.143 A.

.51..“
o . a
ﬁg? Ln...»

 

.. 1:...
:50.“

.1
. .x.
a .r u
. 5%.?-
AS. .333..{.....
“W...n.r.a¢a.?kt

3.32....I

.! ’f
#5533... 2..

3v 2 .s a!!!

a I

.
:in
{hum ”ﬂrﬁﬁg
.5... Ash”-.. ”2
Hi

to!
.Ilar.
1’55...
2...;
.3. 1.1.2: .I

(:32.
SJ... .6931:

1.. J:
a .9...!'....u.ddﬂ.

Jr

:3?! V
1.3).:

 

r,
11—... :7:
4 s) 1...... .

v -.K-lv}..‘
52.9..
C. A:
. 7133......293‘
:2. .t 10...: C 2......
.I :: .II... I. It...
:2

 

a: t. I...)
x...|\ '|b:l .

 

>1L1F.rw 1‘.

THESIS

/.r I‘ "3-)

LVU/

This is to certify that the

dissertation entitled

THE EFFECTS OF SR2+-SUBSTITUTION ON THE REDUCTION
RATES OF YZ' AND ON THE HYDROGEN ATOM ABSTRACTION
MECHANISM IN PHOTOSYSTEM 11

presented by
Kristi L. Westphai

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degree in Chemistry

 

 

ﬂ 7
477 xfy/wé

 

Major professor

/

Date ﬂy”? M 2/07?
J

MS U i: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

 

n mar-mm T
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 CJCIRC/Dateouop65-p. 15

THE EFFECTS OF SRZT-SUBSTITUTION ON THE REDUCTION RATES
OF Yz' AND ON THE HYDROGEN ATOM ABSTRACTION MECHANISM IN

PHOTOSYSTEM II

By

Kristi L. Westphal

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

2000

ABSTRACT

THE EFFECTS OF SRZT-SUBSTITUTION ON THE REDUCTION RATES
OF Yz° AND ON THE HYDROGEN ATOM ABSTRACTION MECHANISM IN

PHOTOSYTEM II
By

Kristi L. Westphal

Photosystem II is a photosynthetic enzyme that catalyzes the light-driven
oxidation of water to molecular oxygen. A total of four sequential light-induced
turnovers of Photosystem II are required to form molecular oxygen. Each stable
intermediate produced by a ﬂash of light is called an Sn-state, n = 0-4. A general
consensus has not yet been reached concerning the mechanism of this reaction
or the structure of Photosystem ll, although much progress has been made over
the last few years. It is generally agreed upon that the oxygen-evolving complex
of Photosystem ll consists of a manganese cluster made up of four manganese
atoms, a redox-active tyrosine called Y2, histidine 190, and Ca2+ and Cl' ions.
Recently, several models have been proposed that directly involve Yz in the
water oxidation chemistry as a hydrogen atom abstractor. Perhaps the most
elusive subject is the role of Ca2+ in PSII. Two main proposals have been set

forth: 1) Ca” plays a structural role and a binding site for CI' and, 2) Ca2+ is

directly involved in the oxygen-oxygen bond forming reaction. One way to
acquire more information on the role of Ca” is to perform biochemical treatments
that perturb this cofactor and then monitor the kinetic efﬁciency of the oxygen-
evolving complex.

Several groups have recently investigated the effects of biochemical
treatments, site-directed mutagenesis, or substitution of essential cofactors on
the kinetics of the stepwise, water-oxidizing chemistry catalyzed by Photosystem
ll. Consistently, these studies showed evidence for a slowing of the ﬁnal,
oxygen-releasing step, 83-)80, of the catalytic cycle. To a degree, some of this
work also showed a slowing of the earlier S-state transitions. To test the roles
that have been proposed for Ca”, PSII membranes were depleted of Ca2+ and
subsequently reconstitution with Sr”. The rates of the S-state transitions in
these samples were monitored by using time-resolved electron paramagnetic
resonance spectroscopy. The results show a slowdown of the last transition in
the cycle, consistent with an earlier report from Boussac and Rutherford
(Boussac, A., Setif, P. and Rutherford, AW. (1992) Biochemistry 31, 1224-1234),
and of the earlier S-state transitions as well. Our observations that both the
lower and higher S-state transition rates decrease suggest that a common
molecular mechanism is at work and that Sr” is less effective than Ca2+ in
supporting it.

The time-resolved electron paramagnetic resonance spectroscopy results
were also analyzed in the context of the hydrogen atom abstraction mechanism.

The retarding effect caused by Sr” substitution is additional evidence to support

this mechanism. Furthermore, the hydrogen atom abstraction mechanism is best
able to explain the unusual characteristics of the S-state transitions: slow rates,

low activation energies, low pre-exponential factors, and small kinetic isotope

effects.

In Memory of My Father, Bruce G. Westphal

ACKNOWLEDGMENTS

I would like to begin by thanking my research advisor, Jerry Babcock, for
his guidance, patience, and support. I couldn’t have asked for a better advisor!

I would like to thank my mother, Diana Westphal, for her words of
encouragement when I needed them most and for her never-ending faith in my
becoming Dr. Westphal.

I would also like to thank my boyfriend, Marcus Sims, for putting up with
me when I knew I was unbearable to be around and for always being able to
make me laugh! Also, for his words of encouragement when l was pulling my
hair out in frustration (especially writing this thingl).

I would lastly like to thank my lab mates and friends for being there when I
needed help or just needed a good laugh (some of you probably didn’t even
know you were supplying the latter!) I would particularly like to thank three of
these people. First, my partner in crime, Heather Sackett, without you I would
have had no one to talk stupid with or to surf the net for questionable web sites!
Second, my second partner in crime and softball buddie, Shannon Haymond,
may we never have another wine night like the ﬁrst! Thirdly, Hans Schelvis, my
desk neighbor for the past ﬁve years. You were always there to answer my
questions and to make the best coffee! I will miss you all dearly.

Thanks to everyone who have made my graduate school experience one

that I will never forget!

vi

TABLE OF CONTENTS

List of Tables ....................................................................................... vii
List of Figures ..................................................................................... viii
Chapter 1 .............................................................................................................. 1
1 Introduction ..................................................................................................... 1
1.1 The Electron Transport Chain in Photosynthesis ..................................... 1
1.2 Photosystem ll ......................................................................................... 3
1.2.1 Polypeptide Composition and Electron Transfer Reactions ................ 3

1.2.2 S-state Cycle in Water Oxidation ........................................................ 6

1.2.3 The Energetics of Water Oxidation ..................................................... 8

1.2.4 The Manganese Cluster ...................................................................... 8

1.2.5 Y0 and Y2 .......................................................................................... 12

1.2.6 A Model for Water Oxidation ............................................................. 13

1.2.7 The Reduction Rates of Y2' ............................................................... 15

1.2.8 Calcium ............................................................................................. 19
1.2.9 Proposed Roles for Ca2+ in Water Oxidation ..................................... 20

1.3 Project Goals ......................................................................................... 22
Chapter 2 ............................................................................................................ 25
2 Electron Paramagnetic Resonance Spectroscopy ....................................... 25
2.1 Basic Principles ...................................................................................... 25
2.2 The EPR Spectrometer .......................................................................... 28
Chapter 3 ............................................................................................................ 33
3 Materials and Methods ................................................................................. 33
3.1 Preparation of PSII Membranes ............................................................. 33
3.2 Calcium Depletion of PSII Membranes .................................................. 35
3.3 Sample Preparation for EPR Experiments ............................................. 35
3.4 Time-Resolved EPR Experiment Setup ................................................. 36
Chapter 4 ............................................................................................................ 39
4 Results of Time-Resolved EPR Experiments ............................................... 39
4.1 S-State Mixed Kinetics ........................................................................... 39
4.2 S-State Resolved Kinetics ...................................................................... 41
4.3 Effects of pH and Isotope Exchange ...................................................... 45
Chapter 5 ............................................................................................................ 53
5 Hydrogen Atom Abstraction .......................................................................... 53
5.1 Electron Transfer ................................................................................... 53
5.2 Proton Coupled Electron Transfer: Sequential or Concerted? ............... 56
5.3 Concerted Electron and Proton Transfer ................................................ 62
Bibliography ........................................................................................ 71

vii

Table 1-1

Table 4-1

Table 5-1

Table 5-2

Table 5-3

Table 5-4

LIST OF TABLES

Half-times for S-state Advance in Various Enzyme
Preparation Types .............................................................. 18

Rates of Y2' Reduction for Ca2+- or Sr2+-Reconstituted
PSII Membranes Determined by Time-Resolved EPR
Spectroscopy .................................................................... 48

Reduction Potentials and pKa Values Used in Determining the
Viability of the ET/PT and PT/ET Pathways ............................. 58

Kinetic and Thermodynamic Parameters for the S-state
Transitions ....................................................................... 61

Kinetic and Thermodynamic Parameters for H-atom
Abstraction Reactions From an Oxygen Atom by an
Oxy-based Radical ............................................................. 63

Kinetic and Thermodynamic Parameters for Hydrogen
Atom Abstraction by Radicals ............................................... 64

viii

Figure 1-1

Figure 1-2

Figure 1-3

Figure 1-4

Figure 1-5

Figure 1-6
Figure 1-7

Figure 2-1

Figure 2-2
Figure 2-3
Figure 3-1

Figure 4-1

Figure 4-2

Figure 4-3

Figure 4-4

Figure 4-5

LIST OF FIGURES

Flow of electrons through the electron transport chain in
higher plants ...................................................................... 2

Schematic representation of the major polypeptides
and redox components of PSII ............................................... 4

A period-four oscillation of oxygen evolution as a
function of ﬂash number in spinach chloroplasts ........................ 7

Redox potentials of the cofactors of PSII .................................. 9

The 9 = 4.1 (a) and multiline (b) EPR signals observed

in PSII membranes ............................................................. 11
Tyrosine Yz oxidation and reduction in the 80 state .................... 14
A model for the S-state cycle in PSII ....................................... 16
Energy level scheme for a free electron as a function of

applied magnetic ﬁeld B ...................................................... 27
Block diagram for a continuous wave EPR spectrometer ............ 29
Continuous-wave EPR spectrum of Y0' ................................... 31
Setup for the time-resolved EPR experiments .......................... 37

Time-resolved EPR steady state kinetic traces (S-states
are mixed) for Y; reduction in salt-washed PSII membranes
reconstituted with (a) CaClz or (b) SrClz .................................. 40

Three-ﬂash S-state resolved EPR kinetic traces ........................ 42

Time-resolved EPR kinetic traces for the S1Yz' -) 8ng
transition triggered by the ﬁrst ﬂash in a three ﬂash experiment....44

Time-resolved EPR kinetic traces for the SzYz' -) 8ng
transition triggered by the second ﬂash in a three ﬂash
experiment ........................................................................ 46

Time-resolved EPR kinetic traces for the Sng' -> SOYz
transition triggered by the third ﬂash in a three ﬂash
experiment ........................................................................ 47

ix

Figure 4-6

Figure 4-7

Figure 5-1
Figure 5-2

Figure 5-3

Time-resolved EPR kinetic traces for the S1Yz'-) SzYz,

SzYz' -> Sng, Sng' -> SOYz transitions in salt-washed,
Srztreconstituted PSII membranes resuspended in buffers

at pH = 5.9, 6.5, 7.1 ............................................................ 49

Time-resolved EPR steady state kinetic traces (S-states
are mixed) for Y; reduction in salt-washed PSII

membranes reconstituted with SrClz in either D20 or H20 ........... 52
Potential energy surfaces of the donor and acceptor .................. 55
Possible reaction paths for proton and electron transfer ............. 57
Calculated activation energies (Ea) for paths 3 and b ................. 60

Chapter 1

1 Introduction

Two to three billion years ago photosynthetic cyanobacteria gained the
capability to extract electrons from water and produce molecular oxygen as a
byproduct. These organisms ﬂourished on an inexhaustible supply of water,
transforming the earth’s atmosphere from a composition of reduced gases to one
rich in 02. Prior to this event, all organisms on earth were adapted to an
anaerobic environment. As the atmosphere became enriched in oxygen, many
organisms became extinct while the others adapted to the new environment. As
a consequence, the highly efﬁcient process of respiration evolved and higher
organisms developed. Today, members of the higher plants, algae, and
cyanobacteria families are all capable of carrying out oxygenic photosynthesis.
1.1 The Electron Transport Chain in Photosynthesis

Photosynthesis in green plants is carried out by two photosystems,
Photosystem l (PSI) and Photosystem ll (PSII) which undergo photochemical
charge separation. The two photosystems are located in the stromal region of
the chloroplast in membranous structures called thylakoids, which are located
within the leaf of the plant. Two additional membranous components are also
involved in the electron transport chain: the cytochrome bef complex (cyt bet) and
adenosine 5’-triphosphate (ATP) synthase. The ﬂow of electrons through these
components is illustrated in Figure 1.1 and is initiated by a light-induced charge
separation in PSII, which takes electrons from water and delivers them to the

mobile electron carrier plastoquinone (PQ). Next, PQ travels to cyt bef, which is

1

ADP + Pi ATP

 

 
 

 

 

 

 

 

 

 

NADPH
A . .
~ 5 ATPZ
PSI synthase
2 H26 .><7__J
02 + 4H+ E.

