained from cells
cultured on C1H3 methionine, and Figure 3D was obtained from cells cultured on C2H3
methionine. In Figures 3C and 3D, the lifetime of P700 has decreased dramatically, as
predicted for samples with A1 as the terminal electron acceptor. The slow phases of 15-17
and 84-96 ms have been replaced by faster biphasic kinetics with lifetimes of 11-13 μs and
69-115 μs (Brettel, K. et al. 1995). These results demonstrate that FA, FB, and FX have been
removed in the apo-FX preparation.
In Figure 4, we present difference (light-minus-dark) FT-IR spectra acquired either
from apo-FX (Figure 4A) or intact PSI (Figure 4B) at room temperature (solid lines) or at 63
10oC (dotted lines). The data in Figure 4A correspond to P700+A1--minus-P700A1 spectra, and
the data in Figure. 4B correspond to P700+FB--minus-P700FB spectra. The spectrum in Figure
4A is in good agreement with previous published time-resolved spectra of P700+/A1,
confirming the accumulation of the A1- species (Sivakumar et al. 2005; Bandaranayake, K.
M. et al. 2006a) and the generation of P700+ and not the P700 triplet, 3P700 (Breton et al. 1999;
Sivakumar et al. 2005).
In Figure 4C (room temperature) and Figure 4D, (-10oC), construction of the double
difference spectrum, A1--minus-A1, was performed by subtraction of data in Figure 4B from
Figure 4A. In construction of these double difference spectra, data in Figure 4A were
divided by a factor of 3 to account for an overall difference in spectral amplitude in the two
PSI preparations. The increased spectral amplitude in Figure 4B, compared to Figure 4A,
may be attributable to an increased efficiency of charge separation in intact PSI, compared to
apo-FX PSI, under the low intensity, continuous, red-filtered illumination used for the FT-IR
measurements.
In the resulting double difference spectra (Figures 4C and 4D), unique vibrational
bands of A1 appear as negative signals; unique vibrational contributions from A1- are positive
bands. The reduction spectrum of FB is not expected to contribute significantly in this
spectral region (Breton et al. 1999). Figure 4 demonstrates that the A1--minus-A1 spectra
acquired at room temperature and -10oC are similar (Figure 4C and 4D), except for an
increased amplitude at ~1620 cm-1 in the room temperature data, which may be due to an
increased contribution from water at the higher temperature (Figure 4C). Previously, the
difference FT-IR A1--minus-A1 spectrum has been reported in apo-FX PSI preparations at
room temperature. Comparison of our data with this previously reported work is difficult
64
due to the use of 2H2O buffers in ref (Hastings et al. 2001). However, the spectra in Figure
4C and 4D do show similarities to the transient spectrum, attributed to A1--minus-A1, in intact
cyanobacterial PSI. The transient spectrum was acquired by microsecond, time resolved,
step-scan FT-IR techniques in 1H2O buffers at 77K (Sivakumar et al. 2005).
In Figure 5, we present difference (light-minus-dark) FT-IR spectra obtained from
apo-Fx PSI samples and attributable to P700+A1--minus-P700A1.
Figure 5A shows the
difference spectrum acquired from the natural abundance, C1H3 phylloquinone control, which
was isolated from cells cultured on C1H3-methionine.
Figure 5B shows the spectrum
acquired from C2H3 phylloquinone labeled samples. The isotope-edited C1H3-minus-C2H3
double difference spectrum (Figure 5C) was constructed by subtraction of Figure 5B from
5A and was multiplied by a factor of ten for presentation. C2H3-induced shifts are observed
as derivative-shaped spectral features.
To appear in the isotope-edited spectrum, the
frequencies must be perturbed by charge separation (for example, by the reduction of A1) and
by isotope-incorporation. Figure 5D shows a control subtraction that was generated by
subtracting one-half of the C1H3-PSI data set from the other half of the data set. The
spectrum will not contain any vibrational bands and allows an estimate of the noise in the
averaged data. Comparison of Figure 5C and Figure 5D demonstrates that the vibrational
frequencies observed in Figure 5C are significant.
Figure 5E is the dark-minus-dark
spectrum, showing that the observed spectra in A and B are dependent on illumination of the
sample.
Both the 134-methyl of chl and the phylloquinone ring C2 methyl group (Figure 1) are
labeled from methionine (Barry, B. A. et al. 1987; Kim et al. 2000b). Chl methyl labeling
has been shown to shift the ester and keto vibrational bands of P700+-minus-P700 (Kim et al.
65
2000a). The ester vibrational bands of P700+-minus-P700 have been previously assigned to
spectral features observed between 1760 and 1700 cm-1, (Kim et al. 2000a) and in Figure 5C,
the expected spectral shifts due to the P700/P700+ ester bands are observed. Therefore, the
isotope-edited spectrum in Figure 5C exhibits contributions from isotope labeling of P700, as
well as from labeling of A1.
To identify isotope shifts due to A1/A1-, the C2H3-spectrum derived from intact PSI
(Figure 6B) was subtracted from the C2H3-edited spectrum derived from apo-FX PSI (Figure
6A), after normalization for the chl ester bands. This normalization accounts for the lower
amplitude of the difference spectrum, acquired from the apo-FX preparation. This lower
amplitude may be due to a smaller amount of charge separation in the apo-FX preparation, as
described above. The resulting isotope-edited spectrum (Figure 6C) will reflect isotopeinduced shifts in vibrational bands of A1 and A1-. Unique vibrational bands of A1 appear as
negative signals, with the corresponding isotope-shifted components appearing as positive
bands. Unique vibrational contributions from A1- are positive bands; the corresponding
isotope-shifted components are negative bands. Figure 6D shows a negative control, in
which no vibrational bands are expected.
To assign spectral features in Figure 6C to normal modes of the phylloquinone singlet
and anion radical, DFT calculations were used to predict the frequencies, intensities, and
C2H3 isotope shifts. The predicted frequencies and intensities are in reasonable agreement
with those reported previously (Bandaranayake, K. M. et al. 2006a). Of particular interest
are the carbonyl and quinone ring stretching vibrations (Table 1), which have frequencies and
intensities that may respond to environmental perturbations, such as hydrogen bonding,
electrostatic interactions, and pi-pi stacking. Table 1 shows that only the CO stretching and
66
CC quinoid ring stretching vibrations are predicted to have vibrational frequencies that are
sensitive to phylloquinone reduction.
Aromatic ring modes are expected to cancel
approximately in the difference spectrum, due to the lack of reduction induced frequency
shift, and are not considered further. The CO antisymmetric stretching vibration is predicted
to be the most intense band both in the singlet and the anion spectrum (Table 1). Although in
principle, two CO bands are expected, the symmetric CO vibrational mode is predicted to be
significantly less intense compared to the antisymmetric mode, and the frequencies are
predicted to be similar (Table 1).
Therefore, the two CO bands may overlap.
Upon
phylloquinone reduction, the CO band is predicted to downshift 174 cm-1. The most intense
quinoid CC stretching vibration is predicted to downshift by 100 cm-1 and to change from a
"symmetric" stretch (predominantly C2C3 stretching) to an "antisymmetric" stretch (mainly
C9C10 stretching). The predicted isotope shifts in Table 1 are accurate only to approximately
5 cm-1 and, therefore, the expected isotope-induced downshifts for these bands are similar
and less than 8 cm-1. The CO and quinoid normal modes are predicted to have similar
intensities both in the singlet and in the anion spectrum, and thus should be observable in the
isotope-edited spectrum.
Based on comparison to these DFT calculations, bands in Figure 6C are assigned to
A1- and A1 in Table 1. For comparison, previous experimental assignments of phylloquinone
singlet and anion radical bands are summarized in Table 2 (Bauscher et al. 1992; Breton et al.
1994; Breton 1997; Bandaranayake, K. M. et al. 2006a). In our isotope-based assignments, a
negative-positive-negative complex spectral feature is observed with frequencies of 1695
(neg.), 1655 (pos.) and 1637 (neg.) cm-1. The data were simulated by the generation of
Gaussian curves (data not shown). These simulations were dominated by two bands centered
67
at 1688(-) cm-1 and 1685(+) cm-1, which we assign to the singlet CO stretching vibration and
its induced isotope shift. The complexity of the observed spectral lineshape may be due to
overlapping contributions from the CO antisymmetric and symmetric stretching vibrations in
each of two A1 molecules (Table 1). Our scaled DFT calculations predict a frequency of
1672 cm-1 for the CO antisymmetric vibrational mode, supporting this assignment.
The singlet quinoid CC stretching mode is predicted in Table 1 with reasonable
intensity at 1610 (scaled) cm-1. Therefore, the negative spectral feature at 1557 cm-1 in
Figure 6C is assigned to the A1 quinoid ring stretching vibration. The corresponding isotopeshifted component is shifted in the predicted direction, but the magnitude of the shift is larger
than predicted (Table 1). This observation suggests significant motion of the methyl group in
these singlet CO and quinoid normal modes. This is supported by the fact that the calculated
mode also contains some methyl group motion.
As discussed above, the CO stretching vibration is predicted to be the most intense
A1- band (Table 1). In Figure 6C, the only positive band, which is assignable to the A1- CO,
is at 1393 cm-1. Therefore, the broad spectral features (Figure 6C) at positive 1393 cm-1 and
negative 1371 cm-1 are assigned to the CO vibration of the anion and its isotope-shifted
component (Table 1). Broadening of these bands may be due to overlapping contributions
from two A1- molecules and from the antisymmetric and symmetric CO bands. The scaled
CO antisymmetric stretching frequency (Table 1) predicts a frequency of 1498 cm-1 for this
mode. We speculate that the lower than expected intensity and frequency for the CO band
(Table 1) may be due to the asymmetric hydrogen bond, an electrostatic interaction, or to the
pi-pi interaction with the quinone binding site tryptophan (see discussion below). In Figure
6C, candidates for the quinoid CC stretching vibrations are not observed in the expected
68
1500-1400 cm-1 spectral region, perhaps because the isotope shifts are small and the bands
have low intrinsic intensity (Table 1).
3.5 Discussion
In this work, we have incorporated C2H3 into the C2 methyl group of the A1 electron
acceptor, and we have used this approach to assign vibrational bands to A1- and A1. This
experiment was performed in apo-FX PSI preparations, in which A1 is the terminal electron
acceptor (Warren et al. 1990; Warren et al. 1993). Previous studies have shown that
inactivation of the rubA gene resulted in PSI complexes lacking the stromal subunits, PsaC,
PsaD and PsaE. Furthermore, this mutant was unable to assemble FX (Shen et al. 2002a).
This work led to the conclusion that removal of the iron-sulfur clusters does not appreciably
alter the orientation of A1- or the distance between A1- and P700+. However, that study did
find that the motion of A1 in the mutants may be greater due to the absence of FX and the
stromal subunits (Shen et al. 2002b). Previous work studying chemically extracted P700-A1
core proteins have concluded that there is a change in the A1 hydrogen bonding environment
(Warren et al. 1990; Warren et al. 1993; Van der Est et al. 1994; Shen et al. 2002a).
However, our approach of using apo-FX samples has the advantage of simplifying the
spectrum, increasing the amplitude of A1- spectral contributions, and avoiding complex
spectral deconvolution techniques, which are required to study A1- in intact PSI, where the
anion radical has a short (microsecond) lifetime (Brettel, K. et al. 2001).
The PSI crystal structure (Figure 2) reveals that the C1O groups of the A1 acceptors
do not participate in hydrogen bonding. However, the C4O groups accept a hydrogen bond
from the backbone NH group of LeuA722 for A1A or LeuB706 for A1B (Fromme et al. 2001;
69
Jordan et al. 2001; Pushkar, Y. et al. 2004; Grotjohann et al. 2005). EPR studies have shown
asymmetric hydrogen bonding to A1- and an asymmetric distribution of spin density (Shen et
al. 2002b; Pushkar, Y. et al. 2004; Teutloff et al. 2004; Pushkar, Y. N. et al. 2005). This
result suggests that the hydrogen bond to C4O is sustained upon reduction of A1. Evidence
for a pi-pi stacking interaction with a nearby tryptophan comes from analysis of the PSI
structure and from EPR and mutational studies (Hanley et al. 1997; Fromme et al. 2001;
Guergova-Kuras et al. 2001; Jordan et al. 2001; Purton et al. 2001).
Our assignments, which are presented in Table 1, are based on DFT calculations as
well as isotope labeling of the phylloquinone molecule. An analysis of these results and
comparison to previous work is presented below.
3.5.1 Comparison to previous A1-/A1 assignments in apo-FX PSI. Our results, based
on phylloquinone labeling, are congruent with previous interpretations of the apo-FX A1/A1spectrum at room temperature (Table 2) (Hastings et al. 2001). In those experiments, a 1680
cm-1 band was assigned to the C1O stretching of A1; the hydrogen bonded C4O group was
assigned at 1643 cm-1. The previously assigned 1680 cm-1 band may correspond to our 1688
cm-1 band. The previous 1643 cm-1 assignment may correspond to the negative 1637 cm-1
band, which we observe in our data. The A1- CO stretching frequencies were assigned at
1455 and 1445 cm-1 in that previous work (Table 2). A band at 1396 cm-1 was observed, but
assigned to a CC stretching vibration of the anion radical. Because our DFT calculations
indicate that the most intense A1- band will be the CO antisymmetric stretching vibration, we
favor the assignment of our 1393 cm-1 band to a carbonyl stretching vibration. Other,
previously reported bands in the 1500-1400 cm-1 region (Table 2) are predicted by our
calculations to be less isotope-sensitive and are not observed here with significant intensity.
70
3.5.2 Comparison to previous A1-/A1 assignments in intact PSI. Recently, A1-/A1
spectra were generated using time-resolved FTIR measurements at 77 K in intact PSI.
Overall, the spectra exhibit similarities to the A1-/A1 data reported here. However, in this
previous time-resolved work, bands at 1655 and 1607 cm-1 were identified as the A1 CO
stretching vibrations (Table 2), consistent with two distinct carbonyl stretching vibrations for
A1. Quinoid CC stretching and benzene aromatic stretching vibrations were assigned at 1651
and 1634 cm-1. Upon anion formation, the CO stretching and CC stretching vibrational
bands were assigned at 1495 and 1414 cm-1, respectively (Sivakumar et al. 2005). These
assignments are in contrast to ours (Table 1), as well as to previous assignments derived from
FT-IR studies of apo-FX PS1 (Table 2 and (Hastings et al. 2001)). These differences could
arise from the deconvolution of the complex, time resolved FT-IR spectra, which was
interpreted to contain chl A0 keto vibrational bands in the 1680 cm-1 region (Sivakumar et al.
2005).
It is also possible that removal of the polypeptides PsaC, PsaD, and PsaE, and FX
causes subtle changes in the binding pocket of A1, which are reflected in the vibrational
frequencies of A1 and A1- in the apo-Fx preparation.
3.5.3 Comparison to previous studies of model phylloquinone and phylloquinone
anion radicals in vitro. Previous electrochemical FTIR studies of phylloquinone in vitro
assigned two singlet CO stretching vibrations at 1658 and 1616 cm-1 (Table 2) and an
aromatic ring stretching vibration at 1596 cm-1. Upon quinone reduction, the anion CO
stretching band was observed at 1488 cm-1, and the quinoid stretching was not assigned
(Bauscher et al. 1992). However, other FTIR studies of phylloquinone (Table 2) have
observed only one singlet CO stretching band at 1661 cm-1, in agreement with our
71
predictions (Table 1). In this second study, the quinoid stretching was assigned at 1618 cm-1,
and the aromatic ring stretching band was assigned at 1597 cm-1 (Breton et al. 1994).
We assign the A1 CO vibration at 1688 cm-1, the A1 quinoid stretching vibration at
1557 cm-1, and the A1- CO vibration at 1393 cm-1 (Table 1). Compared to previous model
compound work (Table 2), these data suggest significant environmental perturbations of A1
and A1- in PSI. This environmental perturbation upshifts the singlet CO stretching vibration
(by ~30 cm-1), downshifts the singlet quinoid ring vibration (by ~40 cm-1), and downshifts
the radical CO vibration (by ~95 cm-1).
3.5.4 Comparison to previous studies of phylloquinone in the bacterial reaction
center. In the bacterial reaction center, phylloquinone can be substituted and can function as
the primary quinone acceptor, QA. In contrast to the PSI A1 acceptor, in the bacterial reaction
center, two hydrogen bond bonds are formed, one to each QA carbonyl oxygen. In R. viridis,
it has been reported that the QA carbonyl oxygens are hydrogen bonded to the peptide NH of
Ala M258 and to a non-heme iron histidine ligand (Deisenhofer et al. 1989).
In R.
sphaeroides, it has been reported that the quinone carbonyl oxygens are hydrogen bonded to
the peptide NH of Ala M258 and to Thr M222 (Breton et al. 1997). The indole group of a
tryptophan is nearly parallel to the QA ring at a distance of 3.1 Å (Deisenhofer et al. 1989;
Breton et al. 1997).
When phylloquinone was substituted into QA binding site in the bacterial reaction
center, (Breton et al. 1994; Breton 1997) two CO bands were observed, with frequencies at
~1650 and 1640 cm-1 (Table 2). These bands were attributed to the asymmetric CO vibration
of the two different carbonyl groups, which have different frequencies due to differential
72
hydrogen bonding (Breton et al. 1994; Breton 1997). The anion CO band was assigned
either at 1438 or 1444 cm-1, ~40 cm-1 lower than frequency compared to in vitro
phylloquinone studies and ~50 cm-1 higher in frequency compared to our assignments for A1(Table 1 and 2).
3.5.5 Rationalization of environmental perturbations to the A1 and A1- vibrational
spectra.
We suggest that observed perturbations, when PSI is compared to model
compounds or to phylloquinone-substituted bacterial reaction centers, are due to a
combination of the asymmetric hydrogen bonding pattern, electrostatic interactions, and the
pi-pi stacking interaction with a tryptophan (Figure 2). We propose that these proteinquinone interactions have differential effects on the anion and singlet CO vibrational bands
(see below).
From the crystal structure, the secondary electron acceptor A1 is known to be in pi-pi
stacking distance with a tryptophan residue. The distance between the aromatic ring system
of the Trp residues, TrpA697 and TrpB677, and the quinone A1A and A1B range between 33.5 Å respectively (Fromme et al. 2001).
Previous ESEEM studies have suggested a
quinone-tryptophan interaction, because it was shown that the unpaired electron on A1- is
coupled to the nitrogen of tryptophan (Hanley et al. 1997; Van der Est et al. 1997). An
increase in the gx value of A1-, compared to phylloquinone in solution, has been interpreted to
imply that A1 exists in a negative electrostatic environment (MacMillan et al. 1997;
Karyagina et al. 2007). Quantum chemical calculation have suggested that a fixed pi-stacked
arrangement, as observed in the PSI structure, may lower the redox potential of the A1
electron acceptor (Kaupp 2002).
A charged aspartate may also modify the quinone redox
potential (Karyagina et al. 2007).
73
It should be noted that in model studies, symmetric hydrogen bonding in protic
solvents had little effect on phylloquinone ring or CO frequencies (Bauscher et al. 1992). In
phylloquinone-substituted bacterial reaction centers, two hydrogen bonds to phylloquinone
produced a splitting of only 10 cm-1 in the CO vibrational bands (Breton et al. 1994; Breton
1997). In addition, DFT calculations have predicted only small splittings and shifts of the
CO vibrational bands (6-14 cm-1), when an asymmetric hydrogen bond to phylloquinone and
the phylloquinone anion radical is included in a simple model system. Therefore, prediction
of the spectral changes due to the introduction of an unusual, asymmetric hydrogen bond
awaits more detailed theoretical approaches based on the PSI structure, which must also
include the effect of electrostatic or Stark interactions on the vibrational spectrum (Suydam
et al. 2003).
