IlllHlH"\llllHHlHlllllllllHlHM(IIIWIHIHIHIIHI

   

THESIS

UNIVE SITY U8

llllllllllllllllllllllllllll | “I Ill

3 1293 0088100

lllslllllllll

 

 

 

 

 

 

 

 

 

 

This is to certify that the
thesis entitled

ESEEM Studies of the Oxygen Evolving
Complex of Photosystem II

presented by

Michelle Mac

has been accepted towards fulfillment
of the requirements for

M q degree in _Chemisic_y_

emit

Major professor

N» mf/fiy 941%”

O-7639 MS U is an Affirmative Action/Equal Opportunity Institution ‘

 

 

f ,7, , “W, _ _.____ _g_ __.._._._____ _, w___}

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to roman thin checkout from your record.
TO AVOID FINES return on or before data due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

___,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

—lml—_

 

 

 

___wMSU Is An Affirmative ActiorVEqual Opportunity Institution

ESEEM STUDIES OF
THE OXYGEN EVOLVING COMPLEX
OF PHOTOSYSTEM 11
By

Michelle Mac

A THESIS

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

MASTER OF SCIENCE

Department of Chemistry

1993

ABSTRACT

ESEEM STUDIES OF THE OXYGEN
EVOLVING COMPLEX OF
PHOTOSYSTEM II

By
Michelle Mac

Photosystem II (PS 11) is the enzyme in photosynthesis which catalyzes the
oxidation of water. Water oxidation chemistry is thought to be performed by a
membrane bound oxygen evolving complex (OEC) containing manganese. The use of a
refined biochemical preparation which increases the concentration of PS II present in a
sample has facilitated study of the Mn multiline EPR spectrum associated with the DEC.
This spectrum has been studied by using magnetic resonance techniques. EPR studies
discussed in this thesis have shown that the reaction center complex (RCC) samples are
spectroscopically equivalent to their predessors. The increased concentration achieved
with the RCCs is essential for EPR analysis. Slight modifications of the published RCC
procedure were made as a result of this increased concentration. It has been shown
previously that substitution of Sr2+ or chelator into calcium depleted BBY samples
causes modifications in the hyperfine splitting seen in the multiline spectrum. These
differences are thought to arise from perturbations of the Mn exchange coupling. These
differences were present in the multiline spectra from the RCC samples presented in this
thesis. ESEEM studies of the RCC samples were also performed. A 4 MHz peak,
thought to arise from coupling to a ligated nitrogen, is present in all spectra recorded.
This indicates a conformational change that does not affect the interactions of the Mn
with its environment. Additional magnetic resonance studies will be performed to
determine the magnitude of these interactions.

 

ACKNOWLEDGEMENTS

I would like to thank both Dr. Jerry Babcock and Dr. John McCracken for their guidance
in this project.

I would also like to thank all the members of the Babcock and McCracken research
groups, and in particular, Kerry Reidy for her encouragement and ”impromptu” quantum
mechanics discussions.

I would especially like to thank Matt Espe for his time, patience and friendship. His
teaching and knowledge were invaluable to me and this thesis could not have been
completed without him.

Lastly, I would like to thank my parents for their love and encouragement

M.M.

ii

TABLE OF CONTENTS
Chapter 1: Introduction ............................................................................... 1
Structure ......................................................................................... 3
Electron Transfer in Photosynthesis ................................................ 7
The Oxygen Evolving Complex ..................................................... 9
Manganese ..................................................................................... 11
Chapter 2: Magnetic Resonance Theory ..................................................... 20
EPR ................................................................................................ 20
ESEEM ........................................................................................... 28
Two Pulse Experiment ........................................................ 28
Nuclear Modulation Effect. ................................................. 32
Chapter 3: Experimental ............................................................................. 36
Materials and Methods .................................................................... 36
EPR and ESEEM Studies ................................................................ 39
Chapter 4: Results and Discussion .............................................................. 41
RCC Characterization ..................................................................... 41
EPR Characterization ...................................................................... 43
ESEEM Studies of the Modified Multiline ...................................... 48
Chapter 5: Future Work ............................................................................. 55
ESE Detected ENDOR. .................................................................. 55
Characterization of the S3 EPR Signal ............................................. 57
PSII Mutants ................................................................................... 58
Bibliography ............................................................................................. 59

m

Figure 1

Figure 2

Figure 3
Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

LIST OF FIGURES

The Z scheme of photosynthesis representing

non-cyclic electron flow in higher plants ........................................ 2
Proposed structure of the PS II/OEC including

polypeptides and co-factors ............................................................ 5
Kok's S-state cycle .......................................................................... 10

Oxygen evolution as a function of flash number
(from reference 25) ......................................................................... 12

The multiline spectrum associated with the 82
state of the CBC .............................................................................. 15

Energy level diagram for an S=1/2, I=ll2 system.
The dashed lines represent allowed EPR transitions
corresponding to the selection rules AM: 3; l and
AM,= 0. The solid lines represent energy differences

between the states. The corresponding energies are
given on the left. .............................................................................. 22

Schematic representation of antiferromagnetic coupling
in the fragment Mn3+-02'-Mn4+ ........................................................ 25

The effect of exchange coupling on the triplet and singlet
states of an S,=l/2, 82: 1/2 system. The degeneracy at

zero field is removed upon the exchange interaction ........................ 27
An inhomogeneously broadened line composed of
individual spin packets ..................................................................... 29

A 2 pulse ESEEM experiment. Inset: Frequencies

corresponding to two wpin packets with respect to

the frequency of the rotating frame. A) Net magneti-

zation along the Z axis. B) Magnetization vector after

the first pulse. C) Procession of individual spin packets

after time 1. D) Location of the magnetization vector

components after the second pulse. E) The formation

of the spin echo .............................................................................. 30-31

iv

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Nuclear modulation effects. Inset: Energy levels
corresponding to an S = 1/2, I = 1/2 system. Solid
lines represent allowed transitions. Dashed lines
represent semi-forbidden transitions. A) Location
of the vector corresponding to transition A at time

1 after the 90° pulse. B) Branching of transitions after

the 180° pulse. C) The echo amplitude is given by the
projection of the vectors on the Y' axis. This is a
function of the angle 0 between the two ....................................... 33-34

Buffers used in the RCC procedure .................................................. 37

The light minus dark multiline spectrum of a) Ca“ re-

constituted RCCs and b) Sr2+ reconstituted RCCs. The

spectra are the result of 30 (a) and 60 (b) signal averaged

scans. Spectrometer conditions were as follows: scan

range, 2000 G; center field, 3400 G; power 10 dB:

modulation amplitude, 20 Gpp; gain 1.6 x 10‘; temperature,

8 K. .................................................................................................. 44

Multiline spectrum from a) Ca2+ reconstituted RCCs

and b) EGTA treated RCCs. Spectrometer conditions

were as in Figure 13. The spectra are the result of

25 (a) and 50 (b) signal averaged scans following

illumination at 200 K. ................................................................ 46-47

Two pulse ESEEM (a) and Fourier cosine transfer (b)

for the Ca2+ reconstituted sample. Measurement con-

ditions for the data shown are: microwave frequency,

8.893 GHz; magnetic field strength. 3400 G; pulse

power, 25 W; pulse width, 15 ns FWI-IH; tau value,

140 ns; sample temperature, 1.8 K; repetition rate,

60 Hz. ............................................................................................ 50

Two pulse ESEEM (a) and Fourier cosine transfer (b)
for the Sr2+ reconstituted samples. Spectrometer conditions
were as in Figure 15 ....................................................................... 52

Two pulse ESEEM (a) and Fourier cosine transfer (b)
for the EGTA treated sample. Spectrometer conditions
were as in Figure 15 ....................................................................... 53

Figure 18 -‘ Davies ESE-ENDOR pulse sequence (top). Energy level
diagram for a coupled S = 1/2, I = 1/2 system. The
diagram on the left shows the level populations at thermal
equilibrium while that on the right shows the level

populations after a selective 180° pulse resonant with
the 1-3 transition has been applied .................................................. 56

\i

Chapter I
Introduction

Although photosynthesis has been studied extensively since the discovery of
oxygen evolution from plants in 1780 by Joseph Priestley, many of the molecular
intricacies of the process remain an enigma. The mechanism of photosynthesis utilizes
photochemical energy from sunlight to disrupt chemical bonds in stable substrates. The
free energy inherent in this process can be trapped by the organism to synthesize vital
amino acids, proteins and other biomolecules. Molecular oxygen is released by higher
plants as a by-product of the reaction. The process can be described with a deceptively
simple reaction:

C02 4- H20 ---------- > (C1120)n + 02
The mechanism behind this reaction however, is complex and requires the interplay of
many proteins and cofactors, many of which have not been completely characterized.

There are two types of pigment-protein complexes, called reaction centers, that
convert light energy into chemically useful energy in higher plants. The first of these,
Photosystem I (PSI) mediates the production of oxidized plastocyanin and reduced
NADP. This NADPH produced is used to fuel the enzymatic cycles associated with the
assimilation of carbon dioxide in the Calvin cycle. The second reaction center, the
Photosystem II/oxygen evolving complex (PS IIIOEC), is associated with water
oxidation and the evolution of oxygen. Both of these systems are represented
schematically in Figure 1. This model, known as the Z-scheme, represents the flow of
electron transport in plants and cyanobacteria.‘ The vertical position of each electron
carrier corresponds to its reduction potential at pH=7 .0.

An interesting aspect of the electron transfer process in photosynthesis is its
vectorial nature. The photosynthetic machinery is located, in higher plants, in an

organelle known as the chloroplast ChlorOpIasts are similar in structure to the

25%. SEE: E .50: «.2520 2.93-5: mes—.3058 $3558.23 .0 052.8 N of. 2 0.53,.—

.Ea.m.?2of 5.9.50 .13 9595830 : 5238.25

 

 

a /
SE .
I .5 t 5

f . . 29:33

.392 on. amen. 2...):

UH! 2. cu: a
El 2— . IE0

a:22 fl. _ .2... So 440

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3

mitochondria of animals in that they contain both an inner and outer membrane. The
inner membrane surrounds an aqueous protein matrix which contains membranous
structures called thylakoid membranes. These closed vesicles separate two chemically
different environments: the stroma. located on the exterior of the thylakoid membranes
and the lumen, located on the interior. These appear on electron micrographs as flattened
sack-like structures and are "stacked" in certain regions of the chloroplast.2 It has been
shown that as much as 85% of the PS II reaction centers are concentrated in these stacked
regions, called grana.2 Photosystem I is not, however, in the same location in the
thylakoids as PS 11. It is found, instead, in the unstacked stromal regions of the
thylakoid. This asymmetric arrangement assists in providing directionality to the
electron flow. It also results in both charge and proton gradients across the
photosynthetic membrane that ultimately provide the free energy necessary for ATP
synthesis.

As seen in Figure 1, PS I and PS II are linked through electron transfer reactions.

