'IIIIIII'III'EL

 

8

HI

ERS

Ill

'Illlflllll

3 1293 01388 9807

s lllllllllll‘llll

 

 

I'.

This is to certify that the

dissertation entitled

PHOTOREDUCTION OF INACTIVE DNA PHOTOLYASE‘
CORRELATED TRIPLET-DOUBLET RADICAL PAIR

POLARIZATION OF TRYPTOPHAN IN THE ENZYME
presentedby

Craig Michael Essenmacher

has been accepted towards fulfillment
of the requirements for

 

 

Ph. D. degreein Chemistry

it ”’73 775

MSUL': an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

 

 

PHOTOREDUCTION OF INACTIVE DNA PHOTOLYASE: CORRELATED
TRIPLET-DOUBLET RADICAL PAIR POLARIZATION OF TRYPTOPHAN IN THE
ENZYME
By

Craig Michael Essenmacher

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1995

ABSTRACT

PHOTOREDUCTION IN DNA PHOTOLYASE: CORRELATED TRIPLET-
DOUBLET RADICAL PAIR POLARIZATION OF TRYPTOPHAN IN THE ENZYME

By

Craig Michael Essenmacher

UV radiation has many harmful effects on DNA, the most significant of which is
the formation of pyrimidine dimers between adjacent thymines that occur in the nucleotide
sequence. All organisms have mechanisms to repair these dimers. One of the most
interesting is that carried out by DNA photolyase, which repairs pyrimidine dimers
through a visible light-initiated photoreaction.

During isolation of the enzyme, one electron oxidation of the flavin produces the
semiquinone form of the chromophore. The enzyme can be photoreactivated to its active
form by electron transfer from an amino acid side chain to the flavin semiquinone. The
electron transfer reaction generates a protein bound radical that can be reduced by an
exogenous electron donor to form the active enzyme.

Time resolved electron paramagnetic resonance was used to detect the transient
radical and to study the photoreactivating reaction. In agreement with previous optical
studies, we have identified the donor to the flavin semiquinone as tryptophan-306 by using
Specific deuteration, l’N labeling, and site directed mutagenesis. Previous studies have
indicated that the flavin semiquinone is reduced by hydrogen atom transfer from trp-306.

Utilizing spin density calculations, we have identified the transient tryptophan radical as a

cation radical, from which we conclude that electron transfer occurs in the
photoreactivating reaction.

We have also found that the transient tryptophan radical is formed initially in a spin
polarized state. Spin polarization results from a non-Boltzman distribution of the electron
spin energy levels. The transient tryptophan radical in photolyase has certain transitions in
emission, which is hallmark of a spin polarized radical. Spin polarization results from
interactions between two radicals in a radical pair. The polarization in photolyase is
distinct from other radical pairs because the transitions alternate between emission and
absorption, and arises from a doublet-triplet radical pair. This pattern has been observed
for doublet-doublet radical pairs and is explained by the presence of an exchange
interaction between the radicals during the observation time by EPR. We propose that
spin polarization in photolyase reflects the same phenomenom but the interaction arises

from a triplet-doublet radical pair.

ACKNOWLEDGEMENTS

I greatly appreciate the financial and emotional support my parents and family gave
to me during my years as a graduate student at Michigan State University. This work
would not have been possible without their continuing support.

I would also like to thank Dr. Aziz Sancar and Dr. Sang-Tae Kim for their
contributions to this project. All of the enzyme used in this study was provided by them.
They provided enzyme to me whenever it was needed, and their hard work made my
research much easier to accomplish. They also had very helpful discussions with me about
the intricacies of the enzyme, and many of the projects could not have been done without
their help.

Dr. Curt Hoganson, Dr. Kurt Wamecke, Dr. Matt Espe, and Dr. Mike Atamian
were also very helpful to me during my graduate career. I would like to thank Dr. Kurt
Wamecke for helpful discussions about electron transfer, and Dr. Matt Espe for his help
with EPR theory and computers. I also appreciate the discussions I had with Dr. Curt
Hoganson concerning photosynthesis and electronics. Dr. Curt Hoganson was the person
responsible for setting up the time-resolved EPR instrumentation, and without his help the
experiments here could not have been performed. His help to me during graduate school
made my work much easier to perform, and it is greatly appreciated. I want to thank Dr.
Mike Atamian training me to use EPR instrumentation, and for his preliminary work on

this project.

iv

Finally, I would like to thank my advisor Dr. G.T. Babcock for his help to me
during my graduate school career. Many times during graduate school I was lost and
confined, but I was allowed to struggle until I was able to solve the problem myself.
Knowing when to help me and when to let me solve the problem myself gave me the
ability to think independently without always relying on others. For this I am deeply
grateful. I would also like to thank him for making me computer lab manager during my
time in his lab. At the time I was told I was the new computer manager, I absolutely did
not want the responsibility because I completely computer illiterate. By the time I left the
lab, I was highly skilled in computer software, programming, and hardware. If I had not
been forced to do this job, I would not have gained the valuable knowledge I have
acquired through this experience. I also appreciate the time that Dr. Babcock spent with
me and his patience to improve my writing skills. Through his help, I have become a
much more proficient writer. Finally, I would like to thank him for making my years at

Michigan State enjoyable and rewarding.

TABLE OF CONTENTS

List of Tables .............................................................................................................. viii
List of Figures ................................................................................................................ ix
1. Chapter 1 Introduction .............................................................................. 1
A Photolyase .............................................................................. 1
B. Electron Spin Polarization ................................................................. 16
C. Correlated Radical Pair Polarization ..................................................... 20
D. References ............................................................................ 29
II. Chapter 2 The Identification of a Transient Tryptophan Radical
in the Photoreduction Reaction of DNA Photolyase
by Time-Resolved Electron Paramagnetic Resonance ................... 31
A. Introduction ............................................................................ 31
B. Materials and Methods ............................................................................ 34
C. Results ............................................................................ 38
D. Discussion ............................................................................ 53
E. References ............................................................................ 62
111. Chapter 3 The trp-306 radical in photolyase is a cation radical
suggesting that photoreduction occurs through electron
transfer rather than hydrogen atom transfer .......................................... 64
A Introduction ............................................................................ 64
B. Materials and Methods ............................................................................ 70
C. Results ............................................................................ 71
D. Discussion ............................................................................ 77
E. References ............................................................................ 96
IV. Chapter 4 The Photoreduction Mechanism in DNA Photolyase:
Correlated Triplet-Doublet Radical Pair Polarization
Arising From Differences in Hyperfine Frequencies .............................. 98
A Introduction ............................................................................ 98
B. Materials and Methods ............................................................................ 99
C. Results ........................................................................... 101
D. Discussion ........................................................................... 102

vi

E. References ........................................................................... 151

V. Chapter 5 Simulation of the Transient Tryptophan EPR Signal

that Occurs during Photoreduction of Inactive Photolyase ................. 153

A. Introduction ........................................................................... 153

B. Materials and Methods ........................................................................... 154

C. Results and Discussion ........................................................................... 166

D. References ........................................................................... 175

VI. Appendix Listing of program to simulate spin polarized EPR spectra ................. 176

vii

LIST OF TABLES

Table 5.1 Parameters used to simulate the spectra in Figure 5.3a and Figure 5.3b ..... 169

viii

Figure 1.1.

Figure 1.2.
Figure 1.3.
Figure 1.4.
Figure 1.5.
Figure 1.6.

Figure 1.7.

Figure 1.8.

Figure 2.1.

Figure 2.2.

Figure 2.3.

LIST OF FIGURES

Mechanism of pyrimidine dimer repair in ultraviolet
induced damaged DNA. PRE is the photoreactivating

enzyme, and pyr<>pyr is a pyrimidine dimer in damaged DNA. ..................... 3
The four isomers of pyrimidine dimers. ....................................................... 5
Possible DNA photolyase chromophores in the enzyme. ................................ 8
Proposed mechanism for pyrimidine dimer repair by DNA photolyase. ....... 10
Three oxidation states of flavin adenine dinucleotide. .............................. 13

Proposed mechanism of photoreactivation of inactive DNA photolyase. ....... 15

Depiction of populations of spin energy levels and EPR

stick diagrams for a boltzman and spin polarized manifold. .......

Diagram of relative populations of a doublet-doublet
radical pair’s energy levels and the corresponding
EPR stick diagram for a) CIDEP, b) CRPP, and c)

interacting CIDEP-CRPP type polarizations. ..............................

Proposed photoreduction mechanism in DNA photolyase. ......

Block diagram and timing sequence for kinetic trace
acquisition. a) shows the experimental setup for

obtaining kinetic traces and b) shows the timing sequence
that was used. The switching of the flashlamp and the
WAAG board are done by a trigger pulse from the CTM-OS

timing card. ............................................................................

Block diagram and the timing sequence for the gated
integrator setup. a) shows a block diagram of the

gated integrator and b) shows the timing sequence that was used.

ix

............ 18

............ 23

............. 33

............ 37

....... 40

Figure 2.4.

Figure 2.5.

Figure 2.6.

Figure 2.7.

Figure 2.8.

Figure 2.9.

Dark stable flavin radical EPR spectrum in DNA

photolyase. Experimental conditions were: 6.3 mW

microwave power, 9.22226 Ghz microwave frequency,

35 us time constant, 3290 G centerfield, 100 G sweep,

2.8 G modulation amplitude. ..................................................... 42

Kinetic trace of the transient radical at 3290 G. The trace in

a) was taken with the flashlamp as the excitation source

and is 600 averages. The trace in b) was taken with the YAG

laser and is 1000 averages. Experimental conditions were:

6.3 mW microwave power, 9.22226 Ghz microwave

fi'equency, 35 us time constant, 3290 G microwave field,

and 2.8 G modulation amplitude. ................................................................. 45

Power dependence of the transient EPR signal

in DNA photolyase. Experimental conditions

were: 9.22226 Ghz microwave frequency, 35 us time

constant, 2.8 G modulation amplitude. Photolyase samples

were excited with the flashlamp. ................................................................. 47

Light saturation plot for the transient radical in
DNA photolyase. Experimental conditions were:
6.3 mW microwave power, 9.22226 Ghz microwave
frequency, 35 us time constant, 3290 G magnetic field,
and 2.8 G modulation amplitude. Light intensity was
measured by exciting the sample with the flashlamp and
placing neutral density filters in front of the cavity. .............................. 50

Transient EPR spectra in DNA photolyase. a) is a

wild type sample and b) contains tryptophan that is

deuterated at the positions indicated by the asterisks

in the inset. a) is the average of 5 scans and b) is the average of
20 scans. Experimental conditions were: 6.3 mW

microwave power, 9.22226 Ghz microwave frequency,

35 us time constant, 3290 G centerfield, 100 G sweep,

2.8 G modulation amplitude, and 1000 second scans. Boxcar
conditions were: 3 averages per point, 1 Hz flash repetition rate,
4 us delay, 48 us aperture, trigger rate of 1 Hz, and

AC-coupling of greater than 10 Hz. ..................................................... 52
Integrated transient spectra for a) wild type and b)
deuterated photolyase samples. ................................................................. 57

Figure 2.10. Proposed reaction mechanism for photolyase that is derived

from the experimental results reported here. .......................................... 60

X

Figure 3.1.

Figure 3.2.

Figure 3.3.

Figure 3.4.

Figure 3.5.

Figure 3.6.
Figure 3.7.

Figure 3.8.

Figure 4.1.

Figure 4.2.

Figure 4.3.

Possible photoreduction mechanisms in photolyase. .............................. 67

Calculated electron spin densities for the cation
and neutral tryptophan radicals. ................................................................. 69

Isotopically labeled samples of DNA photolyase.

Isotopic labeling of the samples is described in

Kim, S.T.; Sancar, A.; Essenmacher, C. and Babcock,

G.T. Proc. Natl. Acad. Sci. USA 1993, 90, 8023.

Experimental conditions were: 6.3 mW microwave power,

9.22226 GHz microwave frequency, 1000 sec scan,

40 scans, 2.8 G modulation amplitude, 35 |J.S time constant,

3220-3350 G microwave field, 3 averages, 1 Hz repetition rate,

4 us delay, 48 us aperture, and AC-coupling of greater than 10 Hz. ....... 73

Tryptophan mutants of DNA photolyase. Samples preparation

is described in Kim, S.T.; Sancar, A.; Essenmacher, C. and
Babcock, G.T. Proc. Natl. Acad Sci. USA 1993, 90, 8023.
Experimental conditions were: 6.3 mW microwave power,

9.22226 GHz microwave fi'equency, 1000 sec scan, 40 scans,

2.8 G modulation amplitude, 35 us time constant, 3220-33 50 G
microwave field, 3 averages, 1 Hz repetition rate, 4 us delay, 48 us

aperture, and AC-coupling of greater than 10 Hz. .............................. 76
Angles used for calculation of hyperfine coupling constants for

B-methylene protons of tryptophan. ..................................................... 80
Tryptophan numbering scheme. ................................................................. 82
Possible geometries for tryptophan B-methylene protons. ................... 88

Proposed mechanism for photoreduction in photolyase
based on the experimental results presented in this chapter. ................... 94

Integrated transient spectra for a)wild type and b) deuterated
photolyase samples. ........................................................................... 104

Magnetic states of triplet-doublet radical pair and the energy levels
and state mixing in the coupled basis set that are formed from
interactions within the radical pair. ................................................... 107

Triplet-doublet state mixing from precessional frequency differences
in the radical pair in the absence of an exchange interaction. ................. 113

Figure 4.4.

Figure 4.5.

Figure 4.6.

Figure 4.7.

Figure 4.8.

presence of an exchange interaction.

Figure 4.9.

Figure 4.10.

Figure 4.11.

Figure 5.1.
Figure 5.2

Figure 5.3.

Figure 5.4.

EPR stick diagrams for net and multiplet efi‘ect electron spin
polarization in the presence and absence of an exchange interaction.

Triplet-doublet state mixing from precessional frequency differences
in the radical pair in the presence of an exchange interaction

with a doublet precursor. .........................................................

Triplet-doublet state mixing in the presence of an exchange

interaction with a doublet precursor at later reaction times. ...........

Hyperfine state mixing in a triplet-doublet radical pair

with a doublet precursor. .........................................................

Triplet-doublet state mixing with a quartet precursor in the

Reaction mechanisms for spin polarization in inactive photolyase.

Possible mechanisms for recombination during DNA

photolyase photoreduction. .........................................................

Mechanism for photoreduction in photolyase that includes doublet

precursor spin polarization. .........................................................
Simulation of EPR spectra that are in thermal equilibrium. ...........

Simulation of a spin polarized radical. .............................................

Comparison of simulated and experimental spectra: (a)
shows the natural abundance spectra and (b) shows the
a-ring deuterated spectra. The top spectrum in each

figure is the experimental spectrum and the lower spectrum is

the simulated one. ....................................................................

Comparison of simulated and experimental absorption spectra
for photolyase. (a) is the natural abundance spectrum and

(b) is the deuterated spectra. The top spectrum in each
figure is the experimental spectrum and the bottom one is

the simulated one. ....................................................................

xii

ooooooooooooooooooooooooooooooooooooooooooooo

..... 119

...... 122

...... 126

...... 130

...... 134

..... 137

...... 146

...... 149

...... 158

...... 160

....... 168

....... 173

Chapter 1

Introduction

Photolyase

Ultraviolet radiation from sunlight has deleterious effects upon cell survival rates in
biological organisms. The most significant target of UV radiation is DNA], and all
organisms have mechanisms to remove the harmful photoproducts that are formed as a
result of irradiation of the cells”. One mechanism that organisms use to reverse the
effects of UV-damage to DNA is photoreactivation. Figure 1.1 shows a diagram of the
basic biological mechanism”. The photoreactivating enzyme (PRE) binds to damaged
DNA (pyr<>pyr), and a subsequent photon of visible light repairs the DNA followed by
dissociation of the enzyme.

The most abundant of the DNA photoproducts that form upon UV-irradiation of
cells is pyrimidine dimers. Pyrimidine is excited to a singlet state with an excited state
lifetime of 10'12 seconds”. The excited state may decay to the triplet or directly react
with a nearby pyrimidine. In DNA, unlike isolated bases, pyrimidine dimer formation can
occur from either the triplet or the excited singlet state”. Four isomers of pyrimidine
dimers are shown in Figure 1.2, but the cis—syn isomer is the predominate product in
double-stranded DNA”.

DNA photolyase is an enzyme that is capable of repairing UV-induced pyrimidine

dimers in DNA6. This class of enzymes has a molecular weight of approximately 50,000.

Figure 1.1. Mechanism of pyrimidine dimer repair in ultraviolet induced damaged DNA.
PRE is the photoreactivating enzyme, and pyr<>pyr is a pyrimidine dimer in damaged
DNA.

PRE + pyr<>pyr

PRE --— pyr<>pyr

PRE + pyr + pyr

Sancar, A. Photolyase:
In Advances in Electron Transfer Chemistry; Mariano, P.E. , Eds,
JAI Press: London, 1992; Vol. 2, pp. 215-272.

Figure 1.1

Figure 1.2. The four isomers of pyrimidine dimers.

O

O
l0 I 0
I H3 HN H
2r x: I:
O NH NH O

cis-syn trans-syn (racemic)
O
H3 l H3
o “N /
rm 0%
CH3
0 CH3 0
trans-anti

cis-anti (racemic)

Figure 1.2

6
The active enzyme contains two chromophores, flavin adenine dinucleotide in its doubly

reduced form"8 and either methenyltetrahydrofolate (MTHF)9 or 8-hydroxy-5-deazaflavin
(8-HDF)‘°’""2. Figure 1.3 shows the molecular structure for each of these species.
Figure 1.3a shows the structure for the doubly reduced flavin chromophore. This
molecule can have one of two ionic forms, the singly protonated anion or the doubly
protonated neutral molecule. The ionic character of the flavin chromophore in the active
form of the enzyme has been proposed to be the doubly protonated neutral form of the
molecule-"8'”, but this assignment will be discussed later in Chapter 3. Figure 1.3b and
Figure 1.3c show the molecular structures for MTHF and 8-HDF. These chromophores
act as additional light harvesting chromophores in the active enzyme to enhance the
absorption cross section of the enzyme, but they are not necessary for the
photoreactivating mechanism to occur9.

Pyrimidine dimer repair by DNA photolyase has been proposed to occur by

13’1"” as shown in Figure 1.4. The photoreactivating enzyme, DNA

electron transfer
photolyase, binds to the pyrimidine dimer. Photoexcitation of either MTHF or the flavin
chromophore results in an excited flavin singlet state. This state can arise from either
direct photoexcitation of the flavin molecule or by photoexcitation of the folate
chromophore followed by electronic energy transfer to the flavin. Electron transfer fiom
the excited singlet flavin species to the pyrimidine dimer results in dimer splitting and
dissociation of the photoreactivating enzyme'3'M’”.

When DNA photolyase is isolated in vitro, one electron oxidation of the flavin

chromophore occurs. This form of the enzyme is inactive in dimer repair and dark

Figure 1.3. Possible DNA photolyase chromophores in the enzyme.

 

Sancar, A. Photolyase: In Advances in Electron Transfer
Chemistry; Mariano, RE Ed.; JAI Press: London, 1992;
Vol. 2, PP. 215-272.

Figure 1.3

Figure 1.4. Proposed mechanism for pyrimidine dimer repair by DNA photolyase.

10

hv

FADH2 1"'FADH2
I I _____> I I
I I I I

pyr pyr pyr pyr

electron
transfer

FIADHZ“
I I
Pyr Pyr
Pyr
dimer

FADH2

Heelis, P.F.', Okamura, T. and Sancar, A. Biochemistry
1990, 29, 5694.