Pi-pi interactions have complex effects on vibrational frequencies, perhaps due to
difficulties in separating the effects of hydrogen bonding, geometry, dispersive, and
electrostatic interactions in many biologically relevant systems (McCarthy et al. 1997;
Plokhotnichenko et al. 1998; Hobza et al. 2000; Hermansson 2002; Sacksteder et al. 2005;
Lee et al. 2006; Saeki et al. 2006). Therefore, the expected effects of pi-pi interactions on
quinone and tryptophan vibrational frequencies and amplitudes must also be addressed in
future studies.
Currently, DNA base stacking provides one possible model system for
assessing the effect of pi-pi interactions. IR studies of poly-uracil/poly-adenine have shown
that shifts of IR bands occur with helix formation. The uracil CO stretching frequency
increased 19 cm-1, and the uracil ring stretching vibration decreased in frequency (Miles et al.
1978). The transition from non-ordered to ordered structures was observed to shift the
carbonyl vibration towards higher frequency (Liquier et al. 1996). Studies of 1,374
dimethyluracil, cytidine, caffeine, inosine and 2’-deoxyadenosine found that stacking of
these molecules also had an effect on the IR spectra. The carbonyl stretching band increased
in frequency, and a decrease in the intensity of the skeletal ring stretching vibrational bands
were observed (Maevsky et al. 1980). These experimental results concerning the effect of pi
stacking on nucleobases are congruent with our A1/A1- assignments.
3.6 Acknowledgments
Supported by NIH GM43273 (B.A.B.) and the Oklahoma Supercomputing Center for
Education and Research (OSCER). The authors thank Dr. J. Golbeck and Dr. R. Agalarov
for performing the transient absorption measurements.
75
3.7 References
Aglarov, R. and B. Klaus (2003). "Temperature dependence of biphasic forward electron
transfer from the phylloquinone(s) A1 in Photosystem I: only the slower phase is
activated." Biochim. Biophys. Acta 1604: 7-12.
Ali, K., S. Santabarbara, P. Heathcote, M. C. W. Evans and S. Purton (2006). "Bidirectional
electron transfer in photosystem I: replacement of the symmetry-breaking tryptophan
close to the PsaB-bound phylloquinone A1B with a glycine residue alters the redox
properties of A1B and blocks forward electron transfer at cryogenic temperatures."
Biochim Biophys Acta 1757: 1623-1633.
Allen, J. P., G. Feher, T. O. Yeates, H. Komiya and D. C. Rees (1988). "Structure of the
reaction center from Rhodabacter sphaeroides R-26: protein-cofactor (quinones and
F2+) interactions." Proc. Natl. Acad. Sci. USA 85: 8481-8491.
Bandaranayake, K. M., V. Sivakumar, R. Wang and G. Hastings (2006). "Modeling the A1
binding site in Photosystem I. Density functional theory for the calculation of "anionneutral" FTIR difference spectra of phylloquinone." Vib. Spectrosc 42: 78-87.
Bandaranayake, K. M., R. Wang and G. Hastings (2006a). "Modification of the
phylloquinone in the A1 binding site in photosystem I studied using time-resolved
FTIR difference spectroscopy and density functional theory." Biochemistry 45: 41214127.
Bandaranayake, K. M., R. Wang, T. W. Johnson and G. Hastings (2006b). "Time-resolved
FTIR difference spectroscopy for the study of photosystem I particles with
plastoquinone-9 occupying the A1 binding site." Biochemistry 45: 12733-12740.
Barry, B. A. (1995). "Tyrosyl radicals in photosystem II." Methods Enzymol. 258: 303-319.
Barry, B. A. and G. T. Babcock (1987). "Tyrosine radicals are involved in the photosynthetic
oxygen evolving system." Proc. Natl. Acad. Sci. USA 84: 7099-7103.
Bauscher, M. and W. Maentele (1992). "Electrochemical and infrared-spectroscopic
characteriziation of redox reaction of p-quinones." J.Phys.Chem. 96: 11101-11108.
Bautista, J. A., F. Rappaport, M. Guergova-Kuras, R. O. Cohen, J. H. Golbeck, J. Y. Wang,
D. Beal and B. A. Diner (2005). "Biochemical and biophysical characterization of
photosystem I from phytoene desaturase and zeta-carotene desaturase deletion
mutants of Synechocystis Sp. PCC 6803: evidence for PsaA- and PsaB-side electron
transport in cyanobacteria." J. Biol. Chem. 280: 20030-20041.
Becke, A. D. (1988). "Density-functional exchange-energy approximation with correct
asymptotic behavior." Phys. Rev. A 38: 3098-3100.
76
Becke, A. D. (1993). "A new mixing of Hartree-Fock and local density functional theories."
J. Chem. Phys. 98: 1372-1377.
Boesch, S. E. and R. A. Wheeler (1995). "Pi-donor substituent effects on calculated
structures and vibrational frequencies of p-benzoquinone, p-fluoranil, and pchloranil." J. Phys. Chem. 99: 8125-8134.
Boesch, S. E. and R. A. Wheeler (1997a). "Pi-Donor Substituent effects on calculated
structures, spin properties, and vibrations of radical anions of p-chloranil, p-fluoranil,
and p-benzoquinone." J. Phys. Chem. A 101: 8351-8359.
Boesch, S. E. and R. A. Wheeler (1997b). "Structures and properties of ubiquinone-1 and its
radical anion from hybrid Hartree-Fock/density functional studies." J. Phys. Chem. A
101: 5799-5804.
Boudreaux, B., F. MacMillan, C. Teutloff, R. Agalarov, F. Gu, S. Grimaldi, R. Bittl, K.
Brettel and K. Redding (2001). "Mutations in both sides of the photosystem I reaction
center identify the phylloquinone observed by electron paramagnetic resonance
spectroscopy." J. Biol. Chem. 276: 37299-37306.
Breton, J. (1997). "Efficient exchange of the primary quinone acceptor QA in isolated
reaction centers of Rhodopsuedomonas viridis." Proc. Natl. Acad. Sci. USA 94:
11318-11323.
Breton, J., C. Boullais, J. R. Burie, E. Nabedryk and C. Mioskowski (1994). "Binding sites of
quinones in photosynthetic bacterial reaction centers investigated by light-induced
FTIR difference spectroscopy: assignment of the interactions of each carbonyl of QA
in Rhodobacter sphaeroides using site-specific 13C-labeled ubiquinone."
Biochemistry 33: 14378-14386.
Breton, J., E. Nabedryk, J. P. Allen and J. C. Williams (1997). "Electrostatic influence of QA
reduction on the IR vibrational mode of the 10a-ester C==O of HA demonstrated by
mutations at residues Glu L104 and Trp L100 in reaction centers from Rhodobacter
sphaeroides." Biochemistry 36: 4515-4525.
Breton, J., E. Nabedryk and W. Leibl (1999). "FTIR study of the primary electron donor of
Photosystem I (P700) revealing delocalization of the charge in P700+ and localization of
the triplet character in 3P700." Biochemistry 38: 11585-11592.
Brettel, K. (1997). "Electron transfer and arrangement of the redox cofactors in Photosystem
I." Biochim. Biophys. Acta 1318: 322-373.
Brettel, K. and J. H. Golbeck (1995). "Spectral and kinetic characterization of electron
acceptor A1 in a Photosystem I core devoid of iron-sulfur centers FX FB and FA."
Photosynth. Res. 45: 183-193.
Brettel, K. and W. Leibl (2001). "Electron transfer in Photosystem I." Biochim. Biophys. Acta
1507: 100-114.
77
Cohen, R. O., G. Shen, J. H. Golbeck, W. Xu, P. R. Chitnis, A. I. Valieva, A. van der Est, Y.
Pushkar and D. Stehlik (2004). "Evidence for asymmetric electron transfer in
cyanobacterial photosystem I: analysis of a methionine-to-luecine mutation of the
ligand to the primary electron acceptor A0." Biochemistry 43: 4741-4754.
Dashdorj, N., W. Xu, R. O. Cohen, J. H. Golbeck and S. Savikhin (2005). "Asymmetric
electron transfer in cyanobacterial Photosystem I: charge separation and secondary
electron transfer dynamics of mutations near the primary electron acceptor A0."
Biophys. J. 88: 1238-1249.
Deisenhofer, J. and H. Michel (1989). "The photosynthetic reaction center from the purple
bacterium Rhodopseudomonas viridis." Science 245: 1463-1473.
Feldman, K. S., D. K. Hester 2nd and J. H. Golbeck (2007). "A relationship between amide
hydrogen bond strength and quinone reduction potential: Implications for
photosystem I and bacterial reaction center quinone function." Bioorg. Med. Chem.
Lett. In press.
Frisch, M. J., G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J.
A. Montgomery, T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyata, J. Fukuda, M. Hasegawa, T. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski,
S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J.
Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J.
A. Pople (2004). Gaussian 03, Revision D.01. Wallingfort, CT, Gaussian, Inc.
Fromme, P., P. Jordan and N. Krauss (2001). "Structure of photosystem I." Biochim Biophys
Acta 1507: 5-31.
Golbeck, J. H. (1994). In The Molecular Biology of Cyanobacteria. Dordrecht, Kluwer
Academic Publishers.
Grafton, A. K. and R. A. Wheeler (1998a). "Vibrational projection analysis: new tool for
quantitatively comparing vibrational normal modes of similar molecules." J. Comput.
Chem. 19: 1663-1674.
Grafton, A. K. and R. A. Wheeler (1998b). "ViPA: A computer program for vector
projection analysis of normal vibrational modes of molecules." Comput. Phys.
Commun. 113: 78-84.
78
Grotjohann, I. and P. Fromme (2005). "Structure of cyanobacterial photosystem I."
Photosynth. Res. 85: 51-72.
Guergova-Kuras, M., B. Boudreaux, A. Joliot, P. Joliot and K. Redding (2001). "Evidence
for two active branches for electron transfer in photosystem I." Proc. Natl. Acad. Sci.
USA 98: 4437-4442.
Hanley, J., Y. Deligiannakis, F. MacMillan, H. Bottin and A. W. Rutherford (1997).
"ESEEM study of the phyllosemiquinone radical A1- in 14N- and 15N-labeled
Photosystem I." Biochemistry 36: 11543-11549.
Hastings, G. and V. Sivakumar (2001). "A fourier transform infrared absorption difference
spectrum associated with the reduction of A1 in Photosystem I: are both
phylloquinones involved in electron transfer?" Biochemistry 40: 3681-3689.
Hehre, W. J., L. Radom, P. v. R. Schleyer and J. A. Pople (1986). Ab initio Molecular
Orbital Theory. New York, John Wiley & Sons.
Helgaker, T. and P. R. Taylor (1995). Modern electronic structure theory, Part II. D. R.
Yarkony. Singapore, World Scientific. 2: 725-856.
Hermansson, K. (2002). "Blue-shifting hydrogen bonds." J. Phys. Chem. A 206: 4695-4702.
Hobza, P. and Z. Havlas (2000). "Blue-shifting hydrogen bonds." Chem. Rev. 100: 42534264.
Holzwarth, A. R., M. G. Muller, J. Niklas and W. Lubitz (2006). " Ultrafast transient
absorption studies on photosystem I reaction centers from Chlamydomonas
reinhardtii. 2: Mutations near the P700 reaction center chlorophylls provide new
insight into the nature of the primary electron donor." Biophys. J. 90: 552-565.
Itoh, S., M. Iwaki and I. Ikegami (2001). "Modification of photosystem I reaction center by
the extraction and exchange of chlorophylls and quinones." Biochim. Biophys. Acta
1507: 115-138.
Joliot, P. and A. Joliot (1999). "In vivo analysis of the electron transfer within Photosystem I:
are the two phylloquinones involved?" Biochemistry 38: 11130-11136.
Jordan, P., P. Fromme, H. T. Witt, O. Klukas, W. Saenger and N. Krauss (2001). "Threedimensional structure of cyanobacterial photoystem I at 2.5 Ǻ resolution." Nature
411: 909-917.
Kamlowski, A., B. Altenberg-Greulich, A. van der Est, S. G. Zeck, R. Bittl, P. Fromme, W.
Lubitz and D. Stehlik (1998). "The quinone acceptor A1 in Photosystem I: binding
site, and comparison to QA in purple bacteria reaction centers." J. Phys. Chem. B 102:
8278-8287.
79
Karyagina, I., Y. Pushkar, D. Stehlik, A. Van der Est, H. Ishikita, E. W. Knapp, B.
Jagannathan, R. Agalarov and J. Golbeck (2007). "Contribution of the protein
environment to the midpoint potentials of the A1 phylloquinones and the FX ironsulfur cluster in Photosystem I." Biochemistry 46: 10804-10816.
Kaupp, M. (2002). "The function of Photosystem I. Quantum chemical insights into the role
of tryptophan-quinone interactions." Biochemistry 41: 2895-2900.
Kim, S. and B. A. Barry (1998). "The protein environment surrounding tyrosyl radicals D•
and Z• in photosystem II: a difference Fourier-transform infrared spectroscopic
study." Biophys. J. 74: 2588-2600.
Kim, S. and B. A. Barry (2000a). "Identification of carbonyl modes of P700 and P700+ by in
situ chlorophyll labeling in photosystem I." J. Am. Chem. Soc. 122: 4980-4981.
Kim, S., J. S. Patzlaff, T. Krick, I. Ayala, R. K. Sachs and B. A. Barry (2000b). "Isotopebased discrimination between the infrared modes of plastosemiquinone anion radicals
and neutral tyrsoyl radicals in photosystem II." J. Phys. Chem. B 104: 9720-9727.
Kim, S., C. A. Sacksteder, K. A. Bixby and B. A. Barry (2001). "A reaction-induced FT-IR
study of cyanobacterial photosystem I." Biochemistry 40: 15384-15395.
Klughammer, C., B. Klughammer and R. Pace (1999). "Deuteration effects on the in vivo
EPR spectrum of the reduced secondary photosystem I electron acceptor A1 in
cyanobacteria." Biochemistry 38: 3726-3732.
Kruip, J., P. Chitnis, B. Lagoutte, M. Rogner and E. J. Boekema (1997). "Structural
organization of the major subunits in cyanobacterial Photosystem I." J. Biol. Chem.
272: 17061-17069.
Lee, C., K. H. Park and M. Cho (2006). "Vibrational dynamics of DNA I. Vibrational basis
modes and couplings." Journal of Chemical Physics 125: 114508.
Lee, C., W. Yang and R. G. Parr (1988). "Development of the Colle-Salvetti correlationenergy formula into a functional of the electron density." Phys. Rev. B 37: 785-789.
Li, Y., A. van der Est, M. G. Lucas, V. M. Ramesh, F. Gu, A. Petrenko, S. Lin, A. N.
Webber, F. Rappaport and K. Redding (2006). "Directing electron transfer within
Photosystem I by breaking H-bonds in the cofactor branches." Proc. Natl. Acad. Sci.
USA 103: 2144-2149.
Liquier, J. and E. Taillandier (1996). Infrared Spectroscopy of Biomolecules. New York,
Wiley-Liss.
MacMillan, F., J. Hanley, L. van der Weerd, M. Knupling, S. Un and A. W. Rutherford
(1997). "Orientation of the phylloquinone electron acceptor anion radical in
Photosystem I." Biochemistry 36: 9297-9303.
80
Maevsky, A. A. and B. I. Sukhorukov (1980). "IR study of base stacking interactions."
Nucleic Acids Res. 8: 3029-3042.
McCarthy, W., J. Smets, L. Adamowicz, A. M. Plokhotnichenko, E. D. Radechenko, G. G.
Sheina and S. G. Stepanian (1997). "Competition between H-bonded and stacked
dimers of pyrimidine: IR and theoretical ab-initio study." Mol. Phys. 91: 513-525.
Miles, H. T. and J. Frazier (1978). "Infrared spectroscopy of polynucleotides in the carbonyl
region in H2O solution: A-U systems." Biochemistry 17: 2920-2927.
Parrett, K. G., T. Mehari and J. H. Golbeck (1990). "Resolution and reconstitution of the
cyanobacterial Photosystem I complexes." Biochim. Biophys. Acta 1015: 341-352.
Parrett, K. G., T. Mehari, P. G. Warren and J. H. Golbeck (1989). "Purification and
properties of the intact P700 and FX containing Photosystem I core protein." Biochim
Biophys Acta 973: 324-332.
Patzlaff, J. S. and B. A. Barry (1996). "Pigment quantitation and analysis by HPLC reverse
phase chromatography: a characterization of antenna size in oxygen-evolving
photosystem II preparations from cyanobacteria and plants." Biochemistry 35: 78027811.
Piccioni, R., G. Bellemare and N. Chua (1982). Methods of polyacrylamide gel
electrophoresis in the analysis and preparation of plant polypeptides. Methods in
Chloroplast Molecular Biology. H. Edelman, R. B. Hallick and N.-H. Chua.
Amsterdam, Elsevier: 985-1014.
Plokhotnichenko, A. M., E. D. Radechenko, S. G. Stepanian and L. Adamowicz (1998).
"Hydrogen-bonded and stacked homo and heterodimers of systems modeling nucleic
acid base pairs." Recent Res. Devel. in Physical Chem. 2: 1087-1104.
Poluektov, O. G., S. V. Paschenko, L. M. Utschig, K. V. Lakshmi and M. C. Thurnauer
(2005). "Bidirectional electron transfer in photosystem I: direct evidence from highfrequency time-resolved EPR spectroscopy." J. Am. Chem. Soc. 127: 11910-11911.
Purton, S., D. R. Stevens, I. P. Muhiuddin, M. C. W. Evans, S. Carter, S. E. J. Rigby and P.
Heathcote (2001). "Site-directed mutagenesis of PsaA residue W693 affects
phylloquinone binding and function in the photosystem I reaction center of
Chlamydomonas reinhardtii." Biochemistry 40: 2167-2175.
Pushkar, Y., D. Stehlik, M. Van Gastel and W. Lubitz (2004). "An EPR/ENDOR study of the
asymmetric hydrogen bond between the quinone electron acceptor and the protein
backbone in Photosystem I." Journal of Molecular Structure 700: 233-241.
Pushkar, Y. N., I. Karyagina, D. Stehlik, S. Brown and A. van der Est (2005). "Recruitment
of a foreign quinone into the A1 site of Photosystem I." J. Biol. Chem. 280: 1238212390.
81
Ramesh, V. M., K. Gibasiewicz, S. Lin, S. E. Bingham and A. N. Webber (2004).
"Bidirectional electron transfer in photosystem I: accumulation of A0- in A-side or Bside mutants of the axial ligand to chlorophyll A0." Biochemistry 43: 1369-1375.
Ramesh, V. M., K. Gibasiewicz, S. Lin, S. E. Bingham and A. N. Webber (2007).
"Replacement of the methionine axial ligand to the primary electron acceptor A0
slows the A0- reoxidation dynamics in photosystem I." Biochim Biophys Acta 1767:
151-160.
Razeghifard, M. R., S. Kim, J. Patzlaff, R. S. Hutchison, T. Krick, I. Ayala, J. Steenhuis, S.
E. Boesch, R. A. Wheeler and B. A. Barry (1999). "In vivo, in vitro and calculated
vibrational spectra of plastoquinone and the plastosemiquinone anion radical." J.
Phys. Chem. B 103: 9790-9800.
Rigby, S. E., M. C. Evens and P. Heathcote (1996). "ENDOR and Special Triple Resonance
Spectroscopy of A1- of Photosystem I." Biochemistry 35: 6651-6656.
Rigby, S. E. J., I. P. Muhiuddin, M. C. W. Evans, S. Purton and P. Heathcote (2002).
"Photoaccumulation of the PsaB phyllosemiquinone in photosystem I of
Chlamydomonas reinhardtii." Biochim. Biophys. Acta 1556: 13-20.
Rippka, R., J. Deruelles, J. B. Waterbury, M. Herdman and R. Y. Stanier (1979). "Generic
assignments, strain histories and properties of pure cultures of cyanobacteria."
Journal of General Microbiology 111: 1-61.