However, due to their different locations in the thylakoid membrane, an additional
membrane bound complex is required to mediate their interaction. Cytochrome b6f (cyt

b6f) is distributed evenly in the chloroplast between the grana (PS II) and the unstacked

regions (PS I) and therefore seems like an ideal charge transfer interface between PS 11
and PS I. The cytochrome is too large to readily diffuse the distances needed for
transportation of electron equivalents between PS 11 and PS I, especially on the timescale
of intersystem electron transfer. The lateral communication between reaction centers is
performed by a small (MW =10.5 kDa) copper protein called plastocyanin.

Structure
A better understanding of the functional aspects of the PS IIIOEC can be obtained
with an in depth discussion of the polypeptides and co-factors associated with this

membrane bound complex. There are less than twenty polypeptides associated with the

4

PS II/OEC, making it a moderately complex system.3 This complexity, however, is
remarkably easy to resolve biochemically. Detergent solubilization methods have
allowed isolation of a number of smaller assemblies which have lead to the construction
of a reasonable working model of the PS II/OEC. This postulated model which shows
relevant polypeptides and co-factors is depicted schematically in Figure 2.3 The electron

transfer pathway in photosynthesis begins with photon absorption by the reaction center
chlorophyll, P630, leading to a charge separation that produces P630+QA'. The

quinone, Q A, reduces another quinone 03. Following two photochemical events, this

quinone is released from the membrane and the electrons are transported to PS I. On the
oxidizing side of PS 11, a complex containing the transition metal manganese, upon
oxidation, provides electrons for a tyrosine molecule which in turn reduces P6304: This
pathway will be described in more detail below.

The PS ’II/OEC contains both extrinsic and intrinsic polypeptides as well as a
number of enzymatically active co-factors (the kinetics of the electron transfer reactions
associated with these will be discussed in the following section). There are three
extrinsic polypeptides, identifiable by their molecular weights: the 17, 23, and 33 kDa
polypeptides. These serve several purposes. First. the 33 kDa protein protects the
manganese complex associated with water oxidation. This Mn ensemble is maintained in
a configuration that optimizes oxygen evolution by this polypeptide. The 33 kDa
polypeptide binds directly to the intrinsic polypeptides, an interaction further stabilized
by the Mn complex! Recent crosslinking studies have determined the stoichiometry of
this polypeptide to be 1:1 with an intrinsic polypeptide (the 47 kDa).5 The 33 kDa
protein also serves an additional function; it isolates the redox active species involved in
water oxidation from the aqueous environment! This is an important function in that the
water oxidizing system is susceptible to spurious reductants and this biochemical barrier
provides a redox shield that allows it to perform its water splitting function successfully.

Unfortunately, this polypeptide is difficult to study quantitatively due to the

.Eofifléo was 82:59:?— wEquE Urn—OE mn— oz. .«0 232:8 nomomoi § 9.53,.—

 

01
N

ha

 

 

 

            

,-....t—
fl

 

 

 

   

 

 

 

 

 

w 2 .5
a m .mmmmmmmamaém...i......" a
- ........ " umammmrummumumumuwmum "m“ I‘m---
m .m.§ammmmmmmmmmmmmmmmmmm mm
" .mmmmmmmmmmmmmmmmmmmmmmmmmm mm
qua-Ala! m mumuwmum"mum“mumuwwmumuw mum «stale!
5....
m Wmmmmm-ommmmmmmm‘ m B

   

 

 

. m o
. . . a n n uuuuuuuuuuuuuu
I O O 0
° . . . .o....o rrrrrrrrrrrrr a.
. . u - o o aaaaaaaaaaaaaa
o o . .u....- IIIIIIIII
I u o o
. . o . . . nnnnnnnnn
m I O I
. . ....... IIIIIIII
n a O I O o
O I O a I I
. . c . o a .
c . . a
D butch-bctbunb D I t.
I

 

rearrangements of the intrinsic polypeptide core that occur upon removal of these soluble
subunits.6 Extraction methods and their effect on the PS II/OEC will be discussed in an
upcoming section.

The remaining extrinsic polypeptides, the 17 and 23 kDa proteins, also function
in binding co-factors needed for water splitting chemistry, in particular the calcium ion
and the chloride ion. These two polypeptides also protect the manganese ensemble from
exogenous reductants.‘ Upon removal of the 17 and 23, oxygen evolution can only be
maintained with increased concentrations of both calcium and chloride.7

The photochemical core of PS II is formed by two membrane spanning intrinsic
polypeptides. These heterodirners (MW approximately 32 kDa) , commonly known as
D1 and D2, bind chlorophylls, pheophytins and quinones that mediate the light driven
charge separation reactions.a Sequence and functional homologies between the L and M
subunits of the bacterial reaction center from Rhodopseudomonas vin'dis and the Dl/D2
heterodirner have led to the suggestion that D1 and D2 each contain five membrane
spanning helices?“ This structural arrangement leads to a pseudo-C2 symmetry for the
PS IIIOEC. This apparent symmetry is very useful in designing experiments to study the
PS II/OEC.ll However, this symmetry is broken by the incorporation of the manganese
ensemble which, in one proposed model , is located off of the C2 symmetry axis}2 The
contrOversy surrounding the organization and location of the manganese complex will be
discussed in a later section.

The remaining intrinsic polypeptides have a wide variety of functions. By
providing binding sites for accessory chlorophylls the 43 and 47 kDa proteins have
mainly a light harvesting function. The 23 kDa polypeptide can be isolated in its pure
form, and in addition to binding chlorophyll a. has been implicated in maintaining the
QB site on D1334“ Conformational changes to D1 are modulated depending on whether
this quinone-binding site is occupied. The smaller (22 and 10 kDa) polypeptides do not

have binding sites for any of the co-factors. However, removal of these two proteins

7

results in increased accessibility of exogenous acceptors like DCBQ to QA'. Therefore.

the 22 and 10 kDa proteins must, in some way, influence the environment around these

quinones. The final two intrinsic polypeptides (4 and 9 kDa MW) each provide a single
histidine ligand to the heme of cytochrome b559.ls

Electron Transfer in Photosynthesis
The initial step in photosynthesis occurs when a photon of light is absorbed by a
light harvesting protein antenna complex (called the LHC) composed of non-covalently
bound chlorophyll a, chlorophyll b and carotenoid molecules (in higher plants and green
algae). In cyanobacteria, this antenna complex is the phycobilisome. These antenna
pigments then transfer energy via exiton interaction and Forster transfer to the reaction

centers. All of the energy absorbed by the LHC is transferred to a specialized monomer
or dimer of chlorophyll a in PS 11 known as P630. This name is derived from the

absorption maximum of this molecule at 680 nm. This energy promotes the P630

molecule into an excited singlet state. This excited state within 3 picoseconds, reduces a
nearby pheophytin molecule (Pheo), forming Phe0' and P5304”. To prevent

recombination, the Phe0‘ quickly (within 300600 ps) reduces a nearby plastoquinone.
Q A36

f The electrons are shuttled out of PS II by electron transfer from Q A' to another
plastoquinone, QB in about 200 us. An interesting change occurs at this point.
Previously, the photosynthetic process had been composed of single electron events. The
movement of electrons out of PS II, however, is a multielectron process. QB‘ remains

tightly bound in its binding site until a second photochemical event reduces it to Q32'

whereupon it becomes protonated and is released from its binding site as plastoquinol.
The plastoquinol is then oxidized by cytochrome b6f and the electrons are transported to

PS I by a plastocyanin molecule. The QBH2 is replaced in its binding site by an

oxidized quinone from the membrane associated quinone pool.l7

8

Recombination between P630+ and Q A' (which occurs with a half time of
approximately 100 [1.8) would be a photosynthetically ”wasteful" process. To prevent this
recombination, either Q A' must be oxidized or P630+ reduced in a time considerably
shorter. The electron transfer between Q A’ and Q3 has been measured to be in the 100-

500 as time range.” This indicates that the reduction of P530+ must occur in the

submicrosecond time range. The reduction, however, was found by Witt and coworkers
to depend on the state of the CBC." To explain this dependence, a redox active donor

was postulated. This donor operates as a charge transfer interface between the CBC and
P630+ in an equilibrium that depends on the net charge of the CBC.18 There has been

spectroscopic evidence for this donor, a tyrosine residue (Y7), although under non-
physiological conditions other complexes may donate to P530+J9 The oxidizing power
generated in producing P630+ must be directed eventually at water. Tyrosine oxidation,
producing the highly oxidizing (Em°=l.0 V) Y; radical is a means by which to achieve
this.”

There is a second redox active tyrosine in PS 11 known as YD+. It has been
identified as Tyr-lol of the D2 polypeptide by site directed mutagenesis experiments.20
It has no known function and in its oxidized form is responsible for the dark stable EPR
signal known as Signal 11.19 The EPR lineshapes of Y5” and Y; are identical which
indicates that the unpaired spin density distribution and the orientation of the tyrosine
phenol ring with respect to the polypeptide backbone are identical for both species.21
However, YD+ is not involved in the electron transfer reactions that lead to water
oxidation. This was confirmed with site~directed mutagenesis experiments by Debus et
a1 where phenylalanine was substituted for both Y2 and Y9.20 In the Yz case.
photosynthetic growth was absent after the deletion whereas in the YD case

photosynthetic growth continued. The two tyrosine moieties are related by an apparent
C2 symmetry. These symmetry related branches, however, are not related functionally.

The Oxygen Evolving Complex
The OEC is the site of water oxidation in the photosynthetic process. The half

reaction corresponding to this process,
2H20 ------- >02+4H++4e‘

has an average reduction potential of 0.93 V at pH=5.0.26 The energy required to drive

this reaction comes from the oxidized P630. The midpoint potential for this reaction
Peso ------- > Peso+ + e'

is 1.17 V at pH=5.O.22 The stoichiometries of these two reactions bring up an interesting

mechanistic question. A single absorbed photon generates only one oxidizing equivalent,

however the water splitting process is a four electron process. The oxidizing power of

four photons needs to be combined in order to facilitate water oxidation. Insight into the

resolution of this paradox began with the now classic 02 flash yield measurements of
Joliot and co-workers}3 These experiments showed that the 02 produced by a sequence
of brief (about 10 as) saturating light flashes showed a damped oscillatory pattern with a
periodicity of four following the third flash. These data were soon reproduced by Kok
and co-workers and a working kinetic model was established.24 This model showed that
the PS II units function independently in accumulating the four oxidizing equivalents
necessary to split water. This so-called S-state cycle is driven by successive
photoactivations of PS 11. A schematic representation of this cycle is shown in Figure 3.
The S represents the water splitting center and the subscripts indicate the number of
oxidizing equivalents accumulated. Oxygen evolution occurs only after the S4 state has
been achieved. This model accurately explains the details of the flash experiments.
Maximal oxygen evolution occurs after the third flash with repeated maxima after
successive four flash intervals. See Figure 4 for similar data on spinach chloroplasts by
Babcock” The oxygen evolution vs. flash pattern eventually dephases and reaches a

steady state value. This pattern can be explained if both So and 81 are dark stable states.

After dark adapting samples it was found that approximately 60% of the centers were in

10

e' e' e' e‘
by ? hvz by ? hv !
So SI 52 53 S4

/\

202+4H+ ZHZO

 

Figure 3: Kok's S-state cycle

11

S 1. To concentrate the DEC in 82 further, Kok and co—workers took dark adapted

samples, subjected them to a brief flash and allowed them to dark adapt a second time.
This treatment resulted in almost no oxygen evolution after the first two flashes and an
even larger maximum after the third?“ The slow dephasing seen is due to the non-zero

probability of ”misses" and ”double hits" resulting in the loss of S-state coherence.