Figure 1.4

1 l
stable“’”. The enzyme can be reactivated to the active form of the enzyme, capable of

dimer repair, by photoexcitation of the inactive enzyme with visible light”. The three
oxidation states of the flavin chromophore are shown in Figure 1.5. Figure 1.5a shows the
oxidized form of the flavin, FAD. Figure 1.5b shows the singly protonated semiquinone
form of the flavin, FADH'. Figure l.5c shows the doubly protonated reduced form of the
enzyme that is proposed to be the active state of photolyase, FADHzms. A model for this
reaction has been proposed and is shown in Figure 1.6”. Photoexcitation of either the
flavin chromophore or MTHF results in an excited doublet state flavin with a lifetime of
10'lo secondsmg. This excited state is formed by direct photoexcitation of FADH‘ or by
excitation of MTHF followed by electronic energy transfer to F ADH'. In this model, the
doublet flavin excited state (”FADH') intersystem crosses to the quartet (4'FADH') which

dB”. Hydrogen atom transfer fi'om an

is postulated to have a lifetime of a microsecon
amino acid donor to 4°FADH' produces a neutral amino acid radical and FADHZB. The
identity of the amino acid donor has been determined to be tryptophan by optical
spectroscopyn'zo; a firrther discussion of the amino acid donor will follow in Chapters 2
and 3. If no exogenous donors are present, the hydrogen atom transfer products
recombine to form the inactive F ADH-containing enzyme13 . In the presence of
exogenous donors, the external donor donates a hydrogen atom to the amino acid to form
the active F ADHz-containing enzymel3 .

Recently, the three dimensional crystal structure for DNA photolyase has been

reported”. The structure shows that trpm, implicated by optical spectroscopy to be the

internal amino acid donor to FADH’ during photoreduction of the inactive enzyme”, is

12

Figure 1.5. Three oxidation states of flavin adenine dinucleotide.

13

H3C g NYC
yelp...
H3C N
3) FAD.
(oxrdrzed)
H3C 113‘ NYC
3.101%
H3C I?

b) FADH-
(paramagnetic)

R H
H3C N NYC
0) FADH2
(reduced-active)

Fig urel .5

14

Figure 1.6. Proposed mechanism of photoreactivation of inactive DNA photolyase.

l5

dark

 

stable I = IOOps 1: = 1118
FADH- 2"‘FADH- 4“FADH-
l hv l .
rntersystem
trpH trpH 8 “PH
H-atom

 
  

Recombination

k= 100 S'1

FA|DH2/ up.

H-atoin donor
trpH

Heelis, P.F.; Okamura, T. and Sancar, A. Biochemistry
1 990, 29, 5694.

Figure 1.6

16
more than 13 A away from F ADH‘Z'. Two tryptophan residues, trp359 and trpm, lie

between FADH‘ and “13306, and these intervening tryptophan residues may be
intermediates during photoreduction of the inactive enzyme considering the distance
between trp306 and FADH' is large”. The crystal structure also shows that trp306 is near
the protein surface and is accessible to the solvent”. This observation is consistent with
the proposed mechanism for photoreduction of photolyase13 that is shown in Figure 1.6.
In order to reduce the inactive FADH-containing enzyme to the active form (FADHz), an
exogenous hydrogen atom donor to trp306 is necessary. Because trp306 is near the protein
surface, an external hydrogen atom donor has access to the internal donor and may be able

to reduce trpgos to form the active enzyme (see Figure 1.6).

Electron Spin Polarization

Electron spin polarization that arises from a pair of interacting doublet radicals has
been observed in photosynthetic reaction centersn”. Spin polarization results in emissive
and enhanced absorptive transitions in the EPR spectrumzw . Transitions in the EPR
spectrum are proportional to the difference in population between the two energy levels
that are involved in a particular transition. In a non-spin polarized EPR spectrum, the
energy difference between an or and a B electron spin energy level is small. As a result, the
difference in population between these two energy levels is also small due to the thermal
equilibration of the levels. Figure 1.7a shows an EPR transition that arises from energy

levels that are in Boltzman equilibria. The transition is low in intensity because the

17

Figure 1.7. Depiction of populations of spin energy levels and EPR stick diagrams for a
boltzman and spin polarized manifold.

18

a) Boltzman equilibrium

IIII ,
IIIII .

A I
E H—>

b) Absorptive spin polarization
4 13
I I I I I I II a

A I
E H—>

c) Emissive spin polarization

MIMI”
4 a

U

A
E H—>|

Figure 1.7

19
population difference between the electron spin energy levels is correspondingly small.

Another characteristic of EPR transitions in a thermally equilibrated EPR spectrum is that
all the transitions are in absorption. Boltzman equilibrium requires that the [3 electron spin
energy level, which is lower in energy, must have a greater population than the higher
energy or electron spin energy level. A transition between the two energy levels will be
absorptive due to the greater population of the lower electron spin energy level.

Electron spin polarized EPR spectra typically have transitions that are enhanced in
intensity compared to a spectrum that is determined by Boltzman equilibriumu'zs. The
transitions may be absorptive or emissive depending on the relative populations of the two
electron spin energy levels. Figure 1.7b shows the relative populations and the
corresponding EPR transition that is expected for an enhanced absorptive spin polarized
EPR spectrum. The population of the lower energy [3 electron spin energy level is much
greater than that of the or electron spin energy level. The result of the large population
difference between the electron spin energy levels in Figure 1.7b is that the transition is
absorptive, and the intensity is much greater than what would be expected for a transition
that is determined by a Boltzman distribution (compare Figure 1.7a to Figure 1.7b).

Figure 1.7c shows the relative populations of the electron spin energy levels for a

spin polarized radical that is in emission. The higher energy or electron spin energy level
has a much larger population than the B electron spin energy level. This large population

difference results in an intense emissive transition. The intensity of the transition is much
larger than the intensity of a transition that arises from a radical whose electron spin

energy levels show a Boltzman population distribution (compare Figure 1.7a to Figure

20
1.7c) because the population difference between the electron spin energy levels is much

larger in a spin polarized manifold24'25.

Correlated Radical Pair Polarization

Photosynthetic reaction centers show electron spin polarization with a polarization
pattern that alternates between emissive and absorptive transitionszz’z‘s’”. The
experimental spectrum has been explained by a Correlated Radical Pair Polarization
(CRPP) mechanism22’26’27. The spin polarized EPR spectrum arises from a pair of doublet
radicals that are interacting through the electronic exchange interactionz‘w. The coupled
basis set of the radical pair has four eigenfunctions for a doublet-doublet radical pair.
These eigenfirnctions are represented as | T. > , | To > , and | T. > for the triplet states and |
So > for the singlet state. State mixing between I So > and | To > occurs because of a
difference in magnetic frequency between the two radicals of the pair. Equation 1.1 shows
the mixed states that occur as a result of the magnetic frequency differences between the

two members of the radical pair.

“So” = | So > + Al To > “To” = | To > - Al So > equation 1.1

“So” is the state that is of primarily singlet character and “To” is predominately triplet

character. The parameter A in equation 1.1 is the amount of state mixing that occurs. A is

21
given in equation 1.2 as the sum of the difference in g-values and the hyperfine coupling

frequencies of the two radicals of the radical pair.

A = gIBH - guBH + 2'." Ark muc - 24, Am, mm equation 1.2

g is the electronic g-value of each radical, B is the Bohr magnetron, and H is the external

magnetic field. A is the nuclear hyperfine frequency coupling constant and m is the
electron spin quantum number.

Figure 1.83 shows the energy level diagram and the resulting EPR transitions for a
doublet-doublet radical pair that was interacting but where the exchange interaction has
decreased to zero during the EPR observation time period“. The states denoted “So” and
“T0” are the mixed states that are formed by the perturbation that is shown in equations
1.1 and 1.2. | TI > and | T- > are the pure triplet states in the coupled basis set of the

radical pair. The splitting AE between the “So” and “To” energy levels is equal to the
perturbation in equation 1.2. AB is shown in equation 1.3 as the sum of the difference in

magnetic frequency and the exchange interaction of the doublet-doublet radical pair”. In
Figure 1.8a, the exchange interaction has decreased to zero so the difference in energy
between “So” and “To” is simply the differerence in magnetic frequencies of the two
radical pair members for the pure CIDEP case in this figurez“.

In photosynthetic reaction centers, the exchange interaction has been predicted to
be a positive quantity”. For neutral radicals, the exchange interaction should be negative

(singlet lies lower in energy than the triplet)25 . When the radical pair is ionic, a positive

22

Figure 1.8. Diagram of relative populations of a doublet-doublet radical pair’s energy
levels and the corresponding EPR stick diagram for a) CIDEP, b) CRPP, and c)
interacting CIDEP-CRPP type polarizations.

23

a) CIDEP
_ _ _ |T+>
/
“'80,, x ‘T ,,
0
4

    

|T>

A__IL
E A

2¢

 

II.

b) CRPP

_ _ _ |T+>
/\

An S‘To,,
_ _ 3
4 'T-> 1
ml l

+ * +|

2 3
c) “interacting” CIDEP-CRPP

____|n>

/ \

An ‘T0,,
_ 3
4 I T? 1
21| I

2|+ r. »

 
 

 

“S as
0

 

 

1
2

>

 

[Tl

 

“S as
0

 

 

1
2

{11>

 

3
Norris, J .R.; Morris, A.L.; Thumauer, MC. and Tang, J.

J. Chem. Phys. 1990, 92, 4239.

Figure 1.8

24
exchange interaction (singlet is higher in energy than the triplet) has been predictedzs’”.

The radical pair in photosynthetic reaction centers are ionic; therefore, the exchange
interaction that arises from the doublet-doublet radical pair in reaction centers should be
positive. Figure 1.8a shows the energy level diagram for a radical pair that was interacting
with a positive exchange interaction that has decreased to zero before the EPR
observation time”. “So” is higher in energy than “To” in the figure because of the positive
exchange interaction“.

The initial population of each state in Figure 1.8a is determined by the precursor
state of the doublet-doublet radical pair due to electron spin conservation”. If the radical
pair is formed from a singlet precursor state, the only states that will have any initial
population are the states that contain singlet character“. Only “So” and “To” contain
singlet character in Figure 1.8a. Therefore, only these states are initially populated. The
pure triplet states in Figure 1.8a, | T. > and | T- >, do not receive any initial population
because these states do not contain any singlet character.

The transitions in Figure 1.83 are determined by the usual EPR spectroscopy

selection rules ( Am; = 0 Am. = j: 1) where m is the nuclear quantum number and m. is the

electron spin quantum number. Because the exchange interaction is zero at the time of
observation by EPR, transitions 1 and 3 in the figure occur at the same external magnetic
field”. The same is true for transitions 2 and 4. The intensity of an EPR transition is
proportional to the product of the probability that a particular transition will occur and the
population difference between the two levels involved in a particular transition. Since the

transitions in Figure 1.8a are to the pure triplet states, | TI > and | T. >, the transitional

25
probability is proportional to the amount of triplet character in the mixed “So” and “To”

states. Transitions from “To” will have a higher transition probability because this mixed
state contains more triplet character than “So”. The population difference is also greater
from transitions that originate from “To” in the CIDEP case that is illustrated in the figure.
Initially with a singlet precursor, “So” will receive more of the initial population because
this state contains more singlet character than “To”. As the radical pair evolves away from
its initial spin distribution, “To” receives more of the population. As a result, the
transitions that arise from the “To” state will show a greater EPR intensity than the
transitions that arise from “So” for the CIDEP case depicted in this figure”.

The low field absorptive peak in the EPR stick diagram arises from transitions 2
and 4 in the energy level diagram. Transition 2 (“80” to | T. >) is emissive, while transition
4 (“To” to | L >) is absorptive. These transitions occur at the same magnetic field, so the
EPR spectrum will only show the net addition of these two EPR transitions. The
transition that originates from “To” (transition 4) will be more intense as discussed above.
The net result is an absorptive peak in the EPR spectrum. The high field emissive peak in
the EPR stick diagram arises from transitions 1 and 3. Transition 1 (“80” to | T. >) will be
absorptive and transition 3 (“To” to | T- >) will be emissive. The resulting EPR spectrum
will be the net addition of these two transitions because they occur at the same magnetic
field. As discussed above, the transition that arises from “To” will be more intense so the
addition of transitions 1 and 3 produces a net emissive peak in the spectrum. The EPR
spectrum shows an absorptive / emissive pattern with no net polarization26'27. There is no

net polarization because the intensity of the low field absorptive peak exactly equals the

26
intensity of the high field emissive peak in the EPR spectrum. This result is typical for

CIDEP electron spin polarization, as well as for CRPP that will be discussed below.

Figure 1.8b shows the energy level diagram and the corresponding EPR stick
diagram for a doublet-doublet radical pair that exhibits Correlated Radical Pair
Polarization (CRPP)“m. The figure depicts a doublet-doublet radical pair that is formed
from a singlet precursor and is weakly interacting during the EPR observation time. The
So and To energy levels in the coupled radical pair basis set mix due to a difference in
magnetic frequency of the two radicals in the pair by a mechanism that was discussed
above for Figure 1.8a. The mixed energy levels are again represented by “So” and “To” in
the figure. At early observation times when CRPP is present, “So” has a greater
proportion of the initial population than “To”. Because the radical pair is formed fi'om a
singlet precursor, only those states that contain singlet character will be populated initially.
| TI > and | T- > are pure triplet states so these states do not receive any of the initial
population. “So” contains more singlet character than “To”, so “So” will have a greater
initial population.

In Figure 1.8a for pure CIDEP, the difference in energy between the “So” and “To”
energy levels is equal to the difference in magnetic fiequency of the two radicals of the
radical pair as is shown in equation 1.2. When CRPP is present, the radicals of the pair
are weakly interacting through the exchange interaction 21. This interaction causes the
singlet to be higher in energy than the triplet for a positive exchange interaction as is the

,7
b262

case for the energy level diagram in Figure 1.8 . The difference in energy between

27
“So” and “To” is equal to the sum of the difference in magnetic frequency of the two

radicals and the exchange interaction as is shown in equation 1.3.

A Q = A + 2] equation 1.3

A is the difference in magnetic frequency of the two radical pair members as given by

equation 1.2, and 2.1 is the exchange interaction between the two radical pair members.
For pure CIDEP, transitions 1 and 3 occur at the same magnetic field, but this is no longer
true for CRPPZZ‘ZG’”. The presence of the exchange interaction during the EPR
observation time causes transitions 1 and 3, as well as transitions 2 and 4, to be split by the
exchange interaction that is present. The difference in magnetic field between transitions 1
and 2, as well as transitions 3 and 4, is still the difference in magnetic frequency of the two
radical pair members (same as for the CIDEP case in Figure 1.8a).

The intensity of all EPR transitions will be equal when CRPP is present as is shown

bzmn. The intensity of an EPR transition is

in the EPR stick diagram in Figure 1.8
proportional to the product of the population difference of the two energy levels and the
transition probability. The low field transition (“80” to | T. >) has a larger population
difference than the transition from “To” to | T+>. The transition probability for these
transitions is proportional to the amount of triplet character in “So” and “To” because the
transition is to the pure triplet. The low field transition has a lower transition probability

than the “To” to I T.» > transition because “So” contains less triplet character. The net

result is that the two transitions will be equal in intensity. The transition that originates

28
from “So” has a larger population difference but a smaller transition probability than the

transition that originates from “To”. These two effects are equal and opposite so the

transitions are equal in intensity26'27.

The two high field transitions are also the same
intensity because of an analogous argument.

Figure 1.8c shows an energy level diagram and EPR stick spectrum for a doublet-
doublet radical pair that shows electron spin polarization through the “interacting”
CIDEP-CRPP mechanism22’26'27. In this figure, there is a positive exchange interaction so
“So” lies higher in energy than “To”. Initially “So” receives more population because the
radical pair is formed from a singlet precursor. At later times, some of the initial
population goes to the triplet as discussed above. The figure shows the case where the
populations of the two mixed states have become accidentally degenerate”. The transition
probabilities have not changed from the pure CRPP case in Figure 1.8b, and the exchange
interaction is still present during the EPR observation time. The result is an EPR spectrum
whose transitions occur at the same magnetic field as the pure CRPP spectrum in Figure
1.8b, but the intensities of the transitions are no longer equa126’27. The population
difference is now equal for all transitions in the figure, so the transition probability is the
only determinative factor for the intensity of the transition. Transitions that originate fi'om
“To” will be more intense than transitions originating fi'om “So” because “To” contains
more triplet character. The spectrum shows an E/A/E/A polarization pattern with net
absorption for the first pair of exchange-split transitions and net emission for the second

pair. This pattern is shown in the EPR stick diagram in the figure and is a consequence of

“To” containing more triplet character than “So”.

29
References

1. Wang, S.Y. Photochem. Photobiol. of Nucleic Acids 1976, Vols. [,1], Academic Press,
NY.

2. Sancar, A; Sancar, G.B.; Annu. Rev. Biochem. 1988, 57, 29.
3. Myles, G. and Sancar, A. Chem. Res. Tax. 1989, 2, 197.

4. Rupert, C.S. J. Gen. Physiol. 1962, 45, 703.

5. Rupert, C.S. J. Gen. Physiol. 1962, 45, 725.

6. Sancar, A Photolyase: In Advances in Electron Transfer Chemistry; Mariano,
P.E., Ed.; JAI Press: London, 1992; Vol. 2, pp. 215-272.

7. Sancar, A; Sancar, G.B. J. Molec. Biol. 1984, 172, 223.
8. Eker, AP.M.; Kooiman, P.; Hessels, J.K.C.; Yasui, A. J. Biol. Chem. 1990, 265, 8009.

9. Johnson, J.L.; Hamm-Alvarez, S.; Payne, G.; Sancar, G.B.; Rajagoplan, K.V.; Sancar,
A. Proc. Natl. Acad Sci. USA 1988, 85, 2946.

10. Eker, AP.M.; Hessels, J.K.C.; van de Velde Biochemistry 1988, 27, 1758.

11. Eker, A.P.M.; Dekker, R.H.; Berends, W. Photochem. Photobiol. 1981, 33, 65.
12. Kiener, A; Husain, 1.; Sancar, A; Walsh, C. J. Biol. Chem. 1989, 264, 13880.
13. Heelis, P.F.; Okamura, T. and Sancar, A. Biochemistry 1990, 29, 5694.

14. Kim, S.T.; Sancar, A; Essenmacher, CE. and Babcock, GT. .1. Am. Chem. Soc.
1992, 114, 4442.

15. Okamura, T.; Sancar, A; Heelis, P.F.; Begley, T.P.; Hirata, Y. and Mataga, N. J. Am.
Chem. Soc. 1991, 113, 3143.

16. Li, Y.F. and Sancar, A. Nucleic Acids Res. 1989, 18, 4885.
17. Payne, G.; Heelis, P.F.; Rohrs, BR. and Sancar, A. Biochemistry 1987, 26, 7121.

18. Yeh, S.-R. and Falvey, D.E. J. Am. Chem. Soc. 1991, 113, 8557.

3O

19. Okamura, TA; Sancar, A.; Heelis, PF; Hirata, Y. and Mataga, N. J. Am. Chem.
Soc. 1989, 111, 5967.

20. Li, Y.F.; Heelis, PF. and Sancar, A. Biochemistry 1991, 30, 6322.

21. Park, H.W.; Kim, S.T.; Sancar, A. and Deisenhofer, J. Science 1995, 268, 1866.

22. Snyder, SW. and Thumauer, M. C. Electron Spin Polarization in Photosynthetic
Reaction Centers: The Photosynthetic Reaction Center 1993, Vol.2 ed. Deisenhofer,
J. and Nonis, J .R. (Academic Press, SanDiego) p. 285.

23. Hoff, AJ. Q. Rev. Biophys. 1984, I7, 153.

24. Jenks, W.S. and Turro, N.J. Rev. Chem. Int. 1990, 13, 237.

25. Adrian, F.J. Rev. Chem. Int. 1986, 7, 173.

26. Norris, J.R.; Morris, A.L.; Thumauer, MC. and Tang, J. J. Chem. Phys. 1990, 92,
4239.

27. Snyder, S.W.; Morris, A.L.; Bondeson, S.R.; Norris, JR. and Thumauer, M.C. J.
Am. Chem. Soc. 1993, 115, 3774.

28. Batchelor, S.N.; Heikkila, H.; Kay, C.W.M.; McLauchlan, K.A. and Shkrob, I.A.
J. Chem. Phys. 1992, 162, 29.

Chapter 2
The Identification of a Transient Tryptophan Radical in the Photoreduction

Reaction of DNA Photolyase by Time-Resolved Electron Paramagnetic Resonance

Introduction

When DNA photolyase is isolated in vitro, one electron oxidation of the active

form of the enzyme to the inactive F ADH‘ form occurs1’2'3"

(see Figure 1.5), as discussed
in Chapter 1. This form of the enzyme is dark stable and catalytically inactive“. Catalytic
activity can be restored in the enzyme by irradiation of photolyase with visible light
causing FADH‘ to be photoreduced to the active form of the enzyme". A proposed
mechanism for photoreduction of photolyase to the active form of the enzyme is shown in
Figure 2.1”. Flash photolysis of the inactive enzyme causes hydrogen atom transfer fi'om
an amino acid donor to the flavin chromophore. Subsequent reduction of the amino acid

donor results in reactivation of the enzymel’u.