Rustandi, R. R., S. W. Snyder, L. L. Feezel, T. J. Michalski, J. R. Norris, M. C. Thurnauer
and J. Biggins (1990). "Contribution of vitamin K1 to the electron spin polarization in
spinach photosystem I." Biochemistry 29: 8030-8032.
Sacksteder, C. A., S. L. Bender and B. A. Barry (2005). "Role for bound water and CH-pi
aromatic interactions in photosynthetic electron transfer." J. Am. Chem. Soc. 127:
7879-7890.
Saeki, M., H. Akagi and M. Fujii (2006). "Theoretical study on the structure and the
frequency of isomers of the naphthalene dimer." Journal of Chemical Theory and
Computation 2: 1176-1183.
Santabarbara, S., P. Heathcote and M. C. Evans (2005a). "Modeling of the electron transfer
reactions in Photosystem I by electron tunnelling theory: the phylloquinones bound to
the PsaA and the PsaB reaction centre subunits of PSI are almost isoenergetic to the
iron-sulfur cluster F(X)." Biochim Biophys Acta 1708: 283-310.
Santabarbara, S., I. Kuprov, W. V. Fairclough, S. Purton, P. J. Hore, P. Heathcote and M. C.
W. Evans (2005b). "Bidirectional electron transfer in photosystem I: determination of
two distances between P700+ and A1- in spin-correlated radical pairs." Biochemistry
44: 2119-2128.
82
Santabarbara, S., I. Kuprov, P. J. Hore, A. Casal, P. Heathcote and M. C. Evans (2006).
"Analysis of the spin-polarized electron spin echo of the [P700+ A1-] radical pair of
photosystem I indicates that both reaction center subunits are competent in electron
transfer in cyanobacteria, green algae, and higher plants." Biochemistry 45: 73897403.
Schlegel, H. B. (1986). "Optimization of equilibrium geometries and transition structures." J.
Comput. Chem. 3: 214-218.
Scott, A. P. and L. Radom (1996). "Harmonic Vibrational Frequencies: An evaluation of
Hartree-Fock, Moller-Plesset, quadratic configuration interaction, density functional
theory, and semiemperical scale factors." J. Phys. Chem. 100: 16502-16513.
Shen, G., M. L. Antonkine, A. Van der Est, I. R. Vassiliev, K. Brettel, R. Bittl, S. G. Zech, J.
Zhao, D. Stehlik, D. A. Bryant and J. H. Golbeck (2002a). "Assembly of photosystem
I. II. Rubredoxin is required for the in vivo assembly of F(X) in Synechococcus sp.
PCC 7002 as shown by optical and EPR spectroscopy." J Biol Chem 277: 2035520366.
Shen, G., J. Zhao, S. K. Reimer, M. L. Antonkine, Q. Cai, S. M. Weiland, J. H. Golbeck and
D. A. Bryant (2002b). "Assembly of photosystem I. I. Inactivation of the rubA gene
encoding a membrane-associated rubredoxin in the cyanobacterium Synechococcus
sp. PCC 7002 causes a loss of photosystem I activity." J Biol Chem 277: 2034320354.
Sivakumar, V., R. Wang and G. Hastings (2005). "A1 reduction in intact cyanobacterial
photosystem I particles studied by time-resolved step-scan Fourier transform infrared
difference spectroscopy and isotope labeling." Biochemistry 44: 1880-1893.
Slater, J. C. (1974). Quantum Theory of Molecules and Solids. New York, McGraw-Hill.
Suydam, I. T. and S. G. Boxer (2003). "Vibrational stark effects calibrate the sensitivity of
vibrational probes for electric fields in proteins." Biochemistry 42.
Teutloff, C., R. Bittl and W. Lubitz (2004). "Pulse ENDOR studies on the radical pair
P700+A1- and the photoaccumulated quinone acceptor A1- of Photosystem I." Appl.
Magn. Reson. 26: 5-21.
van der Est, A. (2001). "Light-induced spin polarization in type I photosynthetic reaction
centres." Biochim. Biophys. Acta 1507: 212-225.
Van der Est, A., C. Bock, J. H. Golbeck, K. Brettel, P. Setif and D. Stehlik (1994). "Electron
transfer from the acceptor A1 to the iron-sulfur centers in Photosystem I as studies by
transient EPR sepctroscopy." Biochemistry 33: 11789-11797.
Van der Est, A., T. Prisner, R. Bittl, P. Fromme, W. Lubitz, K. Mobius and D. Stehlik
(1997). "Time-resolved X-, K-, and W-band EPR of the radical pair state P700+A1- of
83
Photosystem I in comparison with P865+Q- in bacterial reaction centers." J. Phys.
Chem. B 101: 1437-1443.
Vassiliev, I. R., Y.-S. Jung, M. D. Mamedov, A. Y. Semenov and J. H. Golbeck (1997).
"near-IR absorbance changes and electrogenic reactions in the microsecond-to-second
time domain in Photosystem I." Biophys. J. 72: 301-315.
Vosko, S. H., L. Wilk and M. Nusair (1980). "Accurate spin-dependent electron liquid
correlation energies for local spin density calculations: a critical analysis." Can. J.
Phys. 58: 1200-1211.
Warren, P. V., J. H. Golbeck and J. T. Warden (1993). "Charge recombination between P700+
and A1- occurs directly to the ground state of P700 in a Photosystem I core devoid of
FX FB and FA." Biochemistry 32: 849-857.
Warren, P. V., K. G. Parrett, J. T. Warden and J. H. Golbeck (1990). "Characterization of a
photosystem I core containing P700 and intermediate electron acceptor A1."
Biochemistry 29: 6545-6550.
Wasikowski, C. and S. Klemm (1993). XMOL Version 1.3.1, Research Equipment, Inc.
d.b.a. Minnesota Supercomputer Center, Inc.
Wise, K. E., A. K. Grafton and R. A. Wheeler (1997). "Trimethyl-p-benzoquinone provides
excellent structural spectroscopic and thermochemical models for plastoquinone-1
and its radical anion." J. Phys. Chem. A 101: 1160-1165.
Wise, K. E., J. B. Pate and R. A. Wheeler (1999). "Phenoxyl, (methylthio) phenoxyl, and
(methylthio) cresyl radical models for the structures, vibrations, and spin properties of
the cysteine-Linked tyrosyl radical in galactose oxidase." J. Phys. Chem. B 103:
4764-4772.
Xu, W., P. Chitnis, A. Valieva, A. van der Est, Y. N. Pushkar, M. Krzystyniak, C. Teutloff,
S. G. Zech, R. Bittl, D. Stehlik, B. Zybailov, G. Shen and J. H. Golbeck (2003a).
"Electron transfer in cyanobacterial photosystem I: I. Physiological and spectroscopic
characterization of site-directed mutants in a putative electron transfer pathway from
A0 through A1 to Fx." J. Biol. Chem. 278: 27864-27875.
Xu, W., P. R. Chitnis, A. Valieva, A. van der Est, K. Brettel, M. Guergova-Kuras, Y. N.
Pushkar, S. G. Zech, D. Stehlik, G. Shen, B. Zybailov and J. H. Golbeck (2003b).
"Electron transfer in cyanobacterial photosystem I: II. Determination of forward
electron transfer rates of site-directed mutants in a putative electron transfer pathway
from A0 through A1 to Fx." J. Biol. Chem. 278: 27876-27887.
84
Table 1. Assignments of A1 and A1- Vibrational Bands, Based on Isotopic Labeling of
the Phylloquinone Acceptor and DFT Calculations
Singlet
CO antisymmetric
stretching
CO symmetric
stretching
Aromatic ring
stretching
Quinoid
"symmetric"
stretching**
Quinoid
"antisymmetric"
stretching**
Anion radical
Frequency#
C1H3
Frequency
C2H3
Δ
C2H3
Predicted‡
Frequency
Predicted‡*
Δ C2H3
Predicted‡
Intensity
1688
1685
3
1672
0
279
NA∫
NA
----
1666
0
26
NA
NA
----
1584
1
52
1557
1544
13
1610
3
42
NA
NA
----
1571
1
2
Frequency
C1H3
Frequency
C2H3
Δ
C2H3
Predicted‡
Frequency
Predicted‡
CΔ 2H3
Predicted‡
Intensity
Quinoid
NA
NA
---1510
1
35
antisymmetric
stretching
Quinoid symmetric
NA
NA
---1576
1
3
stretching
Aromatic ring
NA
NA
---1585
0
3
stretching
CO antisymmetric
1393
1371
22
1498
1
310
stretching
CO symmetric
NA
NA
---1483
8
17
stretching
#
Frequencies reported in wavenumber (cm-1)
‡
Predicted frequencies (scaled by 0.9614), isotope shifts (Δ 2H3), and intensities from DFT calculations
*Predicted shifts accurate to + 5 cm-1
∫
NA, not assigned
**Although singlet quinoid stretching modes are labeled “symmetric” and “antisymmetric”, the modes are
mixed and highly localized. Thus, the “symmetric” mode contains predominantly C2C3 stretching, whereas the
“antisymmetric” stretch contains predominantly C9C10 stretching.
85
Table 2. Previous Assignments of Phylloquinone Singlet and Anion Radical IR Bands
In vitro phylloquinone vibrational assignments
Description of normal mode
carbonyl stretching
Solvent and sample type
CH2Cl2 (Bauscher et al. 1992)
Singlet (cm-1)
1658, 1616
Anion Radical (cm-1)
1488
quinoid stretching
CH2Cl2 (Bauscher et al. 1992)
1596
aromatic ring stretching
CH2Cl2 (Bauscher et al. 1992)
1596
NA
carbonyl stretching
Film (Breton et al. 1994)
1661
NA
quinone stretching
Film (Breton et al. 1994)
1618
NA
aromatic stretching
Film (Breton et al. 1994)
1597
NA
∫
NA
Bacterial reaction center vibrational assignments
Description of normal mode
carbonyl stretching
Bacterial reaction center sample
Rb. viridis (Breton 1997)
Singlet (cm-1)
1653,1637
Anion Radical (cm-1)
1438
quinoid stretching
Rb. viridis (Breton 1997)
NA
NA
aromatic ring stretching
Rb. viridis (Breton 1997)
1590
NA
carbonyl stretching
Rb. sphaeroides (Breton et al. 1651, 1640
1994)
Rb. sphaeroides (Breton et al. 1608
1994)
Rb. sphaeroides (Breton et al. 1588
1994)
Photosystem I vibrational assignments
quinoid stretching
aromatic ring stretching
1444
1478, 1394-1388
1478, 1394-1388
Description of normal mode
carbonyl stretching
Photosystem I sample
Intact (Sivakumar et al. 2005)
Singlet (cm-1)
1655, 1607
Anion Radical (cm-1)
1495
quinoid stretching
Intact (Sivakumar et al. 2005)
1651, 1634
1414
aromatic ring stretching
Intact (Sivakumar et al. 2005)
1651, 1634
1414
carbonyl stretching
Apo-FX (Hastings et al. 2001)
1680, 1643
1455, 1445
quinoid stretching
Apo-FX (Hastings et al. 2001)
NA
1396, 1381
aromatic ring stretching
Apo-FX (Hastings et al. 2001)
NA
1396, 1381
∫
NA, not assigned
86
Figure 1. The structure of the secondary electron acceptor A1 in photosystem I. Numbering
is according to IUPAC nomenclature. Phy represents the phytol tail bonded to carbon 3.
87
Figure 2. The binding site of A1A left and A1B right in Photosystem I from the PSI crystal
structure at 2.5Å.resolution (Jordan et al. 2001) The tryptophan residues are within πstacking distance (3-3.5 Å) of the quinone ring.
88
Figure 3.
Decay of P700+, as assessed by 820 nm kinetic measurements, in (A) natural
abundance, C1H3-A1 intact PSI, (C) natural abundance, C1H3-A1 apo-FX PSI, (B) labeled
C2H3-A1 intact PSI, and (D) labeled C2H3 apo-FX PSI. The reaction medium contained 50
mM Tris buffer (pH 8.3), 0.04% n-dodecyl-β-D-maltoside, 4 μM DCPIP, and 5 mM sodium
ascorbate. See Materials and Methods for additional experimental details.
89
Figure 4. Light-induced FT-IR difference spectra of (A) natural abundance, apo-FX and (B)
natural abundance, intact PSI. Spectra were recorded either at room temperature (20oC, solid
lines) or at -10oC (dotted lines). The data in (B) were divided by a factor of 3 to correct for a
difference in the amount of stable charge separation. The room temperature A1--minus-A1
spectrum in (C) was constructed by subtracting (B, solid line) from (A, solid line). The
-10°C A1--minus-A1 spectrum in (D) was constructed by subtracting (B, dotted line) from (A,
dotted line). Spectrum (E) is room temperature, dark-minus-dark control, which was derived
from the apo-FX PSI data set. The solid lines in A, B, and E are the average of 21, 22, and 21
spectra, respectively. The dotted lines in A and B are the average of 45 and 27 spectra
respectively. See Materials and Methods for additional experimental details.
90
Figure 5. Light-induced FTIR difference spectra of (A) natural abundance C1H3-A1, apo-FX
PSI and (B) labeled C2H3-A1, apo-FX PSI. The isotope-edited spectrum shown in (C) is a
C1H3-minus-C2H3 double difference spectrum created by subtraction of (B) from (A) and
multiplying by a factor of 10.
Spectrum (D) is a control double difference spectrum
generated by taking one-half of the data in A, subtracting the other half of the data set, and
dividing by √2. Spectrum (E) is a dark-minus-dark control. A, B, D, and E are the average
of 45 spectra. See Materials and Methods for additional experimental details.
91
Figure 6. Isotope-edited FT-IR spectra of PSI. (A) shows the C1H3-minus-C2H3 double
difference spectrum, acquired from apo-FX PSI, (B) shows the C1H3-minus-C2H3 double
difference spectrum, acquired from intact PSI, and (C) shows a subtraction of (A) minus (B),
revealing isotopic shifts in A1- (positive) and A1 (negative) bands. The spectra in A and B
were normalized to give the same intensity in the P700+ ester vibrational bands between 1752
and 1748 cm-1. Spectrum (D) is a control double difference spectrum generated by taking
one-half of the data in A, subtracting the other half of the data set, and dividing by √2. See
Materials and Methods for additional experimental details.
92
CHAPTER 4
LIGHT-INDUCED DYNAMICS IN PHOTOSYSTEM I ELECTRON
TRANSFER
Reproduced with permission from “Light-induced Dynamics in Photosystem I Electron
Transfer,” Bender, S. L. and B.A. Barry., Biophysical Journal, 95; 3927-3934. Copyright
2008
4.1 Abstract
Protein dynamics are likely to play important, regulatory roles in many aspects of
photosynthetic electron transfer, but a detailed description of these coupled protein
conformational changes has been lacking. In oxygenic photosynthesis, photosystem I
(PSI) catalyzes the light-driven oxidation of plastocyanin or cytochrome c and the
reduction of ferredoxin. A chl a/a' heterodimer, P700, is the secondary electron donor, and
the two P700 chl, are designated PA and PB. In this work, we have used specific chl
isotopic labeling and reaction-induced FT-IR spectroscopy to assign chl keto vibrational
bands to PA and PB. In the cyanobacterium, Synechocystis sp. PCC 6803, the chl keto
carbon was labeled from
from
18
13
C-labeled glutamate, and the chl keto oxygen was labeled
O2. These isotope-based assignments provide new information concerning the
structure of PA+. PA+ is found to give rise to two chl keto vibrational bands, with
frequencies at 1653 and 1687 cm-1. In contrast, PA gives rise to one chl keto band at 1638
93
cm-1. Observation of two PA+ keto frequencies is consistent with a protein relaxationinduced distribution in PA+ hydrogen bonding. These results suggest a light-induced
conformational change in PSI, which may regulate the oxidation of soluble electron
donors and other electron transfer reactions.
This work provides novel information
concerning the role of protein dynamics in oxygenic photosynthesis.
94
4.2 Introduction
In oxygenic photosynthesis, Photosystem I (PSI) is a multi-subunit, membrane
protein complex, which carries out the light-induced reduction of ferredoxin and the
oxidation of plastocyanin or cytochrome c (Golbeck et al. 1991). These redox events are
initiated by light-induced electron transfer, which leads to the oxidation of a chlorophyll
dimer, P700, and the reduction of an iron sulfur cluster, FB.(Brettel et al. 2001; Jordan et
al. 2001) Other electron transfer cofactors include the primary chlorophyll donor (eC-2
or eC-3), two A0 chlorophyll and two A1 phylloquinone electron acceptors, and two
additional iron-sulfur clusters, FX and FA. The PSI reaction center contains two large
intrinsic polypeptides, PsaA and PsaB (Jordan et al. 2001). The 2.5 Å resolution crystal
structure reveals a pseudo-C2 symmetric arrangement of the accessory chl molecules and
of the two A0 and the two A1 electron acceptors. These two branches of electron transfer
cofactors are referred to as the PsaA and PsaB branches (Jordan et al. 2001). Studies
have suggested that both branches carry out light-induced electron transfer (Joliot et al.
1999; Guergova-Kuras et al. 2001).
Recent studies have suggested that charge separation begins on an accessory chl
monomer (either eC-2 or eC-3), which donates its electron to an A0 acceptor (Holzwarth
et al. 2006). This primary donor is reduced by P700, a chl a/a’ heterodimer, which is the
terminal electron donor in the PSI reaction center (Webber et al. 2001). P700 is present in
the dark, and P700+ is produced under illumination. P700 is composed of a chl a molecule,
designated PB, and of a chl a’ molecule, designated PA. While the X-ray structure
supports the interpretation that PA has hydrogen bonding interactions with its protein
95
environment, PB does not appear to be hydrogen bonded (Jordan et al. 2001; Grotjohann
et al. 2005).
The soluble copper-containing protein, plastocyanin, acts as the electron donor to
P700+ in plants, algae, and cyanobacteria. Under copper limiting conditions, a cytochrome
c can replace plastocyanin as the electron donor in cyanobacteria (Merchant et al. 1986).
In plants and algae, plastocyanin docks with the luminal side of PSI to form an
interaction complex, in which electron transfer from the reduced copper to P700+ occurs
on the microsecond time scale (Bottin et al. 1985; Hippler et al. 1997). In cyanobacteria,
the cytochrome, but not plastocyanin, has been proposed to form an electron transfer
complex in vivo (Durán et al. 2004). In both cyanobacteria and algae, mutations in a
hydrophobic region near P700+, which contains luminal PsaA and PsaB helices and a
conserved tryptophan dimer, disrupt electron transfer (Sun et al. 1999; Sommer et al.
2002; Sommer et al. 2004).
It has been proposed that conformational rearrangements
may regulate electron transfer in the interaction complex {(Bottin et al. 1985; Sigfridsson
et al. 1997), but see (Drepper et al. 1996)}.
Previously, site-directed mutagenesis and various spectroscopic techniques have
been used to study the interaction of the cation radical, P700+, with its protein environment
(Breton 2001; Brettel et al. 2001; van der Est 2001; Webber et al. 2001; Grotjohann et al.
2005). EPR studies have concluded that the unpaired spin is localized mainly on PB,
while FT-IR studies have concluded that the spin-density is evenly distributed over the
dimer (Breton et al. 1999; Kaess et al. 2001). This discrepancy may be due to unsolved
questions concerning the assignment of the FT-IR vibrational bands, as described below.
96
Recent experimental (Wang et al. 2007) and previous theoretical (Warshel et al.
1989; Gehlen et al. 1994; Parson et al. 2004) studies have underscored the role of protein
dynamics in the control of primary electron transfer in the bacterial reaction center. In
the chl-containing reaction center, photosystem II, an influence of protein dynamics on
the photosynthetic water-splitting cycle has also been suggested by vibrational
spectroscopic experiments on different time scales (Halverson et al. 2003; Barry et al.