Manganese

The first row transition metal manganese has long been associated with the
oxygen evolving complex of photosynthesis. Its many stable oxidation states (+2 to +7)
make it an ideal metal center for the multielectron chemistry that occurs in the OECF‘S
The earliest experiments linking Mn to oxygen evolution were performed in the 1930's
and showed that algae grown on low levels of manganese exhibited a decreased ability to
evolve oxygen.” Later experiments placed the location of manganese directly in PS II.
Cheniae and Martin, using heat shock techniques, demonstrated a loss of oxygen
evolution concomitant with the release of Mn.” Yocum and co-workers found the major
fraction of Mn2+ in highly oxygen evolving thylakoid membranes was in an EPR silent
form.29 This amount, calculated to be 4 per 400 chlorophyll, was resistant to release by
Ca2+ and was released by using acidification processes and subsequently quantified with
EPR measurements. A stoichiometry of four Mn per PS II was determined for optimum
functioning of photosynthetic oxygen evolution. The organization and corresponding
valence states of this manganese ensemble are not known and are currently the subject of
much debate among researchers. Dirners, trimers and tetrarners have all been suggested
as structures for this multinuclear cluster.

A variety of techniques have been employed to determine the oxidation state

changes of this functional Mn ensemble that correspond to S-state transitions.

12

 

 

 

 

1.4 h-
b
a
J
2 u -
)-
o"
O i
U
'3
J
g 0.0 -
§ 0. tvowrtou IN rusmn LIGHT
P \
‘_ l L l L l A l n_ l L
o 0 t: 10 u so

FLASH NUNIER

Figure 4: Oxygen evolution as a function of flash number (from reference 25).

l3

EXAFS: The earliest spectroscopic results were reported by Kirby et al in 1981.30
Manganese X-ray absorption fine structure (XAFS) results for wild type chloroplasts and
chloroplasts treated with alkaline Tris buffers to release Mn and inactivate the water
splitting mechanism were recorded. The differences between these, representing the
active pool of Mn, were shown to be similar to the XAFS from model di-tt-oxo-bridged
Mn systems. Later, using a biochemically refined preparation, Yachandra et al 31
presented extended XAFS (EXAFS) data showing Mir-Mn distances of 2.7 A. This
proximity provides evidence for at least a binuclear Mn cluster. The EXAFS data also
included a Mn-nitrogen or oxygen distance of 1.75-l.8 A and a disordered shell of
oxygens or nitrogens at an average distance of 2.15 A from the Mn. A disadvantage of
using EXAFS as an analytical tool is the uncertainty in the atomic number determination
of ligands coordinated to the metals. This uncertainty (1 10 for atomic number) results
in an ambiguity in ligation as far as nitrogen and oxygen are concerned, two proposed
ligands to the Mn ensemble.32 Recent work by Penner-Hahn in which both improved
preparations (concentration of Mn approximately 1.5 mM) and higher detection
efficiency were used, confirmed two to three Mn-Mn distances of 2.7 A.32 An additional
third backseattering shell of 3.3 A which includes Mn was also confirmed. Again, this

provides evidence for a multinuclear (trimer or tetramer) cluster.

arm Perhaps the greatest amount of information regarding the structure of the Mn
ensemble has come from magnetic resonance techniques, particularly electron
paramagnetic resonance (EPR). EPR can be used to identify paramagnetic species and to
characterize the environment around them. The Mn ensemble in the oxygen evolving
complex is a system which, through the S-states, generates a number of paramagnetic

intermediates, thus maldng it ideal for analysis by EPR. Both continuous wave (cw) and

14

pulsed EPR techniques have been employed to characterize the structure of and ligands
to the Mn ensemble.

The first EPR signal associated with the Mn ensemble was reported by Dismukes
and Siderer in 1980.334 This so-called multiline signal was initially observed in samples
that had been rapidly frozen following a flash of light at room temperature. The EPR
signal consists of 18-20 hyperfine lines separated by 85-90 gauss, is centered at g=l.982
1 0.002 and has a linewidth of 1500-1800 gauss, see Figure 5. It can only be observed at
temperatures less than 35 K. The signal amplitude oscillates with flash number having
maximal intensity after the fast and fifth flashes. This flash induced periodicity is
analogous to the oscillations seen byloliot23 and Kok24 in the S—state experiments. The
multiline signal was thus attributed to the S2 state of the Kok cycle. This was confirmed
with temperature dependent decay experiments performed by Brudvig et al.” The decay
rate following flash illumination was found to be temperature dependent. The half-time
decay of 40 s at 295 K contrasts markedly with the half-time decay of 40 min at 253 K.
Below 240 K a single illuminating flash becomes less likely to produce the multiline
signal. The 40 s half-time in particular provides evidence that the multiline spectrum
results from a relatively stable state rather than a transient intermediate. This 40 s decay
is very similar to that reported for the 82 state of the PS IIIOEC.3647

The advancement of the S states was also found to be markedly temperature
dependent; S1 ----- >82 can occur at temperatures as low as 160 K and 82 ----- >83 occurs

readily only at temperatures above 210 K. Brudvig and co-workers found that a single
intense flash of light became progressively less effective in generating the multiline
signal at temperatures below 240 K and, at lower temperatures (160 K), continuous

illumination was needed.”

Model compounds: Attempts at reproducing the multiline spectrum using synthetic
model compounds have been numerous. The oxo-bridged dimer

Intensity

15

 

800 F I T I
500 -

400 - n

200 H- p

 

 

 

 

 

 

 

 

 

 

-200 -

 

 

 

 

 

-400 r r L L
2500 3000 3500 4000

Field (goose)

Figure 5: The multiline spectrum associated with the S2 state of the OEC.

16

[MannIV(O)2(bpy)4]3+ has been extensively studied as a model for the Mn cluster in

the OEC. The similarity of the EPR spectrum to that of the multiline led early
researchers to the theory that the multiline originated from a mixed-valence, exchange-
coupled cluster of Mn ions. Unfortunately, these suongly coupled dimers did not exhibit
any EPR features beyond the 16 line spectrum expected for an S=1/2 ground state.30
Recently, an asymmetric MnIIIMnIV dimer has been synthesized; it exhibits the weak
antiferromagnetic coupling and the reversible thermally regulated transition from an
S=l/2 ground state to an S=3/2 excited state thought to occur in the Mn complex.38 The
EPR signal from this compound looks similar to that obtained from the Mn complex.
Trinuclear compounds have also been postulated as models for the CBC Mn

complex. The EPR spectra of two of these compounds:
Mn,('dien),(0Ac),(Cto,),o,-H20 (A)
Mn3(dien)3(OAc)21303H20-0.33 CH30H (B)

exhibit the 16 line hyperfine coupling indicative of an effective S=1/2 ground state.39
The evidence for a tetranuclear Mn ensemble has also been convincing.
Dismukes et al using EPR simulations, have postulated that a tetramer composed of three
Mn (III) and one Mn (IV) could model the multiline spectrum!0 This simulation,
however, failed to reproduce the breadth and the number of lines in the experimental
spectrum. Recent theoretical studies have provided additional evidence for a tetrarrrer.‘1
By using six fitting parameters, the shape of the experimental spectrum and its hyperfine
patternsplittings and relative intensities have been described. However, the parameters
used do not give direct information regarding the structure of the tetramer nor do they

provide any information about the coupling schemes between the ions.

Isolation and Characterization: Although the nuclearity of the Mn ensemble remains
unclear, information regarding function and structure can be obtained from

characterization of the environment around the manganese. Improved biochemical

17

preparations and subsequent magnetic resonance analysis have yielded much information
about co-factors and ligands to the OEC.

The role of calcium in PS II has been well documented. Initially observed in
photosynthetic membranes from cyanobacteria calcium requirements have played roles
in higher plant photosynthesis as well.42 Reconstitution of calcium depleted membranes
was found to reverse the inhibition of oxygen evolution.‘3 Depletion of calcium can be
accomplished by salt washing either in the light or dark with 1-2 M NaCl. This treatment
removes the 17 and 23 kDa polypeptides and allows access to the proposed calcium site.
Salt washing in the light is facilitated by the addition of a chelating agent such as
ethylene diamine tetraacetic acid (EDTA). Because illumination was found to have an

effect upon the ease of calcium extraction, the role of calcium in PS 11 was associated
with the S-states of the OEC. The ease of extraction progresses in the order S3 > S2~So

>81.

Calcium effects: The multiline spectrum has been studied to determine the effect of
calcium depletion upon S-state turnover. The S-state transition inhibited by calcium

depletion has been controversial. Boussac and Rutherford provide evidence for
blockage of the S3 ------ > So transition in NaCl washed samples.“ Continuous

illumination of the calcium depleted samples at 200 K resulted in a normal 82 multiline
signal. The intensity of this signal was comparable in amplitude to the signal from
calcium reconstituted samples. Flash experiments showed normal multiline formation
upon the first flash with a typical decrease in amplitude after the second flash. However.
no increase in intensity was observed upon the fifth flash. The typical S-state oscillation

pattern was present upon calcium reconstitution of these samples.

Strontium effects: Reconstitution of these samples using Sr2+ instead of Ca2+ gave a

modified multiline signal.“ The Sr2+ reconstitution resulted in losses. splits, shifts and

l8

redistributions of the hyperfine lines. The average hyperfine spacing decreased from
87.9 G in the Ca2+ reconstituted samples to 71.0 G in the Sr2+ samples. This indicates
that the introduction of Sr2+ into the calcium site perturbs the Mn cluster such that a

slight conformational change occurs.

Chelator effects: Chelator concentration during salt washing also has an effect upon the
multiline spectrum.46 High concentrations of chelator (10 mM) lead to a dark stable,
markedly modified multiline signal. Direct binding of the carboxylic acid functional
groups of the chelator to the Mn is one possible explanation for the modifications
observed in the multiline spectrum. This ligation is not possible in the presence of Ca2+
which provides evidence that Ca2+ may regulate access of a ligand to a binding site on
the Mn.

Nitrogen ligation: Nitrogen, particularly the nitrogen moiety of the amino acid histidine.
has been postulated as a potential ligand to the Mn complex. EPR studies of spinach
grown hydroponically with K15NO3 as the sole nitrogen source were done in 1989 in an
attempt to resolve the nitrogen ligation question.‘7 There were no observable changes in
either the general shape or the linewidth of the hyperfine structure of the multiline
spectrum. The fine structure was therefore attributed to manganese hyperfine

interactions and not to nitrogen superhyperfine couplings.

ESEEM Characterization: Often in EPR, and particularly with transition metal
complexes, the magnitude of the superhyperfine interactions is small compared to the
overall linewidth of the signal. To resolve these couplings better, other techniques must
be used. For large hyperfine interactions, double resonance techniques such as ENDOR
(electron nuclear double resonance) can be applied. The resolution of smaller hyperfine

couplings can be accomplished by using spin echo techniques.