Optical studies of the photoreduction
process in photolyase have suggested that tryptophan-306 is the internal amino acid
donor". Each of the 15 tryptophan residues in the enzyme was individually replaced with
phenylalanine, and the rate of photoreduction was monitored. In the eight mutants that
were able to overproduce photolyase, all except the trp—306 to phe mutant were able to
photoreduce FADH' in spite of the mutations. These results established that trp306 is
involved in photoreduction of photolyase”. In Chapter 1, a proposed mechanism for the
excited states that form in the enzyme during photoreduction of the inactive enzyme was

shown in Figure 1.67. The mechanism proposes that photoexcitation of either F ADH' or

31

32

Figure 2.1. Proposed photoreduction mechanism in DNA photolyase.

33

FADH- h, FADH2
| —> |

trpsosH 4— “P306

recombination

H-atom donor

 

i

FADH2

|
tr13306H

Heelis, P.F.; Okamura, T. and Sancar, A. Biochemistry 1990,
29, 5694.

Figure 2.1

34
MTHF produces an excited doublet flavin state (Z'FADH') and initiates the photoreduction

reaction9. Within 100 ps, 2'FADH' intersystem crosses to the quartet, 4.FADHw. H-atom
transfer from trp-306 to 4'FADH occurs in a microsecond to form the transient species
FADH2---2trp'7. In the presence of exogenous donors, trp' is reduced to form the
catalytically active enzyme. Without exogenous donors, H-atom recombination occurs to
fortn the inactive, FADH-containing enzyme over several milliseconds7.

In this study, the identity of the amino acid donor is confirmed by time-resolved

'0’”. The time-

EPR as trp-306 using isotopic labeling and site directed mutagenesis
resolved EPR spectrum has emissive and absorptive transitions suggesting that the
transient FADH'---trp' species is spin polarized due to interactions between these two
members of the transient species”. Because of the similarity in polarization patterns

between photolyase and the photosynthetic reaction center, a spin polarized mechanism

analogous to the Correlated Radical Pair Mechanism (CRPP)“l3 is proposed.

Materials and Methods

Isolation of photolyase was carried out at the University of North Carolina in the

laboratory of Professor Aziz Sancar and has been described previously""15 .

Isotopic
labeling and site-directed mutagenesis were performed by Sancar as described elsewhere".
Experiments were performed with samples that contained 0.1 mM enzyme, 20% glycerol,
50 mM potassium phosphate (pH 7.0), and 50 mM NaCl. 5 mM potassium ferricyanide

was added to the sample to prevent reduction of the enzyme from FADH' to FADHz. The

35
oxidation state of the flavin was monitored by taking the visible absorption spectrum of

the sample with a Perkin Elmer Lamda 5 double beam UV/visible spectrophotometer. All
experiments were performed at 277 K. EPR measurements were performed on a Bruker
ER200D spectrometer with a TM011 cavity. The sample was held in the cavity with a
300 u] quartz flat cell (Wilmad, Buena, New Jersey). The microwave frequency was
measured with a HP 5245 counter / 5255 frequency converter and the magnetic field was
monitored with a Bruker ER03 5M NMR gaussmeter.

Photoexcitation of the sample was done with either a Xenon flashlamp or a
Quanta-Ray DCR-ll pulsed Nd-YAG laser (Spectra-Physics, Mountain View, California).
The flashlamp is critically damped with a pulse width of 17 us and a lamp energy of 50 J '6.
The laser has a pulse width of 10 ns, 3 pulse energy of 20 m], and an excitation
wavelength of 355 nm. Both excitation sources were used at a repetition rate of 1 Hz.

Kinetic traces were obtained as described previously” and outlined in Figure 2.2.
The signal output from the spectrometer was input into an 8-bit Markenrich aneform
acquisition card (Markenrich Corporation, Duarte, California) that is capable of sampling
from 50 ns to 0.5 ms. Multiple flashes were computer averaged to obtain the kinetic
traces. The timing scheme used for kinetic traces is also shown in Figure 2.2. A
MetraByte CTM-OS timing card (Metrabyte Corporation, Taunton, MA) that is software-
controlled was used to control the timing sequence as shown in Figure 2.2b. The CTM-OS
card provided a pulse at a 1 Hz repetition rate to the waveform acquisition card in order

to trigger the collection of data. A software-controlled delay was set, and after the delay a

36

Figure 2.2. Block diagram and timing sequence for kinetic trace acquisition. a) shows the
experimental setup for obtaining kinetic traces and b) shows the timing sequence that was
used. The switching of the flashlamp and the WAAG board are done by a trigger pulse
fi'om the CTM-OS timing card.

37

a) Kinetic trace setup block diagram

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EPR S‘gnal WAAG data
(sample) > Acquisition ——> computer
card
hv
, trigger
flashlamp

or Nd YAG ‘ trigger CTM-OS

laser timing card

 

 

 

 

 

b) Kinetic trace setup timing sequence

flashlamp
on

 

 

 

 

 

 

off —— __

 

WAAG data acquisition card

011

 

 

 

off

 

Figure 2.2

38
pulse was sent to trigger the excitation source. Data were stored for each flash and

averaged to obtain the kinetic trace.

Time-resolved EPR spectra were recorded17 as shown in Figure 2.3a. The signal
output from the spectrometer was input into a Stanford Research Systems SR250 gated
integrator and boxcar averager (Palo Alto, California). The gated integrator electronically
integrated the signal and sent the averaged signal to a PC to record the signal intensity at a
particular magnetic field position. The field width of 100 G was swept slowly at a rate of
1000 s, and 1000 data points were collected during a sweep. With a 1 Hz excitation rate,
the sample was excited at each data point collected.

Timing of the instrument is outlined in Figure 2.3b. The gated integrator was
triggered internally at 1 Hz. The trigger pulse was also output to the flashlamp to excite
the sample at the same time the boxcar was triggered. A delay and aperture were set to
capture a significant portion of the transient radical. The integrator conditions that were
typically used were a trigger rate of 1 Hz, a delay of 4 us, an aperture of 48 us, and 3
averages per point. AC-coupling of greater than 10 Hz was used to discriminate against
any stable signal that may be present. Typical EPR conditions were 2.8 G modulation
amplitude, 6.3 mW microwave power, 9.22226 GHz microwave frequency, and a time
constant of 35 us.

Results

Figure 2.4 shows a dark stable CW EPR spectrum of inactive photolyase due to

oxidation of the flavin chromophore to the semiquinone form, FADH' (see Figure 1.5).

39

Figure 2.3. Block diagram and the timing sequence for the gated integrator setup. a)
shows a block diagram of the gated integrator and b) shows the timing sequence that was
used.

40

a) Block diagram gated integrator setup

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

computer

 

 

 

 

 

EPR signal gated
F .
(sample) Integrator
Averaged
hv signal
flashlamp 4
trigger
b) Timing sequence
on flashlarnp
off ——
gated integrator
on
off delay I

 

 

 

 

trigger

aperture

Figure 2.3

—L

 

 

trigger

 

41

Figure 2.4. Dark stable flavin radical EPR spectrum in DNA photolyase. Experimental
conditions were: 6.3 mW microwave power, 9.22226 Ghz microwave frequency, 35 us
time constant, 3290 G centerfield, 100 G sweep, 2.8 G modulation amplitude.

42

 

 

ge

l l g

 

 

3240

3260

3280 3300 3320

Gauss

Figure 2.4

3340

43
The spectrum has no well-resolved hyperfine couplings and an apparent g-value of 2.0039.

The absence of resolved hyperfine couplings and the apparent g-value of the dark stable
spectrum in Figure 2.4 are in agreement with previous reports of flavin doublet radicals
that occur in proteins. Flash photolysis of the inactive enzyme whose EPR spectrum is
shown in Figure 2.4 produced an intense EPR transient signal in the g=2 region. The
kinetic trace of this signal is shown in Figure 2.5. Figure 2.5a was obtained by exciting the
sample with a Xenon flashlamp, and Figure 2.5b with a Nd-YAG laser. The kinetic trace
of this signal shows that the rise and fall of the light-induced radical occurs with a half
time of approximately 35 us. The observed half time for radical formation and decay is
most likely the instrument response time due to the use of 100 kHz modulation and does
not represent the true kinetics for the transient radical. Artifacts due to the longer pulse
width of the Xenon flashlamp (17us) are not present as a comparison of Figure 2.5a to
2.5b shows. When the excitation pulse was shortened to 10 ns as in Figure 2.5b, the
kinetics of the transient radical were essentially the same as for a 17 us pulse width.
These results suggest that the transient radical EPR signal is unaffected by the method of
excitation that is used.

Figure 2.6 shows the microwave power dependence for the transient radical's EPR
signal. The signal intensity initially increases linearly with an increase in the square root of
the microwave power. The maximum signal intensity occurs at a microwave power of 6.3
mW. All further experiments were performed at this microwave power in order to obtain
the maximum signal to noise ratio. At higher microwave powers the signal intensity

decreases. Later in this chapter and in chapter 4 evidence will be presented that shows

44

Figure 2.5. Kinetic traces of the transient radical at 3290 G. The trace in a) was taken
with the flashlamp as the excitation source and is 600 averages. The trace in b) was taken
with the YAG laser and is 1000 averages. Experimental conditions were: 6.3 mW
microwave power, 9.22226 Ghz microwave frequency, 35 us time constant, 3290 G
microwave field, and 2.8 G modulation amplitude.

45

 

 

a) kinetic trace at
3290 G with Xenon
flashlarnp

LWW

b) kinetic trace at
3290 G with Nd—YAG
laser with 355 nm
excitation

“\A W“
\/\/~\/\/f\/M\A/‘A

 

' l l l l

 

0.1 0.2 0.3 0.4

Figure 2.5

46

Figure 2.6. Power dependence of the transient EPR signal in DNA photolyase.
Experimental conditions were: 9.22226 Ghz microwave frequency, 35 us time constant,
2.8 G modulation amplitude. Photolyase samples were excited with the flashlamp.

relative intensity

100

80

60

40

20

47

 

 

__

l l

 

 

6 9

square root power

Figure 2.6

15

48
that the transient EPR signal in photolyase is spin polarized due to magnetic interactions

among two radicals in a radical pair. Other spin polarized radical pairs have been shown
to exhibit Torrey oscillations at high microwave powers”. Torrey oscillations arise from a
electron spin phase shift of a radical pair from a singlet to a triplet. This spin phase shift
changes the phase of spin polarization and results in a change of phase of the EPR signal
as well. Signal damping for each phase shift and a constant frequency are characteristics
of Torrey oscillations. In photolyase, these phase shifts at high microwave powers due to
Torrey oscillations are not present. Their absence is probably attributable to our detection

scheme by field modulated EPR, which limits our time resolution to 35 us.

Figure 2.7 shows a light saturation curve for the transient radical when the
flashlamp is used as the excitation source. The curve shows that at firll lamp intensity the
signal intensity is beginning to plateau to a maximum value suggesting that the sample is
light saturated at the typical lamp energies that were used. The transient EPR signal is
linearly dependent upon the light intensity that was used (see Figure 2.7). This linear
dependence suggests that the process that is observed upon photoexcitation of the inactive
enzyme is a one photon reaction.

Using gated integration techniques, a spectrum was obtained for the transient
radical and is shown in Figure 2.8a. The spectrum is an average of five individual scans
taken over a 100 G scan range. The spectrum has three major transitions in a 1:221
intensity ratio and each is split by 15 G (solid arrows). Superimposed on these larger
splittings are smaller ones (less than 5 G) as shown by the dashed arrows in Figure 2.8a.

The apparent g—value for the transient radical in Figure 2.83 is 2.0028 and is typical for an

49

Figure 2.7. Light saturation plot for the transient radical in DNA photolyase.
Experimental conditions were: 6.3 mW microwave power, 9.22226 Ghz microwave
frequency, 35 us time constant, 3290 G magnetic field, and 2.8 G modulation amplitude.
Light intensity was measured by exciting the sample with the flashlarnp and placing neutral
density filters in front of the cavity.

relative signal intens.

100

80

60

40

20

50

 

 

 

l l l l

 

20 40 60 80 100

relative light intens.

Figure 2.7

51

Figure 2.8. Transient EPR spectra in DNA photolyase. a) is a wild type sample and b)
contains tryptophan that is deuterated at the positions indicated by the asterisks in the
inset. a) is the average of 5 scans and b) is the average of 20 scans. Experimental
conditions were: 6.3 mW microwave power, 9.22226 Ghz microwave frequency, 35 us
time constant, 3290 G centerfield, 100 G sweep, 2.8 G modulation amplitude, and 1000
second scans. Boxcar conditions were: 3 averages per point, 1 Hz flash repetition rate, 4
us delay, 48 us aperture, trigger rate of 1 Hz, and AC-coupling of greater than 10 Hz.

 

52

 

 

   

 

  

 

3220

*
*
11 1 *
a) Wild type
r
l I I
b) trp-indole-d5
a l r l r l r l r I r I 1 I
3240 3260 3280 3300 3320 3340 3360

Gauss

Figure 2.8

53
organic radical. Figure 2.8b shows a transient spectrum for a sample that was isotopically

labeled with deuterium on tryptophan molecules in the protein at the positions that are
indicated with asterisks on the tryptophan structure in the inset to the figure. Isotopic
labeling of the sample was performed in the laboratory of Dr. Aziz Sancar at the
University of North Carolina School of Medicine in Chapel Hill, North Carolina and has
been described elsewhere”15 . The spectrum has the same apparent g-value and shows the
same 112:] intensity ratio with a hyperfine splitting of 15 G as the wild type sample in
Figure 2.8a. The smaller couplings that are present in the wild type sample (dashed

arrows in Figure 2.8a) are not resolved in the deuterated sample in Figure 2.8b.

Discussion

The isolated, inactive form of DNA photolyase contains a flavin semiquinone
radical (FADH')5’19 as seen in Figure 1.6 and Figure 2.1. The dark CW EPR spectrum in
Figure 2.4 confirms this assignment”. The spectrum has a g-value (2.0039) that is
consistent with other reported flavin radicals. Optical studies have suggested that the
inactive enzyme can be photoreduced to the active form (see Figure 2.1) by H-atom
transfer from a tryptophan side chain to FADH'7'19’2°. The products of the photoreduction
reaction are a neutral tryptophan doublet radical and a fully reduced triplet flavin
(F ADH2)7'l°’". In the presence of exogenous donors, the tryptophan radical is reduced to

7,21

form the active enzyme . Without donors, the transient tryptophan radical accepts an H-

atom from FADHZ to form the inactive enzyme in a recombination reaction.

54
Photoexcitation of the inactive enzyme produces an intense transient signal in the

g=2 region of the EPR spectrum“). The kinetic trace of the transient radical is shown in
Figure 2.5. The decay kinetics of the radical show a decay phase with a half time of 35 us.
The rise of the transient and the fast decay are not true half times, but rather the response
time of the instrument. The kinetic trace in Figure 2.5 was taken at a magnetic field of
3290 G. Kinetic traces at other field positions also show the same kinetics and ratios of
the fast to slow phase.

Gated integration techniques were used to obtain the EPR spectrum of the
transient radical (see Figure 2.83) that forms following photoexcitation of the enzyme.
The EPR spectrum shows three major couplings in a 122:] ratio. This pattern is consistent
with coupling to two equivalent nuclei, most likely the two B-methylene protons in
tryptophan (see Chapter 3).

To confirm that tryptophan is the transient amino acid radical that is involved in
the photoreduction reaction, a sample with tryptophan that is deuterated on the indole ring
was prepared. The inset to Figure 2.8 shows the tryptophan structure and the asterisks
denote the positions where deuterium replaced hydrogen. The deuterated sample (Figure
2.8b) has a transient EPR spectrum that has the same three major couplings as the wild
type sample, but the smaller couplings are no longer resolved. This result confirms
tryptophan involvement in the photoreduction reaction. Deuteration of the sample
changes hyperfine couplings throughout the transient EPR spectrum, as can be seen by
comparing Figure 2.8a with 2.8b. Because the entire spectrum is changed, the transient

spectrum probably arises from only the tryptophan radical, with no contribution from

55
flavin. If a flavin radical were also observed, some portions of the transient spectrum

would be unaffected by tryptophan deuteration.

Figure 2.9 shows the integrated transient absorption spectrum of the wild type and
deuterated samples whose first-derivative spectra are shown in Figure 2.8. A radical
whose spin energy levels are populated according to a Boltzman distribution shows all
absorptive transitions. Figure 2.8 shows that, for both the wild type and deuterated
samples, some of the transitions are emissive in the transient EPR spectrum. Emissive
transitions are characteristic of an EPR spectrum that is electron spin polarized due to

interactions between two members of a radical pairzw’.

It was shown above that the
transient, spin polarized EPR signal probably arises from a tryptophan radical. The model
in Figure 1.6 proposes that the transient radical that is formed will be trp' and a FADH;
species. In order for spin polarization to occur a radical pair must form. In the photolyase
photoreduction reaction, the most logical choice for the other radical pair member is a
triplet flavin radical because of spin conservation requirements.

The model in Figure 1.6 proposes the following scheme for photoreduction in
photolyase7. After the enzyme is photoexcited, an excited doublet flavin radical 2'FADH‘
intersystem crosses to the quartet. The quartet flavin 4'FADH accepts an H-atom from
trp-306 to form the transient radical whose EPR spectrum is shown in Figure 2.8. The
transient species that forms following hydrogen atom transfer is trp'--FADH2. The spin
polarized character of the transient EPR spectrum in Figure 2.8 suggests that the proposed

mechanism in Figure 1.6 must be modified to include the formation of a radical pair which

is necessary for spin polarization. The transient species that forms following hydrogen

56

Figure 2.9. Integrated transient spectra for a) wild type and b) deuterated photolyase
samples.

 

57

a) \Nlld type

 

 

 

 

b) indole-d5
1 1 l l I l l L J J 1 I l
3240 3260 3280 3300 3320 3340 3360
Gauss

Figure 2.9

58
atom transfer from trp-306 to 4'FADH' is expected to obey spin conservation. Before

hydrogen atom transfer, the overall spin of the two species is a quartet. If spin
conservation is preserved, the quartet character must be preserved after H-atom transfer.
In order to attain an overall quartet character for the radical pair, the flavin species must
be a triplet because of the doublet character of trp—306. Figure 2.10 shows the new
reaction mechanism for photoreduction in photolyase that includes an explanation for the
spin polarized state in photolyase"). Photoexcitation of FADH' produces an excited
doublet flavin species 2‘FADH‘. Within 100 ps intersystem crossing to the quartet occurs
and forms 4'FADH’. Hydrogen atom transfer from 4'FADH‘ produces the transient
3°FADH2--trp' species that is the state that produces the spin polarized EPR spectrum that
is shown in Figure 2.9. Intersystem crossing of 3'FADHz to the ground state (FADHz)
followed by H-atom transfer from an external donor produces the active (F ADHZ-
containing) enzyme. In the absence of external hydrogen atom donors the transient
F ADH2--trp- species recombines in several milliseconds to reform the inactive FADH'-
containing enzyme.

Triplet-doublet radical pair polarization in solution reactions has been reported and
is explained by the Radical Triplet Pair Polarization (RTPM)24’25'26. RTPM predicts a
polarization pattern that is half in emission and half in absorption with a superimposed net
polarization. The polarization pattern in Figure 2.9 is alternating in emission and
absorption with no net polarization. The absence of net polarization rules out RTPM as a

possible explanation of spin polarization that occurs during photoreduction of photolyase.

59

Figure 2.10. Proposed reaction mechanism for photolyase that is derived from the
experimental results reported here.

60

dark stable
2 FADH- 11" 2 *FADH-

 

l —’ l
i trpH trpH
, , intersystem
recombrnatron l crossing
Active
FADH2 4"'FADH-
I l
trpH trpH
H-atom transfer H-atom
transfer
intersytem
FADH2 “Sing 3*FADH2
| <— |

Figure 2.10

61
The alternating polarization pattern in Figure 2.9 is similar to the Correlated

Radical Pair Polarization (CRPP) that is observed for doublet-doublet radical pairs in

photosynthetic reaction centers”:l3

. The alternating pattern is proposed to arise from a
splitting of a transition into an emissive-absorptive doublet by the exchange energy
between the two radical pair members during the observation time by EPR. Non-zero
exchange interactions during EPR observation times have only been observed when the
two radical pair members are protein bound such as in photolyase and photosynthetic
reaction centers. Therefore, spin polarization in photolyase is proposed to arise from an
analogous polarization mechanism to CRPP that is extended to the triplet-doublet case.