2006; De Riso et al. 2006). In PSI, conformational dynamics have the potential to
regulate bidirectionality of electron transfer in the reaction center and also to control the
mechanism of plastocyanin/cytochrome c oxidation.
The effect of light-induced
dynamics on P700 and P700+ can be detected in real time using vibrational spectroscopic
techniques, such as FT-IR spectroscopy.
Previously, evidence for structural
heterogeneity in P700/P700+ has been obtained by methyl ester labeling of chl (Kim et al.
2000).
The light-minus-dark, P700+-minus-P700, FT-IR spectrum is dominated by keto
vibrational bands from chl and chl+, which, in model compounds, are upshifted in
frequency by oxidation (Nabedryk et al. 1990; Breton et al. 1999; Hastings et al. 2001).
Keto vibrational frequencies are expected to reflect changes in polarity, hydrogen
bonding, and electrostatics (Bellamy 1980). Previously keto vibrational bands have been
assigned to PA, PA+, PB, and PB+ in the P700+-minus-P700 spectrum. The observation of
four keto vibrational bands has been attributed to the oxidation-induced change in force
constant, as well as differences in hydrogen bonding interactions between the two halves
of the dimer, PA and PB (Breton et al. 1999; Hastings et al. 2001; Witt et al. 2002; Li et al.
2004; Pantelidou et al. 2004).
However, there has also been controversy over the
97
assignment of the keto FT-IR bands of P700 and P700+ (Breton et al. 1999; Hastings et al.
2001), and the spectrum still contains unassigned bands. Therefore, the development of
methods to specifically isotopically label chl, with minimal interference from protein
labeling, is of particular importance.
In this work, we describe methods to
13
C and
18
O label the chl keto position
(Figure 1) through manipulation of the cyanobacterial chl biosynthetic pathway (Beale
1999; Porra et al. 2001). These isotopically labeled samples were used to assign the 131keto stretching bands of the P700+-minus-P700 FT-IR spectrum.
These data reveal
unexpected cofactor-protein relaxation events, which accompany light-induced electron
transfer in PSI.
4.3 Materials and Methods
For
13
C chl labeling, glutamate-tolerant Synechocystis PCC 6803 cultures were
grown on solid media containing BG-11 (Rippka et al. 1979), 250 μM L-glutamic acid
buffered with 1mM TES-NaOH pH 8.0, 5 mM TES-NaOH, pH 8.0, and 6 mM Na2S2O3
(Barry 1995). Liquid cultures (400 ml) were grown in BG-11, 5 mM TES-NaOH, pH
8.0, and 5 μg/ml kanamycin and were bubbled with sterile air. The cultures were
supplemented with 1 mM L-glutamic acid or L-glutamic-3-13C acid (99% enrichment,
Isotec Inc., Miamisburg, OH), which were buffered with 1 mM TES-NaOH, pH 8.0.
Cultures were harvested at an OD730 nm of 1.1-1.4.
For 18O chl labeling, Synechocystis PCC 6803 cultures were grown on solid media
containing BG-11, 5 mM TES-NaOH, pH 8.0, 6 mM Na2S2O3, and 5 mM glucose..
Liquid cultures were grown as described above, except that the cultures were
98
continuously shaken on a cell culture agitator (New Brunswick Scientific, Edison, NJ).
CO2 was bubbled into the cell cultures at a rate of 1 ml/min for 5 minutes daily.
Immediately following this, 16O2 or 18O2 (99% enrichment, Isotec Inc., Miamisburg, OH)
was bubbled into the cell cultures at a rate of 1 ml/min for 5 minutes. Cultures were
grown a total of 10 days. The OD730 for the cultures ranged from 1.0-1.1.
4.3.1 Purification of PSI. Thylakoid membranes were solubilized (Noren et al.
1991), and trimeric PSI was purified by ion exchange liquid chromatography using a
Mono QTM HR 5/5 column (Amersham Biosciences, Arlington, IL) and an ÄKTA
explorer instrument (Amersham Biosciences, Arlington, IL) (Barry 1995). The fraction
size collected was 0.5 ml, and the visible absorption spectrum of each fraction was
recorded on a Hitachi (San Jose, CA) U-3000 UV-visible spectrophotometer using 1cm
path length cuvettes.
Fractions with absorption maxima greater than 679 nm were
pooled. Purified trimeric PSI was dialyzed overnight in 5 mM HEPES-NaOH pH 7.5,
0.04% dodecyl-ß-D-maltoside (LM) and then concentrated using an Amicon (Bedford,
MA) Ultra 100,000 MWCO centrifugal filter device to a final concentration of 2-3 mg
chl/ml.
4.3.2 FTIR Spectroscopy. FT-IR spectra were collected at -10 °C as described
previously (Sacksteder et al. 2005; Bender et al. 2008).
Samples contained 2 mM
potassium ferricyanide and 2 mM potassium ferrocyanide and were concentrated at room
temperature under a steady flow of nitrogen gas.
Concentration times were 20-30
minutes. Spectral conditions were as follows: resolution, 4 cm-1; zero filling, 1; data
acquisition time, 4.0 min.
Difference spectra (light-minus-dark) were generated by
taking the ratio of single beam spectra collected before and during illumination and
99
converting to absorbance. FT-IR spectra were obtained in the dark or under continuous
illumination with red and heat-filtered light, as previously described (Kim et al. 2001;
Sacksteder et al. 2005; Bender et al. 2008). A 60 min dark relaxation time was used
between successive illuminations.
Spectra were normalized to account for small
differences in sample concentration and path length, using the amplitude of the amide II
band in the infrared absorption spectrum.
All spectra were normalized to 0.5 AU.
Infrared absorption spectra were created by use of a background.
Isotope-edited spectra were generated by subtracting the isotope-labeled
difference spectrum from the natural abundance difference spectrum. For the 18O-isotope
edited spectrum (Fig. 3C), a direct one-to-one subtraction (C, dotted line), as well as an
interactive subtraction (C, solid line), were performed. A subtraction factor of 0.9663
was chosen (Fig. 3C, solid line) in the interactive subtraction to minimize ester
contributions, relative to the keto vibrational bands. For the 13C-isotope edited spectrum
(Fig. 4C), a direct one-to-one subtraction (C, dotted line) as well as an interactive
subtraction (C, solid line) were performed. A subtraction factor of 0.9584 was chosen
(Fig. 4C, solid line) in the interactive subtraction to minimize ester contributions relative
to the keto vibrational bands. In Figs. 3C and Fig. 4C, the
18
O-isotope edited spectrum
was multiplied by 8 and the 13C-isotope edited spectrum was multiplied by 5 to equalize
spectral amplitudes for comparison.
In Figure 5, the interactive subtractions are
displayed. To estimate the level of noise in the isotope-edited spectra, control double
difference spectra (Figs. 3D and 4D) were generated by taking one-half of the control
data set (either 12C or 16O-chl PSI), subtracting from it the second half of the data set, and
dividing by the √2. Dark-minus-dark spectra were generated from data recorded before
100
illumination (Figs. 3E and Fig. 4E). Fig. 6A shows an overlay of Fig. 3A (blue, natural
abundance) and Fig. 3B (black, 18O-labeled); Fig. 6B shows an overlay of Fig. 4A (blue,
natural abundance) and Fig. 4B (black, 13C-labeled).
4.3.3 Assessment of chlorophyll (chl) isotopic enrichment. To determine the
amount of isotope incorporation into chl from cell cultures grown with L-glutamic-3-13Cacid or
18
employed.
O2, matrix-assisted laser desorption/ionization mass spectrometry was
The control samples were cultures grown in the presence of natural
abundance L-12C-glutamic acid or
16
O2. Chl was extracted from PSI samples with a
solution of 80% acetone/20% methanol. The extracted pigments were vortexed,
sonicated, centrifuged, and filtered as previously described (Patzlaff et al. 1996; Bender
et al. 2008). For NMR spectroscopy, chl was HPLC purified (Patzlaff et al. 1996). Chl
was concentrated and dried using a Thermo-Savant SpeedVac concentrator with VLP120
pump and RVT 400 refrigerated vapor trap (Thermo Electron Corporation, Waltham,
MA). Chl samples were stored at -70°C until use. A 4700 Proteomics Analyzer (Applied
Biosystems Foster City, CA) was used for the mass spectral measurements (Bender et al.
2008). Immediately before the measurement, chl was dissolved in a mixture of 80%
methanol, 19% water, and 1% acetic acid; this treatment removes the Mg+2 ion of chl and
generates pheophytin. The matrix used was alpha-cyano-4-hydroxy cinnamonic acid.
The 871.5 m/z peak of pheophytin was used for quantitation. Data were analyzed using
IGOR (Wavemetrics, Lake Oswego, OR) and Excel (Microsoft, Redmond, WA)
software. The data were normalized to the [M+H] peak before the isotope distribution
was calculated. Measurements were repeated twice for 12C-chl and 13C-chl, three times
101
for 18O-chl, and four times for 16O-chl, the error in the measurements was estimated from
the range or standard deviation of the measurements.
4.3.4
18
Mass spectral data from the
O labeling of chl.
subtracted from data derived from the
18
16
O2 control were
O2 labeled sample. Analysis of the remaining
873.5 peak {M+H+2} amplitude showed that 15 ± 6% of chl was labeled from 18O2.
4.3.5 13C labeling of chl. In order to calculate the amount of
13
C incorporation,
mass spectral data derived from the natural abundance sample (Table 1, column 2) were
subtracted from the 13C labeled data (Table 1, column 3). The resulting values are given
in Table S1, column 4. The 872.5 m/z [M+H+1] peak amplitude gives the amount of the
+
1 chl isotopomer, generated by
13
C-glutamate labeling (43 + 8%).
positions are expected to be labeled from
13
Because eight
C glutamate, the probability the 131 keto
position is labeled in the +1 isotopomer was derived by dividing by eight (5%, column
13).
To calculate the amount of simultaneous labeling at two chl positions, generating
the
+
2 isotopomer, the 873.5 [M+H+2] peak amplitude was employed.
The
+
1
isotopomer makes a significant contribution to the +2 isotopomer peak amplitude, which
was factored out using the relative, natural abundance peak intensities (Table 4, column
2). Note that the observed, natural abundance and the predicted isotope distributions are
similar (Table 4, compare column 2 and Table legend). This subtraction shows that 13C
glutamate labeling generated the +2 chl isotopomer in 27 + 7% of the sample (Table 4,
column 6). Because eight positions are expected to be labeled, the probability that the
keto position is labeled in the +2 isotopomer is calculated by division by 2/8 (Table 4,
102
column 13). These calculations were repeated for all observed ion signals between
873.5-877.5 m/z, and the probabilities were summed to estimate the total amount of 13C
label at the keto position.
4.3.6 GC/MS analysis of
determine the amount of
13
13
C isotope incorporation into PSI amino acids. To
C incorporation into amino acids, PSI samples were acid
hydrolyzed, derivatized, and subjected to GC/MS as previously described (Kitson et al.
1996; Sachs et al. 2003). The derivatized amino acids were identified by retention time,
and the characteristic ions for each amino acid were monitored (Table 4) (Kitson et al.
1996). The extent of isotope incorporation was determined by comparison to natural
abundance PSI samples. The analysis was performed in duplicate, and the error in the
measurement was estimated from the range.
4.4 Results
To label the oxygen of the 131 chl keto group, Synechocystis sp. PCC 6803
cyanobacterial cultures were grown in the presence of
18
O2. The formation of the chl a
isocyclic ring is an aerobic process, in which the 131 oxo group (Figure 1) is derived from
molecular oxygen using an oxygenase (Wong et al. 1985; Walker et al. 1989; Schneegurt
et al. 1992; Bollivar et al. 1995, 1996). The amount of
18
O labeling was measured by
mass spectrometry and found to be 15 + 6% (Figure 2A and Materials and Methods).
The keto position of chl (Figure 1) is the only group that is labeled by this protocol
(Wong et al. 1985; Walker et al. 1989; Schneegurt et al. 1992; Bollivar et al. 1995).
Reaction-induced FT-IR spectra were acquired from control and 18O-labeled PSI
samples (Figure 3). The difference spectra, constructed by subtraction of data recorded
103
under illumination and in the dark, are attributable to P700+FB--minus-P700FB (Breton et al.
1999; Bender et al. 2008). Figure 3A shows the difference spectrum from
and Figure 3B shows the difference spectrum from
18
16
O-chl PSI
O-chl PSI. The isotope-edited
spectrum presented in Figure 3C exhibits keto stretching contributions from P700
(negative) and P700+ (positive) (Nabedryk et al. 1990; Breton et al. 1999; Breton 2001;
Hastings et al. 2001). Spectral changes associated with the labeling of the chl keto
oxygen can be identified by an interactive subtraction of control and
18
O-labeled data.
This interactive subtraction minimizes any small contributions from the chl ester
1754(+)/1749(-) cm-1 bands (Kim et al. 2000). Small differences in amplitude in this
region can arise from minor alterations in the amount of stable charge separation (Figure
3). The resulting isotope-edited spectrum (Figure 3C) will have contributions only from
vibrational bands perturbed by P700 oxidation and by the incorporation of the 18O isotope
at the keto position. In the isotope-edited spectrum, natural abundance P700 bands are
negative, and natural abundance P700+ bands are positive.
The isotope-shifted
components for each vibrational band have the opposite sign. Observed spectral features
in Figure 3C are significant compared to the noise, which is estimated from a controlminus-control spectrum (Figure 3D).
To interpret the
18
O isotope-edited spectrum, incorporation of
13
C at the 131
carbon (Figure 1) is advantageous, because 131/13C labeling will also shift spectral bands
assignable to the keto stretching vibration.
In cyanobacterial chl biosynthesis,
aminolevulinic acid, the first universal tetrapyrrole precursor, is formed from glutamate,
which is activated by ligation to a tRNA molecule (Beale et al. 1975; Porra et al. 1983;
Huang et al. 1986; Schön et al. 1986; Rieble et al. 1988). In our experiments, carbon
104
labeling at the chl 131 position was accomplished by growth of a glutamate-tolerant strain
of Synechocystis sp. PCC 6803 in the presence of 1 mM glutamate, which was
13
C
labeled at the third carbon (Table 1). Uptake of this isotope is expected to label the 131
carbon of chl (Figure 1) and seven other chl carbons (Figure 1) (Beale 1999).
The
vibrational bands, which are altered by labeling at the non-keto positions, are expected to
be distinct in frequency from the keto vibrational bands (Boldt et al. 1987). Further, only
a keto vibrational band will be shifted by both
18
O and 131/13C labeling, and CO
vibrational bands will be in common between the two types of isotope-edited spectra.
Gas chromatography-mass spectral analysis on hydrolyzed PSI demonstrated that there
was little scrambling of the 13C glutamate label into other amino acids (Table 2 (Sachs et
al. 2003)). Mass spectrometry (Figure 2B and Table 1) revealed that singly, doubly, and
triply
13
C-labeled chl isotopomers were generated in significant yield (78%).
For
example, the yield of the +1, +2, and +3 13C isotopomers was determined to be 43 + 8%,
27 + 7%, and 8 + 5%. Only 2 + 4% of the +4 isotopomer was detected (Table 2). Using
these measured values, the total probability of labeling at the keto carbon position can be
estimated as 16% (Table 1 and Materials and Methods). NMR spectroscopy on purified,
extracted 13C-labeled chlorophyll in organic solvents (Patzlaff et al. 1996) confirmed the
incorporation of label into the keto position (data not shown).
Reaction induced FT-IR spectra were acquired from these
13
C labeled PSI samples
(Figure 4). Figure 4A is the difference spectrum acquired from 12C-chl PSI and Figure
4B is the difference spectrum acquired from 13C-chl PSI. Figure 4.4C is the 13C isotopeedited spectrum, which exhibits significant features relative to the noise (Figure 4D) and
was produced from an interactive subtraction. In Figure 4C, related natural abundance
105
and isotopomer keto bands, which are expected to be of opposite sign, are labeled with
the same color code. Figure 5 compares the
18
O isotope-edited spectrum (Figure 5A,
solid line) and the 13C isotope-edited spectrum (Figure 5B solid line). Bands assigned to
natural abundance PB and PB+ are filled with vertical lines, and bands assigned to natural
abundance PA and PA+ are filled with horizontal lines (Figure 5).
The assignment of
bands in the 18O and 13C isotope-edited spectra is summarized in Table 3.
Bands at (+) 1718 and (-) 1698 cm-1 are observed in Figure 5A and 5B, consistent
with an assignment to a chl keto vibrational band, which is upshifted by light-induced
electron transfer. Based on the signs, the positive 1718 cm-1 band would be assignable to
P700+; the negative 1698 cm-1 band would be assignable to P700. For these bands, an
isolated, CO harmonic oscillator approximation predicts frequency downshifts of ~40
cm־1 either with
13
C or
18
O labeling.
The isotope-shifted components for the (+) 1718
cm-1 band are observed at (-) 1681 cm-1 in the 18O labeled samples (Figure 5A, red labels)
and (-) 1679 cm-1 in the
13
C labeled samples (Figure 5B, red labels), respectively. The
isotope-shifted components for the (-) 1698 cm-1 band are observed at (+) 1666 cm-1 in
the
18
O labeled (Figure 5A, blue labels) and (+) 1663 cm-1 in the
13
C labeled samples
(Figure 5B, blue labels), respectively. Therefore, the derived 13C and 18O isotope shifts
for the 1718 and 1698 cm-1 bands are 37-39 cm-1 and 32-35 cm-1 (Figure 5) respectively,
in reasonable agreement with the harmonic oscillator predictions for an isolated CO
vibrational mode.
In previous studies, spectral features at (+) 1717 and (-) 1698 cm-1
were assigned to the non-hydrogen bonded, keto band of PB+ and PB, respectively (Breton
et al. 1999; Hastings et al. 2001), and our isotope-based assignments are consistent with
that previous work.
106
Bands at (+) 1653 and (-) 1638 cm-1 are observed in Figure 5A and Figure 5B,
consistent with an assignment to a chl keto vibrational band, which is perturbed by lightinduced electron transfer. Based on the signs, the positive 1653 cm-1 band would be
assignable to P700+; the negative 1638 cm-1 band would be assignable to P700. For these
bands, the harmonic oscillator approximation predicts frequency downshifts of ~40 cm-1.
The isotope-shifted components for the (+) 1653 cm-1 band are observed at (-) 1608 cm-1
(Figures 5A and 5B, black labels). The isotope-shifted components for the (-) 1638 cm-1
band are observed at (+) 1602 (Figure 5A, orange labels) and (+) 1593 (Figure 5B,
orange labels) cm-1. Therefore, the derived 13C and 18O isotope shifts for the (+) 1653 and
(-) 1638 cm-1 bands are 45 cm-1 and 36-45 cm-1, respectively, in reasonable agreement
with the harmonic oscillator predictions.
In previous studies, spectral features at (+)
1653/4 and (-) 1638 cm-1 were assigned to hydrogen bonded, keto bands of PA+ and PA,
respectively (Breton et al. 1999; Hastings et al. 2001), and our isotope labeling work is
consistent with those assignments.
Based on model compounds (Nabedryk et al. 1990), PA, PA+, PB, PB+ would be
expected to give rise to one chl keto vibrational band each. Four keto bands with distinct
frequencies are assigned in Table 3, based on the arguments described above. However,
in each of the isotope-edited spectra presented in Figures 5A and 5B, there is still an
unassigned positive band at 1687 cm-1 and an unassigned negative band at 1644/3 cm-1
(green labels). If the 1687 cm-1 band arises from a chl keto vibration, a harmonic
oscillator treatment predicts a frequency of 1645 and 1648 cm-1 for the
isotopically labeled vibrational band, respectively.