19

Electron spin echo envelope modulation (ESEEM) techniques were used to
observe nitrogen interactions between the Mn complex and a proposed histidine ligand in
both cyanobacteria and spinach thylakoid preparations.‘“9 By studying ammonia
inhibition mechanisms in PS II, Britt and co-workers were able to identify hyperfine and
quadrupole frequencies associated with the 1“N (I=1) of ammonia. Subsequent isotopic
exchange experiments with 15N (1:1/2) were used to confum the nitrogen from ammonia
as the source of these frequencies. Analogous experiments were performed on oxygen
evolving cyanobacteria preparations. Cultures of the cyanobacteria Synechococcus were
grown with KNO3 as their only source of nitrogen. Again, hyperfine and quadrupole
frequencies were determined using ESEEM spectroscopy. Isotope labelling experiments

confirmed nitrogen as the origin of these frequencies.49

Chapter 2
Magnetic Resonance Theory

EPR: Electron paramagnetic resonance techniques are useful in identifying the
paramagnetic species present in a sample. Interactions between a paramagnet and nearby
magnetic nuclei can also be characterized by using this technique. As previously
mentioned, both cw and pulsed EPR techniques have been used on the photosynthetic
system to characterize the ligand environment around the Mn ensemble. Conventional
cwEPR has not been able to account completely for the splittings seen in the multiline
spectrum. In this section, the theory behind cw and pulsed EPR and its application to
the multiline spectrum will be discussed.

An EPR spectrum can be described with a spin hamiltonian composed of several
terms, each describing the interaction of the unpaired electron with its environment”
This hamiltonian, can be written as

X=HEZ+HNZ+HHF+HEI (1)
where H52 represents the electronic Zeeman hamiltonian, HNZ the nuclear Zeeman
hamiltonian, Hm: the isotropic hyperfine hamiltonian, and H51 the exchange interaction
hamiltonian. The hamiltonian operates on the spin only part of the total wavefunction.

The first term of equation 1 represents the interaction of the unpaired electron
with the applied magnetic field. The Zeeman hamiltonian can be explicitly written as

HEZ = gBI-IS (2)
where g is the isotropic electronic g factor, B the Bohr magneton, H the applied magnetic
field and S the spin angular momentum operator. The effect of this term is to split the
degeneracy of the two spin only wavefunctions for the electron (S=l/2 where S in this

context is total spin). The energy corresponding to these levels is given by

21

where Ms represents the quantum number associated with the specific spin state. For a

system with a single unpaired electron. Ms can have two values. 11/2. This gives an

energy separation of

AB 52 - El (4a)
AB = gen (4b)

Transitions between these two energy levels (Zeeman levels) can be induced by an
electromagnetic field of frequency P if the energy of an incident photon, he, is equal to
the splitting between levels. These transitions between energy levels are the resonances
present in an EPR spectrum. The resonance condition for EPR is given in equation 5.

hr = gBl-l (5)

As can be seen from this expression the energy separation depends on both the field and
frequency values. EPR experiments are performed by holding the frequency constant
while varying the magnetic field. Therefore, the energy separation increases as the field
becomes larger. This is evident in the energy level diagram for an S=1/2, I=ll2 system
as presented in Figure 6.

The magnetic field also has an effect on magnetic nuclei. The second term in
equation 1, the nuclear Zeeman interaction, represents the interaction between the nuclear

spin and the applied Zeeman field. It is given explicitly by equation 6:
HNZ = ’ 8n5nHI (6)

where gn is the nuclear g factor, B“ the nuclear magneton, and I the nuclear spin

angular momentum operator. By applying the field along the z axis and assuming
quantization of the nuclear spin along that axis, the nuclear spin quantum number M] can

be substituted in the corresponding energy expression giving
E = - ganH M1 (7)

The effect of this interaction is apparent in the energy level diagram in Figure 6.

22

do. 2: .5 33w

2: 3.325 ”Eggnotg 2.... .335 2.. 303.3 3033...? .333 .3352 3:: 3.8
2.... .c n :24 one . « no2< no.3 oozes—om 2.. o. 3:52.3th 80:55: yr... 333.?
25.352 «2:. 12.26 2.... .895? N: n. .m:um .3 .5.— EEmm... .95. 33cm. 3 0.5”...—

2 3.7.» - :4. -2 an-

 

 

ST _E
NZ> N\< N: thE
. . m2-” .5
S _znzm _. 3.. .2 am- . n
u . Inm
.
. _
I i a..-" E
m... .2..sz + 3. - m .__ an
_z> «2 Six:
2 .2an - I... .2 am
ST _E
3.32.”.— :anoN eomucauoao— 52:3

550.932 2.5%»: vague—fl

23

The magnetic field felt by the unpaired electron is a combination of the applied
field and the fields due to nearby magnetic nuclei (1)0). The third term in the
hamiltonian from equation 1 represents the interaction of the electron with nearby
magnetic nuclei and is termed the hyperfine coupling. The hamiltonian for this
interaction can be given by

HHF = l'rAo S - I (3)

The 2 components of the electronic and nuclear spin angular momentum operators can
be substituted in place of S - I providing the hyperfine interaction is less than the
elecu'onic Zeeman interaction. There are two components to this hyperfine term. The
first. termed isotropic, is represented by equation 8. It results from Fermi contact
coupling, is orientationally independent, and originates from the non-zero probability of
the electron being located at the nucleus of the coupled atom. This can only occur if the
unpaired spin density resides in an s orbital. This interaction energy was given by Fermi
as

Ens = -(81r/3)Itl(0)|2 norm (9)

where “0) is the wavefunction evaluated at the nucleus and #ez and #nz are the
magnetic dipole moments of the electron and the nucleus, respectively. Substitution of
the nuclear and spin angular momentum operator relations for the classical magnetic

dipole moments in equation 9 will yield the isotropic hyperfine hamiltonian. The term
hAu represents the interaction energy between the electron and the nucleus and is equal to

(81/3)gnflng°86| tl(0) lz. A0 is called the hyperfine coupling constant and is measured in
units of gauss or megahertz. The energy corresponding to this interaction can then be
given as

E = hAoMsMI (10)

For an S=1/2. I=1/2 system these energies are:
Ell2,l/2 = hit/4 (10a)

24

E1/2,-1/2 = -hA./4 (10b)
5-1/2.1/2 = -hA./4 (10c)
5—1/2.- 1/2 = hAJ4 (10d)

The effect of this interaction on the energy levels can be seen in Figure 6.

The second part of the hyperfine interaction is the orientationally dependent
dipolar coupling. The dipolar coupling is a through space interaction between the
unpaired electron and neighboring nuclei. The dipolar coupling is anisotropic and
described by the tensor A given in equation 11. The magnitude of the hyperfine tensor A
is dependent on the orientation of the Zeeman field relative to the A-tensor axis system.

AHF = Aiso + Adipolar

Amour: saw I (11)
The hyperfine coupling in the Mn complex has been found to be primarily isotropic,
therefore this dipolar part of the hamiltonian can be neglected.

Referring again to Figure 6, the allowed transitions in an EPR experiment
correspond to the selection rules Ms: 1; 1 and AMI = 0. These transitions can be seen
in Figure 6.

The fourth term in the hamiltonian, the exchange coupling term H51, arises as a
result of the linkage of two molecular fragments, each containing unpaired electrons.Sl
The magnitude of this exchange interaction depends on the bonding characteristics of the
two metal centers. Direct metal-metal bonds result in extremely large couplings.
However, moderate to high oxidation state manganese compounds rarely form metal-
metal bonds. For these types of compounds, a super exchange pathway mediates the
magnetic interaction.

The effective spin on a mixed valent complex results from interactions between
the two fragments. Symmetry requirements constrain electrons on adjacent atoms such
that they can occupy the same space only if their spins are antiparallel. An important

result of this constraint is the phenomenon of antiferromagnetic coupling. Consider an

25

‘2_y2 Z2. PX x2_y2 Z2
02 _L _L
xy yz x2 xy yz
M 3+ 4+
n Mn

Figure 7: Schematic representation of antiferromagnetic coupling in
the fragment Mn3+- 021 Mn“.

26

M-L-M compound in which p electrons from the ligand overlap with d electrons from
the metal to form the M-L bonds.’l Figure 7 shows an Mn3“‘-02‘-Mn4+ fragment which
illustrates this. The formation of a bond between the Mn3+ and the 0 requires partial
transfer of an electron from the px oxygen orbital into the dx2-y2 orbital of the Mn.
The Pauli principle constrains the electron such that this electron must be anti-parallel to
the electrons in the Mn orbitals. The remaining electron from the oxygen is partially
transferred into the dx2-y2 orbital of the W“. The anti-parallel configuration of the
two electrons from the ligand must be maintained during this transfer. In order to
accommodate the Pauli principle, the electrons in the Mn4+ must flip, creating a
coupling of electron spins of the two manganese atoms so that they are anti-parallel to
one another. This results in an effective spin of 1/2 for the system. The strength of the
exchange interaction (J) is usually small compared to bonding energies, but large
compared to the energies associated with the magnetic properties of the electron. The
presence of antiferromagnetic coupling therefore, results in substantial modifications of
quantities associated with electron magnetism, particularly EPR. These modifications
are accounted for in the exchange interaction term of the hamiltonian.

The spin hamiltonian for exchange interaction is given by’0

HEFXBH (512+322) +hAo(Slzllz+Szzlzz) +hJSr 52 (12)
where S] and S; represent the spin operators for electrons 1 and 2 and J is the exchange
integral defined by

J=h'1 <¢A(l)¢5(2) l (e2! r12 ) I ¢B(l)¢A(2) >. (13)
Here 4" is the one-electron orbital on fragment A , 4’3 is the one-electron orbital on
fragment B, and r,2 is the separation of electrons 1 and 2. The corresponding energies
can be obtained from this hamiltonian. For the case I J I >> l A, l , the hyperfine term in

equation 12 can be eliminated. The set of eigenvalues for this zero order hamiltonian are

51,1 = gBH + hJ/4 (14a)
51,0 = + hJ/4 (14b)

27

E1,-1 = -gBH+hJ/4 (14C)
50,0 = -3 hJ/4 (14d)

where the subscripts indicate the states S2 and S2 . These energies are plotted in Figure 8

for J>0.

 

 

 

 

 

ll,1>
+ 1L} |l,0>
ll -l>
LL)
{5111 I0,0>
4
H——P

Figure 8: The effect of exchange coupling on the triplet
and singlet states of an S1=1/2, 82:1/2 system. The degeneracy
at zero field is removed upon the exchange interaction.

28

ESEEM: Flectron spin echo envelope modulation (ESEEM) is a sensitive technique for
measuring the small superhyperfine couplings in systems with large inhomogeneously
broadened EPR lines. Inhomogeneity in an EPR line results from the interaction of the
unpaired electron with varying local fields.50 This variation can be due to anisotropy in
either the g or A terms of the hamiltonian. The electron is subjected to slightly different
local fields such that only a small fraction of the spins are in resonance at a given time.
The EPR spectrum is then composed of a superpostion of these "spin packets” (Figure 9).
Inhomogeneous broadening can also be the result of unresolved hyperfine interactions.