The manifestations of this model and mechanism will be discussed in more detail in

chapters 3 and 4.

10.

11.

12.
13.
14.
15.
16.

17.

62
References

Sancar, A. Photolyase: In Advances in Electron Transfer Chemistry; Mariano,
P.E., Ed.; JAI Press: London, 1992; Vol. 2, pp. 215-272.

Kim. S.T.;Ma1hotra, K.; Sancar, A. In Biological Responses to Ultraviolet
Radiation; Urbach, F., Ed.; Valdenmar: Overland Park, KS, 1992.

Okamura, T.; Sancar, A.; Heelis, PF; Begley, T.P.; Hirata, Y.; Mataga, N. J. Am.
Chem. Soc. 1991, 113, 3143.

Yeh, S.-R. and Falvey, D.E. J. Am. Chem. Soc. 1991, 113, 8557.

. Payne, G.; Heelis, P.F.; Rohrs, B.R.; Sancar, A. Biochemistry 1987, 26, 7121.

Li, Y.F. and Sancar, A. Nucleic Acids Res. 1991, 18, 4885.
Heelis, P.F.; Okamura, T.; Sancar, A. Biochemistry 1990, 29, 5694.
Li, Y.F.; Heelis, P.F.; Sancar, A. Biochemistry 1991, 30, 6322.

Okamura, T.A.; Sancar, A; Heelis, P.F.; Hirata, Y.; Mataga, N. J. Am. Chem. Soc.
1989, 111, 5967.

Essenmacher, C.; Kim, S.-T.; Atamian, M.; Babcock, G.T.; Sancar, A. J. Am. Chem.
Soc. 1993, 115, 1602.

Kim, S.-T.; Sancar, A; Essenmacher, C.; Babcock, G.T. Proc. Natl. Acad. Sci. USA
1993, 90, 8023.

Norris, J .R.; Morris, A; Thumauer, M.C.; Tang, J. J. Chem. Phys. 1990, 92, 4239.
Bittl, R.; Schulten, K.; Turro, N.J. J. Chem. Phys. 1990, 93, 8260.

Sancar, A; Sancar, G.B. J. Mol. Biol. 1984, 172, 223.

Sancar, G.B.; Smith, F.W.; Sancar, A. Nucleic Acids Res. 1983, 11, 677.

Yerkes, C. Ph.D. dissertation 1982, Michigan State University, East Lansing, MI.

Hoganson, C.W.; Babcock, G.T. Biochemistry 1988, 27, 5848.

241

251

26. l

63
18. Gierer, M.; van der Est, A.; Stehlik, D. Chem. Phys. Letts. 1991, 186, 238.

19. Eker, A.M.P.; Kooiman, P.; Hessels, J .K.C.; Yasui, A. J. Biol. Chem. 1990, 265,
8009.

20. Heelis, P.F.; Payne, G.; Sancar, A. Biochemistry 1987, 26, 4634.

21. Malhotra, K.; Kim, S.-T.; Walsh, C.T.; Sancar, A. J. Biol. Chem. 1992, 267, 15406.
22. Jenks, W.S.; Turro, N.J. Rev. Chem. Int. 1990, 13, 237.

23. Adrian, F.J. Rev. Chem. Int. 1986, 7, 173.

24. Blatter, C.; Jent, F.; Paul, H. Chem. Phys. Lett. 1990, 166, 375.

25. Kawai, A; Okutsu, T.; Obi, K. J. Phys. Chem. 1991, 95, 9130.

26. Kawai, A; Obi, K. J. Phys. Chem. 1992, 96, 52.

Chapter 3
The trp-306 radical in photolyase is a cation radical suggesting that photoreduction

occurs through electron transfer rather than hydrogen atom transfer

Introduction

A proposed model for photoreduction of photolyase that includes an explanation
of the spin polarized character of the tryptophan radical is shown in Figure 2.10‘. In this
model, photoexcitation of FADH' produces an excited doublet statez, 2.FADH‘.
Intersystem crossing to the quartet occurs within 100 ps2 and is followed by hydrogen
atom transfer3 from an amino acid side chain donor that has been identified by time-
resolved EPR1 and optical studies as tryptophan-306‘. The transient radical pair that
forms contains a doublet tryptophan radical (trp‘) and triplet flavin (3'FADH2)‘. The
excited triplet flavin intersystem crosses to the ground electronic state and is followed by
reduction of trp-306 by an exogenous donor to form the active (F ADH; containing)
enzymez. In the absence of exogenous donors, FADH2 recombines with trp‘ to form the
inactive FADH' form of the enzymez.

The proposed active form of the enzyme (see Figure 1.3) contains a reduced flavin
adenine dinucleotide in its neutral dihydro form (FADH2)2‘5. Hartman and Rose studied
flavin model compounds and concluded that the anionic, singly protonated form is eight
times more active in dimer repair process than the doubly protonated neutral forrn6. These
results suggest that the active form of photolyase may contain FADH‘ rather than FADHZ
and that photoreduction in the enzyme occurs by an electron rather than hydrogen atom

64

65
transfer reaction mechanism7. Figure 3.1 shows possible reaction mechanisms for

hydrogen atom and electron transfer in photolyase. Figure 3.1a shows the previously

as. Photoexcitation of the inactive

proposed hydrogen atom transfer reaction mechanism
enzyme followed by hydrogen atom transfer from trp-306 to FADH' produces a neutral
tryptophan radical and FADHz. Exogenous donors reduce trpH- to form the active
enzyme that contains FADHz. An alternate reaction mechanism for photoreduction in
photolyase that involves electron transfer instead of hydrogen atom transfer is shown in
Figure 3.1b7. After photoexcitation of the enzyme, electron transfer from trp—306 to
FADH' produces a tryptophan cation radical and the catalytically active F ADH' 7
Reduction of trpH” by exogenous donors produces the active enzyme that contains
FADH‘.

Huckel-McLauchlin spin density calculations have been carried out for the cationic
and neutral forms of the trp radical8 and the results are shown in Figure 3.2. The
tryptophan cation radical has large spin density at the C(2) or-proton and at the C(3) B-
methylene carbon positions, while the neutral tryptophan radical has significant spin
density at the N(1) nitrogen and C(3) B-methylene carbon positions. Isotopic labeling of
tryptophan should allow an identification of the ionic character of the trp radical that
occurs during photoreduction of the enzyme.

In the work presented in this chapter, we show that photoreduction in photolyase

occurs through an electron transfer mechanism as shown in Figure 3.1b, rather than by the

hydrogen atom transfer mechanism that is shown in Figure 3.1a. The implication of the

66

Figure 3.1. Possible photoreduction mechanisms in photolyase.

67

Dark stable

FADH-
l hv
2"‘FADH-
intersystem
crossing
4*
a) Hydrogen FADH b) electron
atom transfer / \ transfer
FADHZ— FADH' — trp+
exogenous
donors
“Active”

FADHZ— up or FADH' —— trp

Figure 3.1

68

Figure 3.2. Calculated electron spin densities for the cation and neutral tryptophan
radicals.

69
a) neutral trp radical

CH2

0.39

0.09

  

0.03
0.04

N 0.30

0.06

0.08

b) cation trp radical

-0.04

0.09

 

-0.01 H

Hoflinan, B.M.; Roberts, J.E.; Kang C .H; Margolhsh, E.
J. Biol-Chem. 1981, 256, 6556.

Figure 3.2

70
results presented here is that the active form of the enzyme contains a flavin in its fully

reduced anionic form (FADH) instead of the previously proposed FADH; species.

Materials and Methods

The enzyme was prepared in the laboratory of Professor Aziz Sancar at the
University of North Carolina and, the preparation of the enzyme has been described
elsewhereg. Incorporation of isotopic labels into the enzyme and preparation of site
directed mutants were carried out as described previously7. 5 mM potassium ferricyanide
was added to the sample to prevent the reduction of the enzyme from the F ADH' form to
FADHz during the course of the experiment. Typical enzyme concentrations were 0.1
mM. The reaction buffer contained 50 mM potassium phosphate buffer pH=7 .0, 50 mM
NaCl, and 20% glycerol.

Photoexcitation of the sample was carried out with a Xenon flashlamp that was
critically damped with a pulse width of 17 us, a lamp energy of 50 J, and a repetition rate
of 1 Hz. It was determined that all flashes were light saturating (see Figure 2.7).

CW EPR measurements were carried out at 273 K on a Bruker ER200D
spectrometer that is fitted with a Varian TM011 mode cavity. The magnetic field was
measured with a Bruker ER035M NMR gaussmeter and the microwave fi'equency was
monitored with a Hewlett Packard HP 5245 counter \ 5255 frequency converter. A
microwave frequency of 9.22 GHz, a microwave power of 6.3 mW, a time constant of 35

us, and a modulation amplitude of 2.8 G was used to perform the experiments.

71
Time-resolved EPR measurements were obtained by the experimental setup

described in Chapter 21°. The field scan range was 100 G and a sweep time of 1000 s was
used. Conditions for the gated integrator were as follows: trigger rate of 1 Hz, 3
averages per point, a 4 us delay, and a 48us aperture. AC-coupling of greater than 10 Hz

was used to discriminate electronically against any stable signals that may be present.

Results

Figure 3.3b is a time-resolved EPR spectrum of the transient tryptophan radical
that forms during photoreduction of a photolyase sample that contains deuterium labeled
tryptophan at the positions indicated by the asterisks on the tryptophan structure in the
inset of the figure. The spectrum has the same apparent g-value (2.0028) as the wild type
sample whose transient spectrum is shown in Figure 3.3a. The transient spectrum in
Figure 3.3b has three large hyperfine couplings (~15 G) that are present in a 1:2:1 ratio
and an absorption spectrum that has an alternating emission and absorption pattern (data
not shown) that is similar to the absorption spectrum for the wild type sample that is
shown in Figure 2.9. The smaller hyperfine couplings of less than 5 G, which are present
in the wild type transient spectrum that is shown in Figure 3.3a, are not resolved in the
isotopically labeled sample whose transient spectrum is shown in Figure 3.3b. The dark
stable flavin spectrum of this isotopically labeled sample shows no apparent differences
from the wild type sample whose dark CW EPR spectrum is shown in Figure 2.4. Figure
3.3c shows a transient spectrum of the tryptophan radical in a photolyase sample that

contains tryptophan that is isotopically labeled with 1’N. The transient spectrum has the

72

Figure 3.3. Isotopically labeled samples of DNA photolyase. Isotopic labeling of the
samples is described in Kim, S.T.; Sancar, A.; Essenmacher, C. and Babcock, G.T. Proc.
Natl. Acad Sci. USA 1993, 90, 8023. Experimental conditions were: 6.3 mW microwave
power, 9.22226 GHz microwave frequency, 1000 sec scan, 40 scans, 2.8 G modulation
amplitude, 35 us time constant, 3220-3350 G microwave field, 3 averages, 1 Hz repetition
rate, 4 us delay, 48 us aperture, and AC-coupling of greater than 10 Hz.

 

73

CH2

it
? \ .
N
H
a) natural abundance trp

b) trp-indole-aC(2),C(5)-d2

c) trp-indole-N(1)-15N

 

3240 3260 3280 3300 3320 3340

Gauss
Figure 3.3

74
same apparent g-value as a wild type sample, and the dark stable CW EPR spectrum of the

flavin semiquinone radical shows no visible change from the wild type sample spectrum in
Figure 3.3a. The transient spectrum shows the same 15 G hyperfine coupling and
alternating emission and absorption pattern as the wild type sample. The smaller
couplings in the wild type sample have changed due to the isotopic substitution with 13N in
this sample and suggest that some of the smaller couplings arise from unpaired spin
density at the nitrogen in the tryptophan structure.

Figure 3.4 shows transient spectra for various site-directed mutant samples.
Figure 3.4b and 3.4c show transient spectra for W306Y and W306F mutants, respectively.
Neither mutant is able to form a transient radical as can be seen by comparing Figures 3.4b
and 3.4c to the wild type sample in Figure 3.4a. Figure 3.4d and 3.4c show transient
spectra for W157F and W418F mutant samples. These mutant samples have transient
spectra that are virtually identical to the wild type sample in Figure 3.4a. Protein
conformational changes due to a mutation at a site that is not involved in photoreduction
can be ruled out as the reason for the absence of transient EPR spectra in the W306Y and
W306F mutants because the W157F and W418F mutants still show a transient spectrum.
All mutants have a dark stable flavin spectrum that is indistinguishable from the wild type
sample. Therefore, it can be concluded that the reason for the absence of a transient EPR
signal in the W306F and W306Y mutants is that trp-306 is the radical that is observed in
the EPR spectrum and that phe and tyr are not able to be oxidized by the flavin doublet

radical, apparently due to a difference in redox potentials of these amino acids.

75

Figure 3.4. Tryptophan mutants of DNA photolyase. Samples preparation is described in
Kim, S.T.; Sancar, A; Essenmacher, C. and Babcock, G.T. Proc. Natl. Acad Sci. USA
1993, 90, 8023. Experimental conditions were: 6.3 mW microwave power, 9.22226 GHz
microwave frequency, 1000 sec scan, 40 scans, 2.8 G modulation amplitude, 35 us time
constant, 3220-3350 G microwave field, 3 averages, 1 Hz repetition rate, 4 us delay, 48
us aperture, and AC-coupling of greater than 10 Hz.

 

76

a) Wild type

b) W306Y mutant

W

e) W306F mutant

d) W157F mutant

“MW

e) W418F mutant

1 l l l l I I l l 1 I

3240 3260 3280 3300 3320 3340
Gauss

Figure 3.4

77
Discussion

The mechanism for photoreduction in photolyase that is shown in Figure 2.10
predicts that the transient radical that forms during the reaction is a neutral tryptophan
doublet radical that is formed by hydrogen atom transfer from trp-306 to FADH‘. The
ionic character of the tryptophan radical in photolyase has been determined by optical
spectroscopy”, but this technique cannot unambiguously distinguish between a cation and
neutral radical in photolyase samples. Figure 3.1 shows two possible reaction schemes for
photoreduction in photolyase. Figure 3. la shows hydrogen atom transfer from trp-306 to
FADH' which produces a neutral tryptophan radical and FADHz'. Figure 3.1b shows an
electron transfer mechanism from trp-306 to FADH' to produce a transient cation
tryptophan radical and FADHJ. A comparison of the time-resolved EPR spectra of
isotopically labeled samples with the transient EPR spectrum of the wild type sample in
Figure 3.3a should determine the mechanism of photoreduction in photolyase. If
hydrogen atom transfer is the mechanism for photoreduction, a neutral tryptophan radical
is predicted. If electron transfer occurs in photoreduction, a cation tryptophan radical
should be present.

Huckel-McLaughlin spin density calculations have been performed for the cation
and neutral tryptophan radicals8 and are shown in Figure 3.2. Spin density calculations
have been used previously and the results have been shown to be quite reliable when they

11.12.13.14

are used in a qualitative manner . The cation and neutral forms of the tryptophan

radical show significant differences in spin density distributions, and these differences

78
allow a determination of the ionic character of the transient radical that forms during

photoreduction by isotopically labeling photolyase samples. The cation radical has large
spin density at the C(2) and C(3) carbons while all the other positions do not show any
significant spin density. The neutral tryptophan radical shows significant spin density at
the C(3) carbon and N(1) nitrogen positions. The hyperfine coupling in an EPR spectrum
is directly proportional to the spin density at that position for nitrogen atoms. Coupling
due to or-protons ( protons that are bonded to a carbon in the aromatic ring) is given by

15.16

the McConnell relation that is shown in equation 3.1.

A... Qp equation 3.1

p is the spin density at the adjacent carbon, Q is the McConnell constant and is -25 G for
or-protons, and Ad is the experimentally observed hyperfine coupling constant. Coupling
due to B-methylene protons ( protons that are bound to a methyl carbon that is bonded to

a ring carbon) is given by equation 3.2.

A9 = 58 p c0320 equation 3.2

p is the spin density at the ring carbon, AB is the observed hyperfine coupling in Gauss,
and 6 is the dihedral angle between a plane that contains the ring carbon, the methyl
carbon, and a perpendicular to the ring plane and a plane that contains the ring carbon, the

methyl carbon, and the B-methylene proton as illustrated in Figure 3.5. Figure 3.6 shows

79

Figure 3.5. Angles used for calculation of hyperfine coupling constants for B-methylene
protons of tryptophan.

80

4- perpendicular to ring plane

H i H

 

B-methylene
carbon

ring carbon
pz orbitals

 

Figure 3.5

Figure 3.6. Tryptophan numbering scheme.

81

82

C(5)

C(6)

 

Figure 3.6

83
the numbering scheme for tryptophan. Both the neutral and cation tryptophan radicals

have significant spin density at the C(3) B-methylene carbon and, as a result, deuteration
of the B-methylene protons will not help in the detemrination of the ionic character of the
transient radical. Specific deuteration of the C(2) or-proton will distinguish between the
cation and neutral tryptophan radicals. Specific deuteration at the C(2) position is not
expected to alter the spin density distribution of the transient radical. Deuterium has a Q
constant ( see equation 3.1) that is one-sixth that of hydrogen. Therefore, if the transient
radical is a cation, implying an electron transfer mechanism, hyperfine couplings that are
due to this proton should be six times smaller when deuterium has replaced hydrogen and
significant changes to the experimental spectrum are expected. If the transient radical is a
neutral tryptophan radical (hydrogen atom transfer), the transient spectrum should remain
largely unchanged because the spin density at C(2) is insignificant.

Figure 2.8b shows a transient EPR spectrum for a sample that was specifically
deuterated at the or-ring proton positions that are indicated by the asterisks on the
tryptophan structure in the inset to the figure. The larger hyperfine couplings (~15 G) that
are present in a 1:2:1 ratio in the wild type sample (solid arrows) in Figure 2.8a are also
observed in the deuterated sample in Figure 2.8b. The smaller hyperfine couplings
(dashed arrows in Figure 2.8a) that are present in the wild type sample are not resolved in
the isotopically labeled sample in Figure 2.8b. The McConnell constant, Q in equation
3.1, is approximately -4 G for an or-deuteron, one-sixth the value of Q for an (Jr-proton.
The difference in energy in magnetic field units between the dashed arrows in Figure 2.8a

is approximately 5 G. Upon isotopic substitution with deuterium, the smaller hyperfine

84
couplings (~5 G) that are present in the wild type sample are no longer resolved in the

deuterated sample. Therefore, the smaller hyperfine couplings (dashed arrows in Figure
2.8a) can be attributed to hyperfine coupling to one or more or-protons that were replaced
with deuterium in the isotopic labeling experiment. The hyperfine coupling of an or-
deuteron in the sample whose spectrum is shown in Figure 2.8b will be less than 1 G
(calculated with Q = -4 G in equation 3.1). A hyperfine coupling of this magnitude (< l
G) is not expected to be resolved because the linewidth of the EPR transitions in
photolyase are at least 3 G. Therefore, the assignment of the smaller hyperfine couplings
to one or more or-protons is consistent with the experimental observation that replacement
of or-protons with deuterium results in the loss of resolution of the smaller hyperfine
couplings in the transient spectrum in photolyase.

A hyperfine coupling due to an or-proton on the tryptophan ring suggests that the
transient EPR spectrum in Figure 2.8a arises from a cation rather than a neutral
tryptophan radical. The calculated spin density distribution for the neutral tryptophan
radical that is shown in Figure 3.2 shows that no significant spin density exists at any of
the or-proton positions. The absence of spin density at these positions rules out a neutral
tryptophan radical as a possible source of the transient EPR spectrum in Figure 2.8a. The
spin density distribution for the cation tryptophan radical is also shown in Figure 3.2 and
shows that a spin density of 0.42 is estimated at the C(2) or-proton position. Hyperfine
coupling due to the or-proton at the C(2) position on the tryptophan cation radical is
calculated using equation 3.1 to be 10 G. the calculated hyperfine coupling value due to

the C(2) or-proton is in agreement with the experimentally observed value of 5 G when

85
consideration is given to the qualitativeness of spin density calculationsmz‘l3 ’14. The

results of the spin density and hyperfine coupling calculations support cation tryptophan
radical formation during photoreduction of photolyase and also suggest that the 5 G
hyperfine coupling that is present in the wild type EPR transient spectrum in Figure 2.8a
arises fiom coupling to the C(2) or-proton.