Figure 1-1 Flow of electrons through the electron transport chain in
higher plants. This ﬁgure was kindly provided by Wanrvick Hillier.

involved in proton translocation, and then the electrons are taken up by a second
mobile electron carrier called plastocyanin (PC) which delivers the electrons to
PSI. Here the electrons are utilized in the terminal step of the electron transport
chain where a molecule of nicotinamide adenine dinucleotide phosphate
(NADP") is reduced to NADPH. In addition to the utilization of electrons for
NADPH production, electron ﬂow within each photosystem and between them
induces a membrane potential through the translocation of protons. The
electrochemical potential produced is then dissipated by ATP synthase in the
production of ATP. The photosynthetic end products, NADPH and ATP, are both
ultimately used by the plant to drive the synthesis of complex biomolecules
essential for growth and reproduction.

1.2 Photosystem ll
1.2.1 Polypeptide Composition and Electron Transfer Reactions

The photochemistry of PSII is performed on a protein scaffold that binds
pigments and cofactors involved in the reaction. A schematic representation of
the primary photochemical events that occur at PSII is depicted in Figure 1-2 (for
reviews see (1, 2)). Light is ﬁrst absorbed by the light-harvesting complex
located peripherally to the PSII core, which consists of non-covalently bound
pigment molecules including chlorophyll a, chlorophyll b and carotenoids (not
shown in ﬁgure). The absorbed photon is then funneled into the reaction center
to a specialized chlorophyll complex referred to as P680. Excitation of P580 results

in formation of the radical pair, P530'+Pheo“, which is stabilized by rapid electron

 

CP43
cyt b-559
/

2
m1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l7

 

 

 

 

 

 

 

 

Figure 1-2 Schematic representation of the major polypeptides and
redox components of PSII. The arrows show the ﬂow of electrons.

transfer to a bound plastoquinone molecule, GA. A second plastoquinone
molecule, 03, oxidizes QA' within 100-200 ps, and after an additional charge-
separation and subsequent protonation, the QBHz plastoquinol (PQ) formed
leaves the reaction center and is replaced by an oxidized quinone from the
plastoquinone pool. The electron hole that is produced at P530 is quickly reduced
in nanoseconds by a redox-active tryosine residue, Yz at D1-161. Subsequently,
Yz is reduced on the microsecond to millisecond time-scale by a Mn cluster
consisting of four Mn atoms. The (Mn)4 cluster and Y2, along with the necessary
cofactors Ca2+ and Cl‘, make up what is called the oxygen-evolving complex
(OEC) of PSII, and it is at this site where the oxidation of water is catalyzed.

In vivo, PSII contains at least 20 polypeptides, but the minimal oxygen-
evolving complex consists of seven major polypeptides: D1, DZ, CP43, CP47,
the or and B polypeptides of cyt b-559, and the extrinsic polypeptide of 33 kDa
(for a recent review see (3)). The D1 and DZ polypeptides, which each contain
ﬁve transmembrane helices, form the photochemical reaction center by binding
P580, Pheo, the non-heme iron, GA, 03, and contain the tyrosine Y2, as well as a
second redox active tyrosine referred to as YD. Both the D1 and DZ proteins
exhibit a very high degree of sequence homology and appear to form a
heterodimer with C2 symmetry. It is estimated that there are 6 chlorophyll 3
molecules and 2 ﬂ-carotene molecules per DI/D2 reaction center. An oxygen-
evolving reaction center requires, in addition to D1 and DZ, two other chlorophyll-
containing polypeptides known as CP47 and CP43. Both peptides contain about

15 bound chlorophyll 3 molecules and 2-3 molecules of B-carotene; they are

believed to act as inner antennae that transfer energy from the light-harvesting
complex to the reaction center P580. Cytochrome b-559, another necessary
component of PSII, consists of two smaller polypeptides known as the a and B
polypeptides, and has been proposed to protect PSII from photooxidative
damage. The seventh major polypeptide of PSII has a molecular weight of 33
kDa and is located extrinsically to PSII. It is referred to as the manganese
stabilizing protein (MSP) because it acts to protect the (Mn). complex from
exogenous reductants and optimizes the complex’s catalytic efﬁciency. Two
other extrinsic polypeptides are associated with PSII and have molecular weights
of 17 and 23 kDa. They are implicated in the efﬁcient production of oxygen by
acting as barriers that prevent Ca2+ from migrating out of the site. Besides the
major polypeptides described above, there are several small peptides associated

with functional PSII complexes whose exact roles are unknown.
1.2.2 S-state Cycle in Water Oxidation

One of the ﬁrst ground-breaking experiments in understanding water
oxidation was performed in 1969 by Joliot and coworkers who studied oxygen
yields from dark-adapted chloroplasts as a function of single-turnover ﬂashes (4).
This work, shown in Figure 1-3, demonstrated that after the third ﬂash and every
fourth ﬂash thereafter, a maximum in oxygen yield occurs. This pattern of oxygen
release indicated that a total of four sequential light-induced turnovers of PSII are
required to form molecular oxygen. Shortly after, Kok and coworkers interpreted
these results by formulating a kinetic model where each stable intermediate

produced by a ﬂash of light is called an S-state (5). Each S-state is designated
6

Relative 02 Yield vs. Flash Number

i

4

4.
O .
12 3-

8

>- - ,
C

>- ~

 

I T T I l

2 216 81D12141618
Flash Number

Independent PSII Units:

4'0" mp
hv hv h v hv
SO—TSl—T 82—> S3"> SK

‘T l 02+ 4H+

Figure 1-3 A period-four oscillation of oxygen evolution as a function of

ﬂash number in spinach chloroplasts. This ﬁgure was kindly provided by
Gerald T. Babcock.

 

.2

Relative 02 Yield vs. Flash Number

 

T I l I I

5316 81D12141618
Flash Number

Independent PSII Units:

49" 2H20
hv hv h v hv
sO—+s,—+ s —+ s3—>s‘<

2
02 + 4H+

Figure 1-3 A period-four oscillation of oxygen evolution as a function of

ﬂash number in spinach chloroplasts. This ﬁgure was kindly provided by
Gerald T. Babcock.

by Sn, where n is the number of oxidizing equivalents stored in the OEC. The 81
state predominates in dark-adapted samples and the 8.; state is a transient
intermediate that spontaneously reverts to the 80 state with the concomitant

release of 02.
1.2.3 The Energetics of Water Oxidation

Water is an extremely stable molecule as evidenced by the fact that it
covers over 75% of the earth’s surface. Thus, in order to oxidize water to
molecular oxygen very high reduction potentials are required. P586, a product of
the initial charge-separation reaction of PSII, is one of the strongest oxidants
known in biology (6) and is sufﬁcient enough to oxidize water (7). Figure 1-4
shows the reduction potentials of the major components of the donor and
acceptor sides of PSII. A total of 3.72 V must be accumulated by the OEC in
order to oxidize water (8). With a reduction potential of about 1.12 V, P530+
oxidizes Y2, which has a reduction potential of about 0.97 V. Therefore, about 4
V of energy is supplied by PSII to oxidize water, which leaves little surplus
energy available to support any potential ﬂuctuations as the OEC proceeds
through the S-state cycle. This weak coupling of the cofactors leads to the highly
efﬁcient conversion of about two thirds of the photon energy into chemical

potential (9-11).
1.2.4 The Manganese Cluster

The PSII complex contains four manganese atoms per P530. From

extended X-ray absorption ﬁne structure (EXAFS) measurements, a model was

8

 

 

 

 

- 1.0—
- 0.8—- Pm" \.
T PhCO
— 0.6 —
- 0.4 —-
E - 0.2— CIA
"a
.5 O __ QBHZ
3.3 hv
a? 0 2 _
Ea
g 0.4 —
°‘ 0.6 - r a
0.8 ~—
HZO“ 32/3,,33/52 Y
1.0 " 80/83 J Z \
1.2— K ”38%de

02 + proton gradient

Figure 1-4 Redox potentials of the cofactors of PSII.

suggested for the local structure of the Mn complex. The model was proposed to
consist of a pair of p-oxo bridged dimers with Mn-Mn distances of 2.7 A linked
together by a mono-,u-oxo bridge with a Mn-Mn separation of 3.3 A arranged in a
C-shaped structure (12, 13). There are two well-established low temperature
EPR signals refered to as the multiline and the g = 4.1 signals that have been
attributed to the 82 state of the (Mn). cluster (14, 15). The multiline signal, shown
in Figure 1-5 (b), displays 18-20 partially resolved hyperﬁne lines and has a
linewidth of about 1500-1800 G. The signal is generated by illumination of 02-
evolving PSII samples at 160-200 K (Dismukes, 1980 #211; Hansson, 1982
#180; Brudvig, 1983 #179). Two other forms of the multiline signal can be
generated by either Cay-depletion (16-18) or ammonia treatment (19-21) of PSII
samples. Both treatments affect the number and splittings of the hyperﬁne lines,
and have therefore been proposed to directly affect the structure of the (Mn)4
cluster.

The Q = 4.1 signal, shown in Figure 1-5 (a), is due to a high-spin, ground
state of the (Mn); cluster (22) and is generated by illumination of PSII samples at
140 K in the presence of high concentrations of cryoprotectant (15, 23). This
signal has a linewidth of 320-360 G and lacks resolved hyperﬁne structure. Upon
warming in the dark, the g = 4.1 signal converts to the multiline signal.

Inhibition of PSII samples by incubation with acetate, which causes
perturbations in the Ca2+ and Cl' sites, traps the OEC in a state in which the EPR
spectrum from Y; is split by a magnetic interaction with the manganese 32 state.

This split EPR signal was ﬁrst discovered by Rutherford and co-workers (24) and

10

 

 

dX"/d H

a) b)

 

 

 

 

 

I r
3000 4000

Field (G)

I
1000 2000

Figure 1-5 The 9 = 4.1 (a) and multiline (b) EPR signals observed in PSII

membranes. This spectrum was kindly provided by Pierre Dorlet.

11

was later identiﬁed as coming from the SzYz' state (17, 25, 26). This signal has
more recently been simulated by the Babcock (27) Britt (28), and Brudvig (29)
groups in order to determine the distance between (Mn). and Yz°. From the
results of the simulation studies, it was estimated that (Mn). and Y; are
separated by a center-to-center, point-dipolar distance of 8-9 A.

The oxidation state of each Mn atom in the cluster has been investigated
by using X-ray absorbance near-edge spectroscopy (XANES). This technique
has been used to detect oxidation state changes of many transition-metal
compounds. The results of the study performed on PSII showed signiﬁcant shifts
in the XANES edge positions on going from Soto S1 and from S1 to $2 and have
been interpreted as manganese-centered oxidations (30-32). Oxidation of the
manganese cluster on the $2 to $3 transition is more controversial with two
groups claiming a signiﬁcant edge shift (30, 32) and one group reporting only a
minimal shift in the edge position upon this transition (31). From XANES studies
on model compounds and PSII, the oxidation states of the four manganese
atoms are suggested to be: 80, II, III, IV, IV; S1, III, III, IV, IV; 82, III, IV, IV, IV; 83,
IV, IV, IV, lV.

1.2.5 Y0 and Y2

There are two redox active tryosine residues found in PSII, YD and Y2,
which are located in position 161 of the D2 and D1 polypeptides, respectively.
The oxidized form of Y0, gives rise to a dark-stable electron paramagnetic
resonance (EPR) signal known as Signal IIS, (5 = slow) due to its slow relaxation

kinetics that are on the order of minutes (for reviews see (33)). It is a stable

12

neutral radical and is believed to be well shielded from solvent (26, 34). The
tyrosine YD does not have a known role in the oxidation of water and will not be
discussed further here.

A second tyrosine residue, Yz', also gives rise to an EPR signal at room
temperature in intact systems called Signal ”Vf (vf = very fast) with fast decay
kinetics on the order of microseconds (35, 36). In contrast to Y0, Y2 has a very
speciﬁc role in water oxidation by functioning as an intermediate cofactor
between P630 and the manganese cluster. In light of experimental evidence and
the role of tyrosine radicals in the general class of metalloradical enzymes (Frey,
1990 #215; Sigel, 1994 #216), it was suggested that Yz functions by using a
hydrogen-atom abstraction mechanism (37, 38). The details of the mechanism

for water oxidation will be discussed below.
1.2.6 A Model for Water Oxidation

A metalloradical mechanism for water oxidation, proposed by the Babcock
laboratory (39, 40), invokes Yz' as a hydrogen atom abstractor from manganese-
bound water on each S-state transition. In this model, the two terminal
manganese atoms are catalytically active and bind the two substrate water
molecules in the So state. Another important feature of this model is charge
neutrality. No charge is accumulated on the (Mn).; cluster, thereby avoiding the
high energy situation of creating a charged species in regions of low dielectric.
Figure 1-6 illustrates the important details of this mechanism for the 80 state.

Following formation of P586, Yz is oxidized in the nanosecond time domain

13

 

Figure 1-6 Tyrosine Yz oxidation and reduction in the So state. The dashed
arrows illustrate the ﬂow of the electron and proton during Yz oxidation.
The solid arrow shows abstraction of a hydrogen atom by Yz' during its
reduction.

14

(41, 42) and simultaneously deprotonates to a nearby histidine residue (43-47),
identiﬁed by mutagenic and kinetic data as D1-H190 (48-52). Subsequently, Yz°
abstracts a hydrogen atom, on the microsecond to millisecond time scale, from a
water molecule ligated to the (Mn), cluster. This sequence of steps is proposed
to occur for each S-state transition as shown in Figure 1-7.