A large number of hyperfine components leads to a decrease in spectral resolution.50
These large interactions can be resolved using double resonance techniques like ENDOR.
A less common cause of inhomogeneous broadening is inhomogeneity in the applied
magnetic field, although this becomes a problem only at large field values. Pulsed EPR
techniques can resolve the structure within a single spin packet by effecting large

rotations of net electron spin magnetization.

Two pulse ESEEM experiment: To explain the theory behind a 2 pulse ESEEM

experiment effectively, a rotating coordinate system will be used. As depicted in Figure
10, this coordinate system is rotating with frequency mo and its axes are denoted with a

prime superscript (e.g. X', Y‘, Z').52 The applied magnetic field, H0, is along the Z axis
and detection is along the Y axis.

Consider two spin packets with characteristic Larrnor frequencies mi and wj (see
inset, Figure 10) which differ from the frequency of the rotating frame. Each of these
spin packets can be considered independently from the others. At thermal equilibrium
there is a net magnetization, Mtotal , along the z axis due to the sum of the individual
magnetic moments of the electrons (see Figure 10a). A short, intense microwave pulse is
then applied which rotates the magnetization 90° such that it lies along the positive Y'

axis (Figure 10b). If the frequency of the rotating frame equals the Larmor frequency of

29

 

 

 

 

 

 

 

 

 

J) J LL

Figure 9: An inhomogeneously broadened line composed of individual
spin packets.

30

Figure 10: A 2 pulse ESEEM experiment. Inset: Frequencies corresponding to two spin
packets with respect to the frequency of the rotating frame. A) Net magnetization along the

Z axis. B) Magnetization vector after the first pulse. C) Precession of individual spin packets
after time t. D) Location of the magnetization vector components after the second pulse. E) The
formation of the spin echo.

31

08
he \\»\\\\\\\\\\\\‘ ‘ e

k“
g \\\\\\\\\ \\\\\~. - -

 

 

time

FiSure l0

 

 

Y!

 

 

32

the magnetization, this vector will remain along the Y' axis. However, if the different
spin packets have different offset frequencies, they will start to dephase. This dephasing
is known as the free induction decay (FID), see Figure 10c. After a time 1, another
microwave pulse is applied. This second pulse rotates the vectors 180° about the X' axis.
The directions and the precessional frequencies of the individual components remain
unchanged (Figure 10d). After time 1- these components refocus along the -Y' axis
(Figure 10e). The building up and decay of this -Y' magnetization is called a spin echo.
The amplitude of this two pulse echo formation is a function of the pulse interval (1') and
the rate of decay of the echo (Tm, or the transverse relaxation time).

Nuclear modulation effect: Interactions that cause mixing of nuclear states such as
anisotropic ligand hyperfine coupling lead to modulations of the spin echo amplitude as a
function of r. The nuclear modulation effect depends on a phenomenon called
”branching of transitions," which in turn depends on all four transitions occurring.
Branching can be described accurately by the diagram in Figure 11. Consider an
energy level diagram for the states S=l/2, I=ll2 (Figure 11, inset). Transitions A and C
are allowed EPR transitions, B and D are previously forbidden transitions. The hyperfine
frequencies are given by «ml and ”N2 Again, the rotating frame (frequency: (no) will
be used in this description. At a time 7 after the 90° pulse, the spin packet corresponding
to transition A (frequency: w A < too) has dephased from the Y' axis of the rotating frame
(Figure 11a). Again, a 180° pulse is applied. This time both transition A and, because of
nuclear state mixing, transition C occur (Figure 1 lb). Transition C has its characteristic
frequency, arc > too. After time 1, the spin packet corresponding to transition A has
rephased along the -Y' axis, however the spin packet corresponding to transition C has
rotated past this point (Figure 11c). The echo amplitude is a measure of the projection of
these vectors along the -Y' axis. The projection is a function of the angle between the

two vectors (0) and can be given by

33

Figure 11: Nuclear modulation effects. Inset: Energy levels corresponding to
an S=l/2, I=l/2 system. Solid lines represent allowed transitions. Dashed lines
represent semi-forbidden transitions. A) Location of the vector corresponding to
transition A at time 1: after the 90° pulse. B) Branching of transitions after the
180° pulse. C) The echo amplitude is given by the projection of the vectors on
the Y' axis. This is a function of the angle 6 between the two.

 

 

 

 

 

 

 

 

m, mi
““2 ”2 l w... 7"\
+12 -12 ‘fi
*' l l
I
I l
| t
| I
12 1'2 1 1 -
B l o D A B C
' (”N2
12 +12 L
A C

A and B allowed transitions
C and D semi-forbidden transitions
1\)

Time-rafter
the90°pulse

 

1
(Q?
o 9

C)

time t
—.’ Yr ‘

Figure 11

35

0 = [wA- we] r (15)
The difference between the two frequencies is given by the hyperfine frequence, air“.
The echo amplitude is then given by the expression

5(21) = C1 + C2 cos [corn (1')] (16)
where C1 and C2 represent transition probability amplitudes.

There are two conditions that preclude nuclear modulation effects, one quantum
mechanical, the other instrumental. As previously mentioned, nuclear state mixing must
occur in order for the branching to take place. The ability to excite all four transitions
coherently is a combination of this quantum mechanical condition and an instrumental
consideration: the microwave pulse must have sufficient energy to excite all four
transitions. In ESEEM a 'hard' pulse, or a pulse containing a wide band of frequencies is
used.

Chapter 3
Experimental

Materials and Methods: The development in 1981 of a PS 11 enriched thylakoid
membrane preparation, free from other electron transfer proteins, yet still functional in
terms of oxygen evolution, facilitated study of the PS II environment.” These soocalled
BBYs allowed direct spectroscopic studies and served as the starting point for the
isolation of smaller membrane free complexes. This preparation relies on the fact that
the thylakoid membranes, with their inherent negative charges, can be "stacked” by using
divalent cations, such as Mg2+. This increased lateral concentration can then be pelleted
out by using selective solubilization methods. This treatment has been shown to remove
95% of the PS I from PS 11. Only the light harvesting complex proteins and the PS
II/OEC remain after completion of this procedure. Oxygen evolution studies performed
on these BBY samples showed an increased oxygen evolution rate per chlorophyll.
Additionally, it was found that the number of chlorophyll molecules per PS 11 had
decreased from 400 ch]! PS 11 in thylakoids to 225-250 chi/PS II in BBYs. This decrease
in the number of chlorophst coupled with the loss of extraneous electron transfer
proteins allowed direct examination of the PS H components, and particularly the Mn
ensemble.

Previously, spectroscopic studies of the Mn complex were performed by using
BBYs. The studies presented here, however, use a procedure that yields PS II
components without the light harvesting complex. These preparations, first characterized
by Ghanotakis et al in 1987, are called RCCs, and contain only the PS II components.“

PS 11 membrane samples from market spinach were isolated by using the method
of Berthold et al53 and diluted to a chlorophyll concentration of 2.5 mg chi/ml by using
buffer A ( see Figure 12 for a list of buffers used in the RCC procedure). These
resuspended membranes were mixed with an equal volume of buffer B and incubated 10

minutes at 4° C in the dark. The non-ionic detergent n-octyl-fi-D-glucopyranoside and

36

Buffer A(pH=6.0)
0.4 M sucrose

50 mM MES
10 mM NaCl

Buffer C (pH=6.0)
1.0 M sucrose

50 mM MES

0.4 M NaCI

5 mM CaClz

37

Buffer B(pH=6.0)

1.0 M sucrose
50 mM MES
0.8 M NaCl
10 mM CaClz

70 mM octyl glucaside

Buffer E(pH=6.0)

0.4 M sucrose
50 mM MES
10 mM NaCl
5 mM CaClz

Figure 12: Buffers used in the RCC procedure.

38

the high salt concentration were included in the buffer to solubilize PS 11 membranes
and to facilitate the removal of the light harvesting complex proteins and the 17 and 23
kDa polypeptides. After the incubation, one part of the solubilized membranes was
mixed with two parts of buffer C. This mixing was followed by centrifugation (90
minutes at 40,000 x g). The pellet, composed of the light harvesting complex and the
solubilized membranes, was discarded. The supematent, containing only PS II ,
components was incubated on ice and an equal volume of 30% w/v polyethylene glycol
(PEG) was added. This PEG precipitation step represents a modification to the original
published procedure.’4 The PEG aggregates the water molecules causing ”clumping" of
the PSII constituents. These are then easily pelleted out using centrifugation. This
solution was centrifuged 30 minutes at 40,000 x g. The resulting pellet was resuspended

in buffer E and subsequently centrifuged 15 minutes at 40,000 x g. The final pellet was
resuspended in buffer A containing 20 mM CaC12 (for the Ca2+ samples) or 20 mM

SrC12 (for the Sr2+ samples). These samples were spun down into 4 mm OD. EPR

tubes, frozen while in the dark and kept at 77 K until EPR experiments could be
performed.

The chelator samples were prepared by procedures similar to those used for the
Ca2+ and Sr2+ samples. Again, market spinach was used to prepare PS II membranes in
the manner of Berthold et al. The same protocol was used except for the PEG
precipitation step. A contamination of the PEG resulted in the appearance of a spurious
radical signal upon EPR analysis. To eliminate this signal the original dialysis step in the
RCC prmdure was followed. Further EPR studies indicated no difference between the
dialde samples and the PEG precipitated samples (data not shown). Following the 90
minute centrifugation, the supernatant was dialyzed for one hour against a buffer
containing 50 mM MES (pH: 6.0) and 10 mM NaCl (Spectrapor No. 6 dialysis tubing
with M,=12,000-14,000 cutoff was used). This dialysis step removes excess NaCl and

sucrose by using a concentration gradient that subsequently causes aggregation of the PS

39

11 components. Centrifugation for 20 minutes at 40,000 x g followed the dialysis step.
The resulting pellet was resuspended to a concentration of 0.25 mg chi/ml by using
buffer A. An equal volume of solution containing 2 M NaCI/buffer A (pH=6.0) was
added to the RCC sample and this solution was incubated 30 minutes at 0° C in room
light. The high salt concentration facilitates removal of the 17 and 23 kDa polypeptides.
Incubation in the light permits S-state turnover and subsequent release of calcium. Tire
chelator ethylene glycol B-bis-(amino ethyl) ether tetraacetic acid (EGTA) was added to
a final concentration of 50 nM and the sample was centrifuged 20 minutes at 40,000 x g.
Buffer A containing 50 uM EGTA was used to resuspend the pellet. Again, the sample
was centrifuged (20 minutes at 40,000 x g). The supernatant was discarded and the pellet
was resuspended in the dark by using a minimal amount of buffer A that contained 10
mM EGTA. The RCC sample was spun down into 4 mm OD EPR tubes and dark
adapted for 30 minutes to ensure that all centers were in $1. The samples were frozen in
the dark and kept at 77 K until EPR studies could be performed.

Throughout both procedures, unless otherwise indicated, precautions were taken
to maintain the temperature below 10° C and to use low light conditions so that maximal

rates of oxygen evolution were retained.