In order to test firrther the ionic nature of the transient tryptophan radical that is
observed during photoreduction of the inactive enzyme, a sample that contained
tryptophan that was isotopically labeled with deuterium at the positions indicated by the
asterisks on the tryptophan structure in the inset to Figure 3.3 was prepared7. The sample
contains deuterium at the C(2) and C(5) positions. The transient EPR spectrum for this
isotopically labeled sample is shown in Figure 3.3b. The transient spectrum has an
apparent g—value of 2.0028 and a dark stable CW EPR spectrum that is indistinguishable
from a wild type sample. The transient spectrum in Figure 3.3b has the same hyperfine
couplings as the transient spectrum in Figure 2.8b that originated from a sample where all
five or-protons were replaced with deuterium. The large 15 G hyperfine coupling is
present in this sample, but the smaller 5 G hyperfine coupling (dashed arrows in Figure
2.8a) that is present in the wild type sample is not resolved when the C(2) and C(5) or-
protons are replaced with deuterium. Since only the C(2) and C(5) or-protons are replaced
with deuterium in the sample whose transient spectrum is shown in Figure 3.3b, the
smaller 5 G hyperfine coupling that is present in wild type samples must arise from
coupling to one or both of the or-protons at these positions. Figure 3.2 shows that neither

the neutral or cation tryptophan radical has significant spin density at the C(5) position;

86
therefore, the 5 G hyperfine coupling in the wild type transient spectrum most likely arises

fi'om coupling to an or-proton at the C(2) position. Hyperfine coupling that arises from a
C(2) or-proton is consistent with tryptophan cation radical formation during
photoreduction of the inactive enzyme.

If the transient radical that forms during photoreduction of inactive photolyase is a
tryptophan cation radical as is predicted by the transient spectra of the photolyase samples
that were isotopically labeled with deuterium (see Figure 3.3a and Figure 3.3b), the large
15 G hyperfine coupling that is present in the wild type transient spectrum in Figure 3.3a
will arise from coupling to the two B-methylene protons at the C(3) carbon. The
hyperfine coupling pattern for the wild type transient spectrum show three transitions that
are split by 15 G and are in a 122:] ratio suggesting that these large hyperfine couplings
arise fi'om two equivalent nuclei. If the 15 G hyperfine coupling in the wild type spectrum
are due to coupling to the two B—methylene protons at the C(3) carbon, the coupling will
be proportional to the dihedral angle between a normal to the tryptophan ring plane and
the B-methylene proton (see equation 3.2 and Figure 3.5). In order to observe identical
hyperfine couplings for the two B-methylene protons the dihedral angle, 6 in equation 3.2,
must be equal for the two protons. The geometry of the four substituents to the B-
methylene carbon is expected to be tetrahedral, and two possible spatial arrangements of
the ligands exist and are shown in Figure 3.5. Figure 3.7a shows a low energy
configuration where the two protons form a 60° angle with the dihedral plane and the
peptide linkage, denoted by R in the figure, lies axial to the tryptophan ring plane. The

geometry for the B-methylene carbon ligands that is shown in Figure 3.7a predicts a spin

87

Figure 3.7. Possible geometries for tryptophan B-methylene protons.

88

a) Low Energy conformation

H 60°

 

 

b) High Energy conformation

HK'
30°

 

R

/ pz orbital

30

 

Figure 3.7

89
density of one at the C(3) carbon (see equation 3.2) when a hyperfine coupling of 15 G is

observed experimentally. The spin density distribution that is shown in Figure 3.2 for the
cation radical predicts a spin density of 0.39 at the C(3) carbon and suggests that the
geometry of the ligands must be different than the one that is shown in Figure 3.7a.
Figure 3.7b shows a high energy geometry of the ligands where the two B-methylene
protons lie on the same side of the pz orbital and form a 30° angle with the dihedral plane.
In this geometry the peptide linkage lies equatorial to the tryptophan ring plane. The
geometry in Figure 3.7b predicts a spin density of 0.34 at the C(3) carbon and is in much
better agreement with the experimentally observed hyperfine coupling of 15 G in the wild
type transient spectrum.

Another possibility for the three transitions that are split by 15 G in the wild type
spectrum in Figure 3.3a is that the three transitions are in a 1:1:1 ratio that arises from the
N(1) nitrogen with a nuclear spin of one. The neutral tryptophan radical has sufficient
spin density at N(1) to produce a hyperfine coupling of 15 G but the cation does not (see
Figure 3.2). Isotopic substitution of 1"N with 1’N in tryptophan should show the origin of
the large 15 G coupling. If the transient radical in photolyase is a cation, as the deuterated
samples (Figure 2.8b and 3.3b) predict, the three line pattern and its coupling should not
significantly change when isotopic substitution of 14N with 15N is performed. If the
transient radical is a neutral tryptophan, the three line pattern should collapse to a two line
pattern with different couplings because of the difference in nuclear spin of these two

isotopes (nuclear spin = 1/2 for 15N).

90
The transient EPR spectrum7 in Figure 3.3c arises from a sample in which l4N

tryptophan is replaced by 15N. The major three line pattern is still present in a 1:221 ratio
as in the wild type sample in Figure 3.3a. The smaller couplings (5 G) in the wild type
sample are affected by this isotopic substitution which suggests that the hyperfine coupling
due to nitrogen is less than 5 G. The small hyperfine coupling due to nitrogen gives
further support for the transient radical being a cation and for the large 15 G coupling
being due to the two [Fl-methylene protons that are in the high energy configuration that is
shown in Figure 3.7b.

The isotopic substitution experiments presented here, in addition to the spin
density distribution calculations in Figure 3.2, have shown that there is a large spin density
at C(2) and C(3) and a much smaller density at the N(1) nitrogen. The results show that
the transient radical that forms during photoreduction of inactive photolyase is most likely
a tryptophan cation radical. The results also show that the geometry of the ligands to the
B-methylene carbon is probably the high energy configuration that is shown in Figure 3.7b
where the peptide linkage is equatorial to the tryptophan ring plane base upon equation
3.2 and a 15 G hyperfine coupling that is observed experimentally.

Optical studies have suggested that tryptophan-306 is the internal H-atom donor to
FADH' during photoreduction of the enzyme”. Site-directed mutagenesis was used to
confirm this assignment”. Figure 3.4 shows various photolyase mutant samples. Figure
3.4b and 3.4c show the W306Y and W306F mutant transient spectra. Neither mutant
exhibits any transient radical formation, but both samples still show a dark stable flavin

EPR spectrum that is indistinguishable from the wild type sample (Figure 2.4). The

91
absence of a transient photo-induced EPR signal in these mutants confirms the assignment

of tip-306 as the internal H-atom donor in photolyase. The transient EPR signals in
Figure 3.4d and 3.4e arise from the W157F and W418F mutants. Both mutants show a
dark stable flavin and transient spectra that are indistinguishable from the wild type sample
as can be seen by comparing Figure 3.4d and 3.4c to Figure 3.4a. This result shows that
any long range effects due to a mutation at a site that is different from the site of
photoreduction does not affect the ability of the enzyme to photoactivate.

The assignment of trp3oe as the internal amino acid donor during photoreduction of
inactive photolyase agrees well with the recently reported crystal structure for the
enzyme“. The structure shows that trp306 is located 13 A away from the flavin
chromophore, and this residue is located near the surface of the protein”. During
photoreduction of inactive photolyase, the tryptophan residue reduces FADH'3. An
exogenous donor subsequently reduces tryptophan to reactivate the enzyme3 . In order for
tryptophan reduction by an exogenous donor to occur, the residue must be accessible to
the solvent. The location of “13306 near the protein surface makes it an ideal candidate as
the internal arrrino acid donor during photoreduction of inactive photolyase because
exogenous donors will have access to the site.

The mechanism for photoreduction that is shown in Figure 2.10 does not account
for the experimental observations here. In order for a tryptophan cation radical to form,
electron transfer from tfpsoé to FADH' must occur not H-atom transfer as previously
suggested. Figure 3.1b outlines a possible mechanism for electron transfer in photolyase.

Following photoexcitation of the enzyme, an excited flavin doublet state forms 2'FADH'.

92
Intersystem crossing to the quartet followed by electron transfer produces the transient

radical pair which is a trp‘+ and F ADH.

The results presented here also suggest that the active form of the enzyme contains
FADH' and not FADHz as shown in Figure 2.10. Hydrogen transfer to FADH' could
occur at later times to produce FADHz, but this is not expected. Hartman and Rose have
studied dimer repair with flavin model compounds‘. Their results show that the anionic
singly protonated flavin is eight times more efficient at dimer repair than the neutral
doubly protonated flavin. Furthermore, the fully reduced form of photolyase has an
optical spectrum'9 that closely resembles the spectrum of the anionic, singly protonated
form of flavin mononucleotide (FMNH') rather than the doubly protonated form
FMNH220. Therefore, FADH‘ is much more likely candidate for the active form of the
enzyme considering model compound work and the presence of FADH‘ after electron
transfer in the photoreduction process in photolyase.

Figure 3.8 shows a proposed mechanism for photoreduction that takes into
account the experimental observations presented here. Photoexcitation of the dark stable
FADH' species produces an excited doublet, 2'FADH‘. In 100 ps, the flavin intersystem
crosses to the quartet, "FADH. Within 1 us electron transfer from trp306 produces a
trpH'+---3'FADH' radical pair. Intersystem crossing to the ground state by the flavin
produces FADH'. Without exogenous donors, the radical pair recombines over several
milliseconds. In the presence of external donors trpH'+ is reduced to form the active

enzyme that contains FADH'

93

Figure 3.8. Proposed mechanism for photoreduction in photolyase based on the
experimental results presented in this chapter.

94

dark stable
I ——>
t113306
recombination
electron
3*FADH' transfer
| ‘—
trp -+
306
donors
“Active”
FAIDH'
trP306

Figure 3.8

2"'FADH-

trP306

intersystem
crossing

4"'FADH

t1'P306

95
This study has shown by isotopic labeling that photoreduction in photolyase occurs

by an electron transfer rather than H-atom transfer mechanism. These results also imply

that the active form of the enzyme is FADH' and not FADHz as previously proposed.

9.

10.
ll.

12.

l3.

14.

15.

16.

17.

96
References

. Essenmacher, C.; Kim, S.T.; Atamian, M.; Babcock, G.T.; Sancar, A. J. Am. Chem.

Soc. 1993, 115, 1602.

Okamura, T.A.; Sancar, A.; Heelis, P.F.; Hirata, Y.; Mataga, N. .1. Am. Chem. Soc.
1989, 111, 5967.

. Heelis, P.F.; Okamura, T.; Sancar, A. Biochemistry 1990, 29, 5694.
. Li, Y.F.; Heelis, P.F.; Sancar, A. Biochemistry 1991, 30, 6329.

. Kim, S.T.; Sancar, A.; Essenmacher, C.; Babcock, G.T. J. Am. Chem. Soc. 1992, 114,

4442.

. Hartman, R.F.; Rose, S.D. J. Am. Chem. Soc. 1992, 114, 3559.

Kim, S.T.; Sancar, A; Essenmacher, C.; Babcock, G.T. Proc. Natl. Acad. Sci. USA
1993, 90, 8023.

. Hoffman, B.M.; Roberts, J .E.; Kang, C.H.; Margoliash, E. .1. Biol. Chem. 1981, 256,

6556.

Sancar, G.B.; Smith, F.W.; Reid, R.; Payne, G.; Levy, M.; Sancar, A. J. Biol. Chem.
1987, 262, 478.

Hoganson, C.W.; Babcock, G.T. Biochemistry 1988, 27, 5848.
McLachlan, AAD. M01. Phys. 1960, 3, 233.

Streitweiser, A, Jr. Molecular Orbital Theory for Organic Chemists 1961 (Wiley,
New York).

Tople, J.A.; Beveridge, D.L.; Dobosh, P.A. J. Chem. Phys. 1967, 47, 2026.
Tople, J .A.; Beveridge, D.L.; Dobosh, PA. .1. Am. Chem. Soc. 1968, 90, 4201.
McConnell, HM. J. Chem. Phys. 1956, 24, 764.

McConnell, H.M.; Chesnut, D.B. Chem. Phys. Letts. 1958, 28, 107.

Payne, G.; Heelis, P.F.; Rohrs, B.R.; Sancar, A. Biochemistry 1987, 26, 7121.

97
18. Park, H.W.; Kim, S.T.; Sancar, A. and Deisenhofer, J. Science 1995, 268, 1866.

19. Heelis, P.F.; Payne, G.; Sancar, A. Biochemistry 1987, 26, 4634.

20. Ghisla, S.; Massey, V.; Lhoste, J.-M.; Mayhew, S.G. Biochemistry 1974, 13, 589.

Chapter 4
The Photoreduction Mechanism in DNA Photolyase: Correlated Triplet-Doublet

Radical Pair Polarization Arising From Differences in Hyperfine Frequencies

Introduction

A current model for the photoreduction / photoactivation process that incorporates
the results in Chapters 2 and 3 is shown in Figure 3.8. Briefly, the model suggests that the
photoreduction reaction is initiated by photoexcitation of the FADH' chromophore and
leads to an electronically excited doublet state (Z‘FADH‘)2. Intersystem crossing to the
quartet ("FADH) is postulated to occur within 100 ps3 and is followed by electron
transfer from a nearby amino acid residue in 1 us. Flash photolysis and time-resolved EPR
studies with site-directed mutants and isotopically labeled samples have identified this

residue as tryptophan-306 in its cationic form'”.

In the absence of exogenous donors,
the transient species that forms following electron transfer (F ADI-1' --2trp3oe'+) recombines
over 7 ms6 to produce the catalytically inactive FADH' form of the enzyme. One electron
reduction of the tryptophan cation radical when exogenous donors are present gives rise
to the active FADH' form of the enzyme”. In an earlier version of the photoreduction /
photoactivation process, electron transfer from the excited doublet state flavin was
postulated.

We showed previously that the transient species that forms following electron
transfer from trp306 in the photoreduction reaction is spin polarized due to magnetic

interactions between a triplet flavin anion and doublet tryptophan cation radical pair’.

98

99
Other triplet-doublet radical pair polarizations have been reported and the resulting

polarization patterns have been explained by the Radical Triplet Pair Mechanism
(RTPM)? Hyperfine dependent polarization in combination with a net intensity for the
spectrum are characteristic for this type of spin polarization”, but photolyase does not
exhibit any net polarization’ and, as we concluded previously, this indicates that RTPM
cannot explain the spin polarization that is observed in the photolyase system. The time-
resolved EPR spectrum of the transient species shows a polarization pattern that is
alternating in emissive and absorptive transitions and is similar to Correlated Radical Pair
Polarization (CRPP) observed in doublet-doublet radical pairs in photosynthetic reaction

centers""l ‘

. Because of the similarity in polarization patterns, we propose an analogous
polarization mechanism for triplet-doublet radical pairs called Correlated Triplet-Doublet
Radical Pair Polarization (CTDRPP)5. We present here a description of CTDRPP and
show the interactions that are responsible for this type spin polarization in triplet-doublet
radical pair systems. Interpretation of the data we have obtained for photolyase within

this model allows us to modify the reaction mechanism in Figure 3.8 so that electron

transfer proceeds directly from the flavin excited doublet state.

Materials and Methods

Preparation of DNA photolyase has been described previously‘z’”.

To prevent
reduction of FADH' to FADH; during the course of the experiment 5mM potassium

ferricyanide was added to the sample. Typical enzyme concentrations were 0.1-0.2 mM.

100
Flash photolysis of the enzyme was performed with a Xenon flashlamp with a 17 us pulse

width and a lamp energy of 50 J. It was determined that all flashes were near saturating.

Time-resolved EPR measurements were made by using a Bruker ER200D
spectrometer at a temperature of 277 K. A 300 pl flat cell (W ilmad, Buena, New Jersey)
was used to hold the sample and was placed in a Varian TM011 mode cavity that was
modified to be compatible with the Bruker spectrometer”. The microwave fi'equency was
measured with a Hewlett-Packard 5245 frequency counter / 5255 fi'equency converter and
the magnetic field was monitored with a Bruker ER03 5M NMR gaussmeter. All
experiments were performed at a microwave frequency of 9.22 GI-Iz, a modulation
amplitude of 2.8 G, a time constant of 35 us, and a microwave power of 6.3 mW, unless
otherwise noted.

Detection of the rise and decay kinetics of the transient radical following flash
photolysis of the flavin radical-containing enzyme has been described previously".
Transient spectra were recorded with an IBM data acquisition card that is interfaced to a
personal computer. A Stanford Research Systems SR250 gated integrator and boxcar
averager (Palo Alto, California) was used to electronically integrate the transient signal
and to trigger the Xenon flashlamp. A 4us delay and 48 us aperture were used to
integrate the transient signal and obtain the amplitude of the signal as a firnction of the
magnetic field. The field was stepped slowly at a rate of 0.1 Gauss/sec and the flashlamp
was pulsed at 1 Hz to obtain the amplitude of the transient at 1000 equally spaced

magnetic field positions.

1 01
Results

Photoexcitation of the dark stable FADH' species produces an intense transient
EPR signal in the g=2 region that we previously assigned to the tryptophan-306 cation
radicals. The kinetic trace of the transient radical was shown in Fig. 2.5. The transient
radical shows a half time of 35 us. The 35 us component is not a true kinetic half time of
the transient radical but is instead the response time of the instrument because of the use
of 100 kHz magnetic field modulation. The decay kinetics of the transient radical are the
same regardless of the magnetic field position (data not shown), suggesting that the
transient radical arises from only the trp radical of the radical pair. This conclusion is in
agreement with previous observations that show that specific deuteration of tryptophan
changes hyperfine couplings throughout the transient EPR spectrum’.

The transient, first derivative EPR spectrum for the radical whose kinetic trace is
shown in Figure 2.5 is shown in Fig. 2.8a. Three major peaks are present (solid arrows in
Fig. 2.8a) in a 1:211 ratio and are split by 15 G. We have previously shown that these
transitions arise from coupling to two B-methylene protons on the transient tryptophan
cation radical”. The fact that the two B-CH2 protons have identical hyperfine couplings
to within our detection limits indicates that their dihedral angles with respect to the C(3)
pz orbital that carries the unpaired spin density are equal and have values of 30° (see
Chapter 3). Smaller splittings (< 5 G) superimposed on the major hyperfine couplings
originate fiom couplings to an or-proton at the C(2) position of the tryptophan ring and to

nitrogen at the N(1) position”. Figure 2.8b shows a spectrum that originates from an

102
enzyme sample in which tryptophan was specifically deuterated at the or-ring positions.

The loss of hyperfine structure confirms the tryptophan origin of the radical and allows us
to assign couplings to the or-protons of the ring in the radical.

The integrated, absorption spectra of the two transient first-derivative spectra in
Fig. 2.8 are shown in Figure 4.1. Figure 4.1a is the transient spectrum for the wild type
sample, while Figure 4.1b is the corresponding spectrum for the tryptophan ring
deuterated sample. Six major transitions can be identified in each spectrum and they
alternate between emission and absorption. The major 15 G splitting that we observe in
the transient, first-derivative in Fig. 2.8 and assign as the -CH2 hyperfine coupling Aflp is
also present in the absorption spectra in Figure 4.1, and is observed as the splitting
between each absorptive-emissive pair of transitions (shown as solid arrows in Figure
4.1a). Each of these pairs is further split into an emissive and absorptive component
(dashed arrows in Figure 4.1a) that are separated by approximately 7.5 G. This splitting is
identified as the exchange coupling, J, below. Thesmaller couplings ( < 5 G) that are
observed in Figure 2.8a are also present in Figure 4.1a as shoulders to the six major

transitions.