Functions for the Ca2+ and CI' ions are also proposed in this model (38).
The CI' ion binds to a Ca2+ atom that is carboxylate-bridged to the (Mn). cluster.
After the formation of S1Yz', Cl‘ migrates to the upper catalytic Mn atom, bringing
with it a proton from the lower position. This ensures that an abstractable
hydrogen atom is available near Yz' in the higher S-states. Upon hydrogen atom
abstraction in the 82 -) $3 transition, the oxo/hydroxyl species is formed. On the
S3 -) 8.; transition, Yz' abstracts the ﬁnal hydrogen atom in concert with the
formation of an oxygen-oxygen bond and the reduction by one electron of the
(Mn). cluster. (37). Molecular oxygen is released as product and then Cl'
migrates back to the Ca2+ site and two water molecules bind to the (Mn)4 cluster
to regenerate the 80 state. This model maintains overall cluster electroneutrality
and minimizes nuclear motion, which are both important in terms of the low
driving force available to oxidize water and the high turnover numbers (up to 200

electrons per second) that the OEC attains.
1.2.7 The Reduction Rates of Y2'

Upon each S—state transition, Y; is formed and is subsequently reduced
by the Mn cluster. Central to understanding the S-state mechanism is the

kinetics of Yz' reduction. The rate of Yz° reduction for each S-state transition was

15

H .
0., .
., \OIIOIe.’ O
.(. ., 3. .
oo o o

This ﬁgure was kindly

Figure 1-7 A model for the S-state cycle in PSII.

provided by Gerald T. Babcock.

16

ﬁrst measured by Babcock in 1976 by using time-resolved electron paramagnetic
resonance spectroscopy (EPR) (53) and more recently by Razeghifard (54).
These rates have also been measured by optical absorption spectroscopic
techniques in the UV region (Dekker, 1984 #212; Renger, 1986 #213; Rappaport,
1994 #191). The optical approach provides better signal to noise, but requires
the use of spectral deconvolution techniques to avoid kinetic interferences from
the acceptor side of PSII. Time-resolved EPR, on the other hand, is more
convenient since it does not require spectral deconvolution, but is more time
consuming because it requires extensive signal averaging to attain adequate
signal to noise. In both cases, the decay rates of Y; were shown to occur on the
us to ms time-scale depending upon the oxidation state of the (Mn). cluster.

The rate of Yz' reduction by the (Mn)4 cluster has also been studied in PSII
samples where the OEC has been chemically, biochemically, or genetically
modiﬁed. The results of these studies are reported in Table 1 and show a
retarding effect on some or all of the S-state transitions. For example, the
depletion/repletion of Ca2+ (55' 56’ or the removal of the 33 kDa protein (57) has
been shown to slow the 839(84)-)So transition. In addition, Razeghifard et al.
showed that Y; reduction was slowed to approximately 6 ms on the 839(84)-)So
transition in an engineered strain of cyanobacteria lacking the Mn-stabilizing
protein (58). All of the above conditions led to a slowdown of the 02 releasing

step in the S-state cycle.

17

 

 

- n 4 ‘-- I n .

Table 1-1 Half-times for S-state Advance in Various Enzyme Preparation
Types - Taken from (58-61)

 

Sample Type Yz'S1-)Y282 Yz'Sz-9Y283 Yz'S3-)YzSo
Thylakoids 65-86 as 140-245 us 750 us;1.3 ms
PSII membranes 30-70 us 55-110 us 1.2-1.4 ms
PSII core particles 75-95 us 225-380 us 4.1-4.6 ms
D61N 240 us 520 us n.d.

-MSP n.d. n.d. 6.0 ms

Cl“ n.d. n.d. 3.3 ms

Br" n.d. n.d. 4.9 ms
N03” n.d. n.d. 15.6 ms
N02“ n.d. n.d. 8.2 ms

I‘ n.d. n.d. 11.7 ms

18

 

Ir

bl

Evidence for perturbations affecting the earlier S-state transitions was also found.
For example, the rates of all the S-state transitions are dependent upon the
degree to which the system is perturbed in the preparation of PSII membranes.
Razeghifard et al. has shown that the treatment of thylakoids to prepare PSII
membranes leads to retardation of all S-state transitions, with the largest effect
seen on the last transition (54, 58). The mutation work done by Hundelt et al. on
D1-Asp61 also demonstrated slower Yz' reduction rates (61). When this
aspartate residue was mutated to the uncharged species, asparagine, oxygen
evolution was retained, but the S-state transitions were retarded. The oxygen-
evolving step was slowed by a factor of 9-10. So not only is the 83-)(84)-)So
transition affected by sample perturbations, but the earlier S-state transitions can

be affected as well.
1.2.8 Calcium

In the early 19803 it was shown that Ca2+ is needed for efﬁcient oxygen-
evolution (62). When the 17 and 23 kDa polypeptides were removed from PSII
by using a high salt treatment, a substantial amount of oxygen-evolution activity
was lost. Successful reactivation of oxygen-evolution activity did not occur upon
reconstitution of these polypeptides, but also required long incubation in the
presence of excess Ca2+. These experiments ultimately led to the conclusion
that the 17 and 23 kDa polypeptides were needed in order to retain Ca2+ in its
native binding site, and without these polypeptides, excess concentrations of

Ca2+ were needed in order to support high rates of oxygen-evolution. Without

19

C3” the system is unable to advance past the Sz'Yz' state, and more than 90-
95°/o of oxygen-evolution activity is suppressed.

A study involving the examination of the ability of other metal atoms to
replace Ca2+ and concurrently support efﬁcient oxygen-evolution has shown that
only Sr2+ can effectively perform this duty, reactivating about 40% of the original
activity (62). Substitution of Sr” for Ca2+ results in a modiﬁed multiline EPR
signal that has narrower hyperﬁne line splittings (63, 64). This result led to the
proposal that the Ca2+ binding site is close to the Mn cluster and the substitution
of Sr2+ for Ca2+ caused perturbations near the Mn cluster due to its larger ionic
radius (1.13 A, compared to 1.0 A for Ca”). Additional evidence to support this
proposal has come from EXAFS studies (65) and FTIR results showing a Ca2+
that is carboxylate-bridged to Mn (66). Although it is generally agreed upon that
Ca2+ is required for efﬁcient oxygen evolution, it is still not known exactly how
Ca2+ participates in the water oxidation chemistry. Some possible roles for Ca2+

have been proposed and are discussed in the following section.
1.2.9 Proposed Roles for Ca2+ in Water Oxidation

Although a general consensus for the function of Ca2+ has not been
reached, several recently proposed models assign a possible role for Ca”. In
general, these roles for Ca” fall into two categories. In one case, direct
involvement of Ca2+ in the formation of the oxygen molecule has been suggested
(67-69). For the second, a more general role for Ca2+ involves assisting in the
catalytic efﬁciency of the OEC (37, 38). These proposed roles will be discussed

along with an experiment devised to test these roles in the function of PSII.

20

In the ﬁrst case, the removal of Ca2+ would be expected to deter the
formation of the O-O bond on the 83-)(84)-)So transition. Pecoraro et al. and
Limburg et al. postulated that Ca2+ presents a hydroxide nucleophile close
enough to a putative Mn"=O species to form the critical O-O bond in the 84 state
(67, 70). Siegbahn and Crabtree have also employed Ca2+ in the O-O bond
forming step to provide a solution to the energetic difﬁculty of forming the reactive
oxyl radical (68). The proposed ﬁve-coordinate, Mn-oxo complex formed in S3 is
unreactive and incapable of forming an O-O bond until it chelates with the
calcium complex to become six-coordinate. This allows the reactive Mn'V-O'
complex to form, which subsequently leads to O-O bond formation. In these
models, O-O bond formation cannot occur without Ca2+ in the site, but the cycle
of events prior to this step is not expected to be inﬂuenced signiﬁcantly by the
depletion of this cofactor. This is especially so if some or all S-state transitions
correspond to pure electron tunneling (71), as there appears to be consensus
that little structural alteration accompanies Ca2+ replacement by Srz+ in the OEC
(13, 65, 72). Likewise, there is no indication of a dramatic change in the driving
force for S—state advance when cofactor substitution is carried out. If Sr2+
substitution did alter potentials signiﬁcantly, oxygen evolution would not be
possible since the potential span between P680*/P580 and water is only about 0.2
V.

For the second class of models, proton-coupled electron transfer provides
the mechanistic basis for each S-state advance (37, 38). Here, small structural

changes will have signiﬁcant effects on kinetics owing to the much greater

21

sensitivity of the proton potential surface to distance (7, 73). Thus, although Ca2+
is necessary for only the higher S-state transitions (38), slight perturbations to its
geometry or replacement by 8er may affect all S-state transitions. This model
places Ca2+ in close proximity to the Mn cluster but does not directly invoke it for
O-O bond formation (38). Calcium’s function, here, is to act as a docking site for
Cl‘, so that Cl' diffusion into the site does not become rate limiting when it is
utilized for charge migration on the S1 to $2 transition, (38). The oxygen-oxygen
bond is formed, in this model, by the concerted reaction between Yz', Mn'V=O,
and Mn'V-OH (37). Siegbahn’s concerns about the energetics of Mn'V-O'
formation disappear in the Hoganson et al. model, as an “energetic boost” is
provided by the concerted O-O bond formation event. Calcium is important for
efﬁcient 02 production, but does not play a central role in product formation.

Strontium replacement for Ca2+ provides an excellent probe to address
these issues. For the ﬁrst class of models, we would expect the substitution of
Sr2+ for Ca2+ to hinder the O-O bond forming step, thereby causing a pronounced
effect on the reduction kinetics of Y; on the 83-)(84)-)So transition. While for the
second class of models, a retardation of all S—state transitions is expected
because, in this model, Ca” plays more of a general role in water oxidation.
1.3 Project Goals

The goal of my thesis research was to investigate the mechanistic
predictions of the above models in regards to the role of Ca2+ in efﬁcient 02-
evolution, and to subsequently gain more information on the hydrogen atom

abstraction mechanism.

22

 

U1

Ci

tr.

IE

th

ei

E)

To investigate the role of Ca2+, the reduction rate of Y; was monitored by
using time-resolved EPR spectroscopy for each S-state transition, S1982,
82-)83, and 83—)(84)-9So, to determine if each transition was affected by Sr2+
substitution or if only the oxygen evolving step was retarded. If the role of Ca2+ is
to assist in the formation of the O-O bond that ultimately leads to the release of
molecular oxygen, then its depletion will hinder the formation of this bond. We
expect, then, that only the rate of the 839(84)-)So transition would be
signiﬁcantly altered, particularly if some, or all, of the earlier S-state transitions
involve pure electron transfer. On the other hand, if Ca2+ does not play a direct
role in bond formation but rather is integral to the active site and plays another
catalytic role by, for example, binding Cl', then its replacement by Sr”, which
affects the integrity of the OEC only slightly, is likely to affect all S-state
transitions.

The second part of my research was to use the results obtained for the
reduction of Y; in Sr2+-substituted PSII membranes to investigate the hydrogen
atom abstraction mechanism proposed for water oxidation. If reduction of Y;
takes place by a pure electron transfer process, any slight perturbations to the
OEC caused by the substitution of Sr2+ would have very minute effects, if any, on
the rate of the electron transfer process. But, if instead, both a proton and an
electron are transferred as a hydrogen atom during Yz' reduction, then we would
expect to see much larger rate alterations, due to the greater dependence of this

type of process on donor and acceptor distance and orientation. Thus, the

23

results of the time-resolved EPR experiments obtained for the Ser-substituted

PSII samples can lend some insight into the mechanism of Yz' reduction.

24

 

k\

k!

T)

SI

Chapter 2

2 Electron Paramagnetic Resonance Spectroscopy
2.1 Basic Principles

Electron paramagnetic resonance (EPR) is based upon the interaction of
radiation of microwave frequency with magnetic moments arising from unpaired
electrons. Therefore, the application of EPR spectroscopy is limited to
paramagnetic systems. Information regarding the properties of the paramagnetic
system is gained by measuring the attenuation versus frequency of a beam of
electromagnetic radiation as it passes through the sample. The lines in the
spectrum represent transitions between energy levels of the radical species, and
their frequency values are a measure of the energy separation of the two levels.
The identity of the radical as well as the nature of its surroundings can be gained
from an analysis of the pattern and shape of the lines obtained.

Electromagnetic radiation is made up of coupled electric (E1) and
magnetic (B1) ﬁelds perpendicular to the direction of propagation, oscillating at
some frequency value. Electromagetic radiation can also be represented as a
stream of particles called photons, which have an energy value equal to hv.
Different energy states arise from the interaction of the unpaired electron spin
moment (n1s = i 1/2) with the magnetic ﬁeld component, B1. This phenomenon,
called the Zeeman effect, lifts the degeneracy of the energy levels for the m8 =
+‘/2 and —‘/2, or as commonly called the or and [3 states, respectively. The
splitting of the energy levels can be varied by changing the static magnetic ﬁeld,

B; absorption occurs when the energy (hv) of the microwave radiation incident on

25

 

the
Fig

the

wh

me
m0

Spe

the sample, equals the splitting of the or and [5 energy levels of the molecule (see
Figure 2-1). This energy value is then used to calculate the Zeeman 9 factor of
the radical species giving rise to the observed transition according to the
equaﬁon:

h V = 916193 (1)
where

h = Plank constant

v = resonant frequency

,6. = Bohr magneton

B = static magnetic ﬁeld

The 9.3 factor for a free-electron is equal to 2.0023. Deviation of the
measured 9 factor from this free-electron value arises from orbital angular
momentum effects and can be used to aid in the identiﬁcation of the radical
species giving rise to the signal.