18
O and
13
C
This comparison supports the
interpretation that the positive 1687 cm-1 band is a keto vibrational band of P700+ and that
107
the 1644/3 cm-1 band is its isotope-shifted component. Therefore, we assign the 1687
cm־1 spectral feature to a second keto vibrational band of PB+ or PA+. Based on previous
site-directed mutagenesis work, we favor the assignment of this band to PA+ (see
discussion below). The observation of two bands for PA+, one at 1687 cm-1 and one at
1653 cm-1, and one band for PA, at 1638 cm-1, suggests that light-induced electron
transfer leads to a structural change in the PSI reaction center. This structural change
results in at least two distinct protein environments for the PA half of the P700 dimer.
4.5 Discussion
In this paper, we employ methods to incorporate
13
C and
18
O isotopic labels
specifically into the chl 131 keto group. Cyanobacterial cultures were used, and
13
C
glutamate and 18O2 were the source of the two labels. The amount of labeling at the keto
position was found to be 15-16%, which is sufficient for detection using FT-IR
spectroscopy. Figure 6 shows a comparison of the 1730-1590 cm-1 region in lightinduced FT-IR spectra, associated with the oxidation of P700. Amplitude changes are
observed when the difference spectra of natural abundance and isotope labeled PSI are
overlaid. In addition, chl labeling occurred with minimal scrambling into amino acid
biosynthetic pathways.
Figure 7A shows the amino acids within 4 Å of the 131 keto group of PA and PB
(Jordan et al. 2001). Ring V of PA is not planar with the rest of molecule, and the 131
keto group of PA is predicted to participate in a hydrogen bond with a conserved
threonine residue, Thr 743 (S. elongatus numbering). There are no analogous hydrogen
bonds predicted on the PB half of the P700 dimer (Figure 7A). Therefore, the keto
108
vibrational bands of PA/PA+ should be distinguishable from PB/PB+ bands. Previous
assignments of the 131 keto stretching vibrations of PB, PA, PB+, and PA+ have been based
on mutagenesis studies, global
2
H incorporation, global
15
N incorporation, and
comparison to model compounds (Nabedryk et al. 1990; Breton et al. 1999; Hastings et
al. 2001). These approaches have led to some discrepancies in the interpretation of the
PSI difference spectrum. Specific isotopic labeling, as performed here, has clarified the
assignments. Four of our isotope-based PB, PB+, PA, and PA+ assignments are consistent
with previous work (Table 3). However, we have identified an additional positive band
at 1687 cm-1, which is a chl keto vibration, based on its
13
C and
18
O isotope-induced
downshifts. The sign of this band indicates that it arises from PB+ or PA+. This positive
band has been observed previously in wild type PSI (Nabedryk et al. 1990; Breton et al.
1999) and either was unassigned or was assigned to a PA+ keto vibration, which was
rationalized to downshift from 1695 cm-1 with oxidation (Hastings et al. 2001).
However, model compound studies predict an oxidation-induced upshift (Nabedryk et al.
1990; Breton et al. 1999).
We favor the assignment of the 1687 cm-1 spectral band to the chl keto vibrational
band of PA+, which upshifts from 1638 cm-1 with light-induced oxidation.
This
interpretation is consistent with previous site-directed mutagenesis studies. In this work,
mutations at the conserved threonine were used to disrupt the hydrogen-bonding network
of PA, and an increase in the amplitude of a positive 1687 cm-1 band was observed (Witt
et al. 2002; Li et al. 2004; Pantelidou et al. 2004). This increase in amplitude was
attributed to an increased population of a more weakly hydrogen-bonded PA+ in the
mutant (Witt et al. 2002). However, there was no significant change in amplitude of the
109
1687 cm-1 band when mutations were used to introduce hydrogen bonds to PB (Breton et
al. 2005).
Based on these previous investigations, we suggest that the 1687 cm-1 band arises
from a state in which the PA+ hydrogen bond to Thr has been weakened by a lightinduced protein structural change. In this interpretation, there are two populations of PA+,
which are generated under illumination, a strongly hydrogen-bonded version, which
contributes to the 1653 cm-1 band, and a less strongly hydrogen-bonded version, which
contributes to the difference spectrum at 1687 cm-1. Observation of only one PA chl keto
band, at 1638 cm-1, demonstrates that the conformational change is light-induced. Thus,
our work suggests a light-induced protein relaxation event, which weakens the hydrogen
bond between PA and Thr, most likely by a change in distance or angle between the 131
keto group of PA and the Thr side chain. To explain the observation of the 1653 cm-1
band, this change in hydrogen bonding must occur in only a subset of PSI reaction
centers.
In our experiments, PSI vibrational spectra were recorded under continuous, low
intensity, red-filtered, illumination. The data acquisition time was 4 minutes. In PSI,
forward electron transfer from the primary donor to the terminal electron acceptor, FB,
occurs with a time constant of ~500 nanoseconds, and P700+FB- recombination events
occur on the tens of milliseconds time scale (Brettel et al. 2001).
Therefore, our
experiment is expected to generate a photosteady state. In this photosteady state, protein
relaxation events, which are coupled to the initial charge separation, will have the
opportunity to accumulate under illumination and to influence the spectrum of P700+.
110
In our previous chl ester labeling experiments, P700+/P700 (Kim et al. 2000) was
demonstrated to have more than four ester frequencies.
Multiple, distributed ester
frequencies are likely to be caused by the protein structural changes under discussion
here. It was suggested that these multiple ester frequencies could reflect a distribution in
keto-enol tautomerization. Such a structural change is expected to be coupled with
alterations in hydrogen bonding, which will also alter the double bond character in the
isocyclic ring of PA (Kim et al. 2000). Comparison to this previous work suggests that
both keto and ester vibrational bands can be used as reporters for light-induced protein
relaxation events in PSI.
In PSI, two rates are observed for oxidation of A1- by FX, which have been
attributed to oxidation of the A1A- and A1B- species (Joliot et al. 1999; Guergova-Kuras et
al. 2001). The rates of the two processes are ~20 and 200 nanoseconds, although the
relative amplitudes of the two kinetic phases differ in different organisms (Brettel et al.
2001; Xu et al. 2003a; Xu et al. 2003b) and references therein}. In the bacterial reaction
center, however, electron transfer to the intermediate acceptor, bacteriopheophytin, is
unidirectional in wild type preparations (Heller et al. 1995; Kirmaier et al. 1999). Our
observation of a PA side structural change raises the question whether this protein
relaxation event alters the rates and amplitudes of the A and B side electron transfer.
From mutation studies, it was concluded that the midpoint potential of P700 decreases by
30-60 mV when a non-hydrogen bonding substitution is made at the conserved Thr
residue (Witt et al. 2002; Li et al. 2004). While it was concluded that the disruption of
the hydrogen bond to PA does not alter the rates or relative amplitudes of A1B- and A1Aoxidation by FX (Li et al. 2004), Thr mutations induced a shift of spin density from PB
111
toward PA (Witt et al. 2002). This result suggests that changes in the strength of the PA+
hydrogen bond can alter electronic coupling in the P700 dimer, and thus influence electron
transfer rate from soluble electron carriers to P700+.
Changes
in
PA+
hydrogen
bond
strength
may
also
influence
cytochrome/plastocyanin oxidation through structural changes in the protein backbone.
Figure 7B shows the luminal PsaA and PsaB helices in close proximity to P700. This
region of PSI has been demonstrated to be important in the facilitation of electron transfer
from plastocyanin and cytochrome c to P700+ (Sun et al. 1999; Brettel et al. 2001;
Sommer et al. 2002; Sommer et al. 2004).
As suggested in Figure 7, changes in the
angle or distance of the Thr-PA+ hydrogen bond are likely to influence the orientation and
conformation
of
these
luminal
PsaA
and
PsaB
helices.
In
the cyanobacterium, Synechocystis, such a change in orientation may alter the rate or
mechanism of cytochrome c oxidation in vivo (Durán et al. 2004). For example, it has
been proposed that a rate-limiting conformation change may occur to regulate oxidation
(Bottin et al. 1985; Sigfridsson et al. 1997). A redox-linked conformational change may
also regulate the binding affinity of the oxidized and reduced forms of the soluble
electron carriers (Drepper et al. 1996).
In photosynthetic reaction centers, the protein environment is an inhomogeneous
matrix, which relaxes after charge separation on many time scales. A complete
understanding of photosynthetic electron transfer necessitates a description of these
coupled protein conformational changes. Our work provides a new, detailed description
of a protein relaxation event, which is coupled to light-induced electron transfer in PSI.
112
We propose that this light-induced change in PA+ hydrogen bonding regulates PSI
electron transfer reactions.
4.6 Acknowledgments.
This work was supported by NIH GM 43273 (B.A.B). The authors thank Dr. L. Gelbaum
for assistance with NMR data acquisition and Prof. Hugo Scheer for helpful discussion.
113
4.7 References
Barry, B. A. (1995). "Tyrosyl radicals in photosystem II." Methods Enzymol. 258: 303319.
Barry, B. A., I. Cooper, A. De Riso, S. H. Brewer, D. M. Vu and R. B. Dyer (2006).
"Time-resolved vibrational spectroscopy detects protein-based intermediates in
the photosynthetic oxygen-evolving cycle." Proc. Natl. Acad. Sci. USA 103:
7288-7291.
Beale, S. I. (1999). "Enzymes of chlorophyll biosynthesis." Photosynth. Res. 60: 43-73.
Beale, S. I., S. P. Gough and S. Granik (1975). "Biosynthesis of delta-aminolevulinic acid
from the intact carbon skeleton of glutamic acid in greening barley." Proc. Natl.
Acad. Sci. USA 72: 2719-2723.
Bellamy, L. J. (1980). Advances in infrared group frequencies: the infrared spectra of
complex molecules. Advances in infrared group frequencies: The infrared spectra
of complex molecules. London; New York, Chapman and Hall. 2: 263-275.
Bender, S. L., J. Keough, S. E. Boesch, R. A. Wheeler and B. A. Barry (2008). "The
vibrational spectrum of the secondary electron acceptor, A1, in photosystem I." J.
Phys. Chem. B 112: 3844-3852.
Boldt, N. J., R. J. Donohoe, R. R. Birge and D. F. Bocian (1987). "Chlorophyll model
compounds: effects of low symmetry on the resonance Raman spectra and normal
mode descriptions of nickel (II) dihydroporphyrins." J. Am. Chem. Soc. 109:
2284-2298.
Bollivar, D. W. and S. I. Beale (1995). "Formation of the isocyclic ring of chlorophyll by
isolated Chlamydomonas reinhardtii chloroplasts." Photosynth. Res. 43: 113-124.
Bollivar, D. W. and S. I. Beale (1996). "The chlorophyll biosynthetic enzyme Mgprotoporphyrin IX monomethyl ester (oxidative) cyclase. Characterization and
partial purification from Chlamydomonas reinhardtii and Synechocystis sp. PCC
6803." Plant Physiol. 112: 105-114.
Bottin, H. and P. Mathis (1985). "Interaction of plastocyanin with the photosystem I
reaction center: a kinetic study by flash absorption spectroscopy " Biochemistry
24: 6453-6460.
Breton, J. (2001). "Fourier transform infrared spectroscopy of primary electron donors in
type I photosynthetic reaction centers." Biochim. Biophys. Acta 1507: 180-193.
114
Breton, J., P. R. Chitnis and M. Pantelidou (2005). "Evidence for hydrogen bond
formation to the PsaB chlorophyll of P700 in photosystem I mutants of
Synechocystis sp. PCC 6803." Biochemistry 44: 5402-5408.
Breton, J., E. Nabedryk and W. Leibl (1999). "FTIR study of the primary electron donor
of photosystem I (P700) revealing delocalization of the charge in P700+ and
localization of the triplet character in 3P700." Biochemistry 38: 11585-11592.
Brettel, K. and W. Leibl (2001). "Electron transfer in photosystem I." Biochim. Biophys.
Acta 1507: 100-114.
De Riso, A., D. L. Jenson and B. A. Barry (2006). "Calcium exchange and structural
changes during the photosynthetic oxygen evolving cycle." Biophys. J. 91: 19992008.
Drepper, F., M. Hippler, W. Nitschke and W. Haehnel (1996). "Binding dynamics and
electron transfer between plastocyanin and photosystem I." Biochemistry 35:
1282-1295.
Durán, R. V., M. Hervás, M. A. De la Rosa and J. A. Navarro (2004). "The efficient
functioning of photosynthesis and respiration in Synechocystis sp. PCC 6803
strictly requires the presence of either cytochrome c6 or plastocyanin." J. Biol.
Chem. 279: 7229-7233.
Gehlen, J. N., M. Marchi and D. Chandler (1994). "Dynamics affecting the primary
charge transfer in photosynthesis." Science 263: 499-502.
Golbeck, J. H. and D. A. Bryant (1991). "Photosystem I." Current Topics in
Bioenergetics 16: 83-177.
Grotjohann, I. and P. Fromme (2005). "Structure of cyanobacterial photosystem I."
Photosynth. Res. 85: 51-72.
Guergova-Kuras, M., B. Boudreaux, A. Joliot, P. Joliot and K. Redding (2001).
"Evidence for two active branches for electron transfer in photosystem I." Proc.
Natl. Acad. Sci. USA 98: 4437-4442.
Halverson, K. M. and B. A. Barry (2003). "Evidence for spontaneous structural changes
in a dark-adapted state of photosystem II." Biophys. J. 85: 2581-2588.
Hastings, G., V. M. Ramesh, R. Wang, V. Sivakumar and A. Webber (2001). "Primary
donor photo-oxidation in photosystem I: a re-evaluation of (P700+ - P700) Fourier
transform infrared difference spectra." Biochemistry 40: 12943-12949.
Heller, B. A., D. Holten and C. Kirmaier (1995). "Control of electron-transfer between
the L- and M-sides of photosynthetic reaction centers." Science 269: 940-945.
115
Hippler, M., F. Drepper, J. Farah and J. D. Rochaix (1997). "Fast electron transfer from
cytochrome c6 and plastocyanin to photosystem I of Chlamydomonas reinhardtii
requires PsaF." Biochemistry 36: 6343-6349.
Holzwarth, A. R., M. G. Muller, J. Niklas and W. Lubitz (2006). " Ultrafast transient
absorption studies on photosystem I reaction centers from Chlamydomonas
reinhardtii. 2: Mutations near the P700 reaction center chlorophylls provide new
insight into the nature of the primary electron donor." Biophys. J. 90: 552-565.
Huang, D. D. and W. Y. Wang (1986). "Chlorophyll biosynthesis in Chlamydomonas
starts with the formation of glutamyl-tRNA." J. Biol. Chem. 261: 13451-13455.
Joliot, P. and A. Joliot (1999). "In vivo analysis of the electron transfer within
Photosystem I: are the two phylloquinones involved?" Biochemistry 38: 1113011136.
Jordan, P., P. Fromme, H. T. Witt, O. Klukas, W. Saenger and N. Krauss (2001). "Threedimensional structure of cyanobacterial photoystem I at 2.5A resolution." Nature
411: 909-917.
Kaess, H., P. Fromme, H. T. Witt and W. Lubitz (2001). "Orientation and electronic
structure of the primary donor radical cation P700+ in photosystem I: a single
crystal EPR and ENDOR study." J. Phys. Chem. B 105: 1225-1239.
Kim, S. and B. A. Barry (2000). "Identification of carbonyl modes of P700 and P700+ by in
situ chlorophyll labeling in photosystem I." J. Am. Chem. Soc. 122: 4980-4981.
Kim, S., C. A. Sacksteder, K. A. Bixby and B. A. Barry (2001). "A reaction-induced FTIR study of cyanobacterial photosystem I." Biochemistry 40: 15384-15395.
Kirmaier, C., D. Weems and D. Holten (1999). "M-side electron transfer in reaction
center mutants with a lysine near the nonphotoactive bacteriochlorophyll."
Biochemistry 38: 11516-11530.
Kitson, F. G., B. S. Larsen and C. N. McEwen (1996). Amino Acids. Gas
Chromatography and Mass Spectrometry, A Practical Guide. San Diego,
Academic Press: 87-94.
Li, Y., M. G. Lucas, T. Konovalova, B. Abbott, F. MacMillan, A. Petronko, V.
Sivakumar, R. Wang, G. Hastings, F. Gu, J. van Tol, L. C. Brunal, R. Timkovich,
F. Rappaport and K. Redding (2004). "Mutation of the putative hydrogen-bond
donor to P700 of photosystem I." Biochemistry 43: 12634-12647.
Merchant, S. and L. Bogorad (1986). "Regulation by copper of the expression of
plastocyanin and cytochrome c-552 in Chlamydomonas reinhardtii." Mol. Cell.
Biol. 6: 462-469.
116
Nabedryk, E., M. Leonhard, W. Maentele and J. Breton (1990). "Fourier transform
infrared difference spectroscopy shows no evidence for an enolization of
chlorophyll a upon cation formation either in vitro or during P700 photooxidation."
Biochemistry 29: 3242-3247.
Noren, G. H., R. J. Boerner and B. A. Barry (1991). "EPR characterization of an oxygenevolving photosystem II preparation from the transformable cyanobacteria
Synechocystis 6803." Biochemistry 30: 3943-3950.
Pantelidou, M., P. R. Chitnis and J. Breton (2004). "FTIR spectroscopy of Synechocystis
6803 mutants effected on the hydrogen bonds to the carbonyl groups of the PsaA
chlorophyll of P700 supports an extensive delocalization of the charge in P700+."
Biochemistry 43: 8380-8390.
Parson, W. W. and A. Warshel (2004). "Dependence of photosynthetic electron-transfer
kinetics on temperature and energy in a density-matrix model." J. Phys. Chem. B
108: 10474-10483.
Patzlaff, J. S. and B. A. Barry (1996). "Pigment quantitation and analysis by HPLC
reverse phase chromatography: a characterization of antenna size in oxygenevolving photosystem II preparations from cyanobacteria and plants."
Biochemistry 35: 7802-7811.
Porra, R. J., O. Klein and P. E. Wright (1983). "The proof by 13C-NMR spectroscopy of
the predominance of the C5 pathway over the Shemin pathway in chlorophyll
biosynthesis in higher plants and of the formation of the methyl ester group of
chlorophyll from glycine." Eur. J. Biochem. 130: 509-516.
Porra, R. J. and H. Scheer (2001). "18O and mass spectrometry in chlorophyll research:
derivation and loss of oxygen atoms at the periphery of the chlorophyll
macrocycle during biosynthesis, degradation and adaptation." Photosynth. Res.
66: 159-175.
Rieble, S. and S. I. Beale (1988). "Transformation of glutamate to delta-aminolevulinic
acid by soluble extracts of Synechocystis sp. PCC 6803 and other oxygenic
prokaryotes." J. Biol. Chem. 263: 8864-8871.
Rippka, R., J. Deruelles, J. B. Waterbury, M. Herdman and R. Y. Stanier (1979).
"Generic assignments, strain histories and properties of pure cultures of
cyanobacteria." Journal of General Microbiology 111: 1-61.
Sachs, R. K., K. M. Halverson and B. A. Barry (2003). "Specific isotopic labeling and
photooxidation-linked structural changes in the manganese-stabilizing subunit of
photosystem II." J. Biol. Chem. 278: 44222-44229.
Sacksteder, C. A., S. L. Bender and B. A. Barry (2005). "Role for bound water and CH-pi
aromatic interactions in photosynthetic electron transfer." J. Am. Chem. Soc. 127:
7879-7890.
117
Schneegurt, M. A. and S. I. Beale (1992). "Origin of the chlorophyll b formyl oxygen in
Chlorella vulgaris." Biochemistry 31: 11677-11683.
Schön, A., G. Krupp, S. Gough, S. Berry-Lowe, C. G. Kannangara and D. Söll (1986).
"The RNA required in the first step of chlorophyll biosynthesis is a chloroplast
glutamate tRNA." Nature 322: 281-284.
Sigfridsson, K., S. Young and O. Hansson (1997). "Electron transfer between spinach
plastocyanin mutants and photosystem 1." Eur. J. Biochem. 245: 805-812.