EPR and ESEEM studies: Dark EPR spectra of the Sr2+ and Ca2+ samples at 8 K were
recorded by using a Bruker ER200D spectrometer fitted with an Oxford Instruments
ESR-900 cryostat. A Hewlett Packard 5245L electronic counter with a 5255A frequency
converter plug in and a Bruker ER035M gaussmeter were used to measure microwave

frequency and magnetic field, respectively. A rectangular TE102 cavity was used.

Dark adapted samples (5 minutes) were illuminated at 210 K in an ethanol/solid
C02 mixture by using a 750 W projector. Illumination time was 8 minutes. Following

illumination, the samples were frown in the dark and the light spectra were recorded at 8

K. The spectra from both the dark and the illuminated Ca2+ sample correspond to an

4o

addition of 30 accumulations. The spectra from the Sr2+ sample correspond to an
addition of 60 accumulations. Signal averaging was accomplished by using an IBM PC
fitted with a data analysis program written in house. The spectra presented in the
following section, for both the Ca2+ and the Sr2+ samples, represent the difference
between the illuminated spectrum and the dark adapted spectrum. Subtraction was
carried out by using an IBM clone equipped with a subtraction program written in house.
Spectrometer conditions accompany each figure.

Only the light spectrum of the chelator sample was recorded. The sample was
dark adapted for 5 minutes and then illuminated according to the protocol above. This
spectrum represents the addition of 30 accumulations. Spectrometer conditions
accompany each figure.

ESEEM spectra of the illuminated samples were recorded at 1.8 K by only using
a home-built ESEEM spectrometer previously described.” Data were collected by using
only the two pulse (90-r-180) microwave pulse sequence. The collection of ESE
envelopes was controlled by an Apple McIntosh 11 computer. The Fourier transform of
the ESE envelope was carried out with a modified version of the dead time
reconstruction procedure that was developed by Mirnsft6 For two pulse experiments the
r was varied from approximately 190 as to 4 us for each data set containing 1024 points.
Other experimental conditions accompany each figure.

Chapter 4

Results and Discussion

RCC Characterization: The activity of enzymatic proteins present in biochemical
pathways is often dependent on their location. Catalytic activity in a membrane bound
complex may be decreased or lost upon its release from that membrane. This loss of
activity may be concomitant with structural and, or, electronic rearrangements.
Therefore, the presence or absence of a membrane bound complex can have direct effects
on the function of that enzyme. This is seen in the transition from BBYs to RCCs.
BBYs are composed of the membrane bound components and light harvesting proteins
of PS 11. In preparing RCCs from BBY membranes, the integrity of the membrane as
well as the LHC is lost upon treatment of the samples with buffer containing the
detergent, octyl-glucoside and a high concentration of NaCl. The elimination of these
components directly affects subsequent biochemical manipulations. First, the calcium
depletion procedure for BBYs requires extensive incubation of the membranes in the
light. This incubation has to occur in the presence of chelators and in a high (2 M) salt
buffer. This was found to be unnecessary for the RCC samples. The solubilization of
the LHC and the removal of the 17 and 23 kDa polypeptides during the detergent
washing facilitated spontaneous calcium release. Exclusion of calcium from subsequent

buffers ensures maximum release of the cation. Reconstitution is easily accomplished
with addition of either 20 mM CaC12 or, for strontium substituted samples, 20 mM

SrClz.

Additionally, the dark adaptation needed for BBY samples is unnecessary for the
RCCs. The BBY procedure includes a step whereby the membranes are incubated for 30
minutes at 0° C prior to storage.“ RCC samples were centrifuged directly into the EPR
tubes at 0° C in the dark without an additional dark adaption period. The centrifugation

resulted in a highly concentrated sample. w hich proved to be ideal for spectroscopic

4|

42

analysis. The EPR spectra of the RCCs not subjected to the dark prior to storage
appeared similar to those obtained from BBYs. These spectra will be discussed further in
an upcoming section.

As a direct consequence of the increased concentration of PS II that results from
disruption of the membrane in BBYs and elimination of the light harvesting chlorophyll
protein complexes, the absorption cross section of the RCC samples decreased. This
effect was manifested practically as RCC samples were found to need additional
illumination time (8 minutes instead of 5 minutes) to obtain maximum signal intensity.“

The use of the RCC preparation provided the increased concentrations of PS II
needed for spectroscopic analysis. However, as a result of these increases, redefinition of
the biochemical preparation procedure was needed.

The Mn complex and other'components of the PS II/OEC had previously been
studied by using samples that had been isolated according to the BBY procedure. These
samples, containing both PS 11 components and the LHC, present some spectroscopic
difficulties. The intensity of the multiline signal depends directly on the concentration
of PS II, and specifically, the number of spins present in the sample. It is difficult to
prepare BBYs at a concentration sufficient to generate the multiline spectrum without
extensive signal averaging. The development and refinement of the RCC procedure has
alleviated the concentration problems previously encountered, although PS II
concentration is still low as an absolute basis in RCCs. Because these samples contain
only PS II constituents, higher concentrations are attainable and, theoretically, this would
indicate an increase in multiline signal intensity. However, complete spectrosc0pic
characterization of the multiline signal from RCCs needs to be done before comparisons
between the two procedures can be made. Additionally, the effect on the multiline of
signal in RCCs of chelators, calcium depletion and reconstitution, or strontium
substitution was not known. Spectroscopic studies of the multiline and modified

multiline signals from RCC preparations are presented here. The similarities of these

43

spectra to those obtained from BBYs show that RCC preparations can be used in place of

BBYs without the appearance of any spectroscopic anomalies.

EPR characterization: The difference spectrum from the 82 state of the Ca2+

reconstituted RCC sample is presented in Figure 13a. This spectrum is the result of the
difference between the spectrum recorded after illumination and the spectrum recorded
after dark adaptation. This technique is used to subtract out signals which appear in both
spectra and allows the multiline signal to be viewed in its entirety. The signal is centered
at approximately g=2 and consists of 16-18 hyperfine lines split by an average of 92
gauss. The average hyperfine coupling was determined by direct measurement of the
splittings between peaks off of the spectrum. This spectrum is similar to that obtained by
Boussac and Rutherford for BBY samples which showed 16 lines and hyperfine
couplings of 90 gauss.“~‘° Figure 13b shows the difference spectrum for the Sr2+
substituted samples. The signal is centered at approximately g=2 and consists of 16- 18
hyperfine lines split by 71 gauss. Reconstitution with Sr2+ instead of Ca2+ results in
losses, shifts, splittings and redistributions of the amplitude of the different lines. These
modifications are similar to those seen by Boussac and Rutherford in Sr2+ reconstituted
BBYs.“-‘° The hyperfine splitting in this case was found to be approximately 71 gauss.
The overall shape of the signal, however, as well as its breadth was comparable to that
seen in BBYs. These similarities indicate that the RCC samples can be substituted for
the BBY samples in spectroscopic studies.

NaCl (1.2 M)/EGTA (50 nM) washing of PSII components results in the
removal of Ca2+ and the inhibition of oxygen evolution.“ This treatment precludes the
appearance of the chelator modified multiline signal, which has only been detected by
using high concentrations of chelator (10 mM) in the light. It is generated after

illumination at 200 K and is attributed to a perturbation of the 82 state. Characterization

of this modified multiline signal was carried out by Boussac et al in BBY samples!“S

MW

2500 3000 3500 4000

 

Figure 13: The light minus dark multiline spectrum of a) Ca2+ reconstituted RCCs and

b) Sr2+ reconstituted RCCs. The spectra are the result of 30 (a) and 60 (b) signal
averaged scans. Spectrometer conditions are as follows: scan range, 2000 G; center field,

3400 G; power, 10 dB; modulation amplitude, 20 Gpp; gain, 1.6 x 106; temperature, 8 K.

45

These researchers found a multiline signal with peak amplitudes approximately half of
those found in the normal multiline signal. The modified multiline from RCC has not
been characterized. Figure 14a shows the multiline signal from Ca2+ reconstituted RCC
samples after 8 minutes of illumination at 200 K. The corresponding chelator modified
multiline is presented in Figure 14b. The decrease in amplitude from the calcium to the
chelator is obvious. Both spectra include the large feature at g=2 attributed to the YD+
radical. This feature was subtracted out in previous spectra. The modifications are
thought to arise from direct binding of the carboxylic acid substituent groups from the
chelator to the Mn of the OEC. In the presence of calcium this interaction cannot occur,
which has led to the suggestion that calcium plays a role in controlling ligand binding to
Mn.“5

Other spectroscopic features seen previously in BBY samples have been observed

in EPR spectra from RCC preparations during the study described in this thesis. These
include the EPR signal from g=2.0046 YD+ (Signal 11), from the iron quinone(g=l.9)

and from cytochrome b559 (gz=3.0, gy= 2.1, data not shown).57 These features
correspond in amplitude and fine structure to those seen from BBY samples.

The EPR results from the RCC samples indicate that they are spectroscopically
equivalent to the BBY samples. All spectral features in the Ca2+ reconstituted, Sr2+
substituted and chelator modified specua are present. These RCC samples are free of
spurious proteins and, therefore, are 2.5 times (75 uM as compared to 30uM in BBY) as

concentrated as the BBYs.

46

Figure 14: Multiline spectra from a) Ca2+ reconstituted RCCs and b) EGTA treated
RCCs. Spectrometer conditions are as in Figure 13. The spectra are the result of 25 (a)
and 50 (b) signal averaged scans following illumination at 200 K.

EPR iNTENSlTY

47

 

 

 

 

 

 

 

 

 

 

 

a) Ca2+
b) EGTA *
l
2400 zone 3po Jenna

GAUSS

 

48

ESEEM Studies of the Modified Multiline

Pulsed EPR studies of the multiline signal have focused on the signal from either
spinach chloroplasts or cyanobacteria in the presence of calcium.‘“9 The effects of
strontium substitution or chelator modification have not been explored. The
modifications seen in the hyperfine structure of the continuous wave EPR experiments
indicates a reorganization of spin densities and, or, exchange interactions in both the
strontium and chelator samples. Changes in spin density may affect the interactions of
the paramagnetic center with its environment. Both direct ligations and hydrogen bonds
are affected by these changes. Britt and co-workers by using ESEEM techniques have
recently identified a 4.5 MHz frequency that is attributed to a superhyperfine interaction
of nitrogen with the Mn of the OEC.49 This nitrogen has been proposed to be the
nitrogen moeity of histidine.‘9 The isotropic hyperfine coupling of a directly ligating
nitrogen to the Mn of a dimanganese (III,IV) di-u-oxo bridged bipyridine model
compound was determined to be less than 1 gauss." This coupling is too small to be seen
with conventional continuous wave EPR. It is expected that the coupling of a nitrogen to
the Mn ensemble of the OEC would resemble the coupling in the dimanganese
compound. Researchers have been unsuccessful in resolving the nitrogen hyperfine and
quadrupole parameters associated with this observed frequency. Analysis of these
parameters can lead to identification and characterization of the interacting species.
Additional double resonance and multi-frequency experiments will need to be performed
to identify these parameters (see Future Work section, Chapter 5).

Both strontium and EGTA treatments result in a modification of the continuous
wave multiline spectrum, as described above. Interactions between the paramagnet and
the strontium or the carboxylate groups of the chelator seem to be the source of these
changes. Small structural changes can also affect the hyperfine interaction through the

Fermi contact term of the hamiltonian and the magnitude of the exchange interaction. In

49

this study, the effect of strontium and chelator on the proposed nitrogen coupling will be
addressed by using ESEEM on RCC samples.