In addition to the interaction of an electron with an external magnetic ﬁeld,
there is another interaction called nuclear hyperﬁne coupling. The electron
experiences a second ﬁeld generated by magnetic nuclei in its vicinity. These
local ﬁelds 3.003. add vectorially to the external ﬁeld to produce the total ﬁeldBerr
at the electron being probed. The neighboring nuclei may possess an intrinsic
spin angular momentum with possible nuclear spin quantum number 1 values of

0, 1/2, 1, 3/2, 2, etc., with corresponding nuclear spin states given by 21+ 1.

26

E = + 1/2 geﬁeB

    

 

 

B = O resonance
ﬁeld Br
B ——+

Figure 2-1 Energy level scheme for a free electron as a function of
applied magnetic ﬁeld B. Absorption occurs when the incident radiant
energy equals the energy-level separation.

27

Th
SP
hr
50

of

The result of electron interaction with neighboring nuclear-dipole moments is a
splitting of the resonance, referred to as hyperﬁne splitting. The nuclear
hyperﬁne interaction splits each of the electron Zeeman energy levels into 21 + 1
sublevels. Hyperﬁne interactions assist in the study of paramagnetic species by
offering more detailed information in addition to g-factor information gained from
interaction of the electron with the external magnetic ﬁeld.
2.2 The EPR Spectrometer

A block diagram of a typical EPR spectrometer is shown in Figure 2-2.
The source of microwave radiation is usually a klystron and the frequency
typically used in most EPR spectrometers is approximately 9.5 GHz. The
klystron is a vacuum tube that can produce microwave oscillations centered on
small frequency ranges. The frequency of the radiation is determined by
mechanical tuning of the klystron cavity and by the voltage applied to the
klystron. Microwaves are transmitted from the klystron to the cavity through
waveguides of a particular size. The most common is the “X band” which uses
12.7 x 25.4 mm OD rectangular brass pipe for the frequency range of 8.2-10.9
GHz. The frequency of the source is tuned to the appropriate resonant frequency
of the cavity, which is determined by the dimensions of the cavity. In order to
obtain the highest EPR signal sensitivity, the cavity should hold the sample at a
maximum of B1 (the magnetic component of the microwave radiation) and at a
minimum of E1 (the electric component of the of the microwave radiation), and
have B1 perpendicular to the static magnetic ﬁeld B. The source of the static

magnetic ﬁeld B is an electromagnet and the magnitude of B is measured and

28

Field Modulator

  
   

Klystron

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

lsolator Directiona ‘
Attenuator . coupler Resonant
.:':::::3:‘: - «'g;‘77:?f.§2§=?5£4=“ *i ""3155; t O f, Ampliﬁer
AO’ J cavity
‘v
5:3: I! T
[Will Computer
.1153 "’ Scope
. . as; Electromagnet Recorder
Monitoring Detector ,3,-

 

microwave frequency and power ,_

Figure 2-2 Block diagram for a continuous wave EPR spectrometer.
Adapted from Electron Paramagnetic Resonance, J.A. Weil, J.R. Bolton,
and J.E. Werlz, 1994, John Wiley 8. Sons. Reprinted by permission of John
Wiley 8. Sons, Inc.

29

controlled by a Hall-effect detector. Stability in the static magnetic ﬁeld B is
important in order to get accurate measures of the energy level separation AE.

The microwave radiation that is produced at the klystron travels through
the isolator, which prevents any microwave energy from reﬂecting back to the
klystron, to the attenuator where the microwave power incident on the sample is
adjusted. The circulator directs the microwave power to the cavity and
simultaneously directs the power reﬂected from the cavity to the detector. Most
often, the frequency of the microwave power is held constant while the magnetic
ﬁeld B is varied until the resonant condition is met. At resonance, energy is
absorbed by the sample and a change in the power level reﬂected from the cavity
is produced at the detector. Electron paramagnetic resonance spectroscopy
employs a technique to increase the signal-to-noise ratio called phase sensitive
detection where the magnetic ﬁeld B is modulated at 100 kHz. Iy modulating B
the noise-contributing components of frequencies close to or below the
modulation frequency are limited. The amplitude of the detected signa! will be
proportional to the slope of the absorption peak. The output signal appears as
the ﬁrst derivative of the absorption signal where the inﬂection points of the
absorption curve have maximum amplitudes in the ﬁrst derivative signal and at
the resonant ﬁeld the amplitude is zero.

For time-resolved EPR experiments, the magnetic ﬁeld B is set to a
resonant ﬁeld value of the species under study, in this case Yz'. For these
experiments, the ﬁeld value chosen is the low-ﬁeld peak, as indicated in Figure 2-

3 by the arrow, and the amplitude of the signal is monitored over time. This ﬁeld

30

 

 

 

Figu

ﬁeld setting tortime-resolved experim ents

I

 

I ' I ' I ' I ' I ' I

3300 3320 3340 3360 3380 3400
Field (Gauss)

Figure 2-3 Continuous-wave EPR spectrum of Yo'.

31

value minimizes interference from chlorophyll absorption (Babcock, #222). The
principal g-values of the tyrosine radical in PSII have been measured by using
high-ﬁeld EPR spectroscopy and have been reported as gx=2.0074, gy=2.0045,
and gz=2.0021. The maximum difference between 9)( and 92 corresponds to
about nine Gauss at X-Band. The overall bandwidth of Y; is much greater than
this as the result of hyperﬁne coupling (Babcock, 1997 #223). Thus, we are
conﬁdent that setting the ﬁeld at the low-ﬁeld peak provides a homogenous
sampling of all g-values. Using this technique, the formation and decay of Y;

can be monitored as the S-states are cycled and its lifetime can be determined.

32

Chapter 3

3 Materials and Methods
3.1 Preparation of PSII Membranes

Oxygen-evolving PSII membranes can be isolated by detergent
solubilization of the chloroplast thylakoid membranes (74). These particles,
called BBYs, consist of the PSII core with the Mn cluster, Ca2+, Cl', the extrinsic
polypeptides, and the light harvesting complex, and are nearly devoid of PSI.
Further reﬁnement to make core particles with oxygen evolving capacity is
possible, where most of the light harvesting complex is stripped away from the
OEC (75). Both of these preparations have made possible signiﬁcant advances
in the study of the OEC due to the high rates of oxygen evolution activity that can
be achieved.

Photosystem ll membranes were prepared from fresh market spinach
according to the method of Berthold et al. (74) with the modiﬁcations described in
(76). The ﬁnal product is capable of high rates of oxygen evolution and contains
very little contamination from PSI. The procedure used in these experiments is
as follows:

0 About 750 g of market spinach was deveined and washed. The sample was

kept at 4° C throughout the isolation procedure.

0 The spinach was homogenized in a blender and ﬁltered through a 6-layer
cheesecloth. For every 300 g of spinach, 250 mL of grinding buffer

containing 0.4 M NaCl, 2.0 mM MgCl2'6H20, 1.0 mM EDTA, and 2.0 g/L BSA,

33

20.0 mM N-2-hydroxyethylpiperazine-N-Z-ethanesulfonic acid (HEPES), pH
7.5 was used.

The ﬁltrate was centrifuged in a GSA rotor until 1,000 RPM was reached in
order to remove large fragments. The supernate was centrifuged for 12 min.
at 10,000 RPM.

The pellet was resuspended by using a paint brush into the ﬁrst resuspension
buffer containing 0.15 M NaCl, 0.40 mM MgCl2'6H20, and 20.0 mM 2-(N-
morpholino)ethanesulfonic acid (MES), pH 6.0 homogenized in a glass-teﬂon
homogenizer, and centrifuged for 12 min. at 10,000 RPM.

The pellet was resuspended by using a paint brush into the second
resuspension buffer containing 15.0 mM NaCl, 0.50 mM MgCl2'6H20, 1.0 mM
ascorbic acid and 50.0 mM MES, pH 6.0 and homogenized.

The chlorophyll concentration was determined by measuring the absorbance
values at 645 nm and 663 nm in 80% acetone and then adjusted to 2 mg/mL.
A 25% Triton X-100 solution was added dropwise to the stirring sample in the
dark to reach a detergent to Chl ratio of 30:1.

After a 30-minute incubation period in complete darkness, the sample was
centrifuged for 30 minutes at 20,000 RPM.

The pellet was resuspended by using a paint brush into a buffer called SMN
containing 15.0 mM NaCl, 0.40 M sucrose, and 50.0 mM MES, pH 6.0
homogenized, and centrifuged for 30 min. at 20,000 RPM.

The pellet was resuspended by using a paint brush into SMN buffer to a ﬁnal

chl concentration of about 4 mg/mL.

34

The ﬁnal PSII membrane sample was saved in 1 mL aliquots and stored at -

80° C.

3.2 Calcium Depletion of PSII Membranes

Salt-washed membranes were prepared as in Boussac and Rutherford (63) as

follows:

PSII membranes were diluted to a chl concentration of 1 mg/mL and
incubated in room light for 30 minutes at 4°C in a SMN buffer containing 0.30
M sucrose, 2.4 M NaCI, and 25 mM MES, pH 6.5 at a 1:1 ratio.

Following incubation 50 pM EGTA was added. The sample was centrifuged
for 20 min. at 20,000 RPM.

The pellet was resuspended in a SMN buffer containing 0.30 M ultra-pure
sucrose (Aristar, BDH Laboratory Supplies, United Kingdom), 30 mM NaCl,
50 pM EGTA, and 25 mM MES, pH 6.5.

After homogenization, 50 uM EGTA was added and the sample was
centrifuged for 20 min. at 20,000 RPM.

The pellet was resuspended in a SMN buffer containing 0.30 M ultra-pure

sucrose, 15 mM NaCl, and 25 mM MES, pH 6.5, to a ﬁnal chl concentration of

2 mg/mL.

3.3 Sample Preparation for EPR Experiments

Reconstitution of the Cay-depleted samples was carried out by adding

CaClz or SrClz to a ﬁnal concentration of 15 mM. Prior to ﬂash illumination 100

pM PPBQ in DMSO and 1 mM ferricyanide were added to the sample. For the

pH studies, the salt-washed PSII membranes were resuspended in the same

35

buffer, but at a pH of 5.9, 6.5 or 7.1. For the istope exchange studies, salt-
washed PSII membranes were exchanged into a MN buffer containing 15 mM
NaCI, 25 mM MES (pD 6.5) in 99.9% D2O with a ﬁnal addition of 15 mM SrCl2.
The preparation of the sample was identical to that above, except that the ﬁnal
washing steps were performed in a D2O-MN buffer. The rate of isotope
exchange has been shown to occur in a matter of minutes or less, well within the
time it takes to carry out the ﬁnal washing steps (77-80).
3.4 Time-Resolved EPR Experiment Setup

EPR spectroscopy was performed at room temperature with a Bruker ESP
300E spectrometer equipped with a TM011 cavity. The spectrometer was tuned
to the low-ﬁeld peak of the stable YD radical (D2-Y160) and the kinetic traces
obtained for Y2 oxidation and its subsequent reduction were recorded at this ﬁeld
value. The experiments under steady state ﬂashing light conditions were carried
out by using a 350 pL capacity suprasil quartz ﬂat cell (Wilmad, Buena, NJ).
Each sample received 100 ﬂashes before being replaced by a fresh sample. For
the S-state resolved studies, a home-built, continuous-ﬂow system that is
depicted in Figure 3-1 was used to provide fresh samples to the ﬂat cell mounted
in the cavity. Each sample received a train of three ﬂashes before being
replaced. A pump (Gilson, lnc., Middleton, W!) was used to control the ﬂow rate
so that a fresh, dark-adapted sample was supplied for each ﬂash sequence.
Following ﬂash illumination, the sample was recycled to a reservoir that was held
at 4 °C and kept in absolute darkness for 15 minutes. Each experiment was

completed in approximately ﬁve hours with about 2000 scans averaged.

36

 

 

HI—— ﬂat cell

 

 

 

 

 

~1 r ‘— EPR cavity

 

pump

 

 

 

sample reservoir

Figure 3-1 Setup for the time-resolved EPR experiments. Flowing the
sample allows the formation and decay of Y2 to be monitored for each 8-
state transition. Sample averaging is possible by recycling the enzyme.

37

Photoexcitation of the sample at 532 nm was accomplished by using a
Quanta-Ray DCR-11 pulsed Nd-YAG laser (Spectra-Physics, Mountain View,
CA). A personal computer equipped with a Metrabyte WAAG board and a
Metrabyte CTM-05 timing board was used to trigger both the EPR spectrometer
and the laser (81). The EPR conditions for each experiment are given in the
ﬁgure legends. A least-squares ﬁtting routine (Origin 5.0, Microcal Software, Inc.,
Northampton, MA) was used to analyze the kinetic traces.