Sommer, F., F. Drepper, W. Haehnel and M. Hippler (2004). "The hydrophobic
recognition site formed by residues PsaA-Trp651 and PsaB-Trp627 of
photosystem I in Chlamydomonas reinhardtii confers distinct selectivity for
binding of plastocyanin and cytochrome c6." J. Biol. Chem. 279: 20009-20017.
Sommer, F., F. Drepper and M. Hippler (2002). "The luminal helix l of PsaB is essential
for recognition of plastocyanin or cytochrome c6 and fast electron transfer to
photosystem I in Chlamydomonas reinhardtii." J. Biol. Chem. 277: 6573-6581.
Sun, J., M. Hervas, J. A. Navarro, M. A. De la Rosa and P. R. Chitnis (1999). "Oxidizing
side of the cyanobacterial photosystem I: mutational analysis of the luminal H
loop of the PsaB subunit." Photosynth. Res. 62: 241-250.
van der Est, A. (2001). "Light-induced spin polarization in type I photosynthetic reaction
centres." Biochim. Biophys. Acta 1507: 212-225.
Walker, C. J., K. E. Mansfield, K. M. Smith and P. A. Castelfranco (1989).
"Incorporation of atmospheric oxygen into the carbonyl functionality of the
protochlorophyllide isocyclic ring." Biochem. J. 257: 599-602.
Wang, H., S. Lin, J. P. Allen, J. C. Williams, S. Blankert, C. Laser and N. W. Woodbury
(2007). "Protein dynamics control the kinetics of initial electron transfer in
photosynthesis." Science 316: 747-750.
Warshel, A., Z. T. Chu and W. W. Parson (1989). "Dispersed polaron simulations of
electron transfer in photosynthetic reaction centers." Science 246: 112-116.
Webber, A. N. and W. Lubitz (2001). "P700: the primary electron donor of photosystem
I." Biochim. Biophys. Acta 1507: 61-79.
Witt, H., C. Teuloff, J. Niklas, E. Bordignon, D. Carbonera, S. Kohler, A. Labahn and W.
Lubitz (2002). "Hydrogen bonding of P700: site-directed mutagenesis of threonine
A739 of photosystem I in Chlamydomonas reinhardtii." Biochemistry 41: 85578569.
Wong, Y. S., P. A. Castelfranco, D. A. Goff and K. M. Smith (1985). "Intermediates in
the formation of the chlorophyll isocyclic ring." Plant Physiol. 79: 725-729.
118
Xu, W., P. Chitnis, A. Valieva, A. van der Est, Y. N. Pushkar, M. Krzystyniak, C.
Teutloff, S. G. Zech, R. Bittl, D. Stehlik, B. Zybailov, G. Shen and J. H. Golbeck
(2003a). "Electron transfer in cyanobacterial photosystem I: I. Physiological and
spectroscopic characterization of site-directed mutants in a putative electron
transfer pathway from A0 through A1 to Fx." J. Biol. Chem. 278: 27864-27875.
Xu, W., P. R. Chitnis, A. Valieva, A. van der Est, K. Brettel, M. Guergova-Kuras, Y. N.
Pushkar, S. G. Zech, D. Stehlik, G. Shen, B. Zybailov and J. H. Golbeck (2003b).
"Electron transfer in cyanobacterial photosystem I: II. Determination of forward
electron transfer rates of site-directed mutants in a putative electron transfer
pathway from A0 through A1 to Fx." J. Biol. Chem. 278: 27876-27887.
119
Amount of 13C chl labeling from 13C-labeled glutamate
Table 1.
1
m/z
871.6
(molecular ion)
872.6
(+1-13C
isotopomer)
873.6
(+2-13C
isotopomer)
874.6
(+3-13C
isotopomer)
875.6
(+4-13C
isotopomer)
876.6
(+5-13C
isotopomer)
877.6
(+6-13C
isotopomer)
1
m/z
871.6
(molecular ion)
872.6
(+1-13C
isotopomer)
873.6
(+2-13C
isotopomer)
874.6
(+3-13C
isotopomer)
875.6
(+4-13C
isotopomer)
876.6
(+5-13C
isotopomer)
877.6
(+6-13C
isotopomer)
2
Isotope
distribution
for natural
abundance
†
chl*
3
Isotope
distribution
for 13Clabeled chl
4
C-labeled
chl minus
natural
abundance chl
(column 3minus-2)
5
Predicted
+ 13
1 C
isotopomer
‡
distribution
6
Measured
+ 13
2- C
isotopomer
contribution
(column 4minus-5)
7
Predicted
+ 13
2 C
isotopomer
‡
distribution
100
100
0
0
0
0
63.7 ± 7.7
106.3 ± 2.1
42.6 ± 8.0
42.6 ± 8.0
0
0
23.6 ± 0.7
77.3 ± 3.5
53.7 ± 3.5
27.1 ± 6.1
26.6 ± 7.0
26.6 ± 7.0
5.7 ± 0.4
40.6 ± 1.0
34.9 ± 1.1
10.0 ± 1.9
24.9 ± 2.2
16.9 ± 4.9
2.9 ± 0.5
18.5 ± 1.0
15.6 ± 1.1
2.4 ± 0.5
13.2 ± 1.2
6.3 ± 1.7
1.1 ± 0.8
6.6 ± 0.3
5.5 ± 0.9
1.2 ± 0.2
4.3 ± 0.9
1.5 ± 0.4
1.1 ± 0.5
2.2 ± 0.4
1.1 ± 0.6
0.5 ± 0.4
8
Measured
+ 13
3- C
isotopomer
contribution
(column 6minus-7)
9
Predicted
+ 13
3- Cisotopomer
‡
contribution
10
Measured
+ 13
4- C
isotopomoer
contribution
(column 8minus-9)
11
Predicted
+ 13
4- Cisotopomer
‡
contribution
12
Total
13
C-labeling
for each
isotopomer
13
Probability of
13
C
incorporation
into the keto
position§
0
0
0
0
0
0
0
0
0
0
42.6 ± 8.0
5.3
0
0
0
0
26.6 ± 7.0
6.6
8.0 ± 5.4
8.0 ± 5.4
0
0
8.0 ± 5.4
3.0
6.9 ± 2.1
5.1 ± 3.5
1.8 ± 4.3
1.8 ± 4.3
1.8 ± 4.3
0.9
2.8 ± 1.0
1.9 ± 1.3
0.9 ± 1.6
1.2 ± 2.3
ND¶
0
13
± 0 .7
± 0.2
0
*
Values reported as percentages (%); †Predicted isotope distribution from the Isotopident tool
(http://education.expasy.org/student_projects/isotopident/). 871.6 m/z, 100%; 872.6 m/z, 64.4%;
873.6 m/z, 21.4%; 874.6% m/z, 4.9%; 875.6 m/z, 0.9%; 87.6 m/z, 0.1%; ‡Based on measured
isotope distribution in column 2; §Based on the statistical probability of labeling at 8 possible
carbon positions; ¶ND, not detected, <1%.
120
Table 2.
Isotopic labeling of PSI amino acids as determined with GC/MS
Amino Acid
[13C] Glu*
Glutamic Acid†
16.0 ± 0.1
Proline
14.0 ± 0.5
Serine
ND§
Aspartic Acid‡
ND
Alanine
ND
Glycine
ND
Valine
ND
Leucine
ND
Isoleucine
ND
Threonine
ND
Phenylalanine
ND
Lysine
ND
*
Values reported as percentages (%); †Glutamic acid and deamidated glutamine; ‡Aspartic acid
and deamidated asparagine; §ND, not detected, < 1%.
121
Table 3. Chlorophyll keto assignments for PA and PB in Photosystem I,
based on 13C and 18O keto labeling
18
13
Natural
Abundance*
1718
O keto
Labeled*
1681
C keto
Labeled*
1679
Assignment
1698
1666
1663
PB
1687
1643
1644
PA+
1653
1608
1608
PA+
1638
1602
1593
PA
PB+
*Frequencies in wavenumbers (cm-1)
122
Figure 1. (A) Structure of chlorophyll a displaying the 8 carbons (red) that will be
isotopically labeled from glutamic-13C-3 acid. The keto group of chl is boxed. The keto
oxygen (blue) will be isotopically labeled from 18O2.
123
Figure 2.
MALDI mass spectrometry showing incorporation of
cyanobacterial chl.
18
O and
13
C into
In (A), the cyanobacteria were cultured in the presence of
16
O2
(white) or 18O2 (black). In (B), the cyanobacteria were cultured in the presence of natural
abundance glutamic acid (white) or L-glutamic-3-13C-acid (black). The normalized 871.5
m/z peak corresponds to the [M+H] ion of pheophytin, which was generated by acid
treatment of the chl sample immediately before the measurement. See Supplemental
Materials and Methods for additional information.
124
Figure 3. Light-induced FT-IR difference spectra, associated with the oxidation of P700
in PSI. In (A), the samples were derived from cyanobacteria cultured in the presence of
16
18
O2. In (B), the samples were derived from cyanobacteria cultured in the presence of
O2. In (C, solid line), the isotope-edited spectrum,
16
O-minus-18O, was generated by
interactive subtraction of (B) minus (A) and multiplication by a factor of 8. A subtraction
factor of 0.9663 was used to minimize ester contributions to the isotope-edited spectrum.
In (C, dotted line), the isotope-edited spectrum was generated by a direct one-to-one
subtraction and multiplication by a factor of 8. (D) shows a control double difference
spectrum, in which no vibrational bands are expected. This control was generated by
taking one-half of the data set in (A), subtracting it from the other half of the data set in
125
Figure. 3 continued
(A), and dividing by the square root of two. (D) was also multiplied by a factor of 8 for
comparison to (C). (E) is a dark-minus-dark control spectrum. 69 spectra were averaged
in A-C and E. Each tick mark on the y axis represents 5x10-3 absorbance units.
The
color-coding scheme for band labels is described in the paper. See Material and Materials
for additional information.
126
Figure 4. Light-induced FT-IR spectra, associated with the oxidation of P700 in PSI. In
(A), the samples were derived from cyanobacteria cultured in the presence of natural
abundance glutamate. In (B), the samples were derived from cyanobacteria cultured in
the presence of L-glutamic-3-13C-acid. In (C, solid line), the isotope-edited spectrum,
12
C-minus-13C, was generated by interactive subtraction of (B) minus (A) and
multiplication by a factor of 5. A subtraction factor of 0.9584 was used to minimize ester
contributions to the isotope-edited spectrum. In (C, dotted line), the isotope-edited
spectrum was generated by a direct one-to-one subtraction and multiplication by a factor
of 5. (D) shows a control double difference spectrum, in which no vibrational bands are
127
Figure 4. continued
expected.
This control was generated by taking one-half of the data set in (A),
subtracting it from the other half of the data set in (A), and dividing by the square root of
two. (D) was also multiplied by a factor of 5 for comparison to (C). (E) is a dark-minusdark control spectrum. 64 spectra were averaged in A-C and E. Each tick mark on the y
-3
axis represents 5x10 absorbance units. The color-coding scheme for band labels is
described in the paper. Material and Material and Methods for additional information.
128
Figure 5. Isotope-edited FT-IR spectra showing the effects of 18O and 13C chl labeling
on P700+/P700 keto vibrational bands. (A) is the
12
C-minus-13C spectrum.
16
O-minus-18O spectrum, and (B) is the
In (A) and (B), control difference spectra, in which no
vibrational bands are expected, are also presented as the dotted lines. Natural abundance
bands assigned to PB and PB+ are shaded with vertical lines. Natural abundance bands
assigned to PA and PA+ are shaded with horizontal lines. The natural abundance band
(filled) and its corresponding isotope-shifted component (unfilled) are labeled with the
same color label. In (A), 69 spectra were averaged and in (B), 64 spectra were averaged.
129
Figure 5. continued
In (A) and (B), the sample size was 64 μg chl. Each tick mark represents 5x10-4 AU. See
Materials and Methods for additional information.
130
Figure 6. Comparison of the 1730-1590 cm-1 region in light-induced FT-IR spectra,
associated with the oxidation of P700. (A) is an overlay of Figure 3A (natural abundance,
blue) and Fig. 3B (18O-labeled, black).
(B) is an overlay of Fig ure4A (natural
abundance, blue) and Fig. 4B (13C-labeled, black).
represents 2x10-3 absorbance units.
Each tick mark on the y axis
See Material and Material and Methods for
additional information.
131
Figure 7.
(A) The protein environment of PA and PB in P700. The hydrogen bonding
network of PA with Thr 743, H2O and Tyr 603 is represented by dashes. (B) The luminal
helices of PsaB and PsaA near P700. The figure is reproduced from the 2.5 Å resolution
crystal structure of PSI from S. elongatus (PDB file accession number 1JB0 (Jordan et al.
2001)) using Swiss-Pdb viewer (v3.7; www.expasy.ch/spdbc) with numbering according
to S. elongatus, and rendered using POV-Ray (v3.5; www.povray.org).
132
CHAPTER 5
PROBING ELECTRON TRANSFER IN PSI
5.1 Abstract
In oxygenic photosynthesis, photosystem I (PSI) catalyzes the light-driven
oxidation of plastocyanin or cytochrome c and the reduction of ferredoxin. Studies have
underscored the importance of protein interactions in efficient electron transfer from
plastocyanin to PSI; however specific characterization of these interactions has not been
forthcoming.
In this work we have used deuterium exchange methods and isotope
labeling of tryptophan to investigate light-induced protein dynamics associated with
stable charge separation in PSI. Reaction-induced FT-IR spectroscopy and isotope-edited
spectra reveal protein relaxation events that may be linked to structural conformational
changes. The deuterium exchange isotope-edited spectra reveal light-induced
perturbations to amide I (C=O) of the protein backbone, and perturbations of aspartate
and glutamate residues. Isotope-based assignments for tryptophan may provide new
information concerning this amino acid’s involvement with electron transfer reactions in
PSI. The protein dynamics described in this work may be associated with the oxidation
of soluble electron carriers or other electron transfer reactions in PSI. These studies
provide unique information concerning electron transfer in oxygenic photosynthesis.
133
5.2 Introduction
Photosystem I is a protein complex responsible for the light-induced oxidation of
plastocyanin and the reduction of ferredoxin (Golbeck 1994). It is one of two protein
complexes involved in light-induced charge separation across the thylakoid membrane in
plants, green algae, and cyanobacteria. Three polypeptides are responsible for binding the
electron cofactors in PSI. Subunits PsaA and PsaB form a heterodimeric core and are
responsible for binding chl, phylloquinone and an iron-sulfur cluster. The PsaC subunit
of PSI, an extrinsic subunit located on the stromal side of the thylakoid membrane, is
responsible for binding two iron-sulfur clusters (Jordan et al. 2001).
The primary donor in PSI is believed to be a chlorophyll monomer, which reduces
A0, a second chlorophyll monomer. (Holzwarth et al. 2006) The electron from A0- is
transferred to A1, a phylloquinone, and then to the iron-sulfur cluster FX. These electron
cofactors are bound by the PsaA and PsaB subunits. FA, a second iron-sulfur cluster
oxidizes FX and is itself oxidized by the terminal electron acceptor FB. FA and FB are
bound to the PsaC subunit. (Jordan et al. 2001) The primary donor is reduced by P700, a
chlorophyll heterodimer comprised of chl a and chl a’. P700 is the terminal electron donor
of PSI and is reduced by plastocyanin or cytochrome c, depending on copper availability
(Wood 1978; Merchant et al. 1986; Ho et al. 1988).
Plastocyanin, a 10kDa soluble copper protein, is responsible for the reduction of
the photoxidized dimer, P700, in PSI and is reduced by the cytochrome b6f complex. The
crystal structures of oxidized and reduced forms of plastocyanin reveal two conserved
regions that are believed to be involved in recognition and binding to PSI.
In
134
cyanobacteria, a single oriented collisional mechanism has been reported for the
reduction of PSI by plastocyanin. This process involves electrostatic interactions without
formation of a kinetically detectable electron-transfer complex (Hervas et al. 1994;
Hervas et al. 1995; Hervas et al. 1996). In iron-deficient conditions, cytochrome c will
act as the reducing agent for PSI (Wood 1978). Studies have shown that cytochrome c
and PSI may form a transient complex before electron transfer and do not follow the
same kinetics as plastocyanin (Hervas et al. 1995). In these studies, second order kinetics
were observed for reduction of P700+ by cytochrome c compared to pseudo-first order
kinetics for plastocyanin reduction of P700+.
The docking site for the soluble electron carriers are made up of two helices from
the PsaA and PsaB subunits, designated in the crystal structure of S. elongates as
transmembrane helices I and J (Fromme et al. 2001; Jordan et al. 2001). Sequence
alignments of PsaB subunits from various organisms have shown several amino acids
residues in the loop region of helix J to be conserved. Mutation studies have suggested a
role in the extra-membrane loop in the interaction with soluble electron donor proteins
(Sun et al. 1999a; Sun et al. 1999b). Furthermore mutation studies of conserved
tryptophan residues located in the luminal loop regions of PsaA and PsaB have shown
these residues to be important in efficient reduction of P700+ from plastocyanin or
cytochrome c (Sun et al. 1999b; Sommer et al. 2002; Sommer et al. 2004). In contrast to
cyanobacteria, eukaryotic organisms require the subunit PsaF for efficient electron
transfer from plastocyanin to P700+. In eukaryotes, it is believed that this subunit is
responsible for coordinating and binding plastocyanin (Bottin et al. 1985; Haehnel et al.
1994).
135
Vibrational spectroscopy has underscored the importance of protein dynamics
associated with light-induced electron transfer in PSI (Kim et al. 2000; Kim et al. 2001;
Bender et al. 2008a). These experiments have demonstrated specific protein relaxation
events near P700 that may be linked to interactions with soluble electron carriers.
Furthermore, isotope labeling of tyrosine residues in PSI also showed perturbations
associated with CH-π interactions. These perturbations may be associated with protein
dynamics around P700 (Sacksteder et al. 2005). To investigate protein dynamics that may
be associated with electron transfer in PSI, deuterium was used as a molecular probe of
the system.
In this chapter, difference FT-IR spectroscopy was performed on 2H2O buffer
exchanged PSI samples. PSI samples containing all the polypeptides or PSI samples
missing the PsaC, PsaD, and PsaE subunits were compared.
Isotope-edited spectra
reveal protein perturbations consistent with light-induced structural changes of the PsaA
and/or PsaB subunits. These findings identify specific conformation changes that may be
important in interactions with soluble electron carriers and in efficient reduction of P700+.
To further characterize the structural dynamics during electron transfer in PSI, tryptophan
residues were isotopically labeled. The results reported here may be consistent with
perturbations of tryptophan residues, which may be associated with structural dynamics
controlling the oxidation of plastocyanin/cytochrome c6.
5.3 Materials and Methods
5.3.1 PSI purification for 1H2O and 2H2O buffer exchanged samples Wild-type
Synechocystis PCC 6803 liquid cultures (15L) were grown and supplemented with 5mM
136
glucose. Trimeric PSI samples were purified as previously described (Noren et al. 1991;
Kim et al. 2001). Briefly, thylakoid membranes were solubilized with 1% dodecyl-ß-Dmaltiside, centrifuged for one hour at 100,000 x g, and loaded onto a Q-sepharose
column. A linear gradient was applied, and PSI trimer fractions with an absorbance ≥679
absorbance units (AU) were pooled and were exchanged into 1H2O or 2H2O buffers
containing 5 mM HEPES-NaOH pH 7.5, 0.04% dodecyl-ß-D-maltiside (LM) by dialysis.
The p2H of the 2H20 buffer is reported as the uncorrected meter reading (Kim et al. 2001),
because the small solvent isotope effects on weak acid pKa values are essentially
cancelled by the 2H2O change in response of the pH electrode (Schowen et al. 1982).
Three rounds of dialysis were performed over three days. The total volume of buffer
exchange was 1.5 L, which correspond to 500 ml exchanged three times. Trimeric PSI
was concentrated using an Amicon (Bedford, MA) Ultra 100,000 MWCO centrifugal
filter device to a final concentration of 2 mg chl/ml.