Figure 15a is the time domain and Figure 15b the Fourier transform of a 2 pulse
ESEEM experiment performed on the Ca2+ reconstituted sample. Three features are
apparent in the Fl‘ spectrum. Two of these, the peaks at 14.553 and 29.249 MHz, can be
attributed to interactions with protons in the immediate environment of the paramagnet
(”IF 14.48 at 3400 G). The third feature, a peak at 4.110 MHz is similar in intensity and
frequency to peaks seen in both BBYs and cyanobacteria experiments performed by
DeRose and co-workers and Britt and co-workers.‘“9

The 4 MHz feature was initially observed in ESEEM experiments that established
ammonia binding to the Mn of the OEC. The source of the signal at that time was
unknown. It was observed in BBYs with and without ammonia present. It was
postulated that the signal arose from a component within the OEC, possibly a manganese
ligand. The identity of this ligand was not confirmed, although nitrogen was suggested.
It was not until further ESEEM experiments were performed in 1990 by DeRose and co-
workers that a source for the signal was determined!9 In these experiments
cyanobacteria were grown on a medium in which the only source of nitrogen was from
added K15NO3. A control experiment was run whereby only K14NO3 was used. The 4
MHz peak was seen only in the control experiment. This led to the conclusion that it
must arise from a nitrogen ligand to the Mn of the OEC.‘9 This origin of this nitrogen is
controversial. The proposed Mn binding regions on the D1 and D2 polypeptides have
several nitrogen containing amino acid residues that are conserved among cyanobacteria
and higher plants.’9 Of these amino acids (lysine, arginine and histidine) the only one
that is known to coordinate to manganese containing proteins is histidine.49 This
suggests histidine as the source of the ligating nitrogen.

The substitution of strontium for calcium in the RCC samples led to a modified

multiline signal. These modifications were observed in both the number of peaks and

50

 

 

 

 

 

A) rrr—fiTIIIIIleIITIIIII'+

P
r. .l
_- .J

G)

U

:3 f J

4.)

.H .-

.. I i

O.

5 II

o .-

.C

U h

ti)
LJIIILIJLLJJillgl r_r

 

 

J
O 1 2 3 4

tau (usec)

3)

 

V
J

 

INTENSITY (0.0.)

 

 

 

 

mm 0.41)

Figure 15: Two pulse ESEEM (a) and Fourier cosine transfer (b) for the Ca2+ reconstituted
sample. Measurement conditions for the data shown are: microwave frequency, 8.893 GHz',
magnetic field strength, 3400 G; pulse power. 25 W; pulse width, 15 ns FWHH', tau value,
140 ns; sample temperature, 1.8 K; repetition rate. 60 Hz.

51

their splittings and intensities. Modifications of this type are thought to arise from
conformational changes that result in redistribution of electron spin density. If these
structural changes are of sufficient magnitude they will perturb the interactions between
the paramagnet and ligating magnetic nuclei. This perturbation should be apparent in the
ESEEM spectrum. The 2 pulse time domain (Figure 16a) and Fourier transform (Figure
16b) spectra from the Sr2+ RCC samples are identical in appearance to the spectra from
the Ca2+ samples. The peak at 4.106 MHz correponds to the peak attributed to nitrogen
seen in the Ca2+ samples. The ESEEM spectra from the EGTA treated samples are
presented in Figure 17. The Fourier transform (Figure 17b) of the time domain data
exhibits a peak in the 4 MHz range. These data are very similar to those seen for the
calcium and strontium samples. This indicates that the electronic interaction of the
nitrogen with the Mn has been unaffected by the substitutions. There are several possible
explanations for this. First, the substitution of strontium or EGTA could affect only the
electronic interactions within the Mn cluster without affecting the interaction of the
ligand. This structural rearrangement causes changes in the Mn hyperfine interactions
that are apparent in the continuous wave EPR spectrum but do not affect the nitrogen
couplings seen in the ESEEM spectrum. A second possible explanation is that
substitution of strontium induces a conformational change that does not affect the site of
nitrogen ligation. Therefore the interactions of the nitrogen and the manganese are not
affected. A final explanation for these similarities is that there is a structural change
upon strontium or chelator substutition that results in the modified multiline signal,
however there is a simultaneous change in either the bond angle or orientation of the
nitrogen that offsets this conformational change. The exact explanation for this
phenomenon requires additional spectroscopic techniques that will be discussed in the
following section.
The use of a refined biochemical preparation which increases the concentration of

PS II present in a sample has facilitated study of the Mn multiline EPR spectrum

52

 

 

 

 

 

 

 

 

 

 

 

 

[—firtlrrr—fijlrrrlr
r— Ii
8
3 r
H
v" )-
H
or
e -
t0
0 P
r;
U P
0.)
i-
t
O
tau (usec)
B) W ,
=i
s
it .
i
E
H
h
bani .1 La
0 to 20 so no

Figure 16: Two pulse ESEEM (a) and Fourier cosine transfer (b) for the Sr2+ reconstituted
samples. Spectrometer conditions were as in Figure 15.

53

 

A) W I r r i r r r r 1 r I r I 1 l r r r I ‘

 

 

 

 

 

 

 

 

 

 

 

Q)
U h-
3
U
-H >-
H
O.
E r
to
o P
1:
UL
OJ
F
i-
Fl
0
r tau (usecl
B) r r'Yr' Y'r' '
r 1
F 1
=1 .
3 P
>' 4
._ P
N
”z"
w b
E
H
F
, 4
A 1 1L ALIA
o 10 20 so 40

FREWY 0“!)

Figure 17: Two pulse ESEEM (a) and Fourier cosine transfer (b) for the EGTA treated
sample. Spectrometer conditions were as in Figure 15.

54

associated with the 82 state of the OEC. This spectrum. which is thought to arise from

exchange coupling of two, three or four mixed valence Mn atoms. has been studied by
using magnetic resonance techniques. EPR studies discussed in this thesis have shown
that the RCC samples are spectroscopically equivalent to their predessors. The increased
concentration achieved with the RCCs is essential for EPR analysis. Slight modifications
of the published RCC procedure were made as a result of this increased concentration. It
has been shown previously that substitution of Sr2+ or chelator into calcium depleted
BBY samples causes modifications in the hyperfine splitting seen in the multiline
spectrum. These differences are thought to arise from perturbations of the Mn exchange
coupling. These differences were present in the multiline spectra from the RCC samples
presented in this thesis. ESEEM studies of the RCC samples were also performed. A 4
MHz peak. thought to arise from coupling to a ligated nitrogen, is present in all spectra
recorded. This indicates a conformational change that does not affect the interactions of
the Mn with its environment. Additional magnetic resonance studies will be performed

to determine the magnitude of these interactions.

Chapter 5
Future Work

Quantitative analysis of the proposed nitrogen ligations to the Mn cluster of the
OEC has not been possible by using continuous wave and spin echo EPR techniques.
Neither the hyperfine coupling parameters nor the nuclear quadrupole parameters have
been determined for this interaction. These parameters can be used to characterize the
interacting species more completely. From this information, additional knowledge can
be gained about the structure of the OEC. To resolve the peaks corresponding to these
types of transitions other techniques must be employed. Pursuit of these techniques is the

topic of this chapter.

ESE detected ENDOR: The development of pulsed-EPR schemes that use electron
spin echoes to determine ENDOR frequencies by Mims‘° and Davies‘51 has allowed
further study of systems in which small hyperfine interactions are poorly resolved using
conventional ESEEM techniques. The theory of ESE detected ENDOR is based upon
the double resonance technique developed for pulsed NMR spectroscopy. The pulse
sequence for the Davies ESE-ENDOR is given in Figure 18. In this experiment a 1800
pulse of microwave radiation is applied to the demon spins followed by a radio
frequency pulse of duration 1‘. This is followed by the conventional Hahn pulse
sequence (900-1-1800). The spin echo appears at time 1 after the final 180° pulse. The
effect on the spectrum can be understood with the aid of Figure 18 for an S=1/2, 1:1/2
system. At thermal equilibrium the population of the lower spin states is slightly higher
than the population of the upper states (this is shown with the relative lengths of the bars
in Figure 18). Assuming a microwave pulse frequency that is resonant with only one
EPR transition, the initial 180° pulse inverts the populations of the levels corresponding
to that EPR transition. If the rf pulse is resonant. with either NMR transition. then the

population difference between these two levels will be equalized and the electron spin

55

56

 

 

 

 

 

 

 

 

 

 

T 4+— ‘U+"J-—D.
i I i . l i . g
macaw: ' I 5‘ -! /\
.-- ___. . ‘___/
chze:
Pup, \\(l“\\n\\/\
j‘J J J \l" J
:4) l ! I
‘3)! I ' i l: l
32>l J l j

 

 

 

 

 

lt>[ ] E
M Al,
or ‘ V7:

Figure 18: Davies ESE-ENDOR pulse sequence (top). Energy level diagram for a
coupled S=l/2. I=ll2 system. The diagram on the left shows the level populations at
thermal equilibrium while that on the right shows the level populations after a selective
[80' pulse resonant with the l-3 transition has been applied.

57

echo amplititude will disappear. An ENDOR spectrum can be generated by plotting the
ESE amplitude as a function of rf pulse frequency.62 Recent studies have shown that the
14N couplings in the protein Stellacyanin can be resolved in the 1-60 MHz region. The
ESE-ENDOR spectrum from this protein compares well with the continuous wave
ENDOR spectrum}53 The advantage in using the ESE-ENDOR method over
conventional ENDOR rests in the data acquisition time period. The ESE-ENDOR
spectrum was collected in 12 minutes while the ENDOR spectrum took 1.5 hours to
acquire. The application of this technique to the PS II/OEC system should provide a
more sensitive and systematic means by which to measure hyperfine interactions. There
are additional advantages to using this method. It provides both increased sensitivity and

a systematic means of collecting data.

Characterization of the S 3 EPR signal: Boussac et al“ reported an EPR signal
attributed to the S3 state of the OEC. This signal. found in calcium depleted BBYs. is

centered around g=2 and has a width of approximately 164 gauss. It is generated by
illumination of the samples at 273 K. followed by freezing while the sample is still under
illumination. It was later shown that EDTA, EGTA or citrate were required in order to
observe this signal. The source of this signal is still controversial. It has been suggested
that this signal results from the interaction of the manganese ensemble with a nearby
oxidized histidine residue.“ It is unknown whether this oxidation occurs as a part of the
normal S state transitions or whether it is a direct consequence of the calcium depletion.
Recently, however, Hallahan et al“ proposed that the S3 state signal arose from Yz, a
tyrosine radical. This was based on the observation that. at 50 K. the signal associated
with the tyrosine radical increased relative to that of the S3 signal. This was immediately
challenged by Boussac and Rutherford67 who cite power saturation artifacts as the source
of this Yz- signal increase. The assignment of the split 83 signal to a histidine radical

remains unproven and controversial.