Oxygen evolution activity was measured with a Clark type electrode under
continuous illumination with saturating white light from two projector lamps of 250
W each that are ﬁltered through a CuSO4 solution. Oxygen evolution rates for
intact BBYs measured at a concentration of 1 mg/mL chl in SMN buffer, pH 6.0
were typically between 650-750 umol O2/mg chl'h while salt-washed, Ca2+ and
Sr” reconstituted samples measured at a concentration of 10 pg/mL chl in SMN

buffer, pH 6.5 were typically 450-550 and 200-300 pmol O2/mg Chl'h,

respectively. Ferricyanide and DCBQ, at concentrations of 600 pM and 300 pM,

respectively, were used as the electron acceptors for these measurements.

38

Chapter 4

4 Results of Time-Resolved EPR Experiments
4.1 S-State Mixed Kinetics

The formation and decay of Y2° was monitored under steady state ﬂashing
light conditions by setting the magnetic ﬁeld to the low-ﬁeld peak of the stable YD
radical (see Figure 2-3) and inducing the oxidation and reduction of Y2 by using
532 nm laser ﬂashes. Figure 4-1 shows the kinetic EPR traces of the decay of
Y; in salt-washed PSII membranes reconstituted with either Ca2+ (trace a) or Sr2+
(trace b). The number of spins was determined by comparing the amplitudes of
each kinetic trace to one obtained under the same experimental conditions for a
tris-washed sample, which generates one Yz' spin per reaction center (82). Using
this procedure, we estimate that 0.5 spin per reaction center for the Ca2+-
reconstituted sample was produced on each ﬂash, which is in agreement with
values reported by this laboratory (81) and reports on intact PSII membranes by
other labs (36, 58). The Sr2+-reconstituted sample, however, showed a 20%
increase in spins due to the slowdown of Y2' reduction in the Sig->81 transition,
which is sufﬁciently slowed in the Sr2+ sample that its detection is now possible
(see below). The lower apparent Yz' spin count in O2-evolving samples relative to
tris-washed material may be due to broadening of its EPR signal by the (Mn).
cluster. In Cay-depleted samples, Y; is in close association with Mn, which

broadens its EPR signal at lower temperatures (25, 27, 83). Similar effects could

39

Mixed S-states
a) CaCI2

 

b) SrClQ

Signal Amplitude

 

|5msl

 

 

Time

Figure 4-1 Time-resolved EPR steady state kinetic traces (S-states are
mixed) for Y,’ reduction in salt-washed PSII membranes reconstituted with
(a) CaCl2 or (b) SrCl2. Experimental conditions: Time constant 40 us,
modulation amplitude 4 G, gain 1 x 105, microwave power 3 dB, frequency
9.78 GHz, magnetic ﬁeld locked at low ﬁeld peak (3470 6), room
temperature, ﬂash frequency 0.66 Hz. 2000 events were averaged.

40

operate in O2-evolving samples and cause an apparent decrease in spin
concentration.

Each kinetic trace in Figure 4-1 shows a biphasic decay. For the Ca”-
reconstituted sample, the fast phase, which is approximately 50% of the total
amplitude, is attributed to the reactions S1Yz' —)S2Yz and S2Y2' -)S3Yz, while the
slow phase, which is approximately 35% of the total amplitude, is produced from
the Sng' -)(S4)-)SOY2 reaction. The remaining 15% of the total amplitude is
attributed to the decay of Y; in damaged centers that behave like tris-washed
PSII membranes. A laser ﬂash artifact is detected on each ﬂash and does not
completely cancel with subtraction of an off-resonance trace obtained at 3400 G.
This should not greatly interfere with the rate determination of the decaying
phases of the kinetic traces. The Ca2+-reconstituted sample (trace a) shows half-
times of ~300 us and ~4 ms for the fast and slow phase, respectively. The Sr”-
reconstituted sample (trace b) has a greater amplitude due to contributions to the
fast phase by So-)S1. The half-times are ~900 us and ~15ms for the fast and
slow phase, respectively. Comparisons of the two kinetic traces shown in Figure
4-1 show that Sr2+-reconstitution apparently slows the rate of Y2' reduction on all
S-state transitions and also shows an increase in the signal amplitude due to the
additional detection of So-)S1.

4.2 S-State Resolved Kinetics

In order to study the effects of SET-reconstitution on the rates of Y2“

reduction in more detail, we resolved the S—state transitions individually as

demonstrated in Figure 4-2. To perform this experiment, a sample that was

41

a)CaCb

 

WM

 

 

 

 

 

 

o
'o
3
a
E
< b) SrCl2
E
a SnYz. H Sntle
.0.)
\Yz —’ p680.
U s,—>s2 s,—>s3 s,—>s,,
T T T Time
ﬂash 1 ﬂash 2 ﬂash 3 I———‘I

50 ms

Figure 4-2 Three-ﬂash S-state resolved EPR kinetic traces. Flash one
induces the formation of Yz'. Decay of the signal represents the reduction

of Y,’ by the (Mn). cluster.

42

poised in the 81 state by dark adaptation was ﬂowed into the ﬂat cell and given
three saturating ﬂashes. The ﬂow rate was adjusted so that each sample
received three ﬂashes spaced 100 ms apart before it was replaced by the next
dark-adapted aliquot. The ﬁrst ﬂash initiated the transition S1Y2' -)S2Y2, while
the second ﬂash generated primarily S2Yz' 98ng. The ﬁnal ﬂash caused Yz'
reduction on the Sng' ->(S4)->SOY2 transition, with a small contribution from the
other states that is caused by double hits and misses. Figure 4-3 shows kinetic
traces and best ﬁts for the ﬁrst ﬂash given to salt-washed PSII membranes
reconstituted with either Ca2+ (trace a) or Srz+ (trace b). Both traces show
biphasic behavior with the fast phase being attributed to the decay of Yz° upon
the S1Y2' --)S2Yz transition and the slow phase to reduction of Y2' in damaged
centers. The reduction of Y; on the ﬁrst ﬂash for the Ca2+-reconstituted sample
(trace a) has a half-time of ~200 ps. The Sr2+-reconsitituted sample displayed
slower reduction kinetics on the SrYz’ -)S2Yz transition (trace b) with a half-time
of ~900 us. The spin counts for each of the above kinetic traces were 0.40 and
0.90 spins for the Cay-reconstituted and the SrZI-reconstituted samples,
respectively. The greater number of spins from the Sr2*-reconstituted sample
was in part due to the additional detection of the So-)Sr transition, to Yz sites in
centers deﬁcient in 02 evolution, and to YD' generation on the ﬁrst ﬂash. The
remaining increase is most likely due to S-state dependent effects caused by
magnetic interaction between the tyrosyl radical and the manganese cluster that

produce small differences in the Y; lineshape. These later effects will not

43

 

 

 

 

 

 

1st Flash
S1 8'2
(D is.
.0 .
3
t
E.
E
<
6
c
.9
(I)
5ms
l J
l I

 

 

Time

Figure 4-3 Time-resolved EPR kinetic traces for the S1Y2' 9 S2Y2 transition
triggered by the ﬁrst ﬂash in a three ﬂash experiment. Salt-washed PSII
membranes reconstituted with (a) CaCl2 or (b) SrCl2 were used.
Experimental conditions: Time constant 20 us in (a) and 40 us in (b),
modulation amplitude 4 G, gain 1 x 105, microwave power 3 dB, frequency
9.78 GHz, magnetic ﬁeld locked at low ﬁeld peak (3470 G), room
temperature. 4000 events were averaged.

44

substantially alter the fast kinetics that are the primary focus of this study and will
not be discussed further.

The kinetic traces for the second transition are shown in Figure 4-4. The
S-state advance, S2Yz' 98%, showed a slow down in the SrZI-reconstituted
sample that was similar to that of the ﬁrst ﬂash. The half-times calculated for the
Cay-reconstituted and Sr2+-reconstituted samples were ~450 us and ~1.3 ms,
respectively. The spin count, 0.50, is nearly the same for both sample types.
Figure 4-5 shows the kinetic traces for Ca2+- and Sr2+-reconstituted samples on
the S3Y2'9(S4)9SOY2 transition. Due to mixing of the S—state transitions by the
third ﬂash, the decays also had a small contribution from the earlier and faster 8-
state transitions. In addition, the amplitude of these traces increased, compared
to the second ﬂash, to a spin count of about 0.55. The half-times were ~5.0 ms
and ~18 ms for Ca2*- and Sr2*-reconstituted samples, respectively. The results
of the steady-state and S-state resolved experiments are presented in Table 4-1.
4.3 Effects of pH and Isotope Exchange

The effects of pH on the reduction rates of Y; in the salt-washed, Sr”-
reconstituted samples were investigated to determine if a pH optimum for Y2°
reduction becomes apparent in any of the S-state transitions upon metal
substitution. The effects of pH were studied by carrying out the S-state resolved,
three-ﬂash experiment at pH values of 5.9, 6.5, and 7.1 in Ser-reconstituted
samples. The results of this experiment, shown in Figure 4-6, indicate that there
is no substantial effect of pH on the kinetics of Y2' reduction by the Mn cluster in

the 5.9 to 7.1 range. A similar lack of pH effects in the physiological range have

45

 

 

 

 

2nd Flash
82-» S3
0)
'o
:3
:0:
E.
E
<1:
5
c
F.”
a) b) SrCI2
5 ms
l l
I I

 

 

 

Time

Figure 4-4 Time-resolved EPR kinetic traces for the S2Y2' 9 S2Yz transition
triggered by the second ﬂash in a three ﬂash experiment. Salt-washed PSII
membranes reconstituted with (a) CaCl2 or (b) SrCl2 were used.
Experimental conditions: Time constant 20 us in (a) and 40 us in (b),
modulation amplitude 4 G, gain 1 x 105, microwave power 3 dB, frequency
9.78 GHz, magnetic ﬁeld locked at low ﬁeld peak (3470 G), room
temperature. 4000 events were averaged.

46

 

 

 

 

3rd Flash

83 80
a)
'o - ﬂ
3 I
E I
o- .
E
<(
E
C
.9
(D b) SrCI2

5ms
I 1'

 

 

 

Time ‘

Figure 4-5 Time-resolved EPR kinetic traces for the Sng' 9 $er2 transition
triggered by the third ﬂash in a three ﬂash experiment. Salt-washed PSII
membranes reconstituted with (a) CaCl2 or (b) SrCl2 were used.
Experimental conditions: Time constant 20 us in (a) and 40 us in (b),
modulation amplitude 4 G, gain 1 x 105, microwave power 3 dB, frequency
9.78 GHz, magnetic ﬁeld locked at low ﬁeld peak (3470 6), room
temperature. 4000 events were averaged.

47

Table 4-1 Rates of Y2' Reduction for Ca23 or Sr”-Reconstituted PSII
Membranes Determined by Time-Resolved EPR Spectroscopy.

 

S-state spin .11/2 (slow) 31/2 (fast)
Steady-state
Ca2+ 0.50 4 ms (35%) 300 ps (50%)
Sr2+ 0.60 15 ms (35%) 900 ps (50%)
s,-> 32
Ga” 0.40 40 ms (15%) 200 ps (85%)
Sr2+ 0.90 40 ms (15%) 900 ps (85%)
52 -> 53
Ca2+ 0.50 40 ms (15%) 450 ps (85%)
Sr2+ 0.50 40 ms (15%) 1.3 ms (85%)
$3 —> 80
Ga” 0.55 40 ms (15%) 5 ms (85%)
s? 0.55 40 ms (15%) 18 ms (85%)

* The rates here are reported as half-times to be consistent with the Photosystem ll literature.

48

 

 

 

 

 

 

 

0 pH 7.1
‘O
3
E.
E I
, | I
_ I .r
g I! II ‘ I ‘ I) 1' ‘ .
i I I. I.
g), I ‘I t‘ .‘
Time
a) pH 6.5
“O
B
'5.
E
< ‘I
E I 'i“ I ‘i I!) ii
Time
pH 5.9

 

Signal Amplitude

 

I 1 ‘ I—-'"-’l

Time

 

Figure 4-6 Time-resolved EPR kinetic traces for the S1Yz'9 S2Y2, S2Y2' 9
5ng, Sng' 9 $erz transitions in salt-washed, Sr2*-reconstituted PSII
membranes resuspended in buffers at pH = 5.9, 6.5, 7.1. Experimental
conditions: Time constant 40 us, modulation amplitude 4 G, gain 1 x 105,
microwave power 3 dB, frequency 9.78 GHz, magnetic ﬁeld locked at low
ﬁeld peak (3470 G), room temperature. 2000 events were averaged.

49

been reported earlier for the unperturbed, O2-evolving system (84).

In addition, a hydrogen/deuterium isotope effect study was performed.
Deuterium kinetic isotope effects on Y2“ reduction have been carried out in
oxygen-evolving PSII membranes by using optical (60) and EPR (85) methods.
These experiments were performed in light of the proposal that Y; reduction is
accompanied by proton transfer from substrate water. This proposal suggests
the possibility of signiﬁcant kinetic H/D isotope effects, but the results of both the
optical and EPR studies showed moderate deuterium kinetic isotope effects on
the reduction of Yz'. The kH/ko values obtained range from 1.3 to 2.9, depending
on the S-state transition, which suggests that the reduction of Y; is not severely
limited by proton transfer in the intact system.