5.3.2 PsaC, PsaD and PsaE removal. The stromal subunits PsaC, PsaD, and
PsaE along with the terminal electron acceptors FA and FB were removed as previously
described (Parrett, K. G. et al. 1989; Parrett, K. G.; et al. 1990). Briefly PSI samples
were incubated in 6.8 M Urea, 10 mM Tris and 5 mM glycine-pH 10 for 70 minutes.
Afterwards, an equal volume of tris buffer was added to stop the reaction. Samples were
dialyzed into 1H2O or 2H2O buffers as described above. Afterwards, samples were
concentrated to 2 mg/ml.
5.3.3 Growth conditions and PSI purification for 2H5-tryptophan.
Feedback
inhibition of the chorismate biosynthetic pathway was performed by adding 0.25 mM
tyrosine, 0.25 mM tryptophan (control) or
2
H5-tryptophan (98% purity, Isotec,
137
Maimisburg, OH), and 0.50 mM phenylalanine to the BG-11 media (Srinivasan et al.
1959; Smith 1962; Barry et al. 1987; Poulson et al. 1991). Synechocystis cells were
grown in 1L BG-11 buffered with 1mM TES NaOH- pH 8.0. Cultures were grown for 11
days and harvested with an OD reading of 1.3-1.5. PSI was purified as previously
described.(Bender et al. 2008a)
5.3.4 FTIR Spectroscopy. FT-IR spectra were collected at -10 °C as described
previously (Sacksteder et al. 2005; Bender et al. 2008b). Samples contained 3 mM
potassium ferricyanide and 3 mM potassium ferrocyanide and were concentrated at room
temperature under a steady flow of nitrogen gas.
Concentration times were 20-30
minutes. Spectral conditions were as follows: resolution, 4 cm-1; zero filling, 1; data
acquisition time, 4.0 min.
Difference spectra (light-minus-dark) were generated by
taking the ratio of single beam spectra collected before and during illumination and
converting to absorbance. FT-IR spectra were obtained in the dark or under continuous
illumination with red- and heat-filtered light, as previously described (Kim et al. 2001;
Sacksteder et al. 2005; Bender et al. 2008b). A 60 min dark relaxation time was used
between successive illuminations. Sixty difference spectra were averaged for 1H2O and
2
H2O PSI intact samples; forty-five difference spectra averaged for 1H2O and 2H2O PsaC,
PsaD and PsaE removed PSI samples. Fifty difference spectra were averaged for 1H5-trp
PSI samples, and forty-nine difference spectra were averaged for 2H5-trp PSI samples.
Spectra were normalized for small differences in sample concentration and path length.
The amplitude of the amide I band in the infrared absorption spectra was used for
normalization of the 1H2O and 2H2O buffer exchanged PSI samples. The amide II band
in the infrared absorption spectra was used for normalization of the 1H5-trp and 2H5-trp
138
PSI samples. The infrared absorption spectra were created by normalizing dark scans to
backgrounds not containing protein.
Isotope-edited spectra were generated by subtracting the isotope difference
spectra from the control difference spectra. A direct 1 to 1 subtraction was used for all
isotope-edited spectra reported here. Control double difference spectra were created by
subtracting ½ of the control data from the second half of the control data set and
multiplying by the square root of 2.
Dark spectra were obtained by subtracting
subsequent dark spectra from one another.
5.4 Results
PSI samples were dialyzed for three days into 2H2O containing 5mM HEPES pH
7.5 buffer. In contrast to our earlier solvent exchange protocol, which gave 20% labeling
(Kim et al. 2001), here a dialysis method was used to increase the amount of solvent
isotope exchange (Rath et al. 1998; Kim et al. 2001; Jenson et al. 2007). The amount of
exchange was measured to be 42 + 5% for samples containing all extrinsic subunits, and
55± 16% for the PsaC, PsaD and PsaE removed samples. The exchange was assessed by
FT-IR spectroscopy (Rath et al. 1998; Kim et al. 2001).
In Figure 1, difference FT-IR spectra, associated with P700+FB--minus-P700FB, in
1
H2O-exchanged samples (Figure 1A) and 2H2O-exchanged samples (Figure 1B) are
presented. Figure 1C is the 2H2O isotope-edited spectrum generated by subtracting the
difference spectrum in Figure 1B from the difference spectrum in Figure 1A. The
isotope-edited spectrum in Figure 1C will contain frequencies associated with electron
transfer reactions and isotope-exchange. For a frequency to be observed in the isotope-
139
edited spectrum, the frequency must be sensitive to both of these conditions. Bands
corresponding to the protein in the dark will appear negative, with the isotope shifted
bands appearing positive.
Those frequencies corresponding to the protein under
illumination will appear as positive bands, with isotope shifts appearing as negative
bands. Differences arise in the amplitudes of bands in the isotope-edited spectra
presented here, compared to previous 2H2O exchange experiments (Kim et al. 2001).
This may be due to differences in the hydration of the samples and differences in the
amount of exchange.
To better assess the protein dynamics observed for the 2H2O isotope-edited
spectrum in Figure 1C, the stromal subunits PsaC, PsaD and PsaE were removed. Figure
2 shows the resulting gel from PSI samples that were subjected to urea treatment (lane 1)
and from intact PSI samples (lane 2). Three bands are missing in lane 1 compared to lane
2, these bands have previously been assigned to the subunits PsaC, PsaD, and PsaE
(Kruip et al. 1997). Removal of the PsaC subunit will also result in the loss of the two
terminal electron acceptors, FA and FB. The resulting PSI samples will contain FX as the
terminal electron acceptor, and the photoaccumulated species will be P700+ and FX-. The
samples were 2H2O buffer-exchanged as outlined in the Materials and Methods section.
The exchange was measured by FT-IR spectroscopy and was found to be 55 ± 16% (Rath
et al. 1998; Kim et al. 2001).
In Figure 3, the light-minus dark FT-IR spectra, associated with P700+FX--minusP700FX are presented. The spectrum in Figure 3A corresponds to 1H2O-exchanged PSI,
and the spectrum in Figure 3B is from 2H2O-exchanged PSI. The 2H2O-exchanged
isotope-edited spectrum is shown in Figure 3C. The 2H2O-exchanged isotope-edited
140
spectrum was generated by subtracting the spectrum in Figure 3B from the spectrum in
Figure 3A. The spectrum in Figure 3C shows significant spectral features compared to
the noise, as assessed by a dark-minus-dark (Figure 3E), or a control double difference
spectrum (Figure 3D) in which no vibrational bands are expected. Similar spectral
features are observed in the stromal subunit removed isotope-edited spectrum (Figure 3C)
and the intact isotope-edited spectrum (Figure 1C).
A comparison of the isotope-edited spectra of the intact samples (Figure 4A) and
samples with the stromal subunits removed (Figure 4B) is shown in Figure 4. Figure 4A
contains the isotope-edited spectrum of PSI samples containing all stromal subunits and
FB as the terminal electron acceptors. Frequencies above 1700 cm-1 are most likely due
to COOH stretching vibrations of aspartic acid or glutamic acid (Chirgadze et al. 1975;
Barth 2000; Wolpert et al. 2006). Evidence for these assignments comes from the
derivative shaped band observed at (+) 1275 and (-) 1267 cm-1 in the 2H2O isotope-edited
spectrum in Figure 4A. This frequency can be associated with COH bending of aspartic
acid and glutamic acid This stretching mode is expected to shift to lower frequencies in
2
H2O buffer, and can be assigned to the derivate-shaped band at (-) 1048 and (+) 1038
cm-1.
Isotope-shifted bands corresponding to amide I C=O stretching are observed in
(+) 1663 and (-) 1658 cm-1 in Figure 4A. Other frequencies observed in the 1700-1600
cm-1 region of the spectrum may be associated with glutamine or asparagine and basic
amino acids such as lysine and arginine. Frequencies in Figure 4A observed at (+) 1696
cm-1, (-) 1670 cm-1, (+) 1678 cm-1, and (+) 1641 cm-1 may be assigned to these amino
acids. The complex spectral banding pattern in this region may be the result of multiple
141
amino acid side chain contributions. Therefore, it is not possible to assign frequencies to
specific amino acids in this region of the isotope-edited spectrum.
Unique bands observed in the isotope edited spectra of Figure 4A are derivative
shaped bands at (-) 1541 cm-1 and (+) 1550 cm-1; (+)1588 cm-1 and (-)1597 cm-1. These
bands are of interest, because few amino acid side chain groups contribute to this region
of the spectrum and are expected to be sensitive to deuterium exchange. The lower
frequency pair, (-) 1541 cm-1 and (+) 1550 cm-1, may be from deprotonated aspartate or
glutamate in the dark state of PSI. Model compound studies have shown an increase in
frequency in 2H2O of the asymmetric COO- stretching of both amino acids (Chirgadze et
al. 1975; Tamm et al. 1997; Barth 2000; Wolpert et al. 2006). These bands shift to higher
frequencies ((+) 1588 cm-1 and (-) 1597 cm-1) under illumination. This shift to higher
frequency suggests a conformational change associated with charge separation. The
symmetric stretching of the COO- group of aspartate and glutamate is also sensitive to
deuterium exchange. However model compounds report a frequency shift of 2-3 cm-1;
this small isotope-shift may be below the spectral resolution of the spectrum presented
here.
Figure 4B contains the isotope-edited spectrum associated with PSI samples with
the stromal subunits removed and containing FX as the terminal electron acceptor. The
terminal electron acceptor has changed from FB in Figure 4A to FX in Figure 4B.
Frequencies observed in both isotope-edited spectra in Figure 4 are most likely due to the
donor side of PSI, which has not changed in Figure 4B compared to Figure 4A.
Therefore, similar bands observed in both isotope-edited spectra in Figure 4 can be
assigned to perturbations of the PsaA and PsaB subunits. Frequencies at (+) 1725 cm-1,
142
and (+) 1726 cm-1 are observed in the isotope-edited spectra depicted in Figure 4A and
Figure 4B respectively. These bands are most likely due to the COOH antisymmetric
stretching of aspartic acid and/or glutamic acid residues located in the PsaA and/or PsaB
subunits. From model compounds, these frequencies are expected to shift to lower
frequency in both isotope-edited spectra. The isotope-shifted bands may contribute to the
complex spectral features observed from 1720-1680 cm-1.
Similar frequencies are also observed in the 1700-1600 cm-1 region of both
isotope-edited spectra in Figure 4. Frequencies at (+) 1696/4 cm-1, (-) 1670 cm-1, (+)
1678 cm-1, and (+) 1641 cm-1 may be assigned to the side chains of glutamine,
asparagine, arginine and/or lysine located in the PsaA and/or PsaB subunits. Isotope
shifted bands corresponding to amide I C=O stretching, are observed at (+) 1663 and (-)
1658 cm-1 in both isotope-edited spectra. These results suggests protein relaxation of the
PsaA and/or PsaB subunits associated with electron transfer. Frequencies at (-) 1541 and
(+) 1550/1 cm-1 observed in both isotope-edited spectra are assigned to aspartate and
glutamate COO- antisymmetric stretching. These bands are believed to shift to (+) 1588
and (-) 1597 cm-1 during light-induced electron transfer in PSI.
To further probe specific protein relaxation during stable charge separation
tryptophan was used as a molecular probe of the system. The tryptophan dimer located
on the luminal side of PSI has been reported to be essential for efficient electron transfer
from plastocyanin to P700 (Sun et al. 1999b; Sommer et al. 2002; Sommer et al. 2004).
Light induced difference spectra were generated from samples containing 2H5-trp labeled
PSI to investigate tryptophan perturbations during electron transfer. Figure 5 shows the
difference spectra from PSI samples containing 1H5-Trp (Figure 5A) or PSI samples
143
containing 2H5-Trp (Figure 5B). An isotope-edited spectrum was created by subtracting
the spectrum in Figure 5B from the spectrum in Figure 5A. The resulting isotope-edited
spectrum is depicted in Figure 5C. Only vibrational frequencies that are perturbed during
electron transfer and sensitive to 2H5-Trp incorporation are expected to contribute to the
spectrum in Figure 5C. Vibrational frequencies observed in Figure 5C are above the
noise as shown in Figure 5D and 5E. To assign the isotope shifted bands of tryptophan in
PSI, absorbance spectra of 1H5-Trp and 2H5-Trp were collected (data not shown).
Vibrational bands associated with isotope incorporation and perturbations of
tryptophan are expected between 1660-1200 cm-1 based on model compounds and UVRR
studies (Maruyama et al. 1995; Hu et al. 1997; Barth 2000; Wolpert et al. 2006). A
derivative shaped band at (-) 1659/ (+) 1654 cm-1 can be assigned to amide I stretching or
to scrambling of the label into other amino acids. From the model UVRR compound
studies, vibrational modes associated with the indole ring are expected to contribute to
the spectrum from ~1560-1200 cm-1 (Hu et al. 1997).
Figure 5C contains multiple
vibrational frequencies in this region; however model compound studies suggest larger
spectral features in the 1480-1300 cm-1 region of the spectrum in association with
tryptophan vibrations.
The spectral features observed in the 2H5-Trp isotope-edited
spectrum may be from isotope scrambling or assigned to vibrational frequencies of the
tryptophan dimer. The π-stacking interaction of the conserved tryptophan dimer located
near P700 may affect the vibrational spectrum; π-π interactions may shift vibrations to
unique frequencies.
To resolve these discrepancies, hydrolyzed PSI samples were
subjected GC/MS analysis to quantify the amount of isotope enrichment into amino acids
other than tryptophan. From this analysis, less the 1% label scrambling was found in the
144
thirteen amino acids monitered using this procedure (data not shown).
UVRR
absorbance spectra from 1H5-Trp PSI and 2H5-Trp PSI samples showed multiple
tryptophan vibrational frequencies shifted in the labeled PSI samples (data not shown).
These results indicate that the vibrational frequencies observed in Figure 5C are from
tryptophan.
5.5 Discussion
Previous studies have illustrated the importance of protein relaxation due to
electron transfer in photosynthesis (Kim et al 2000; Kim et al/ 2001; Sacksteder et al.
2005; Bender et al. 2008a; Bender et al. 2008b). In the bacterial reaction center, it has
been demonstrated that protein relaxation events occur on the same time scale as electron
transfer (Wang et al. 2007). In PSI, studies have demonstrated heterogeneity of the
vibrational frequencies of the terminal electron donor P700. These studies have suggested
light-induced structural changes during charge separation (Kim et al. 2000; Kim et al.
2001; Bender et al. 2008a).
Furthermore, hydration studies on PSI have shown a
profound effect of hydration levels on the rate of P700+ reduction and its interaction with
the protein environment (Sacksteder et al. 2005). These protein dynamics may influence
the oxidation of soluble electron carriers.
To probe the protein dynamics that may be associated with these interactions, we
have employed a dialysis method for deuterium solvent exchange (Jenson et al. 2007). A
three-day dialysis, which included exchange of the buffer solution daily, was employed
(see Materials and Methods). The hydration levels of the FTIR samples were kept
constant, with a ratio of the OH stretching at 3300 cm-1 to amide 1 between 1.1-1.5.
145
The isotope-edited spectra presented here are of intact PSI (Figure 4A), and PSI
samples with the stromal subunits removed (Figure 4B). All the samples have undergone
the three-day dialysis into either 1H2O HEPES pH 7.5 buffer and 0.04% dodecyl-ß-Dmaltiside or 2H2O HEPES pH 7.5 buffer and 0.04% dodecyl-ß-D-maltiside. In Figure
4A, the photoaccumulated species are P700+ and FB-. Vibrational frequencies are expected
from amino acids perturbed by these photoaccumulated species and sensitive to
deuterium exchange. The photoaccumulated species associated with Figure 4B is P700+
and FX-. Amino acids perturbed by these photoaccumulated species and are sensitive to
deuterium exchange are expected to contribute to the spectrum in Figure 4B. Similar
bands observed in the 1800-1200 cm-1 region of both spectra are expected to be from
puerturbations associated with the oxidation of P700+. These perturbations may be
associated with the PsaA/PsaB heterodimer that binds P700, A0, A1 and FX.
Bands
assignable to aspartate and glutamate anitsymmetric COO- stretching are observed in
both spectra. The carboxylate vibrational bands shift to higher frequencies during charge
separation, indicating structural changes around these residues. Bands assignable to
amide I C=O stretching vibrations are also observed in both spectra. The amide I
stretching may be associated with a light-induced structural change. These structural
changes are most likely associated with the PsaA and/or PsaB subunits since they are
observed in both spectra.
Previous mutation studies of in the H loop of PsaB have suggested basic amino
acid residues located in this region are important for electrostatic interactions with
plastocyanin (Navarro et al. 2000; Durán et al. 2006).
Furthermore, mutations to
conserved residues including tryptophan 622 (Synechocystis numbering) in the luminal J
146
loop of the PsaB subunit, demonstrated the role of the loop in the interaction with soluble
electron carriers. This study underlined the importance of the luminal J loop, which
includes tryptophan and aspartic acid residues, in the interaction of PSI with plastocyanin
and cytochrome c6 (Sun et al. 1999b). The perturbations observed in the tryptophan
isotope-edited spectrum (Figure 5C) may be from a protein conformational change in this
conserved luminal loop region.
Figure 6A shows the tryptophan residues located in the PsaA, PsaB, PsaC, PsaD,
and PsaE subunits. Three tryptophan residues are found in PsaD and PsaE and may also
be perturbed during charge separation in PSI. Figure 6B shows the protein structure of
the PsaA and PsaB subunits with the acidic amino acids highlighted (Jordan et al. 2001).
The two large loops of PsaA and PsaB are believed to be important with the interaction of
the soluble electron donor plastocyanin (Sun et al. 1999a; Sun et al. 1999b; Fromme et al.
2003). Both loops contain glutamate and aspartate, and may be perturbed during P700
oxidation.
Two conserved tryptophan residues (Trp 655 and Trp 631) on the luminal side of
PSI have been shown to be important in electron transfer from plastocyanin to P700+
(Sommer et al. 2004). These tryptophan residues are within π-stacking distance, and may
contribute to the 2H5-Trp isotope-edited spectrum in Figure 5C. Near these conserved
tryptophan molecules are two aspartate residues (Asp 652 and Asp 628). Figure 7 shows
these residues as well as P700. The deuterium studies presented here indicate acidic
amino acids perturbed during electron transfer in PSI. Furthermore these studies indicate
a protein conformational change in PsaA and/or PsaB which may be a result of the
luminal loop regions. Both of the aspartate residues shown in Figure 7 may be perturbed
147
during P700+ formation, and may be important in electron transfer from plastocyanin to
P700+. Further studies are needed to indicate the precise role these aspartate residues and
tryptophan residues may play in light-induced electron transfer in PSI.
5.6 Acknowledgements
This work was supported by NIH GM 43273 (B.A.B). The authors would like to thank
Professor Roseann Sachs for performing the PSI acid-hydrolysis.
148
5.7 References
Barry, B. A. and G. T. Babcock (1987). "Tyrosine radicals are involved in the
photosynthetic oxygen evolving system." Proc. Natl. Acad. Sci. USA 84: 70997103.
Barth, A. (2000). "The infrared absorption of amino acid side chains." Prog. Biophys.
Mol. Biol. 74: 141-173.
Bender, S. L. and B. A. Barry (2008a). "Light-induced protein dynamics in Photosystem
I." Biophys. J. 93: 3927-3934.
Bender, S. L., J. Keough, S. E. Boesch, R. A. Wheeler and B. A. Barry (2008b). "The
vibrational spectrum of the secondary electron acceptor, A1, in photosystem I." J.
Phys. Chem. B 112: 3844-3852.
Bottin, H. and P. Mathis (1985). "Interaction of plastocyanin with the photosystem I
reaction center: a kinetic study by flash absorption spectroscopy " Biochemistry
24: 6453-6460.