Discussion

Electron spin polarization arises fi'om the magnetic interaction between two

members of a radical pair and the spin polarized spectrum shows emissive and absorptive

transitions that are enhanced in intensity due to a non-Boltzman population difference

103

Figure 4.1. Transient absorption spectra for a)wild type and b) deuterated photolyase
samples. The spectra were obtained by numerically integrating the transient first-
derivative EPR spectra in Figure 2.8a and Figure 2.8b.

ma

 

104

 

a) WIld type

 

 

 

b) indole-d5

l l I l l l l a l l l a J

 

3240 3260 3280 3300 3320 3340 3360

Gauss
Figure 4.1

105

'6’". We have shown that the transient tryptophan

between the magnetic energy levels
radical in Figure 2.8 is spin polarized; the most obvious manifestation of the non-
thermalized character of the transient radical is the presence of emissive transitions in the
time-resolved spectrum’. The absorption spectrum of the transient radical that is shown in
Figure 4.1a clearly shows the emissive character of certain transitions and supports the
spin polarization mechanism that has been proposed’. The isolated, inactive enzyme
contains a doublet flavin radical, FADH', that abstracts an electron from trp—306 when the
enzyme is photoexcited"‘8. The transient, spin polarized EPR signal in Figure 2.5 that
arises from photoexcitation of the inactive enzyme has been shown to originate from only
a doublet tryptophan-306 cation radical’- In order for spin polarization to occur in
photolyase, a radical pair must form following photoexcitation of the enzyme. A likely
candidate for the other radical pair member is a triplet FADH' species because spin
conservation is expected following electron transfer. Any other choice for the other
radical pair member will not allow electron spin conservation or spin polarization to occur.
Therefore, we have proposed that the second member of the radical pair in photolyase is a
triplet FADH‘ species that is not detectable by room temperature EPR”. Consistent with
this interpretation is the fact that previous studies of triplet flavin radicals have shown that
their detection by EPR at room temperature is difficult”. The identification of excited
flavin intermediates in the photoreduction reaction of the inactive enzyme by optical
spectroscopy gives further support for triplet flavin radical formation in inactive
photolyase during photoreduction 1’3"”).

The energy level diagram for the coupled basis set of a triplet-doublet radical pair

and the individual magnetic states that lead to these energy levels are shown in Figure 4.2.

106

Figure 4.2. Magnetic states of triplet-doublet radical pair and the energy levels and state
mixing in the coupled basis set that are formed fi'om interactions within the radical pair.

107

Q3/2

Id stilt

 

rat. Q-s/z
\ 3..“ TI +D-l/2 ,T0+D1,2
\,\ \ T. + D“. . T0 + D...
/ / T-l + 13.112
2] / 7

Dl/2

I

g 1 H
D-l/2
Radical separation —————>

Kawai, A; Okutsu, T. and Obi, K. J. Phys. Chem.
1991, 95, 9130.

Figure 4.2

~ 108
Figure 4.2 shows the magnetic substates of the triplet and doublet radicals that lead to the

coupled basis energy levels. The TH and Du; form the quartet state with an electron spin
S2 of 3/2 (Q30) while the T4 and D4,; levels form the Q30 state in the coupled basis set of
the triplet-doublet radical pair. The Q10, QM, D.m, and D-m states in the coupled basis
set are formed by the magnetic states of the triplet and doublet radicals as is shown in
Figure 4.2. Figure 4.2 also shows the energy level diagram for a coupled triplet-doublet
radical pair. The four quartet states are separated by the Zeeman energy, gBH, where g is
the electronic g value, B is the Bohr magnetron, and H is the magnetic field. The doublet
states are also separated by the Zeeman energy, and the difference in energy between the
doublet and the quartet (difference between D10 and Qm) is the exchange interaction
energy 2] between the doublet and triplet radicals of the radical pair.

Net polarization in triplet-doublet radical pairs. Triplet-doublet radical pairs in non-
biological solution systems show electron spin polarization that is explained by the Radical
Triplet Pair Mechanism (RTPM)7‘8’9. The spin-polarized EPR spectrum of a triplet-
doublet radical pair that exhibits RTPM shows a polarization pattern that is dependent
upon the nuclear spin quantum and a superimposed net absorptive or emissive
component”. Net emissive polarization arises from state mixing between the -3/2 quartet
(QM) and the 1/2 doublet (DI/2) States in the coupled basis set of the triplet-doublet
radical pair, while net absorptive polarization arises from state mixing between the Q30
and D.1/2 states (see Figure 4.2). State mixing occurs because of a perturbation from the

zero field splitting energy that is shown in equation 4.18.

109
11...: ms“2 - 1 / 3542) + ns,; - $.32) (equation 4.1)

D and E are the zero field splitting parameters, and S is the electron spin operator. The
subscript T denotes the total spin and If) is the zero field axis system. The mixed state that

forms as a result of state mixing is shown in equation 4.2

"Q-3/2" = IQ-3/2> + AID|Q> (equation 4.2)

where ”Q32" is a linear combination of the Q30 and D12 states and A is given by equation

4.3

A = ( <Q.3/2 I Hm I D1/2> )/ (EQ - ED ) (equation 4.3)

where Hzfi is given by equation 4.1 and EQ - ED is the energy difference between the QM
and Dm states. For experiments performed at X-band fiequencies, the energy difference
between Q-” and D1,; is greater than 6000 G. In order for any significant state mixing
between the Q30 and Du) states to occur, A must be significant. The zero field splitting
energy is typically 600 G while the difference in Zeeman energy between the Q30 and Dm
states is on the order of 6000 G. These values give a value for A of 0.1. When the two
radicals of the radical pair diffuse to a region of high exchange energy, the energy
difference between the two states, EQ - ED, is reduced to a value that is much closer to the

zero field splitting energy Hzf, which allows significant state mixing in equation 4.2 to

l 10
occur. Figure 4.2 shows that, in the region of high exchange energy, the Q30 state is

converted to Dm and D112 is converted to the Q31; state. When the two radicals of the pair
separate to a region of zero exchange, only the quartet states are able to cross back to
separated triplet and doublet radicals”. The doublet states cannot cross back to the
region of separated radicals and result in triplet quenching. Triplet quenching produces
separated doublet radicals with enhanced or spin resulting in net emissive polarization in
the time-resolved EPR spectrum”. For absorptive polarization, the states involved are
Q3/2 and Dan states; the arguments then run parallel to those above for emissive
polarization”.

Hyperfine dependent component of RTPM. Hyperfine dependent spin polarization in
RTPM arises from state mixing between the Q10 and D1,; states and the Q40 and Dan
states in the coupled basis set of the radical pair8 (see Figure 4.3). Mixing between these
states is induced by the perturbation shown in equation 4.4 and arises from the different

precessional frequencies of each member of the radical pair.

0) = Ag + 2i A; 111.; - 21 A; m“ (equation 4.4)

0) is the sum of the difference in g-values (in frequency units) of the two radicals and the
sum of the differences in hyperfine frequencies where A is the hyperfine coupling constant
and ms is the nuclear spin quantum number. Equation 4.5 shows the mixing for the

eigenfunction that is represented by "Qt/2" in Figure 4.3.

111
"erz" = | Q1/2 > + Al Dm> (equation 4.5)

where A is shown in equation 4.6

A. = ( <QI/2 I (D I D1/2 > ) / ( EQ - ED) (equation 4.6)

where 0) is given by equation 4.4 and the denominator to equation 4.6 is the difference in
energy between the doublet and the quartet. The other mixed states in Figure 4.3 are
formed by analogous equations to equation 4.5 with 0) (see equation 4.4) being the
perturbation that causes the state mixing. The result is a spin-polarized, transient
spectrum that is half emissive and half absorptive, similar to the Radical Pair Mechanism
for doublet-doublet radical pairs'6"7.

RTPM does not explain polarization patterns in photolyase. Spin polarization in
photolyase cannot be explained by RTPM because this mechanism does not predict the
polarization pattern that is observed in photolyase’. The polarization pattern in Figure
4.1a alternates between emissive and absorptive transitions and is inconsistent with RTPM
where a transient spectrum that is half in emission and half in absorption is predicted for
the hyperfine dependent component. The absence of net polarization in Figure 4.1a also
precludes RTPM as a possible mechanism for spin polarization in photolyase and
differences between non-biological and protein samples contributes to the absences. Net

emissive polarization, as discussed above, results from state mixing between the Q30 and

Dm states”. Because the difference in energy between these states at X-band is large, a

112

Figure 4.3. Triplet-doublet state mixing from precessional fiequency differences in the
radical pair in the absence of an exchange interaction.

113

 

 

 

Figure 4.3

114
large exchange interaction is necessary to mix these states effectively. Previous triplet-

doublet radical pair polarizations occurred in solution reactions where diffusion of each
radical pair member to a region of high exchange energy is possible”. In photolyase, the
radical pair is protein bound, thus preventing any diffusion from occurring. Without
diffusion, the radical pair remains at a fixed separation throughout the reactions. The high
exchange energy that is needed for effective state mixing is not attainable with protein
bound radical pairs, and this results in no net polarization in photolyase.

Correlated Radical Pair Polarization. Doublet-doublet radical pairs that are transiently
present in the photosynthetic reaction centers show an alternating enrissive and absorptive

polarization pattern that is similar to the pattern in Figure 4.1a'0'".

Electron spin
polarization in reaction centers is explained by the Correlated Radical Pair Polarization
(CRPP) mechanism‘o’". CRPP predicts four transitions that are alternating in emission
and absorption for a doublet-doublet radical pair when there are no resolved hyperfine
couplings. Figure 1.8b shows the energy levels for the coupled basis set of the doublet-
doublet radical pair. State mixing between the singlet and zero triplet energy levels
produces the eigenfunctions "So" and "To" in the figure by the perturbation 0) (A!) in
Figure 1.8b) that is shown in equation 4.4. In order for CRPP to occur in a doublet-
doublet radical pair, the two radicals must have a non-zero exchange interaction energy
during the observation time by EPR“). Figure 1.8b illustrates pure CRPP for a doublet-
doublet radical pair when no hyperfine coupling is observed. In the figure a positive

exchange interaction (singlet is higher in energy than the triplet) and a singlet precursor

are assumed. Initially when the precursor state is a singlet only the "So" and "To" energy

115
levels are populated because these eigenfirnctions are the only ones that have any singlet

character. The "S0" energy level receives more of the initial population because it has
more singlet character than the "To" energy level. The four transitions that arise from a
CRPP mechanism with a singlet precursor and a positive exchange interaction are shown
at the bottom of Figure 1.8b. The transition that occurs at the lowest magnetic field is the
So—rT- transition and, as the figure illustrates, this transition will be in emission. The other
three transitions arise from the energy levels that are depicted in the figure. The result is a
spin polarized EPR spectrum with equally intense lines and an E/A/E/A pattern (E denotes
emission and A absorption). The two low field transitions (So—)T. and To—rT.) arise from
one radical of the radical pair and the difference in energy between these two anti-phase
transitions is the exchange interaction between the two members of the radical pair. The
two high field transitions (To—)T- and So—-)T.) also form an anti-phase E/A pattern with
the difference in energy being the exchange interaction between the two radical pair
members. These two transitions arise fiom the other member of the radical pair. The
difference in energy between each of these anti-phase doublet E/A pairs is the difference in
precessional frequencies of the two radicals in the radical pair and is given by (0 (A!) in the
figure) in equation 4410'". If the sign of the exchange interaction or the precursor state is
changed an opposite phase A/E/A/E pattern is predicted by CRPP").

The equality in the intensity of the EPR transitions for pure CRPP as shown in
Figure 1.8b arises from spin conservation of the radical pair"). For a singlet precursor, the
amount of singlet character in "So" and "To" determines how much of the initial population

each eigenfirnction will receive. The T. and T- energy levels do not receive any of the

116
initial population because these energy levels do not contain any singlet character. The

intensity of an EPR transition is proportional to the product of the population difference
between the two energy levels involved in the transition and the transition probability for a
transition to occur. The transition probability for a transition is proportional to the
amount of triplet character that is contained in each eigenfunction because the transitions
that are observed in CRPP are to the pure triplet state eigenfirnctions T. and T-. For the
"So" eigenfunction, the population difference is greater than "To" but the transition
probability is smaller. The opposite is true for ”To" eigenfunction. The net result is that
all of the transitions are equal in intensity for a spin polarized CRPP spectrum'o’“.

Figure 1.8c shows a CIDEP-CRPP spectrum that is observed when the radical pair

10'". At later times, the

has had time to evolve away fiom its initial population distribution
population distribution amongst the "So" and "To" energy levels is no longer determined by
the precursor state of the radical pair. Figure 1.80 shows the case when the populations
are accidentally equal, but numerous other distributions are also possible. The equality in
populations of "So" and "To" makes the population differences with the pure triplet state
eigenfunctions equal. This redistribution in population does not affect the transition
probabilities to the pure triplet states and the intensity of each transition reflects this fact.
The two low field transitions (So—>T. and To—rT.) are still split by the exchange
interaction energy, but the resulting emissive and absorptive transitions are no longer
equal in intensity. The So—rT- transition has a lower transition probability than the To-—>T.

transition and the result is an E/A pattern that shows net absorption. The two high field

transitions (To—)T. and So—>T.) show an E/A pattern with net emission. The overall

117
spectrum still shows no net polarization because the amount of net absorption that is

observed for the low field radical is exactly offset by the amount of net emission that
occurs for the high field radical. In the absence of an exchange interaction during the
observation time by EPR, the polarization pattern in Figure 1.8c would show an A/E
pattem'which is formed by net absorption in the two low field transitions and net emission
in the two high field transitions'o’".

The spin polarized spectra in Figure 1.8b and Figure 1.8c are observed because of
state mixing between the So and To eigenfunctions due to a difference in the g-values of
the two radicals of the radical pair"). Equation 4.4 shows that state mixing can occur
because of a difference in the g-values and/or a difference in hyperfine coupling
frequencies of the two radicals of the radical pair. When state mixing occurs primarily
because of a difference in the g-values of the two radicals of the radical pair, net effect

16,17

spin polarization occurs . If state mixing occurs because of a difference in hyperfine

coupling fi'equencies of the two radicals, multiplet effect spin polarization results"’".
Figure 4.4 illustrates net and multiplet effect spin polarization in the presence and absence
of the exchange interaction during the observation time by EPR for doublet-doublet
radical pairs with a singlet precursor. Figure 4.4a shows net effect spin polarization when
the exchange interaction was positive but is zero at the observation time by EPR. An A/E
pattern results with absorption arising from one radical and emission from the other”.
When a positive exchange interaction is present during the observation time by EPR, afour

line E/A/E/A pattern is seen as is shown in Figure 4.4bm. The two low field transitions

show net absorption while the two high field transitions show net emission"). Figure 4.4c

118

Figure 4.4. EPR stick diagrams for net and multiplet effect electron spin polarization in
the presence and absence of an exchange interaction.

H9

 

 

 

 

 

 

 

 

 

 

a) Net effect
2] = 0
A
E ¢ AI > I
b) Net effect
2] > O
A ,
E I 2.1
c) Multiplet effect
2] = 0
A l
E 4‘47 AI .>
d) Multiplet effect
2] > 0
A
E 21

Figure 4.4

 

 

 

120
shows a spin polarized spectrum that arises from multiplet effect spin polarization from

one radical that has hyperfine coupling to an I=1 nucleus. The spectrum shows an A/E
pattern that arises from coupling to the m1 = 1 nucleus of the radical. The low field
absorptive transition arises from the m1 = 1 transition, while the high field emissive
transition arises fi'om the m; = -1 transition. The m1 = 0 transition is not spin polarized
and, consequently, this transition cannot be observed in the spin polarized EPR
spectrum‘6’17. Figure 4.4d shows multiplet spin polarization when an exchange interaction
is present during the observation time by EPR. The low field E/A pair arises from the m1
= 1 transition which has been split by the exchange interaction. This exchange split pair
shows net absorption analogous to the m1 = 1 transition in Figure 4.4c when the exchange
interaction is zero. The center E/A pair arises from the m1 = 0 transition and shows no net
polarization, and the high field E/A pair arises from the mt = -1 transition and shows net
emission. The entire spectrum shows no net polarization as is required in a CRPP
spectrum'o’”.
A mechanism analogous to CRPP occurs in photolyase. The polarization pattern for
photolyase that is shown in Figure 4.1 is an alternating emissive and absorptive pattern
that is similar to patterns that have been observed in photosynthetic reaction centerslo’”.
The similarity in polarization patterns of photolyase and reaction centers suggests that spin
polarization in photolyase occurs by an analogous mechanism to CRPP, but in a triplet-
doublet manifolds. Figure 4.5 shows the energy level diagram and transitions for the
coupled basis set of the radical pair for a triplet-doublet system that exhibits Correlated

Triplet-Doublet Radical Pair Polarization (CTDRPP). The spin polarized stick spectrum

that is shown at the bottom of the figure is shown for both radicals that show a positive

121

Figure 4.5. Triplet-doublet state mixing from precessional frequency differences in the
radical pair in the presence of an exchange interaction with a doublet precursor.

122

 

 

2J>0

 

'Qm

 

 

,m
p

w+2J

 

G‘an99
CGQ-l/299

 

/
— - " 'Q-3/2
D
l
|

2J1

 

Figure 4.5

123
exchange interaction, have no observable hyperfine coupling, and originate from a doublet

precursor. State mixing between the D1,; and Q m unmixed energy levels produce the
“DI/2” and “Q10” levels in the coupled basis set by the perturbation that is shown in
equation 4.4. The mixed eigenfunctions, “DI/2” and “ 112”. are formed by linear
combinations of the unmixed energy levels as shown in equations 4.5 and 4.6. The “Du/2”
and “Q-m” mixed eigenfirnctions are formed by an analogous mechanism to the “Qt/2” and
“DI/2” state mixing that is described above. Initially, with a doublet precursor only the
energy levels that contain doublet character are populated. Therefore, only the mixed
eigenfimctions receive any of the initial population as illustrated in Figure 4.5. The
population will be larger for the “DI/2” and “D-m” energy levels because they contain more
doublet character than the “Qt/2” and “Q40” eigenfirnctions. The four transitions that arise
from the energy level diagram in Figure 4.5 are shown at the bottom of the figure. The
two transitions labeled A and B arise from one member of the radical pair and form an
E/A exchange split doublet. In the absence of an exchange interaction, transitions A
(“D-m” —> Q-3/2) and B (“Qt/2” —) Q30) would be equal in energy. When the exchange
interaction is present A and B are split by the exchange interaction as illustrated in Figure
4.5. The transitions that are labeled C and D also form an E/A pair that is formed by a
splitting fi'om the exchange interaction of the radical pair, and these two transitions arise
form the other radical pair member. The difference in energy between each exchange split
pair is given in equation 4.4 as the difference in precessional frequencies of the two radical

pair members. The resulting spin polarized spectrum is one that shows four transitions

that are equal in intensity and has an E/A/E/A pattern. At later times the precursor state

124
no longer determines the population distribution of the radical pair, and each transition in

the exchange split doublet will no longer be equal in intensity. Figure 4.6 shows that at
later times the radical pair has more quartet character, but the transition probabilities to
the pure quartet states do not change. Transition A is lower in intensity than transition B
and this exchange split pair shows net absorption. Net emission is observed for the other
exchange split pair that includes transitions C and D. The resulting spin polarized
spectrum still shows no net polarization because the amount of net absorption that is
observed for transitions A and B is balanced by the amount of net emission for transitions
C and D.