58

The'S3 signal has recently been generated in this laboratory. however complete

spectroscopic analysis of the signal has not been performed. ESEEM experiments can
provide information about the small hyperfine couplings that may be masked by the
broad featureless signal. The signal, which has been determined to be centered at g=2
with a linewidth of 148 gauss. appears to be inhomogeneously broadened. Spectroscopic
studies in which both ENDOR and ESEEM are used will be undertaken to further resolve
this signal.

PS II Mutarm: In conjuction with Professor Lee McIntosh's lab in the Plant Research
Laboratory the characterization of cyanobacteria lacking PS I will be performed . Site
directed mutagenesis techniques have been employed to remove the gene necessary to
assemble the PS 1 components of photosynthesis. By eliminating this reaction center,
purification of the PS II system has been simplified resulting in preparations that have
increased oxygen evolving activity. Preliminary EPR data shows a multiline spectrum
similar to that obtained from BBYs. ESEEM experiments have been performed on this
PS I minus mutant and the 4 MHz peak associated with nitrogen has been observed. An
additional lower frequency peak has also been seen. although no assignment has been
made thus far.

Further mutations of this PS I minus sample are planned. The D1 polypeptide has
been identified as a potential site of Mn ligation by several groups.12 Mutations in this
polypeptide. and in particular of Asp-170. have resulted in loss of photoautotrophic
growth. This loss occurred when Asp-170 was replace with 14 out of 16 residues or
when Asp-242 was replaced with Gln, Ala or Val.“ Evidence that these or other amino
acid residues ligate either the Mn ensemble or the calcium co-factor must come from
spectroscopic analysis. Spectroscopic data. however. has only been reported for
mutations constructed at Asp-170 of the D1 polypeptide. Complete spectroscopic

characterization of the mutants obtained from Professor McIntosh's lab is planned.

Bibliography

10

ll

12

13

14

15

Bibliography

Hill. R. and Bendall F. (1960) Nature 186, 136.

Staehelin LA. (1986) in Encyclopedia of Plant Physiology, New Serices, Volume
19, Photosynthesis III,(L.A. Staehlin and CJ. Amtzen,eds.) Springer-Verlag.
Berlin. 476.

Babcock. G.T., Barry, B.A., Debus. R.J.. Hoganson. C.W., Atamian. M..
McIntosh.L.. Sithole. I., and Yocum, CY. (1989) Biochemistry, 28, 9557.

Ghanotakis, D.F.. Topper. IN. and Yocum, CF. (1984) Biochim Biophys Acta,
767. 524.

Enami. 1., Kaneko. M.. Kitamua. N.. Koike. H.. Sonoike. K., Inoue. Y.. and
Katoh. S. (1991) Biochim Biophys Acta, 1060, 224.

Hoganson, C.W., Babcock. G.T., and Yocurn. CF. (1989) Photosynth. Res, 22,
285.

Nakatani, H.Y. (1984) Biochim. Biophy. Res. Commun, 120, 299.
Nanba. O. and Satoh, K. (1987) Proc. Natl. Acad. Sci. U.S.A., 84, 109.

Diesenhofer. 1., Epp. 0., Miki, K., I-Iuber. R.. and Michel. II. (1985) Nature,
318, 618.

Sayre, R.T.. Andersson. B.. and Bogorad. L. (1986) Cell, 47. 601.

Debus. R.J.. Barry, B.A., Sithole. I.. and Babcock. G.T. (1988) Biochemistry,
27. 9071.

Rutherford. AW. (1989) Trends. Biochem. Sci., 14(6), 227.

Ghanotakis. D.F.. Demetriou, D.M., and Yocum, CF. (1987) BiochintBiophys.
Acta. 891, 15.

Bowlby, N.R., Ghanotakis, D.F.. Yocum, C.F.. Petersen. J.. and Babcock. G.T.,
(1988) Light Energy Transduction in Photosyntheisi: Higher Plant and Bacterial
Models (Stevens. S.E. Jr., and Bryant. D.A., Eds.) p. 215, The American Society
of Plant Physiology Publishers, Baltimore. MD.

Babcock. G.T., Widget, W.R., Cramer, W.A., Oertling, W.A., and Metz, LG.
(1985) Biochemistry. 24, 3638.

59

16

17

18

19

20

21

22

23
24

25

26

27
28
29

30

31

32
33

34

60

Nuijs. AM. Van Gorkom, I-I.J.. Flijter, 1.1.. and Duysens L.N.M. (1986)
Biochim. Biophys. Acta. 848, 167.

CrofIS. AR. and Wraight. CA. (1983) Biochim. Biophys. Acta,726, 149.

Witt. H.T..Schlodder. E.. Brettel. K., and Saygin, O. (1986) Photosynth. Res,
10. 453.

Barry, B.A. and Babcock. G.T. (1987) Proc. Natl. Acad. Sci. U.S.A., 84, 7099.

Debus, R1, Barry. B.A., Babcock. G.T., and McIntosh. L. (1988) Proc. Natl.
Acad. Sci. (1.5.4., 85, 427. ‘

Barry, B.A. and Babcock. G.T. (1988) Chem Scr 28A, 117.

Klimov, V.V., Allakhverdiev, S.I.. Demeter. S.. Krasnovskii, A.A. (1980) Dokl.
Akad. Nauk. SSSR, 249. 227.

Joliot. P.. Barbieri. 6.. and Chabaud. R. (1969) Photochem. Photobiol., 10, 309.
Kok. B., Forbush, 8., and McGloin, M. (1970) Photochem Photobiol., 11,457.

Babcock. GT. (1973) PhD thesis. Department of Chemistry. University of
California. Berkeley. Lawrence Berkeley Laboratory Report LBL-2172.

Babcock. G.T. (1987) The Photosynthetic Oxygen Evolving Process. In: New
Comprehensive Biochemistry (Ed.: Amesz. J.) Elsevier, Amsterdam. 125.

Pirson. A. (1937) 2. Bot. 31. 193.
Cheniae. G.M. and Martin. LP. (1966) Brookhaven Syrup. Biol. 19, 406.

Yocum, C.F., Yerkes. C.T.. Blankenship, R.E.. Sharp, RR. and Babcock. GT.
(1981) Proc. Natl. Acad. Sci. USA. 78. 7507.

Kirby, J.A., Robertson. AS. Smith. J.P.. Thompson. A.C., Cooper. S.R..and
Klein, M., (1981) .I. Am. Chem. Soc., 103, 5529.

Yachandra. v.1c, Guiles. 9.1)., McDermott. AE., Britt. rum. Dexheimer, 3.1...
Sauer. 1c. and Klein M. (1986) Biochim. Biophys. Acta, 850, 324.

Penner-I-Iahn. J. (1990) J. Am. Chem. Soc..112, 2549.
Dismukes. G.C. and Siderer. Y. (1980) FEBS Lett., 121. 78.
Dismukes. G.C. and Siderer, Y. (1980) Proc. Natl. Acad. Sci. U.S.A., 78, 274.

35

36
37
38

39

41

42

43

45

47

4s

49

50

51

52

61

Brudvig, G.W.. Casey, J.L.. and Sauer. K. (1983) Biochim. Biophys. Acta, 723,
366.

Forbush. B., Kok. B.. and McGloin. M. (1971) Photochem. Photobiol., II, 457.
Inoue. Y. and Shibata. K. (1978) FEBS Lett, 85, 193.

Larson. E.. Haddy, A.. Kirk, M.L.. Sands. R.H.. Hatfield, W.E.. and Pecoraro. V.
(1991) J. Am. Chem. Soc.,112. 2549.

Bhula. R., Gainsford. 6.1.. and Weatherburn, DC. (1988) J. Am. Chem. Soc.,
110. 7550.

Dismukes. G.C.. Ferris, K., and Watnick. P. (1982) Photochem. Photobiol.,3,243.

Bonvoisin, J., Blondin. G., Girerd. 1.1.. and Zimmerman. J.L.(l992) Biophys. J.,
61. 1076.

England. RR. and Evans, EH; (1981) FEBS Lett., 143, 175.
Ghanotakis. D.F., Babcock. G.T., and Yocum, CF. (1984) FEBS Lett, 167, 127.
Boussac. A. and Rutherford. AW. (1990) FEBS Lett., 277. 69.

GhanotaKiS. D.F., Babcock. G.T., and Yocum, CF. (1984) .Biochim. Biophys.
Acta,,765, 388.

Boussac. A., Zimmerman. J.L. and Rutherford. A.W.(l990) FEBS Lett., 277, 69.
Andreasson, LE. (9189) Biochim. Biophys. Acta .973, 465.

aria. 3.1)., Simmerman, 1.1... Sauer,1(., and Klein. M. (1989)]. Am. Chem. Soc.,
111. 3522.

DeRose. V.J.. Yachandra. V.K., McDermott. A.E., Britt. R.D., Sauer. K., and
Klein. M. (1990) Biochemistry, 30, 1335.

Wertz. LE. and Bolton. LR. (1986) Electron Spin Resonance: Elementary Theory
and Applications. Chapman and Hall. New York.

Palmer, G. (1980) Electron Paramagnetic Resonance in: Methods for
Determining Metal Ion Environments in Proteins (Eds: Darnell, D.W. and
Wilkins. R.G.) Elsevier, North-Holland. 153.

Schweiger, A. (1991) Angew. Chem. int. Ed Engl., 30, 265.

53

54

55
56
57
58

59

61
62
63

65

66

67

68

62

Berthold, D., Babc06k. G.T. and Yocum, CF. (1981) FEES Lett., 134, 231.

Bowlby. N.R.(1989) PhD thesis.Department of Chemistry, University of
Michigan , Ann Arbor, MI

McCracken, J.L.
Mirns, W.B. (1984) J. Magn. Res, 59, 291.
Miller. A. F. and Brudvig, G.W. (1991)Eiochim. Biophys. Acta, 1056, l.

Britt, RD. (1988) PhD thesis. Depamnent of Physics,University of California
Berkeley, Lawrence Berkeley Laboratory.

Gingrich, J.C.. Gasparich, G.E., Sauer, K. and Bryant, DA. (1990) Photosyn.
Res.,24, 137.

Mims, W.B. (1965)Proc. Roy. Soc. A, 283, 452.
Davies. ER. (1974) Phys. Lett., 47A, 1.
Mehring, M., Hofer, P., 611199. A. (1986) Phys. Rev. 44.33. 3523.

Roberts, J.E., Brown, T.G., Hoffman, B. M.. and Peisach, J. (1980) J. Am. Chem.
Soc., 102. 825.

Boussac, A. and Rutherford, AW. (1988) Biochemistry, 27, 3476.

Boussac. A., Zimmerman, J.L., Rutherford, AW. and Lavergne, l. (1990), J. Am.
Chem. Soc,347.303..

Hallahan. B.J., Nugent. J.H.A.. Warden, J.T.. and Evans. S.M.C.W. (1992)

Biochemistry, 31, 4562.

Boussac, A. and Rutherford, A.W. (1992) Biochemistry, in press.

Diner, B.A., Nixon, P.J.. and Farchans, J.W. (1991) Current Opinion In Struct.
Eiol.,1, 546.

 

 

Armenian
nuts uvfltuam
xuvuan

-——-u——

 

 

"I7'111111111111111ls