Chirgadze, U. N., O. V. Fedorov and N. P. Trushina (1975). "Estimation of amino acid
residue side-chain absorption in the infrared spectra of protein solutions in heavy
water." Biopolymers 14: 679-694.
Durán, R. V., M. Hervas, B. De la Cerda, M. A. De la Rosa and J. A. Navarro (2006). "A
laser flash-induced kinetic analysis of in vivo Photosystem I reduction by site
directed mutants of plastocyanin and cytochrome c6 in Synechocystis sp. PCC
6803." Biochemistry 45: 1054-1060.
Fromme, P., P. Jordan and N. Krauss (2001). "Structure of photosystem I." Biochim.
Biophys. Acta 1507: 5-31.
Fromme, P., A. Melkozernv, P. Jordan and N. Krauss (2003). "Structure and function in
Photosystem I: interaction with its soluble electron carriers and external antenna
systems." FEBS Letters 555: 40-44.
Golbeck, J. H. (1994). In The Molecular Biology of Cyanobacteria. Dordrecht, Kluwer
Academic Publishers.
Haehnel, W., T. Jansen, K. Guase, R. B. Kloegen, B. Stahl, D. Michl, B. Huvermann, M.
Karas and R. G. Hermann (1994). "Electron transfer to plastocyanin to
Photosystem I." The EMBO Journal 13: 1028-1038.
Hervas, M., J. A. Navarro, A. Diaz, H. Bottin and M. A. De la Rosa (1995). "Laser-flash
kinetic analysis of the fast electron transfer from plastocyanin and cytochrome c6
149
to Photosystem I. Experimental evidence on the evolution of the reaction
mechanism." Biochemistry 34: 11321-11326.
Hervas, M., J. A. Navarro, A. Diaz and M. A. De la Rosa (1996). "A comparative
thermodynamic analysis by laser-flash absorption spectroscopy of Photosystem I
reduction by plastocyanin and cytochrome c6 in Anabaena PCC 7119,
Synechocystis PCC 6803, and spinach." Biochemistry 35: 2693-2698.
Hervas, M., J. M. Ortega, J. A. Navarro, A. Diaz, M. A. De la Rosa and H. Bottin (1994).
"Laser flash kinetic analysis of Synechocystis PCC 6803 cytochrome c6 and
plastocyanin oxidation by Photosystem I." Biochim. Biophys. Acta 1184: 235-241.
Ho, K. K. and D. Krogmann, W, (1988). "Electron donor to P700 in cyanobacteria and
algae. An instance for unusual genetic variabilitiy." Biochim. Biophys. Acta 766:
310-316.
Holzwarth, A. R., M. G. Muller, J. Niklas and W. Lubitz (2006). " Ultrafast transient
absorption studies on photosystem I reaction centers from Chlamydomonas
reinhardtii. 2: Mutations near the P700 reaction center chlorophylls provide new
insight into the nature of the primary electron donor." Biophys. J. 90: 552-565.
Hu, X. and T. G. Spiro (1997). "Tyrosine and tryptophan structure markers in
hemogloben utlraviolet resonance raman spectra: mode assignments via subunitspecific isotope labeling of the recombinant protein." Biochemistry 36: 1570115712.
Jenson, D. L., A. Evans and B. A. Barry (2007). "Proton-coupled electron transfer and
tyrosine D of photosystem II." J. Phys. Chem. B 111: 12599-12604.
Jordan, P., P. Fromme, H. T. Witt, O. Klukas, W. Saenger and N. Krauss (2001). "Threedimensional structure of cyanobacterial photoystem I at 2.5A resolution." Nature
411: 909-917.
Kim, S. and B. A. Barry (2000). "Identification of carbonyl modes of P700 and P700+ by in
situ chlorophyll labeling in photosystem I." J. Am. Chem. Soc. 122: 4980-4981.
Kim, S., C. A. Sacksteder, K. A. Bixby and B. A. Barry (2001). "A reaction-induced FTIR study of cyanobacterial photosystem I." Biochemistry 40: 15384-15395.
Kruip, J., P. Chitnis, B. Lagoutte, M. Rogner and E. J. Boekema (1997). "Structural
organization of the major subunits in cyanobacterial Photosystem I." J. Biol.
Chem. 272: 17061-17069.
Maruyama, T. and H. Takeuchi (1995). "Effects of hydrogen bonding and side-chain
conformation on the raman bands of tryptophan-2,4,5,6,7-d5." J. Raman
Spectrosc. 26: 319-324.
150
Merchant, S. and L. Bogorad (1986). "Regulation by copper of the expression of
plastocyanin and cytochrome c-552 in Chlamydomonas reinhardtii." Mol. Cell.
Biol. 6: 462-469.
Navarro, J. A., M. Hervas, J. Sun, B. De la Cerda, P. R. Chitnis and M. A. De la Rosa
(2000). "Negatively charged residues in the H loop of PsaB subunit in
Photosystem I from Synechocystis sp PCC 6803 appear to be responsible for
electrostatic repulsions with plastocyanin." Photosynth. Res. 65: 63-68.
Noren, G. H., R. J. Boerner and B. A. Barry (1991). "EPR characterization of an oxygenevolving photosystem II preparation from the transformable cyanobacteria
Synechocystis 6803." Biochemistry 30: 3943-3950.
Parrett, K. G., T. Mehari and J. H. Golbeck (1990). "Resolution and reconstitution of the
cyanobacterial Photosystem I complexes." Biochim. Biophys. Acta 1015: 341-352.
Parrett, K. G., T. Mehari, P. G. Warren and J. H. Golbeck (1989). "Purification and
properties of the intact P700 and FX containing Photosystem I core protein."
Biochim. Biophys. Acta 973: 324-332.
Poulson, C. and R. Verpoorte (1991). "Roles of chorismate mutase, isochorismate
synthase and anthranilate synthase in plants." Phytochemistry 30: 377-386.
Rath, P., W. J. DeGrip and K. J. Rothschild (1998). "Photoactivation of rhodopsin causes
an increased hydrogen-deuterium exchange of buried peptide groups." Biophys. J.
74: 192-198.
Sacksteder, C. A., S. L. Bender and B. A. Barry (2005). "Role for bound water and CH-pi
aromatic interactions in photosynthetic electron transfer." J. Am. Chem. Soc. 127:
7879-7890.
Schowen, K. B. and R. L. Schowen (1982). "Solvent isotope effects of enzyme systems."
Methods Enzymol. 87: 551-606.
Smith, L. C. (1962). "Control of 3-deoxy-D-arabino-heptulosonic acid 7-phosphate
synthesis by phenylalanine and tyrosine." J. Biol. Chem. 237: 3566-3570.
Sommer, F., F. Drepper, W. Haehnel and M. Hippler (2004). "The hydrophobic
recognition site formed by residues PsaA-Trp651 and PsaB-Trp627 of
photosystem I in Chlamydomonas reinhardtii confers distinct selectivity for
binding of plastocyanin and cytochrome c6." J. Biol. Chem. 279: 20009-20017.
Sommer, F., F. Drepper and M. Hippler (2002). "The luminal helix l of PsaB is essential
for recognition of plastocyanin or cytochrome c6 and fast electron transfer to
photosystem I in Chlamydomonas reinhardtii." J. Biol. Chem. 277: 6573-6581.
Srinivasan, P. R. and D. B. Sprinson (1959). "2-keto-3-deoxy-D-arabo-heptonic acid 7phosphate synthetase." J. Biol. Chem. 234: 716-722.
151
Sun, J., M. Hervas, J. A. Navarro, M. A. De la Rosa and P. R. Chitnis (1999a).
"Oxidizing side of the cyanobacterial photosystem I: mutational analysis of the
luminal H loop of the PsaB subunit." Photosynth. Res. 62: 241-250.
Sun, J., W. Xu, M. Hervas, J. A. Navarro, M. A. De la Rosa and P. R. Chitnis (1999b).
"Oxidizing side of the cyanobacterial Photosystem I. Evidence for interaction
between the electron donor proteins and a luminal surfance helix of the PsaB
subunit." J. Biol. Chem. 274: 19048-19054.
Tamm, L. K. and S. A. Tatulian (1997). "Infrared spectroscopy of proteins and peptides
in lipid bilayers." Quarterly Reviews of Biophysics 30: 365-429.
Wang, H., S. Lin, J. P. Allen, J. C. Williams, S. Blankert, C. Laser and N. W. Woodbury
(2007). "Protein dynamics control the kinetics of initial electron transfer in
photosynthesis." Science 316: 747-750.
Wolpert, M. and P. Hellwig (2006). "Infrared Spectra and Molar Absorption Coefficients
of the 20 Alpha Amino Acids in Aqueous Solution in the Spectral Range of 1800500 cm-1." Spectrochim. Acta, Part A 64: 987-1001.
Wood, P. M. (1978). "Interchangeable copper and iron proteins in algal photosynthesis.
Studies on plastocyanin and cytochrome c-552 in Chlamydomonas." Eur. J.
Biochem. 87: 9-19.
152
Figure 1. Light-induced FT-IR difference spectra of intact PSI. (A) The P700+FB--minusP700FB spectrum from PSI dialyzed with 1H2O HEPES-NaOH pH 7.5 buffer. (B) P700+FB-minus-P700FB spectrum from PSI dialyzed with 2H2O HEPES-NaOH pH 7.5 buffer. The
spectrum in (A) is overlaid as a dotted line. (C) The isotope-edited spectrum generated
by a direct one to one subtraction of the spectrum in (B) from the spectrum in (A). (D)
Control double difference spectrum generated by subtracting ½ the data set in (A) from
the second half of the data set. (E) Dark minus dark spectrum of PSI dialyzed with 1H2O
HEPES-NaOH pH 7.5 buffer. Spectra in (A) (B) and (E) are an average of 64 spectra.
Each mark on the y-axis represents 2x10-3 Absorbance units (AU).
See Materials and
Methods for additional information.
153
Figure 2. SDS-Urea page gel of PSI samples. Lane 1 is of PSI samples that have been
treated with 6.8M urea to remove the PsaC, PsaD, and PsaE subunits. Lane 2 contains
PSI samples containing all extrinsic subunits (PsaC, PsaD, and PsaE) as shown by the
arrows.
154
Figure 3. Light-induced FT-IR difference spectra of PSI samples treated with 6.8M urea
to remove the PsaC, PsaD and PsaE subunits. (A) The P700+FX--minus-P700FX spectrum of
dialyzed with 1H2O HEPES-NaOH pH 7.5 buffer.
(B) The P700+FX--minus-P700FX
spectrum of dialyzed with 2H2O HEPES-NaOH pH 7.5 buffer. The spectrum in (A) is
overlaid as a dotted line. (C) The isotope-edited spectrum generated by a direct one to
one subtraction of the spectrum in (B) from the spectrum in (A). (D) Control double
difference spectrum generated by subtracting ½ the data set in (A) from the second half
of the data set. (E) The Dark minus dark spectrum of PSI dialyzed with 1H2O HEPESNaOH pH 7.5 buffer. Spectra in (A) (B) and (E) are an average of 50 spectra. Each mark
on the y-axis represents 2x10-3 (AU). See Materials and Methods for more information.
155
Figure 4. The isotope-edited spectra of deuterium exchanged PSI. (A) P700+FB--minusP700FB isotope edited spectrum. (B) P700+FX--minus-P700FX isotope edited spectrum. The
dotted spectra overlayed in (A) and (B) are the control double difference spectra in which
no vibrational bands are expected. Each mark on the y-axis represents 2x10-4 (AU). See
Materials and Methods for additional information.
156
Figure 5. Light-induced FT-IR difference spectra of intact PSI. (A) The P700+FB--minusP700FB spectrum of PSI isolated from Synechocystis cultures grown with 0.25 mM 1H5tryptophan.
(B) The P700+FB--minus-P700FB spectrum from PSI isolated from
Synechocystis cultures grown with 0.25 mM 2H5-tryptophan. (C) The isotope-edited
spectrum generated by a direct one to one subtraction of the spectrum in (B) from the
spectrum in (A). (D) Control double difference spectrum generated by subtracting ½ the
data set in (A) from the second half of the data set. (E) The dark minus dark spectrum of
PSI isolated from Synechocystis cultures grown with 0.25 mM 1H5-tryptophan.. Spectra
in (A) (B) and (E) are an average of 50 spectra. Each mark on the y-axis represents 2x103
(AU). See Materials and Methods for more information.
157
Figure 6. The 2.5 Ǻ resolution crystal structure of PSI depicting the tryptophan residues
in the PsaA, PsaB, and PsaC subunits of PSI (A) and acidic amino acids found in the
PsaA and PsaB subunits of PSI (B). The figure is reproduced from the 2.5 Å resolution
crystal structure of PSI from S. elongatus (PDB file accession number 1JB0 (Jordan et al.
158
2001)) using Swiss-Pdb viewer (v3.7; www.expasy.ch/spdbc), and rendered using POVRay (v3.5; www.povray.org).
159
Figure 7. The protein environment around P700. The two conserved tryptophan residues
and aspartate residues located in the luminal loop region that may be perturbed during
electron transfer in PSI are shown. The luminal loop regions of PsaA and PsaB are also
shown as ribbons. The figure is reproduced from the 2.5 Å resolution crystal structure of
PSI from S. elongatus (PDB file accession number 1JB0 (Jordan et al. 2001)) using
Swiss-Pdb
160
Figure 7 continued
viewer
(v3.7;
www.expasy.ch/spdbc),
and
rendered
using
POV-Ray
(v3.5;
www.povray.org).
161
CHAPTER 6
CONCLUSION
Photosystem I (PSI) is a protein complex involved in light-induced electron
transfer across the thylakoid membrane. Electron transfer in PSI involves a series of
oxidation reduction reactions of cofactors bound to the subunits PsaA, PsaB and PsaC
(Jordan et al. 2001) Protein dynamics have shown to play a major role in photosynthetic
electron transfer reactions (Warshel et al. 1989; Gehlen et al. 1994; Halverson et al. 2003;
Parson et al. 2004; Barry et al. 2006; De Riso et al. 2006; Wang et al. 2007). The studies
presented here have underscored the role of protein dynamics and cofactor interactions in
the control of electron transfer in PSI. The protein relaxations observed in these studies
have the potential to control the bidirectionality inherent to PSI, and may control the
interactions with soluble electron donors and acceptors.
Characterization of the secondary electron acceptor A1 has shown the influence of
noncovalent interactions on the vibrational spectrum. Hydrogen-bonding, π-stacking and
electrostatic interactions observed in previous studies have the potential to modulate the
redox properties of the secondary electron acceptor (Pushkar, Y. et al. 2004; Pushkar, Y.
N. et al. 2005; Feldman et al. 2007; Bender et al. 2008). The work presented here has
shown the effect of these interactions on the vibrational spectrum of A1 in comparison to
DFT calculations of the quinone molecule.
162
Novel experiments incorporating various isotopes into the keto group of chl have
given greater insight into the oxidation of P700+. These studies suggested a light-induced
conformational change in PSI, which may regulate the oxidation of soluble electron
donors or other electron transfer reactions. Deuterium exchange studies revealed a lightinduced conformational change in PSI, which may be related to the previous protein
relaxations identified by isotope labeling of P700.
Furthermore, these studies have
characterized aspartate or glutamate perturbations associated with the PsaA subunit
and/or the PsaB subunit. These novel perturbations indicate structural changes associated
with the loop regions of these two subunits. To further investigate protein dynamics
associated with stable charge separation, tryptophan residues were isotopically labeled in
PSI.
The isotope-edited spectrum of 2H5-Trp PSI reveal vibrational frequencies that
can be associated with amide I perturbations as well as the indole ring of tryptophan.
These results indicate unique vibrational frequencies of tryptophan that may be indicative
of the conserved tryptophan dimer found on the luminal side of PSI (Sun et al. 1999;
Jordan et al. 2001; Fromme et al. 2003; Sommer et al. 2004). These results combined
with the deuterium-exchange results emphasize the importance of protein relaxation
events that accompany electron transfer reaction in PSI.
The studies presented here underscore the importance of protein dynamics and
cofactors interactions in oxygenic photosynthesis. The results signify specific protein
relaxation events that accompany stable charge separation in PSI. Further studies are
needed to specify the role of these protein dynamics; however, the studies presented here
163
are indicative of protein conformational changes that may control the oxidation of soluble
electron carriers or other electron transfer reactions in PSI.
164
6.1 References
Barry, B. A., I. Cooper, A. De Riso, S. H. Brewer, D. M. Vu and R. B. Dyer (2006).
"Time-resolved vibrational spectroscopy detects protein-based intermediates in
the photosynthetic oxygen-evolving cycle." Proc. Natl. Acad. Sci. USA 103:
7288-7291.
Bender, S. L., J. Keough, S. E. Boesch, R. A. Wheeler and B. A. Barry (2008). "The
vibrational spectrum of the secondary electron acceptor, A1, in photosystem I." J.
Phys. Chem. B 112: 3844-3852.
De Riso, A., D. L. Jenson and B. A. Barry (2006). "Calcium exchange and structural
changes during the photosynthetic oxygen evolving cycle." Biophys. J. 91: 19992008.
Feldman, K. S., D. K. Hester 2nd and J. H. Golbeck (2007). "A relationship between
amide hydrogen bond strength and quinone reduction potential: Implications for
photosystem I and bacterial reaction center quinone function." Bioorg. Med.
Chem. Lett. 17: 4891-4894.
Fromme, P., A. Melkozernv, P. Jordan and N. Krauss (2003). "Structure and function in
Photosystem I: interaction with its soluble electron carriers and external antenna
systems." FEBS Letters 555: 40-44.
Gehlen, J. N., M. Marchi and D. Chandler (1994). "Dynamics affecting the primary
charge transfer in photosynthesis." Science 263: 499-502.
Halverson, K. M. and B. A. Barry (2003). "Evidence for spontaneous structural changes
in a dark-adapted state of photosystem II." Biophys. J. 85: 2581-2588.
Jordan, P., P. Fromme, H. T. Witt, O. Klukas, W. Saenger and N. Krauss (2001). "Threedimensional structure of cyanobacterial photoystem I at 2.5A resolution." Nature
411: 909-917.
Parson, W. W. and A. Warshel (2004). "Dependence of photosynthetic electron-transfer
kinetics on temperature and energy in a density-matrix model." J. Phys. Chem. B
108: 10474-10483.
Pushkar, Y., D. Stehlik, M. Van Gastel and W. Lubitz (2004). "An EPR/ENDOR study of
the asymmetric hydrogen bond between the quinone electron acceptor and the
protein backbone in Photosystem I." Journal of Molecular Structure 700: 233241.
Pushkar, Y. N., O. Ayzatulin and D. Stehlik (2005). "Transient and pulsed EPR study of
17
O-substituted methyl-naphthoquinone as radical anion in the A1 binding site of
photosystem I and in frozen solution." Appl. Magn. Reson. 28: 195-211.
165
Sommer, F., F. Drepper, W. Haehnel and M. Hippler (2004). "The hydrophobic
recognition site formed by residues PsaA-Trp651 and PsaB-Trp627 of
photosystem I in Chlamydomonas reinhardtii confers distinct selectivity for
binding of plastocyanin and cytochrome c6." J. Biol. Chem. 279: 20009-20017.
Sun, J., W. Xu, M. Hervas, J. A. Navarro, M. A. De la Rosa and P. R. Chitnis (1999).
"Oxidizing side of the cyanobacterial Photosystem I. Evidence for interaction
between the electron donor proteins and a luminal surfance helix of the PsaB
subunit." J. Biol. Chem. 274: 19048-19054.
Wang, H., S. Lin, J. P. Allen, J. C. Williams, S. Blankert, C. Laser and N. W. Woodbury
(2007). "Protein dynamics control the kinetics of initial electron transfer in
photosynthesis." Science 316: 747-750.
Warshel, A., Z. T. Chu and W. W. Parson (1989). "Dispersed polaron simulations of
electron transfer in photosynthetic reaction centers." Science 246: 112-116.
166