Deuterium kinetic isotope effect studies on the reduction rate of Y2' in salt-
washed, Sr2+-reconstituted PSII membrane fragments were conducted in this
work in efforts to test suggestions that perturbations to the OEC, caused by Sr”-
substitution, hinders proton transfer during Yz' reduction. The rate of proton
transfer is dependent upon donor and acceptor distance to a much greater
degree than electron transfer, and therefore is expected to rate limit Y2' and lead
to large deuterium kinetic isotope effects (86). In order to test this theory, salt-
washed PSII membrane fragments were exchanged into a MN buffer containing
15 mM NaCI, 25 mM MES (pD 6.5) in 99.9% D2O with a ﬁnal addition of 15 mM
SrCl2. The preparation of the sample was identical to that of the experiments
performed above on samples in H2O, except that the ﬁnal washing steps were

performed in a D2O-MN buffer. The rate of isotope exchange has been shown to

50

occur in a matter of minutes or less, well within the time it takes to carry out the
ﬁnal washing steps (77-80). Analysis of the ﬂash-induced kinetic traces that are
shown in Figure 4-7 for a steady state experiment shows a kinetic isotope effect

similar to that reported for the intact system (60, 85, 87) with a maximal value of

about 1.4.

51

Signal Amplitude

 

 

 

 

 

.3 . ,
u‘ . ' ‘ K
v. 7‘ T. .
A ‘.‘~. <3. If s. ‘-
If .. t a -
'I'? '5' ..\.:l‘ :_ :' .‘
.' J ‘;.’ ..‘#{:."I ' ‘I" ‘
' ' ('1' :-T',.."u.‘: I. n , 2 [:0
' ' .r- T " 'Sj . Ii: .13;
.' ‘Iv, ' - _| '1‘.‘
'. _“." ate 1.,
'.-\ ‘- -
C

 

Time

Figure 4-7 Time-resolved EPR steady state kinetic traces (S-states are
mixed) for Y,’ reduction in salt-washed PSII membranes reconstituted with
SrCl2 in either D20 or H2O. The experimental conditions were the same as
in Figure 4-2.

52

Chapter 5

5 Hydrogen Atom Abstraction
5.1 Electron Transfer

The exact function of Ca2+ in efﬁcient water oxidation by PSII is currently
unknown and was the focus of the time-resolved EPR experiments presented in
Chapter 4. The results of these experiments suggest that Ca2+ takes on a more
generalized role that remains constant throughout the S-state cycle since the
kinetics of Y2' reduction on each S-state transition is retarded by a comparable
extent when Ca2+ is replaced by Sr”. Thus, the direct involvement of Ca2+ in the
oxygen-oxygen bond forming step in water oxidation is unlikely. In addition,
similar results have been presented where one or more of the S-state transitions
are slowed upon point mutation or biochemical treatment of PSII (Table 1-1).
Therefore, in order to show by what mechanism the reduction of Y2' is altered
upon Sr2+-substitution in PSII membranes, we must also take into account the
results shown in Table 1-1. To do this, we look at how electron transfer reactions
occur, and speciﬁcally, how the rates of electron transfer between a donor and
acceptor can be affected.

Electron transfer reactions and their rate characteristics have been
extensively studied (for a review see (88)). Two parts, electronic and nuclear,
determine the rate of electron tunneling. The electronic part arises from the
dependence on the coupling strength between the electron donor and acceptor
wavefunctions; as the distance, R, between donor and acceptor increases, the

rate decreases exponentially owing to the uncoupling of the wavefunctions. The

53

nuclear part depends on the driving force, AG, and the reorganization energy, 1..
The driving force term is the ground state free energy difference between the
donor and acceptor while the reorganization energy is the amount of energy
needed to reorganize the nuclear coordinates for electron tunneling. These
parameters are deﬁned pictorially in Figure 5-1. A kinetic expression for the
estimation of rates of intraprotein, non-adiabatic electron tunneling has been

formulated by Page et al. (71) as follows:

log k.t = 13 — ((1.2 — 0.8p)(R - 3.8)) - 3.1 ((AG + 102/1]

where ket represents the exergonic electron tunneling rate (s'1), p is the speciﬁc
packing density of the inter-cofactor protein volume, R is the edge-to-edge
distance between the donor and acceptor (A), AG is the driving force (eV), and
I. is the reorganizational energy (eV). We can use this expression to predict the
electron transfer rate for the reduction of Y2' by the Mn cluster to gain a better
understanding of the mechanism of this process.

In order to use this kinetic expression for the electron transfer between the
Mn cluster and Yz', we use values to substitute into the equation as follows. The
dipolar distance between the Mn cluster and Y2' has been reported as
approximately 8 A from magnetic-resonance studies (27-29, 83, 89). The driving
force for this reaction is approximately 60 meV (90) and the reorganizational
energy is taken to be about 0.8 eV, which is a reasonable estimate for a protein

environment (91-93). The packing density is determined by the van der Waals

54

 

 

 

 

 

Reaction Path

Figure 5-1 Potential energy surfaces of the donor and acceptor. AG is the
free energy difference between donor and acceptor, 7. is the reorganization
energy, and R is the distance between donor and acceptor.

55

radii of the atoms between the donor and acceptor molecules and has an
average value of about 0.75 in the proteins analyzed by Page, et al (71). When
these values are substituted into the equation, rates on the ns time scale are
predicted, which are orders of magnitude faster than what has been
experimentally observed for Y; reduction in O2-evolving systems. Therefore, we
conclude that the rate limiting step in the overall 8,, 9 Sn.) process is not the
tunneling of the electron (71 ).

As Y; is reduced it becomes protonated (94, 95). Thus, a more precise
approach to studying this process is within the context of a proton coupled
electron transfer (PCET) reaction. We expect that this approach will give
additional insight into the rates that we and others (54, 58, 60, 87) have
measured.

5.2 Proton Coupled Electron Transfer: Sequential or Concerted?

For PCET reactions, three possible mechanistic pathways can be
considered, as shown in Figure 5-2. In order to test the viability of each pathway,
the energetics of each must be taken into account. The values that were used to
determine the energetics of paths a and b are listed in Table 5-1. The ﬁrst path
involves the transfer of the electron, followed sequentially by the transfer of the
proton, ET/PT (path a). For the process under investigation here, the transfer of
an electron results in the reduction of Yz° to form the tyrosinate (state II). The
reduction potential of Y'IY‘ is about 0.68 V aqueous solution (a = 80) (96) and the
potential required to oxidize water is about 0.93 V (8); therefore the transfer of an

electron from substrate water/Mn cluster to Yz' is an uphill process with a barrier

56

H+

w

 

   

   

 

 

(ll)YZO MnTHZO (F0) . (IV) YZOH Mn+OH
“- (a) (Y. S...)

A {I
e e

e.t. e.t.
W (a) (b) W
(F0) (F0)

(FC) (FC)

 

(I) Ygo Mn H20 “5"” - (Ill)Yz'OH+ MnOH-
H
(Y: 8.)
(F0)

Figure 5-2 Possible reaction paths for proton and electron transfer. The
four species that are shown are identiﬁed as follows: (I) the reactant state,
(IV) the product state, and the two charged intermediate states, (II) and (III).
In path (a), electron transfer is followed by proton transfer, while in path
(b), proton transfer is followed by electron transfer. In path (c), the electron
and proton transfer in a concerted process.

57

Table 5-1 Reduction Potentials and pKa Values Used in Determining the
Viability of the ETIPT and PTIET Pathways of Figure 5-1.

 

 

 

 

 

 

 

Redox couple Reduction potential Reference
Fwy/P680 E... = 1.12 a: 0.05 v (6) ”
Yz'le Em = 0.97 :t 0.02 V (90)
81/80 Em s 0.70 V (90)
82/81, 83/32 Em = 0.90-0.95 V (90)
Y'IY‘ Em = 0.68 V (aqueous) (7)
02/H2O Em = 0.93 V (aqueous) (8)
Molecule pKa value Reference
“YT _.- -2 (aq ueous) W” __..__ WW- ( 7)
Mn(H2O) 2 7 (38)

 

58

of at least 8 kcaI/mol. Furthermore, the difference in reduction potentials
between the two cofactors will increase in the low dielectric of a protein
environment, resulting in an even larger barrier. We can estimate the effect of

the protein dielectric on these potentials by using the Born model:

AG = 14.397/8r

where AG 0!) is the electrostatic penalty for introducing a charged species with
radius r (A) in a medium with a homogenous dielectric constant 8 (97). Assuming
an average radius of 3 A for the tyrosine aromatic ring and a dielectric constant
of about 8-10, which has been recently suggested for PSII (98), the Y°/Y’
potential will decrease from an aqueous value of 0.68 V to a value of 0.26 V.
This leads to an activation barrier of at least 16 kcal/mol as illustrated in Figure 5-
3 (path a), which includes the energy that would be regained from the stabilizing
interaction between the positive and negative charge produced. Activation
energies have been measured for each of the S-state transitions (see Table 5-2)
and are much lower than this estimated barrier. Thus, electron transfer followed
by proton transfer is not supported experimentally and this ﬁrst mechanistic
possibility can be discarded.

The second possibility is the transfer of a proton followed by the
sequential transfer of an electron, PT/ET, Figure 5-2 (path b). Energetic
considerations must also be taken into account to test the viability of forming the

cationic tyrosine radical (state Ill). The pKa’s of the proton donor and acceptor

59

Path 8:

 

 

 
  
 
 
  

Yzo‘ Mn+H,o t 9- YZOH Mn+0H"
p. .
(I
YZO- Mn+H,o
e.t. /
Ea(et)2 16 £133:
028.)
YZO- Mn H20 “25"“)
Path b: +
+ YZOH Mn 0H-
Y;OH MnOH
/ A
Ea(pt)_>_ 12 $3!
e.t.
(Y5 5") (Y. S...)

 

 

YZO- Mn H20

p.t.

- YZOHT MnOH‘

Figure 5-3 Calculated activation energies (Ea) for paths a and b. This ﬁgure

kindly provided by Gerald T. Babcock.

Table 5-2 Kinetic and Thermodynamic Parameters for the S-state
Transitions - Taken from (7, 59, 60)

 

 

Reaction Ea (kcal M") A (s") KIE

)r,°so—>)rzs1 1.2

Yz'S19YZS2 3.0 4.0x10‘5 1.3; 2.9

Yz'szevzsg 9.0 5.4x1o9 1.3

v;s,—>vzso 5.0T 8.9x105i 1 4-1.6
92* 2.9x10‘“

Ea = activation energy
A = Arrhenius preexponential factor
KIE = kinetic isotope effect

Yz'le = reduction potential of Yz'le
Sn+1/Sn = reduction potential of SWISn

IT>280K
*T<280K

61

are now important in determining the energetics of the process. The pKa of a
Mn-l-I2O complex is expected to be 2 7 (99), while the pKa of Yz° measured in
water is —2 (100). Thus, the ApKa for the proton transfer is -9, which as Figure 5-
3 (path b) illustrates, results in a barrier of at least 12 kcallmol, with this value
reaching greater than 23 kcallmol in the lower dielectric protein environment.
Consequently, the transfer of a proton prior to that of the electron leads to the
formation of a barrier at least 3 kcallmol higher than has been observed for the
S-state transitions, which allows us to discard this possibility.

The general characteristics of the S-state transitions are summarized in
Table 5-2 and, in many respects, are remarkable. Renger and coworkers have
carefully characterized the activation energy EE, and pre-exponential factor A for
the $1982, $2983, and 839(84)9So transitions (60). These values are
substantially lower than those typically found for an enzyme catalyzed reaction;
textbook values are usually taken to be about 10-12 kcallmol and 1013 M"s‘1 for
E. and A, respectively (101, 102). The Iow A values, in particular, are striking.
The low rates for Y; reduction and the small kinetic isotope effects that have
been measured are also unusual considering the small distance between donor
and acceptor (27-29) and that proton tunneling has been suggested in Y;
reduction (59). Whether the concerted ETPT mechanism can explain these
characteristics now becomes the issue at hand.
5.3 Concerted Electron and Proton Transfer

Typical kinetic parameters that are observed for H-atom abstraction by

radicals are reported in Tables 5—3 and 5-4. Considering ﬁrst the Ea values, the

62

Table 5-3 : Kinetic and Thermodynamic Parameters for H-atom Abstraction
Reactions From an Oxygen Atom by an Oxy-based Radical

 

 

Reaction Ea log A KIE k Ref
(kcal M") (M‘s‘) (M‘s‘)
4-OMe-Phenol + PhO' n.d. n.d. n.d. 4.75 x 105 (103, 104)
4-H-Phenol + PhO' 4.9a n.d. n.d. 3.5 x 105 (103, 104)
4-t-butyl-Phenol + PhO' 4.8 5.5 n.d. 51.3 (Mahoney,
1969 #210)
3,5-dimethyl-Phenol + PhO' 6.8 6.4 n.d. 16.3 (104)
or-tocopneroi + PhO' 2.0 10.0 1.17 3.1 x 10° (105)
a-naphthol + PhO' 2.2 8.9 n.d. 2.3 x 107 (105)
ubiquinoI-O + PhO' 3.5 10.5 n.d. 9.1 x 107 (105)
o-tocopnerol + (C5H5)2CO' n.d. n.d. 1.09 5.1 x 109 (Evans,
1992 #217)
a-tocopherol + (CH3)3CO' n.d. n.d. 1.31 3.8 x 109 (Evans,
1992 #217)
(CH3)3CO-H + .OC(CH3)3 2.6 6.4 n.d. n.d. (106)

 

3 Calculated by the BEBO method.

63

Table 5-4: Kinetic and Thermodynamic Parameters for Hydrogen Atom
Abstraction by Radicals.

Kinetic and Thermodynamic Parameters for Hydrogen Atom Abstraction by
Radicals.