The polarization pattern in Figure 4.1a shows six transitions that are in an
E/A/E/A/E/A pattern with no net polarizations. The polarization patterns in Figures 4.5
and 4.6 for CTDRPP predict four transitions when a net effect (difference in g-values
dominate equation 4.4) is observed. Therefore, the polarization pattern in photolyase
cannot arise from net effect CTDRPP. The transient spectrum in Figure 4. la has been
shown to originate from a tryptophan cation radical (see Chapter 2) because the triplet
flavin radical is undetectable at room temperature in our experiments’. If the transient
absorption spectrum in Figure 4.1a originates from net effect spin polarization as outlined
in Figure 4.3b, the transient spectrum would show net polarization. Figure 4.3b shows
that one radical of the radical pair is in absorption, while the other is in emission when net
effect spin polarization is observed. In photolyase, the triplet flavin radical 3'FADI-f is
undetectable and only a doublet TrpH'+ is observed’. The absence of net polarization in

photolyase rules out net effect CTDRPP as a mechanism because only half of the spectrum

125

Figure 4.6. Triplet-doublet state mixing in the presence of an exchange interaction with a
doublet precursor at later reaction times.

on 1111a

126

 

2J>0

 

 

 

“‘— ‘an

D
w J l

w+2J

I
IB-

 

 

 

“Q1/2,,

C‘Q- In”
C
-3/2
D

6‘D-1/2”
A

2]

A l I
C

Figure 4.6

127
(the emissive or absorptive halt) would be seen. Other evidence that supports multiplet

rather than net effect spin polarization in photolyase are the g-values of the triplet flavin
and doublet tryptophan radicals. The apparent g-value for the transient tryptophan radical
is 2.0028. Triplet flavin radicals are expected to have a g-value of 20039”. State mixing
between the “Qt/2” and “DI/2” energy levels by the perturbation in equation 4.4 is required
in order to observe CTDRPP. The difference in energy between each exchange split
doublet is equal to the perturbation 0) in equation 4.4 (see Figure 1.8b for analogy to
CRPP). The difference between each doublet in photolyase (denoted Am; in the figure) in
Figure 4.1a is 15 G. In order for net effect CTDRPP to occur in photolyase, the
difference in g-values of the two radical pair members must be 15 G. In photolyase the
difference in g-values between trpH'+ and 3'FADH‘ is only 2 G. Therefore, spin
polarization in photolyase must arise from multiplet rather than net effect CTDRPP.

The polarization pattern in Figure 4. la is E/A/E/A/E/A with no net polarization. It
has been shown previously that this pattern arises from two equivalent nuclei that are the
B-methylene protons on the tryptophan cation radical (see chapter 3). Figure 4.7 shows
an energy level diagram for a spin polarized mechanism that explains the polarization
pattern in photolyase. The figure shows the pure quartet energy levels, OM and Q30, are
initially unpopulated when the precursor state is a doublet. The four mixed states; “ 1,2”,
“Dm”, “Qt/2” and “D40”, are formed by the perturbation in equation 4.4, and the linear
combination of pure states that the mixed states are composed of is shown in equations
4.5 and 4.6. The mixed states are the only eigenfirnctions that receive any of the initial

population because they are the only states that have any doublet character. Each of the

128
six zeeman energy levels in Figure 4.7 are firrther split into three sublevels that arise from

hyperfine coupling to the two B-methylene protons on the tryptophan cation radical. The
triplet-doublet radical pair whose energy level diagram is shown in Figure 4.7 also has a
positive exchange interaction (doublet lies higher in energy than the quartet). The six solid
arrows in the figure show the six transitions that arise in photolyase. The bottom of the
figure shows the stick diagram EPR spectrum for the transitions in the energy level
diagram. Transitions A and B arise from the m; = 1 transition and are split into an E/A
pair that are separated in energy by the exchange interaction between the two members of
the triplet-doublet radical pair. The absorptive transition is higher in intensity because of a
rephasing of the radical pair from pure doublet to a state with partial quartet character.
Transitions C and D arise from the m1 = 0 transition, are equal in intensity, and also are
split by the exchange interaction of the radical pair. Transitions E and F arise from the m;
= -1 transition and show net emission. The total spin polarized spectrum shows no net
polarization as required by CTDRPP and CRPP rules. The six transitions that are denoted
by the dashed arrows are not observed because these transitions arise from the triplet
member of the radical pair which is not observable by EPR at room temperature in
photolyase.

The diagram in Figure 4.7 predicts that the splitting between the transitions in the
first E/A pair in the transient spectrum in Figure 4.1a (denoted as 2] in the figure) is the
exchange interaction between the doublet tryptophan and triplet flavin radicals in
photolyase. This suggests that the exchange interaction in photolyase is 7-8 G. As

discussed above, the state mixing between the quartet and doublet eigenfirnctions is

129

Figure 4.7. Hyperfine state mixing in a triplet-doublet radical pair with a doublet
precursor.

130

 

.oublel

 

 

Figure 4.7

13 1
induced by the perturbation in equation 4.4. The mixing in photolyase is primarily due to

a difference in hyperfine coupling frequencies (multiplet effect) of the radical pair members
rather than a difference in g-values (net effect). Multiplet effect CRPP has yet to be
reported, but these effects are common in other spin polarization mechanisms such as
RTPM and RPM. The perturbation in equation 4.4 is approximately equal to the
difference in energy between the exchange split doublets (Al-[p in Figure 4.1a) and is
approximately 15 G in photolyase. The results here are the first reported case of a CRPP-
type polarization in a triplet-doublet system and the first time that CRPP-type polarization
is driven by hyperfine fiequency differences rather than a difference in g-values.
Differences from CRPP in Photolyase. Even though spin polarization in

photolyase and the photosynthetic reaction centers is very similar, there are some

11,21

1

differences. Deuteration of reaction centers increases the amount of spin polarization
while deuteration in photolyase causes a loss in polarization by a factor of 2-35. An
explanation of this seemingly contradictory result lies in the method of deuteration that
was used. Reaction center deuteration consisted of incubating the sample in D20 and
causes a large number of protons to be exchanged with deuterium“. Linewidth
broadening in reaction center EPR signals has been proposed to arise primarily from
unresolved hyperfine coupling to protons”. Deuterium has an intrinsic hyperfine coupling
that is smaller than hydrogen, and as a result the linewidth will be smaller in samples that
have the exchangeable protons replaced by deuterium. Considerable overlap occurs
between the absorptive and emissive component of the exchange-split pair. A smaller

linewidth will enhance the intensity of each transition in the exchange-split pair because

132
less overlap occurs”. In photolyase, specific deuteration of the or-ring protons on the

tryptophan ring was perfonned'". As a result, fewer protons are replaced by deuteriums
than in the reaction center case. Deuteration of tryptophan will reduce the difference in
hyperfine fi'equencies between the two members of the radical pair. The main factor in
producing spin polarization in photolyase is this difference in hyperfine frequencies.
Deuteration of tryptophan reduces the difference in hyperfine frequencies of the two
radical pair members, therefore, less spin polarization will result. In reaction centers, spin
polarization mainly occurs through g-value differences of the two radicals, so linewidth
reduction effects dominate". In photolyase, deuteration affects the difference in hyperfine
coupling more strongly than linewidth considerations. The effect of deuteration in
photolyase is that less state mixing occurs as seen in equation 4.4. Because less state
mixing occurs, the polarization in photolyase will decrease as well.

Determination of the Precursor State in Photolyase. The energy level diagram in
Figure 4.7 correctly predicts the polarization pattern for photolyase that is shown in
Figure 4.1a and suggests that this diagram is an accurate representation of the energy
levels of the triplet-doublet radical pair in photolyase. The diagram in Figure 4.7 is for a
triplet-doublet radical pair that shows hyperfine coupling to two equivalent nuclei with a
nuclear spin of 1/2, has a positive exchange interaction, originates from a doublet
precursor, and mixes quartet and doublet states of the triplet-doublet radical pair through
a difference in hyperfine coupling frequencies of the two radical pair members. An
alternate energy level diagram which is shown in Figure 4.8 can also account for the
polarization pattern for photolyase that is shown in Figure 4.1a. Figure 4.8 is an energy

level diagram for a triplet-doublet radical pair that shows hyperfine coupling to two

133

Figure 4.8. Triplet-doublet state mixing with a quartet precursor in the presence of an
exchange interaction.

134

2]<0I
Q3/2 - .

 

 

 

 

 

 

i\
A \\
\E \\\
\C \\\
\\\
“Q10” ' \\\
\\\
\\
(”+21 “DI/2”

 

 

 

Figure 4.8

135
equivalent spin = 1/2 nuclei and mixes states through a difference in hyperfine coupling

frequencies of the two radical pair members (see equation 4.4). The polarization pattern
that is predicted from the energy level diagram in Figure 4.8 is shown at the bottom of the
figure as an E/A/E/A/E/A pattern, and this pattern is the same as the one predicted by the
energy level diagram in Figure 4.7. The difference between Figure 4.7 and Figure 4.8 is
the sign of the exchange interaction and the precursor state of the triplet-doublet radical
pair. Figure 4.8 shows the energy levels for a triplet-doublet radical pair that has a
negative exchange interaction (doublet lies lower in energy than the quartet) and originates
from a quartet precursor. Even though the sign of the exchange interaction and the
precursor state are different for Figures 4.7 and 4.8, both figures correctly predict the
polarization pattern that occurs in photolyase following flash photolysis of the inactive
enzyme.

Figure 4.9a shows a possible mechanism for the formation of a triplet-doublet
radical pair during photoreduction of inactive photolyase when the radical pair forms from
a quartet precursor. Flash photolysis of the inactive enzyme (FADH‘ containing) produces
an excited doublet state, 2'FADH’. Intersystem crossing to the quartet occurs to produce
4'FADH. Electron transfer fi'om an amino acid side chain that has been identified
previously as tryptophan-306 forms a triplet flavin anion radical 3'FADI-I‘ and a doublet
cation radical trp’. The triplet-doublet radical pair that forms produces a spin polarization
pattern by the mechanism that is discussed above and shown in Figure 4.8.

An alternate mechanism for spin polarization in photolyase is shown in Figure
4%. The mechanism in Figure 4% is triplet-doublet radical pair formation that arises

from a doublet precursor. Flash photolysis of inactive photolyase produces an excited

136

Figure 4.9. Reaction mechanisms for spin polarization in inactive photolyase.

137

a) Quartet precursor triplet-doublet radical pair

FADH-

I hv
2*FADH-
I intersystem crossing

4"'FADH-

I electron transfer

3"FADH- — 2trp-*

b) Doublet precursor triplet-doublet radical pair
FADH-

‘ hv
”FADH-
I electron transfer

1*FADH- — Ztrp.+

I intersystem crossing

3"‘FADH' — 2trp~

Figure 4.9

138
doublet state 2'FADH'. Electron transfer from trp-306 to 2'FADH' produces an excited

singlet anion radical l'FADH‘ and trp'i. The excited singlet "FADH' intersystem crosses
to the triplet 3'FADl-I' and spin polarization results from the triplet-doublet radical pair as
shown in Figure 4.7.

The two mechanisms for spin polarization in photolyase that are shown in Figure
4.9 show triplet-doublet radical pair formation, but the precursor state for the spin
polarized state is different in each case. Quartet precursor spin polarization as shown in
Figure 4.9a arises fiom the excited flavin quartet precursor state 4‘FADH and suggests
that the exchange interaction between the two radical pair members must be negative in
order to observe an E/A/E/A/E/A polarization pattern (see Figure 4.8). Doublet precursor
spin polarization, as shown in Figure 4%, arises from a singlet-triplet radical pair
precursor state l'FADI-I‘ -- trp‘+ and suggests that the exchange interaction between the
doublet and triplet radicals must be positive in order to observe the polarization pattern
that is seen experimentally in photolyase (see Figure 4.7). The difference between Figure
4.9a and Figure 4% is the sequence in which intersystem crossing of the flavin moiety
occurs in relation to electron transfer from trp—306. For a quartet precursor, intersystem
crossing must precede electron transfer, while the reverse must be true in order to observe
doublet precursor spin polarization. At the present time insufficient experimental evidence
exists to distinguish between these two mechanisms, but the arguments that follow suggest
a reasonable choice and allow us to postulate a mechanism for photoreduction in

photolyase.

139
Comparison of electron transfer rates with the photosynthetic reaction center. A

comparison of the electron transfer rates of photolyase and the bacterial reaction center
provide evidence for the doublet precursor mechanism in Figure 4% rather than the
quartet precursor one in Figure 4.9a. Optical data have been interpreted to indicate that
the excited flavin doublet state, with a lifetime of 100ps, intersystem crosses to the
quartet, and electron transfer follows in a microsecond”. The polarization pattern of the
transient radical in photolyase that is shown in Figure 4.1a suggests that the rate of
electron transfer must be 100 ps rather than the optical data prediction of a microsecond.
Transient spin polarized spectra in the bacterial reaction center, which exhibit CRPP, set
the value of the exchange interaction between the bacterial pheophytin and the chlorophyll
dimer to 7 .5 Gu’n. This value is similar to the exchange interaction in photolyase which
we estimate to be 7-8 G. Because the exchange interaction is similar in these two proteins
we expect that the rates of electron transfer will also be similar. The rate of electron

transfer, as given by Marcus theory”, is shown in equation 4.7.

kct = 21t/h V112 F C (equation 4.7)

k... is the rate of electron transfer, VR2 is the electronic coupling, and FC is the Franck-
Condon factor. The similarity in the exchange interaction of both proteins suggests that
the electronic coupling between the donor and acceptor is also very similar because both
the exchange interaction and the electronic coupling are dependent upon the distance that

separates the donor and acceptor. The FC term in equation 4.7 could differ between the

140
two proteins, because this term depends on the difference in free energy of the reactants

and the reorganization energy of the reaction. Moser et. al. have shown that the FC term
in equation 4.7 can modulate the rate, but not to any great extent. These authors
suggest” that even though the rate can theoretically change by 5 orders of magnitude with
a change in FC, such large rate changes are not observed experimentally. The electron
transfer rate in nature is modulated instead by varying the distance between the two
reactants, which results in a change in the electronic coupling of the two reactants. We
cannot calculate the rate of electron transfer in photolyase because we do not know the
free energy change or reorganization energy of the reaction. The findings of Moser et.
a1.” indicate that the electron transfer rate in photolyase should be similar to the rate
between BPh and BChl in reaction centers because similar exchange interactions in each
system implies that the distance between the donor and acceptor is also similar. The rate

25.26 In

of electron transfer in bacterial reaction centers has been reported to be 3 ps
photolyase, the spin polarization data in Figure 4.1a imply that the rate should be on a ps
time scale owing to the similarity in exchange interactions, while the optical data suggests
a rate of lusz. The similarity in exchange interactions insinuate that the rate of electron
transfer in photolyase should be much faster than the optical data prediction. If the rate of
electron transfer in photolyase were a microsecond as the optical data suggest, the rate of
electron transfer with similar distances would have to decrease FC in photolyase by more
than 5 orders of magnitude. This result is contrary to what has been observed previously

for electron transfer reactions“. If the rate were on a ps time scale for photolyase, the FC

term would not have to decrease the rate by such a large factor. Therefore, an electron

141
transfer rate on a ps time scale is much more likely in photolyase. A ps time scale electron

transfer rate also implies that electron transfer must precede intersystem crossing in
photolyase. Optical studies of photolyase show two species with lifetimes of 100 ps and 1
I152. We have shown above that the species with a 100 ps lifetime must decay as a result
of electron transfer from trp-306 to FADH'. The slower decaying species has to arise
from intersystem crossing of 1'FADI—I' to 3'FADI-l' if the 100 ps species arises fi'om
electron transfer. The mechanism in Figure 4% for doublet precursor spin polarization
shows that electron transfer precedes intersystem crossing of the flavin radical and is in
agreement with the electron transfer rate in photolyase being on a picosecond time scale.
The mechanism in Figure 4.9a for quartet precursor spin polarization cannot account for a
rate of electron transfer that is faster than the rate of intersystem crossing, and the
discussion above implies that the correct mechanism for photoreduction in photolyase is
the one in Figure 4% where the triplet-doublet radical pair is formed from a doublet
precursor.

Calculated energies of doublet and quartet. Further evidence that supports the
mechanism given in Figure 4% is the relative energies of the quartet and doublet states.
Theoretical calculations27 have shown that a quartet state should be higher in energy than
a doublet. If the flavin excited quartet state 4'FADH is higher in energy than the excited
doublet flavin state, intersystem crossing to the quartet as shown in Figure 4.9a is not
expected. This gives strong support for the mechanism in Figure 4% where intersystem
crossing from the excited flavin singlet to the triplet occurs rather than intersystem

crossing fiom an excited doublet to the quartet as proposed in Figure 4.9a.

142
Other evidence that supports the formation of a triplet-doublet radical pair in

photolyase that arises fi'om a doublet precursor is the sign of the exchange interaction that
is necessary in order to observe an E/A/E/A/E/A pattern as in Figure 4.1a for photolyase.

Figure 4.8 shows that, when the triplet-doublet radical pair originates from a quartet
precursor, a negative exchange interaction (doublet lies lower in energy than the quartet)
is necessary in order to observe a polarization pattern that is similar to the one that is
observed in photolyase. Figure 4.7 shows a polarization pattern that arises fi'om a triplet-
doublet radical pair that originates from a doublet precursor and implies that a positive
exchange interaction (doublet lies higher in energy than the quartet) is needed to observe
this pattern. Neutral radical pairs in solution reactions are expected to have a negative
exchange interaction in accordance with Hund’s rule9. Adrian has shown that it is
theoretically possible to have a positive exchange interaction in cases where the two
radical pair members are ionic”. This result has been proved experimentally by work in
which the authors showed that all ionic radical pairs that were used had a positive
exchange interaction”. The photosynthetic reaction center, which also has an ionic radical
pair, has been shown to have a positive exchange interaction“. These results suggest that
the exchange interaction in photolyase is most likely positive because the radical pair
members are ionic (see Chapter 3). A positive exchange interaction implies that the
triplet-doublet radical pair in photolyase must arise from a doublet precursor as shown in
Figure 4% because Figure 4.7 shows that a positive exchange interaction requires a
doublet precursor in order to show a polarization pattern that is in agreement with the

pattern that is observed in photolyase.

143
Rationalization of data with previous optical studies. The mechanism for spin

polarization in photolyase shown in Figure 4% is inconsistent with the mechanism that
was proposed previously from optical data18 (Figure 3.8) . This implies that electron
transfer from tryptophan to the excited flavin occurs when the flavin is in an excited
doublet rather than an excited quartet state (compare Figure 3.8 with Figure 4%). The
optical data used to assign electron transfer to the excited quartet state has a low signal to
noise ratio and makes differentiation of the two mechanisms difficult”. Furthermore,
detection of a tryptophan radical was not searched for optically until 4ps after the
excitation of the enzyme18 owing to experimental limitations. Both mechanisms (Figure
3.8 and Figure 4%) predict that the transient tryptophan radical will be present at this
time and the two mechanisms cannot be distinguished based on this data.

Explanation of kinetics in photolyase. The kinetic trace for the transient tryptophan
radical in photolyase that is shown in Figure 2.5 reflects the mechanism that was discussed
above. The instrument-limited rise of the transient radical corresponds to the quenching
of l'FADH' and the appearance of doublet precursor polarization. Figure 4.7 shows the
relative populations of each state following singlet quenching. The only states with any
initial population will be the states that have doublet character. These states are the D1,;
and D4,; states and also the Q12 and Q-l/Z states due to state mixing as in equation 4.4.
The decay of the transient radical could be due to either recombination of the radical pair
or quenching of 3'FADH to the ground singlet state as shown in Figure 4.10. Figure 4.10a
shows triplet quenching of the triplet-doublet radical pair preceding recombination. The
excited triplet flavin can only be quenched through the doublet states29 of the radical pair,

resulting in an excess of quartet character, which manifests itself as quartet precursor spin

144
polarization (an A/E/A/E/A/E polarization pattern). The state that results from triplet

quenching can recombine over several milliseconds to form the dark stable inactive form
of photolyase. Figure 4.10b shows a mechanism in which the radical pair recombines
directly from the excited triplet flavin state. In this mechanism the radical pair recombines
by electron transfer to form a state that contains a diamagnetic tryptophan species and an
excited 3.FADH radical that intersystem crosses to the ground state to form the inactive
form of the enzyme. The mechanism in Figure 4.10b should also show quartet precursor
spin polarization because only the doublet states of the triplet-doublet radical pair are
expected to recombine29 when the product state is a doublet (FADH'). Both mechanisms
predict a second polarization that is opposite in phase to the one that is observed in Figure
4.13, but instrumental limitations have prevented us from observing this second
polarization. Direct detection EPR measurements (faster time resolution) should allow us
to observe this second phase in the future. The difference between the mechanism in
Figure 4.103 and Figure 4.10b is the time at which triplet quenching and recombination
will occur. Recombination of the doublet trp radical and the flavin species has been
reported to occur over several millisecondsz. Excited flavin triplet lifetimes have been
measured in a photolyase sample that was catalytically active and are reported to be
approximately 1 p330. Activity of the enzyme is not expected to influence the lifetime of
the excited triplet flavin radical which suggests that the lifetime of the radical will be the
same in the inactive form of the enzyme. The lifetimes predicted above suggest that

intersystem crossing precedes recombination of the radical pair as is shown in Figure

4.10a.

145

Figure 4.10. Possible mechanisms for recombination during DNA photolyase
photoreduction.