 

 

Reaction E. (kcal M") log A (M"s") KIE k (M"s") Ref
CH3CH2-H + 'CH3 11.59 11.7 4.04 2.57 x 105 (107)
H3C-H +‘c113 14.51 11.8 3.61 7.41 x 103 (107)
SiH3-H + 'CH3 6.98 11.9 3.10 1.20 x 108 (107)
NH2-H + 'CH3 9.72 10.8 5.20 2.82 x 105 (107)
SH-H + 'CH3 2.50 11.6 3.85 1.44 x 1010 (107)
SiH3-H + 'CH2CH3 7.24 11.7 1.08 5.89 x 107 (107)
PhS-H + 'CH2Ph 1.5-1.7 9.1-9.4 n.d. 1.3 x 108 (108)
(CH3)3Sn-H + 'CH3 3.23 9.4 2.3 1.06 x 107 (109)
(CH3)gSn-H + 'CH2CH3 3.8 9.1 1.9 2.3 x 106 (109)
CchH2-H + 'OOCH3 14.9 11.5 n.d. n.d. (110)
CH3-H + 'OOCH3 18.48 11.3 n.d. n.d. (110)

64

tabulated data span a broad, almost 20 kcal M'1 range. However, for reactions in
which a hydrogen atom is abstracted from an oxygen atom by an oxy-based
radical, which is the situation we postulate in PSII, the situation simpliﬁes and
becomes more homogeneous. The Ea values measured for these types of
reactions are most often in the range of 2—7 kcallmol, and are unusually small
compared to the values shown in Table 5-4 for other H-atom abstraction
reactions, especially those in which C-H or N-H bonds are the hydrogen-atom
donor. This phenomenon has been rationalized by Zavitsas and Chatgilialoglu
as being due to triplet repulsion in the transition state (TS) (106). As the parallel
spins on the donor and acceptor molecules move closer together during
formation of the TS, a repulsive energy term becomes more signiﬁcant to the
overall activation energy barrier for the reaction. This term is much smaller for H-
atom abstraction between RO' and —O-H bonds, thereby allowing the oxygen
atoms to form a tighter transition state, which lowers the activation energy
barrier. On the other hand, as Table 5-4 shows, H-atom abstraction between two
carbon atoms, for example, occurs with much higher activation energies.
According to Zavitsas and Chatgilialoglu, this is due to the stronger triplet
repulsion experienced by carbon compared to oxygen. In the case of PSII,
reduction of Y2’ has been proposed to occur via hydrogen atom abstraction by the
tyrosine radical from Mn-ligated water or hydroxide ligands. The low activation
energies that have been measured in PSII ﬁt well in this context as reﬂecting
what has been measured in other systems in which H-atom transfer between an

oxygen-based radical and an -0-H bond occurs. Other theoretical treatments

65

have reached similar conclusions with respect to low Ea values, when hydrogen
bonding allows relatively close approach of reactant species in a hydrogen-atom
transfer process (81, 124).

Textbook values for the Arrhenius pre-exponential factor, A, for hydrogen-
atom abstractions are generally low; A s 108'5 M'1s'1 is often cited as a typical
number (101, 102). These values are for second order processes, but with
effective PSII concentrations of 1-10 M, the values reported below for ﬁrst-order
PSII processes become essentially directly comparable to the second-order
solution reactions. The fact that the S-state transitions also proceed with low A
factors provides support for our conclusion that the PSII reactions proceed along
the concerted pathway in Figure 5-2. We can gain physical insight into the
meaning of these values for A as follows. The classic interpretation of an A
factor is as an “attempt frequency”. A low A value implies an unfavorable entropy
of activation, which indicates that the system can position itself to enter the
transition state only infrequently. For pure electron tunneling at the short Yz'Sn
distance and for pure proton transfer much higher, typically 1013 M‘s", A values
are expected (71, 111), which indicates a much higher attempt frequency. The
fact that these larger A values are not found in PSII reinforces our conclusion
(112) that the sequential pathways, a and b in Figure 5-2, are not likely in the 8-
state advances.

Figure 5-2 rationalizes the A-values in terms of the Franck-Condon (FC)
factors for proton and electron transfer and follows closely arguments that have

been advanced by Cukier (113). A rate constant expression for concerted

66

PCET can be formulated in analogy to the Marcus-Levich pure ET rate constant

as(114x

2

e-( ).\'+:L\(;0+&'ﬁ, .cin' )2 /4/1.Sk”'l‘

 

 

kerpr :

 

71 2.87 2” 2W"

The parameters in Eq. (1) that come from the coupling of the reactant and

product charge distributions to the surrounding medium are the solvent

reorganization (As) and reaction-free energies (AG0); V6, is the electronic

coupling. These are the standard parameters that determine an ET rate
constants value (115, 116). The new ingredients are Franck-Condon factors of
the initial (ﬁnal) state proton vibronic wavefunctions 1,2,1 (12,). The initial states are
summed over the equilibrium initial state proton distribution, pin». The “effective”
activation energy appearing in the exponent of Eq. (1) involves the energetic
difference of the proton eigenstates, am - an, and arises from the requirement of
overall energy conservation between the initial and ﬁne! electron-proton states.
The tunneling of the proton is manifested in the Franck-Condon (FC) factors in
Eq. (1). These factors tend to be small as they are determined by the overlaps of
proton wavefunctions localized around the initial and ﬁnal proton equilibrium
positions, respectively. Thus, concerted PCET is limited by a “Franck-Condon
drag” that is a reﬂection of the requirement that both the electron and the proton
are tunneling species. As tunneling is a difﬁcult process, and in concerted PCET
there are two tunneling species, the tendency is to obtain relatively small rate
constants. However, concerted PCET still can be the dominant reaction channel

if the sequential processes are rate limited. As discussed above, both ET/PT

67

and PT/ET sequential processes are indeed rate limited and that leaves the
concerted process as the viable channel. Use of reasonable choices for all the
parameters in Eq. (1) does lead to rate-constant values compatible with the

magnitudes displayed in Table 2. An important feature of Eq. (1) is that it does

not, strictly speaking, have the form k = Aexp(—Ea/RT). Nevertheless, over

typical temperature ranges accessible to biological studies, rate constants based
on Eq. (1) will approximately obey this form. Thus, we may consider that
effective activation energies and pre-exponential factors can be deﬁned for
concerted PCET reactions. Evaluation of these factors for appropriate parameter
ranges produce values for E, and A consonant with those in Table 2. In
particular, unfavorable entropy of activation factors (low A values) are to be
expected for this mechanism. These considerations provide additional support
for the notion that the S-state transitions occur by concerted hydrogen-atom
transfer.

The rates and kinetic parameters for overall hydrogen-atom transfers can
now be rationalized. If the energetics are not too unfavorable, a sequential
pathway is more likely, as the Franck-Condon factors do not hinder the reaction

appreciably. This appears to be the case in the bacterial reaction center, where

the proton-coupled Q2 to Q‘B reaction follows a sequential route (117). If,
however, the energetics are too unfavorable, as we have argued is the case for
the S-state transitions in the OEC, then the concerted pathway is favored. This

was also the conclusion drawn by Roth et al. (118) and by Sjodin et al. (119) for

their work on small-molecule model compounds. In these two studies, the

68

authors concluded that the electron transfer/deprotonation process occurred via
a concerted mechanism by using similar arguments as those presented above.
Thus, for the concerted reaction, the activation energies are generally lower than
for the sequential process, but the Franck-Condon drag discussed above lowers
the A factors and slows the overall rate considerably. The net result is that rates
on the order of 103-105 8'1 are observed for both the S-state advances and for
RO' + R'OH 9 ROH + R0" in small organic systems.

In concerted PCET, a substantial isotopic effect can occur, as the
deuteron FC factors are smaller than those for a proton. However, these FC
factors are strongly dependent on the distance between the ﬂanking heavy atoms
that form the hydrogen/deuterium/tritium (H/DfT) bond, with the FC factor
increasing for decreasing heavy-atom separation. This increase is more
dramatic for T vs D vs H, and studies conducted by Krishtalik on model
compounds have shown that the optimum tunneling distance for H, D, and T are
different (120, 121). This result leads to the conclusion that it is more favorable
for heavier isotopes to spend more energy against the repulsive forces of
bringing the ﬂanking atoms together, since tunneling is more difﬁcult than for
lighter isotopes. The net result is a decrease in the heavy-atom separation for
which tunneling is most effective in the series H, D, T, and this reduces the
isotope effect. Furthermore, as the heavy-atom separation decreases, the FC
factors saturate to values that are not strongly dependent on distance (122).
Thus, once the heavy-atom separation is sufﬁciently small, the distinction

between H. D, and T tunnel factors disappears. The rate constants that result

69

from averaging over the heavy-atom separations then lead to kinetic isotope
effects that are much more modest than would othenrvise be predicted. Thus,
small isotope effects for hydrogen-atom abstractions are expected and are
conﬁrmed by theoretical calculations with the result of a proton/deuteron rate
ratio of no more than a factor of 2, well in agreement with experimental values of
1.7 and 2.3 reported by the lngold and Mayer groups, respectively (105, 118).

In conclusion, small disturbances of the cluster caused, for example, by
the substitution of the larger Sr2+ ion, could cause a change in the OEC
geometrically and/or electrostatically. Any small change upon the substitution of
Sr”, or any of the other modiﬁcations mentioned above, can lead to modiﬁcation
of the activation barrier for hydrogen atom transfer. Since the donor and
acceptor species are intimately coupled in these processes, any change that
leads to a higher barrier will slow the rate of hydrogen atom transfer. This
behavior can explain the reduction in rates seen upon depletion of Ca2+ or
substitution of Sr” in the OEC, as well as explain the results of the other
modiﬁcations to the intact system. For PSII, these effects will be seen on all S-
state transitions by causing a slowdown of the reduction of Y2' by the substrate

water/Mn cluster.

70

BIBLIOGRAPHY

71

10.

11.

12.

Bibliography

Debus, R. J. (1992) Biochim Biophys Acta 1102, 269-352.
Barber, J., and Kt‘Jthrandt, W. (1999) Curr. Opin. Struc. Biol. 9, 469-475.

Debus, R. J. (2000) in Metal Ions in Biological Systems (Sigel, A., and
Sigel, H., Eds.) pp 657-711, Marcel Dekker, Inc., New York.

Joliot, P., Barbieri, G., and Chabaud, R. (1969) Photochem. Photobiol. 10,
309-329.

Kok, B., Forbush, B., and McGolin, M. (1970) Photochem. Photobiol. 11,
457-475.

Klimov, V. V., Allakhverdiev, S. I., Demeter, S., and Krasnovskii, A. A.
(1979) Oak]. Acad. Nauk. SSSR 249, 227-230.

Tommos, C., and Babcock, G. T. (2000) Biochim. Biophys. Acta 1458,
199-219.

Krishtalik, L. I. (1986) Biochim. Biophys. Acta 849, 162-171.

Duysens, L. N. M. (1958) in Photochemical Apparatus, Brookhaven
Symposium in Biology pp 18, Brookhaven National Laboratory, Upton, NY.

Ross, R. T., and Calvin, M. (1967) Biophys. J. 7, 595-614.

Diner, B. A., and Babcock, G. T. (1996) in Oxygenic Photosynthesis: The
Light Reactions (Ort, D. R., and Yocum, C. F., Eds.) pp 213-247, Kluwer
Academic Publishers, Dordrecht.

Yachandra, V. K., Sauer, K., and Klein, M. P. (1996) Chem. Rev. 96,
2927-2950.

72

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

Cinco, R. M., Robblee, J. H., Rompel, A., Fernandez, C., Yachandra, V.
K., Sauer, K., and Klein, M. P. (1998) J. Phys. Chem. 102, 8248-8256.

Zimmermann, J. L., Boussac, A., and Rutherford, A. W. (1986)
Biochemistry 25, 4609-4615.

Zimmermann, J.-L., and Rutherford, A. W. (1984) Biochim. Biophys. Acta
767, 160-167.

Sivaraja, M., Tso, J., and Dismukes, G. C. (1989) Biochemistry 28, 9459-
64.

Hallahan, B. J., Nugent, J. H. A., Warden, J. T., and Evans, M. C. W.
(1992) Biochemistry 31, 4562-73.

van Vliet, P., Boussac, A., and Rutherford, A. W. (1994) Biochemistry 33,
12998-13004.

Beck, W. F., de Paula, J. C., and Brudvig, G. W. (1986) J. Am. Chem.
Soc. 108, 4018-4022.

Andreasson, L.-E., Hansson, 0., and von Schenck, K. (1988) Biochim.
Biophys. Acta 936, 351-360.

Boussac, A., Rutherford, A. W., and Styring, S. (1990) Biochemistry 29,
24-32.

Haddy, A., Dunham, W. R., Sands, R. H., and Aasa, R. (1992) Biochim.
Biophys. Acta 1099, 25-34.

Casey, J. L., and Sauer, K. (1984) Biochim. Biophys. Acta 767, 21-28.

Boussac, A., Zimmermann, J.-L., Rutherford, A. W., and Lavergne, J.
(1990) Nature 347, 303-306.

Gilchrist, M. L., Jr., Ball, J. A., Randall, D. W., and Britt, R. D. (1995) Proc.
Natl. Acad. Sci. USA. 92, 9545-9549.

73

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

Tang, X. S., Randall, D. W., Force, D. A., Diner, B. A., and Britt, R. D.
(1996) J. Am. Chem. Soc. 118, 7638-7639.

Dorlet, P., Di Valentin, M., Babcock, G. T., and McCracken, J. L. (1998) J.
Phys. Chem. B 102, 8239-8247.

Peloquin, J. M., Campbell, K. A., and Britt, R. D. (1998) J. Am. Chem.
Soc. 120, 6840-6841.

Lakshmi, K. V., Eaton, S. 8., Eaton, G. R., Frank, H. A., and Brudvig, G.
W. (1998) J. Phys. Chem. B 102, 8327-8335.

Ono, T.-a., Noguchi, T., lnoue, Y., Kusunoki, M., Matsushita, T., and
Oyanagi, H. (1992) Science 258, 1335-1337.

Roelofs, T. A., Liang, W. C., Latimer, M. J., Cinco, R. M., Rompel, A.,
Andrews, J. C., Sauer, K., Yachandra, V. K., and Klein, M. P. (1996) Proc.
Natl. Acad. Sci. USA 93, 3335-3340.

luzzolino, L., Dittmer, J., Dorner, W., Meyer-Klaucke, W., and Dau, H.
(1998) Biochemistry 37, 17112-17119.

Barry, ,8. A., and Babcock, G. T. (1987) Proc. Nat. Acad. Sci, USA 84,
7099-7103.

Diner, B. A., Force, D. A., Randall, D. W., and Britt, R. D. (1998)
Biochemistry 37, 17931-17943.

Blankenship, R. E., Babcock, G. T., Warden, J. T., and Sauer, K. (1975)
FEBS Lett. 51, 287-293.

Warden, J. T., Blankenship, R. E., and Sauer, K. (1976) Biochim. Biophys.
Acta 423, 462-478.

Hoganson, C. W., and Babcock, G. T. (1997) Science 277, 1953-1956.
Tommos, C., and Babcock, G. T. (1998) Accounts Chem. Res. 31, 18-25.

74

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

Babcock, G. T. (1995) in Photosynthesis: from Light to Biosphere (Mathis,
P., Ed.) pp 209-215, Kluwer Academic Publishers, Netherlands.

Hoganson, C. W., Lydakis-Simantiris, N., Tang, X.-S., Tommos, C.,
Warncke, K., Babcock, G. T., Diner, B. A., McCracken, J., and Styring, S.
(1995) Photosynthesis Res. 46, 177-184.

Brettel, K., Schlodder, E., and Witt, H. T. (1984) Biochim. Biophys. Acta
766, 403-415.

Meyer, 8., Schlodder, E., Dekker, J. P., and Witt, H. T. (1989) Biochim.
Biophys. Acta 974, 36-43.

Rappaport, F ., and Lavergne, J. (1997) Biochemistry 36, 15294-15302.

Schilstra, M. J., Rappaport, F., Nugent, J. H. A., Barnett, C. J., and Klug,
D. R. (1998) Biochemistry 37, 3974-3981.

Christen, G., and Renger, G. (1999) Biochemistry 38, 2068-2077.

Christen, G., Seeliger, A., and Renger, G. (1999) Biochemistry 38, 6082-
6092.

Hays, A.-M., Vassiliev, I. R., Golbeck, J. H., and Debus, R. J. (1999)
Biochemistry 38, 11851 -1 1865.

Diner, B. A., Nixon, P. J., and Farchaus, J. W. (1991) Curr. Opin. Struct.
Biol. 1, 546-554.

Roffey, R. A., Kramer, D. M., Govindjee, and Sayre, R. T. (1994) Biochim.
Biophys. Acta 1 185, 257-270.

Chu, H.-A., Nguyen, A. P., and Debus, R. J. (1995) Biochemistry 34,
5839-5858.

Hays, A. M. A., Vassiliev, l. R., Golbeck, J. H., and Debus, R. J. (1998)
Biochemistry 37, 1 1352-11365.

75

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

Mamedov, F., Sayer, R. T., and Styring, S. (1998) Biochemistry 37,
14245-14256.

Babcock, G. T., Blankenship, R. E., and Sauer, K. (1976) FEBS Lett. 61,
286-289.

Razeghifard, M. R., Klughammer, C., and Pace, R. J. (1997) Biochemistry
36, 86-92.

Dekker, J. P., Demetrios, F. G., Plijter, J. J., Van Gorkon, H. J., and
Babcock, G. T. (1984) Biochim. Biophys. Acta 767, 515-523.

Boussac, A., Setif, P., and Rutherford, A. W. (1992) Biochemistry 31,
1224-34.

Razeghifard, M. R., Wydrzynski, T., Pace, R. J., and Burnap, R. L. (1997)
Biochemistry 36, 14474-8.

Razeghifard, M. R., and Pace, R. J. (1997) Biochim. Biophys. Acta 1322,
141-150.

Tommos, C., Hoganson, C. W., Di Valentin, M., Lydakis-Simantiris, N.,
Dorlet, P., Westphal, K., Chu, H.-A., McCracken, J., and Babcock, G. T.
(1998) Curr. Opin. Chem. Biol. 2, 244-252.

Karge, M., Irrgang, K. D., and Renger, G. (1997) Biochemistry 36, 8904-
8913.

Hundelt, M., Hays, A.-M., Debus, R. J., and W., J. (1998) Biochemistry 37,
14450-14456.

Ghanotakis, D. F., Babcock, G. T., and Yocum, C. F. (1984) FEBS Lett.
167, 127-130.

Boussac, A., and Rutherford, A. W. ( 1988) Biochemistry 27, 3476-3483.

Boussac, A., and Rutherford, A. W. (1988) Chem. Scripta 28A, 123-126.

76

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

Latimer, M. J., Derose, V. J., Mukerji, I., Yachandra, V. K., Sauer, K., and
Klein, M. P. (1995) Biochemistry 34, 10898-10909.

Noguchi, T., Ono, T., and lnoue, Y. (1995) Biochimica Et Biophysica Acta-
Bioenergetics 1228, 189-200.

Pecoraro, V. L., Baldwin, M. J., Caudle, M. T., Hsieh, W.-Y., and Law, N.
A. (1998) Pure & Appl. Chem. 70, 925-929.

Siegbahn, P. E. M., and Crabtree, R. H. (1999) J. Am. Chem. Soc. 121,
117-127.

Limburg, J., Vrettos, J. S., Liable-Sands, L. M., Rheingold, A. L., Crabtree,
R. H., and Brudvig, G. W. (1999) Science 283, 1524-1527.

Limburg, J. Szalai, V. A, and Brudvig, G. W. (1999) J. Chem. Soc,
Dalton Trans. 9, 1353- 1361.

Page, C. C., Moser, C. C., Chen, X., and Dutton, P. L. (1999) Nature 402,
47-52.

Riggs-Gelasco, P. J., Mei, R., Ghanotakis, D. F., Yocum, C. F., and
Penner-Hahn, J. E. (1996) J. Am. Chem. Soc. 118, 2400-2410.

Scheiner, S. (1994) Acc. Che. Res. 27, 402-408.

Berthold, D. A., Babcock, GT, and Yocum, CF. (1981) FEBS Lett. 134,
231-234.

Ghanotakis, D. F., Demetriou, D. M., and Yocum, C. F. (1987) Biochim.
Biophys. Acta 891, 15-21.

Ghanotakis, D. F., Topper, J., Babcock, G. T., and Yocum, C. F. (1984)
Biochim. Biophys. Acta 767, 524-531.

Hillier, W., Messinger, J., and Wydrzynski, T. (1998) Biochemistry 37,
16908-16914.

77

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

Tommos, C., McCracken, J., Styring, S., and Babcock, G. T. (1998) J. Am.
Chem. Soc. 120, 10441-10452.

Qian, M., Dao, L. A., Debus, R. J., and Burnap, R. L. (1999) Biochemistry
38, 6070-6081.

Hillier, W., and Wydrzynski, T. (2000) Biochemistry in press.
Hoganson, C. W., and Babcock, G. T. (1988) Biochemistry 27, 5848-5855.

Babcock, G. T. (1987) in New Comprehensive Biochemistry (Amesz, J.,
Ed.) pp 125-158, Elsevier, Amsterdam.

Lakshmi, K. V., Eaton, S. S., Eaton, G. R., and Brudvig, G. W. (1999)
Biochemistry 38, 12758-12767.

Rappaport, F., BIanchard-Desce, M., and Lavergne, J. (1994) Biochim.
Biophys. Acta 1 184, 178-192.

Lydakis-Simantiris, N., Ghanotakis, D. F., and Babcock, G. T. (1997)
Biochim. Biophys. Acta. 1322, 129-140.

Cha, Y., Murray, C.J., & Klinman, JP. (1989) Science 243, 1325-1330.

Haumann, M., Bogershausen, 0., Cherepanov, D., Ahlbrink, R., and
Junge, W. (1997) Photosynth. Res. 51 , 193-208.

Barbara, P. F., Meyer, T. J., and Ratner, M. A. (1996) Journal of Physical
Chemistry 100, 13148-13168.

MacLachlan, D. J., Nugent, J. H. A., Warden, J. T., and Evans, M. C. W.
(1994) Biochim. Biophys. Acta 1188, 325-334.

Vass, |., and Styring, S. (1991) Biochemistry 30, 830-839.

Winkler, J. R., and Gray, H. B. (1992) Chemical Reviews 92, 369-379.

78

92.

93.

94.

95.

96.

97.

98.

99.

100.

101.

102.

103.

104.

McLendon, G., and Hake, R. (1992) Chemical Reviews 92, 481-490.

Moser, C. C., Keske, J. M., Warncke, K., Farid, R. S., and Dutton, P. L.
(1992) Nature 355, 796-802.

Un, S., Tang, X.-S., and Diner, B. A. (1996) Biochemistry 35, 679-684.

Berthomieu, C., Hienerwadel, R., Boussac, A., Breton, J., and Diner, BA.
(1 998) Biochemistry 37, 10547-10554.

Clark, W. M. (1960) Oxidation-Reduction Potentials of Organic Systems,
The \Mlliams & Wilkins Company, Baltimore, VA.

Rich, P. R., Meunier, B., Mitchell, R., and Moody, A. J. (1996) Biochim.
Biophys. Acta 1275, 91-95.

Mulkidjanian, A. Y. (1999) Biochim. Biophys. Acta 1410, 1-6.

Caudle, M. T., and Pecoraro, V. L. (1997) J. Am. Chem. Soc. 119, 3415-
3416.

Dixen, W. T., and Murphy, D. (1976) J. Chem. Soc. Faraday Trans. ll 72,
1221-1230.

Segel, l. H. (1975) Enzyme Kinetics, Behavior and Analysis of Rapid
Equilibrium and Steady-State Enzyme Systems, John VVIley & Sons, New
York.

Chance, 8., DeVault, D. C., Frauenfelder, H., Marcus, R. A., Schrieffer, J.
R., and Sutin, N. (1979) Tunneling in Biological Systems, Academic Press,
New York.

Mahoney, L. R., and DaRooge, M. A. (1975) J. Am. Chem. Soc. 97, 4722-
4731.

Mahoney, L. R., and DaRooge, M. A. (1972) J. Am. Chem. Soc. 94, 7002-
7009.

79

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

Foti, M., lngold, K. U., and Lusztyk, J. (1994) J. Am. Chem. Soc. 116,
9440-9447.

Zavitsas, A. A., and Chatgilialoglu, C. (1995) J. Am. Chem. Soc. 117,
10645-10654.

Arthur, N. L., and Bell, T. N. (1978) Rev. Chem. Intenned. 2, 37-74.

Franz, J. A., Bushaw, B. A., and Alnajjar, M. S. (1989) J. Am. Chem. Soc.
111, 268-275.

Chatgilialoglu, C., lngold, K. U., and Scaiano, J. C. (1981) J. Am. Chem.
Soc. 103, 7739-7742.

Mims, C. A., Mauti, R., Dean, A. M., and Rose, K. D. (1994) J. Phys.
Chem. 98, 13357-13372.

Silverrnan, D. N. (2000) Biochim. Biophys. Acta 1458, 88-103.

Westphal, K. L., Tommos, C., Cukier, R. l., and Babcock, G. T. (2000)
Current Opinions in Plant Biology 3, 236-242.

Cukier, R. I. (1999) J. Phys. Chem. A 103, 5989-5995.

Cukier, R. |., and Nocera, D. G. (1998) Annu. Rev. Phys. Chem. 49, 337-
369.

Ulstrup, J. (1979) Charge Transfer Processes in Condensed Media,
Springer, Berlin.

Marcus, R. A., and Sutin, N. (1985) Biochim. Biophys. Acta 811, 265.

Okamura, M. Y., Paddock, M. L., Graige, M. S., and Feher, G. (2000) in
press.

Roth, J. P., Lovell, 8., and Mayer, J. M. (2000) J. Am. Chem. Soc. ASAP.

80

119.

120.

121.

122.

Sjodin, M., Stenbjorn, S., Akermark, 8., Sun, L., and Hammarstrom, L.
(2000) J. Am. Chem. Soc. 122, 3932-3935.

Krishtalik, L. I. (2000) Biochim. Biophys. Acta 1458, 6-27.
Krishtalik, L. I. (1979) J. Electroanalyt. Chem. 100, 547-561.

Cukier, R. |., and Zhu, J. (1997) J. Phys. Chem. 101, 7180-7188.

81

   

1((111)(11111111111111).1