146

a) Recombination from the ground state

2( 3‘FADH’ trpsf) - 4( 3‘FADH- trp‘“)

\triplet quenching

FADH' + ttp~+

I recombination
FADH- + trp

b) Recombination from the excited state

2( 3'FADH- trp.+) - 4( 3‘FADH- trp.+)

\recombination

3‘FADH- + tl‘p
I triplet quenching

FADH- + trp

Figure 4.10

147
New mechanism for photoreduction in photolyase. The previously proposed

mechanism for photolyase18 shown in Figure 3.8 is inconsistent with the formation of
doublet precursor polarization. If electron abstraction from trp306 occurs after the
excited flavin intersystem crosses to the quartet as is shown in Figure 3.8, the radicals
formed would be 3°FADH' and Ztrp” due to conservation of spin. The polarization pattern
would arise from a quartet precursor (4'FADH) which is inconsistent with an
E/A/E/A/E/A pattern with a positive exchange interaction.

A different reaction scheme shown in Figure 4.11 can account for the polarization
pattern that is shown in Figure 4.1a. If electron transfer from “13306 to 2'FADH occurs
before any intersystem crossing by the flavin, the radical pair members are 2trp'+ and
l'FADH' because of spin conservation. The singlet flavin (1°FADI-1') is quenched to form
a triplet-doublet radical pair (2trp’ and 3'FADH') that shows a doublet precursor
polarization pattern (Figure 4.1a). Quenching of the triplet in doublet state radical pairs
produces excess quartet states and a quartet precursor polarization results.
Recombination occurs over several milliseconds to reform the dark stable inactive form of
the enzyme (F ADH').

We have proposed a new triplet-doublet radical pair mechanism that is analogous
to CRPP for doublet-doublet radical pairs. Figure 4.11 shows the complete mechanism
for the photoreduction reaction in photolyase. The presence of the exchange interaction
causes an alternating emissive and absorptive pattern. The close similarity between the
polarization patterns in photolyase and the photosynthetic reaction centers and the fact

that both reactions involve radical pairs that are protein-bound suggest that the mechanism

148

Figure 4.11. Mechanism for photoreduction in photolyase that includes doublet precursor
spin polarization.

INK?

149

FADH- —— trp dark stable
I hv
2"FADH- —'— tip

I electron transfer
l'FADH' —— trp-+

I singlet quenching

2(3'FADH-__— t1'p+) <_>4(3‘1:: ADH-—— trp‘“)

I recombination

FADH- — trp inactive
enzyme

Figure 4.1 l

150
that we are proposing is not a special but rather a general case of spin polarization in

proteins. The phenomena we observe is probably present in other proteins that function
through electron transfer mechanisms. As more proteins and their reactions are

discovered, we expect to see other cases of CRPP and CTDRPP surface.

8.

9.

10

ll.

12.

13.

14.

15.

16.

17.

1 5 1
References

. Kim, S.T.; Sancar, A.; Essenmacher, C. and Babcock, G.T. Proc. Natl. Acad. Sci. USA

1993, 90, 8023.

Heelis, P.F.; Okamura, T. and Sancar, A. Biochemistry 1990, 29, 5694.

. Okamura, T.A.; Sancar, A; Heelis, P.F.; Hirata, Y. and Mataga, N. J. Am. Chem. Soc.

1989, 111, 5967.

. Li, Y.F.; Heelis, PF. and Sancar, A. Biochemistry 1991, 30, 6322.

. Essenmacher, C.; Kim S.T.; Atamian, M.; Babcock, GT. and Sancar, A. J. Am.

Chem. Soc. 1993, 115, 1602.

Heelis, P.F.; Payne, G. and Sancar, A. Biochemistry 1987, 26, 4634.
Blattler, C.; Jent, F. and Paul, H. Chem. Phys. Lett. 1990, 166, 375.
Kawai, A.; Okutsa, T. and Obi, K. J. Phys. Chem. 1991, 95, 9130.

Kawai, A. and Obi, K. J. Phys. Chem. 1992, 96, 52.

. Norris, J .R.; Morris, A.L.; Thumauer, MC. and Tang, J. J. Chem. Phys. 1990, 92,

4239.

Snyder, SW. and Thumauer, M.C. Electron Spin Polarization in Photosynthetic
Reaction Centers: The Photosynthetic Reaction Center 1993, Vol.2 ed. Deisenhofer,
J. and Norris, J .R. (Academic Press, SanDiego) p. 285.

Sancar, A. and Sancar, G.B. J. Biol. Chem. 1984, 172, 223.

Sancar, G.B.; Smith, F.W.; Reid, R.; Payne, G.; Levy, M. and Sancar, A. J. Biol.
Chem. 1987, 262, 478.

Yerkes, C. Ph.D. dissertation 1982, Michigan State University, East Lansing, MI.
Hoganson, CW. and Babcock, G.T. Biochemistry 1988, 27, 5848.
Jenks, W.S. and Turro, N.J. Rev. Chem. Int. 1990, 13, 237.

Adrian, F.J. Rev. Chem. Int. 1986, 7, 173.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

152
Kim, S.T.; Heelis, P.F.; Okamura, T.; Hirata, Y.; Mataga, N. and Sancar, A.

Biochemistry 1991, 30, 11262

Lhoste, J .M.; Haug, A. and Hemmerich, P. Biochemistry 1966, 5, 3290.
Heelis, PF. and Sancar, A. Biochemistry 1986, 25, 8163.

Bock, C.; Stehlik, D. and Thumauer, M.C. Israel J. Chem. 1988, 28, 177.

Snyder, S.W.; Morris, A.L.; Bondeson, S.R.; Norris, J .R. and Thumauer, M.C. J.
Am. Chem. Soc. 1993, 115, 3774.

Marcus, RA. and Sutin, N. Biochim. Biophys. Acta 1985, 811, 265.

Moser, C.C.; Keske, J .M.; Wamcke, K.; Farid, RS. and Dutton, P.L. Nature 1992,
355, 796.

Woodhouse, 1H. and Dziewonski AM. J. Geophys. Res. 1984, 89, 5953.

Ito, E. and Yamada, H. High-Pressure Research in Mineral Physics 1982, ed.
Akimoto, S. and Manghnani, M.H. ( Center for Academic Publications, Tokyo) pp.
405-419.

Hoijtink, G.J.; Velthorst, NH. and Zandstra Mol. Phys. 1960, 3, 543.

Batchelor, S.N.; Heikkila, H.; Kay, C.W.M.; McLauchlan, K.A. and Shkrob, I.A. J.
Chem. Phys. 1992, 162,29.

Schulten, K. J. Chem. Phys. 1984, 80, 3668.

Heelis, P.F.; Rolley-Williams, C.A.; Hartman, RF. and Rose, S.D. J. Photochem.
Photobiol. B: Biol. 1994, 23, 155.

Chapter 5
Simulation of the Transient Tryptophan EPR Signal that Occurs during

Photoreduction of Inactive Photolyase

Introduction

Inactive DNA photolyase can be photoreduced to produce an active enzyme that is
capable of repairing UV-damaged DNAL”. Photoreduction of inactive photolyase
produces a transient tryptophan radical that exhibits electron spin polarization”, as
discussed in Chapter 4. A current model for photoreduction of the inactive enzyme that
accounts for spin polarization of the tryptophan radical as discussed in Chapter 4 is shown
in Figure 4.11 and is discussed briefly below. Photoexcitation of the dark stable FADH‘
radical produces an excited flavin doublet radical 2'FADH‘. Electron transfer fi'om
tryptophan-306 to 2.FADH' produces an excited singlet flavin anion species (liFADH')
and a tryptophan cation radical (trpH‘+). Intersystem crossing of l'FADH‘ to the triplet
produces a triplet-doublet radical pair that exhibits electron spin polarization. 3 'F ADI-1'
intersystem crosses to the ground singlet state to form a diamagnetic FADH‘ species that
recombines over several milliseconds with trpH‘+ to reform the inactive, FADH' containing
enzyme when exogenous donors are absent. When exogenous donors are present, trpH“
is reduced to form the active FADH containing enzyme3 (see Chapter 3).

During the photoreduction reaction of inactive photolyase an intense EPR signal is
observed transiently, and we have identified the origin of this signal as a tryptophan cation
radical5 (see Chapter 3). The EPR spectrum of this transient signal is shown in Figure 4.1,

153

154
radicals (see Chapter 3). The EPR spectrum of this transient signal is shown in Figure 4.1,

and shows that some transitions are emissive which suggests that the tryptophan radical in
photolyase is spin polarized. In Chapter 4 a possible mechanism for spin polarization in
photolyase was proposed. State mixing between the energy levels of the doublet and
quartet is responsible for the observed spin polarization. State mixing in the basis set of
the triplet-doublet radical pair is induced through a difference in hyperfine coupling
frequencies of the two radical pair members, as discussed in Chapter 4. The presence of
an exchange interaction between the two radical pair members causes each transition to
split into an emissive-absorptive pair where each transition in the pair is split in energy by
the exchange interaction. The resulting EPR spectrum shows an alternating emission and
absorption pattern with no net polarization.

In this chapter the experimentally observed spectrum in Figure 4.1 will be simulated in
order to test the validity of the model for spin polarization in photolyase that was
proposed in Chapter 4. A description of the program that was used to simulate the
spectrum and the parameters that were used in the simulation will be presented. The
results show that the simulated spectrum fits quite well with the experimentally observed

spectrum and give support for the model that was suggested in Chapter 4.

Materials and Methods

The simulation program was written in the C++ programming language and uses the

Object Windows Library (Borland International Inc, Scotts Valley, CA) and the

155
Microsoft Windows 3.1© Application Programming Interface (Microsoft Corporation,

Redmond, WA) for certain portions of the program. The program requires the installation
of Microsoft Windows 3. l© in order to run the program. Compilation of the source code
was done with Borland C++ 3.1 for Windows and Application Frameworks (Borland
International Inc., Scotts Valley, CA), and this code is shown in Appendix A. The
simulation routine is a modified version of a BASIC program that was written by Dr. Curt
Hoganson to simulate EPR spectra“. The original simulation program was converted to
C++ and spectral parameters that are necessary in order to introduce spin polarization
were added to the simulation routine.

Even though the program is able to calculate EPR spectra for crystals, partially
oriented membranes, and powder samples; the only case that will be discussed here will be
the solution one because there is not enough experimental data to warrant any other type
of simulation at this time. The flow chart and EPR stick diagrams that are shown in
Figure 5.1 show the procedure that the simulation program uses to generate an EPR
spectrum for a radical whose energy levels are populated according to a Boltzman
distribution6’7. The user must enter the g—value, microwave frequency, magnetic field
limits and linewidth for the experimental spectrum that is to be simulated. If there is
hyperfine coupling present in the spectrum, the user must also enter the coupling constant
and nuclear spin for each coupled nucleus in the experimental spectrum. The program
determines the magnetic field at which resonance occurs in the absence of hyperfine

coupling by equation 5.1.

156
H = (hv) / (gB) 2 equation 5.1

where H is the magnetic field, h is Planck’s constant, g is the electronic g-factor, B is the
Bohr magnetron, and v is the microwave frequency. The transition resonant from

equation 5.1 is split by hyperfine coupling according to equation 5.2

H = (hv - hAoSI) / (gB) (equation 5.2)

where H, h, g, v, and B are the same as in equation 5.1, and A0 is the hyperfine coupling
constant, S is the electron spin quantum number, and I is the nuclear spin quantum
number. The program applies equation 5.2 for each coupled nucleus in the EPR spectrum
and assigns equal intensity for each line that results from the splitting. The EPR stick
diagram is shown for the case of hyperfine coupling to a spin 1/2 nucleus in Figure 5.1.
The resulting EPR stick diagram has a first-derivative gaussian lineshape with a linewidth
applied to it that is specified by the user, and the result is a simulated EPR spectrum with
spectral parameters that were input by the user.

Figure 5.2 shows EPR stick diagrams of the simulation procedure when spin
polarization is introduced into the calculation. The user must enter the precursor state, the
exchange coupling between the doublet and triplet radicals, and the largest ratio of the
transitions in an exchange-split doublet. The precursor state determines whether the
polarization pattern begins in emission or absorption. As discussed in Chapter 4, a
doublet precursor with a positive exchange interaction will begin in emissions. If either

the precursor state or the sign of the exchange interaction is reversed, the pattern will

157

Figure 5.1. Simulation of EPR spectra that are in thermal equilibrium.

158

3) Simulation of the center field transition

g-value, microwave frequency, linewidth

l

 

 

A0 and nuclear spin

b) Simulation of hyperfine coupling

 

 

 

A0

 

Figure 5.1

159

Figure 5.2. Simulation of a spin polarized radical.

160

a) therrnalized radical with coupling to nuclear spin =1

 

 

 

I

I

I

I precursor, 2J
I

I

I b) spin polarized radical
I

I

I

I

 

 

 

 

0) Spin polarized radical
with CIDEP ratio

 

 

 

Figure 5.2

161
begin in absorption“. Figure 5.2a shows an EPR stick diagram for a thermalized doublet

radical that shows hyperfine coupling to one spin == 1 nucleus. The pattern has three lines
that are equal in intensity and split by the hyperfine coupling constant A0. The magnetic
field positions once spin polarization is introduced into the spectrum is shown in Figure
5.2b. The exchange coupling splits each line in the thermalized spectrum into two lines as

shown below in equation 5.38.

H = H. i (2]) / 2 (equation 5.3)

where H. is the magnetic field position for a transition in the thermalized spectrum, H is
the new field position in the spin polarization spectrum, and 2] is the exchange interaction
between the two radicals of the radical pair. Equation 5.3 is applied to each transition in
the thermalized spectrum. The phase of each transition is determined by the sign of the
exchange interaction and the precursor state as discussed above. For example, in Figure
5.2 the pattern is E/A/E/A/E/A because the precursor state is a doublet and the exchange
interaction is positive”. For the case in Figure 5.2b, the lower energy transition in the
exchange split pair is in emission and the higher energy transition is in absorption”.

Figure 5.2c shows how the simulation program assigns intensities to each transition in
the spin polarized spectrum. According to equation 4.4, the amount of state mixing that
occurs in a triplet-doublet radical pair is proportional to the difference in precessional
frequencies of the two radical pair members“. Therefore, the transitions that are farthest

away from center magnetic field of the spectrum which is calculated by equation 5.1 will

162
show the largest ratio of transitions in the exchange split pair. The program first

calculates from equation 5.2 the transition that is farthest away from the center of the
spectrum. For a doublet precursor with a positive exchange interaction, all transitions that
are at a lower magnetic field than the center field will have intensities that obey equations

5.4 and 5.5

1.: I./ {(R — 1) r [(H — H.) / (H. - H,)] + 2} (equation 5.4)

II. = I, - I. ' (equation 5.5)

where I. is the intensity of the lower field transition in the exchange split pair, and It. is the
intensity of the higher field transition. I. is the relative intensity of the transition that was
calculated in the thermalized spectrum, H is the field value of the transition that is
calculated by equation 5.1, H. is the field value of the farthest transition from the center
field Hg, and R is the ratio that is entered by the user. For transitions that are at a higher
field value than the center frequency the equations for I. and II. are reversed.

Equations 5.4 and 5.5 are based upon the assumption that there is a linear relationship
between the amount of net polarization that occurs in an exchange split pair and the
difference in precessional frequencies of the two radical pair members. Equation 4.4
shows that the amount of state mixing in triplet-doublet radical pairs is proportional to the
difference in precessional frequencies of the two radical pair memberss. The amount of

state mixing is proportional to the net polarization that is observed in an exchange split

163
doublet. It is well known from spin polarized doublet radicals in solution that the center

10"“12’”, and equations 5.4 and 5.5

exchange split doublet should show no net polarization
assume that the ratio of the two transitions will vary linearly from one at the center of the
spectrum to the ratio that the user enters at the far edges of the spectrum where equation
4.4 will have the largest value. The ratio that the user enters is an entirely empirical
parameter because the ratio cannot be calculated at the present with the current data that
are available. In order to calculate the ratio of the two transitions in an exchange split
doublet, it is necessary to have kinetic data on the lifetime of the triplet-doublet radical
pair”. At the present time we do not know the true kinetics of spin polarization in
photolyase because of the use of 100 kHz modulation. Consequently, we cannot calculate
the ratio of the two transitions in an exchange split doublet until the true kinetics of spin
polarization in photolyase are known.

The simulation program whose source code is shown in Appendix A assumes that the
only contribution to the difference in precessional frequencies of the two radical pair
members is a difference in hyperfine coupling frequencies (see equation 4.4) of the two
radicals. This assumption is reasonable for spin polarization in photolyase if a comparison
of the g-values of the two radical pair members is considered. The excited triplet flavin
anion radical is expected to have a g—value that is close to the dark stable doublet flavin
radical whose g-value is 2.0039”. The tryptophan doublet cation radical whose
absorption spectrum is shown in Figure 4.1a has an apparent g-value of 2.0028. The

difference in magnetic field units between these two g-values is only 2 G. The difference

in hyperfine coupling frequencies of the two radical pair members is at least 15 G which is

164
much larger than the g-value difference of the two radicals". Further evidence for

hyperfine coupling differences rather than g-value differences are driving spin polarization
in photolyase is the polarization pattern for photolyase that is shown in Figure 4.1a. The
first exchange split doublet shows net absorption, the second shows no net polarization,
and the third doublet shows net emission“. The overall spectrum shows no net
polarization because the amount of net absorption in the first exchange split doublet is
offset by the amount of net emission in the third exchange split doublet. The polarization
pattern of the exchange split doublets in Figure 4.1a is typical of a spin polarization that
originates from a difference in hyperfine coupling frequencies (multiplet effect) rather than
a difference in g-values (net effect) of the two radical pair members. If spin polarization in
photolyase arises from a net effect, the polarization pattern would show either net
emission or net absorption. In Chapter 2, we showed that the transient EPR signal in
photolyase arises from only one member of the radical pair. In order to observe net effect
spin polarization, one radical of the radical pair must show net emission and the other net

‘0’". The absence of net polarization in photolyase excludes net effect spin

absorption
polarization as a possible mechanism because the transient EPR signal would have to show
net polarization that arises from one member of the radical pair. Therefore, spin
polarization in photolyase must arise fi'om a multiplet effect as the simulation program in
Appendix A assumes.

Another assumption of the simulation program in Appendix A is that the initial

population distribution of nuclear spin states is equal for all nuclei in the other radical pair

member. The manifestation of this assumption is in the difference in hyperfine coupling

165
fi'equencies of the two radical pair members as shown in equation 4.4. The transition that

occurs at the lowest magnetic field in Figure 4.1a has a hyperfine coupling of
approximately 15 G. In order to calculate the hyperfine coupling frequency difference for
this particular transition, the difference in frequency must be taken for each nuclear spin
state in the other radical of the triplet-doublet radical pair”. For example, if the other
radical pair member has hyperfine coupling to a single proton the frequency difference for

the low field transition in Figure 4.1a will be as shown in equation 5.6.

A 0). = 15 G - (1/2)nI/2Ao + (1/2)n.1/2Ao (equation 5.6)

where A to. is the difference in precessional fi'equencies of the two radical pair members
for the low field transition, 11 is the number of radicals that show hyperfine coupling to the
particular nuclear spin state that is shown as a subscript, and A0 is the hyperfine coupling
constant in Gauss for this proton. The simulation program in Appendix A assumes that
nm = n.1,; and the result is that the low field transition precesional fiequency difference is
equal to 15 G. The program assumes for hyperfine couplings to the simulated radical that
the frequency difference is simply the hyperfine coupling that is observed for the simulated
radical, and the other radical does not contribute to this frequency difference as is shown
in the above example. At the present time, no theoretical work has been done on the
initial population distribution of nuclear spin states in spin polarized systems, and as a
result, the only practical approach that the simulation program can take at this time is that

a random statistical distribution of nuclear spin states is formed.

166

Results and Discussion

Figure 5.3a shows the experimental time—resolved EPR spectrum (top) of a photolyase
sample that contains natural abundance tryptophan. Directly below this spectrum is the
spectrum that is obtained with the simulation program that is shown in Appendix A.
Figure 5.3b shows the experimental and simulated spectra for a photolyase sample that
contains tryptophan which is isotopically labeled with deuterium on all the or-ring carbons.
The parameters that were used to simulate the EPR spectra are shown in Table 5.1. The
simulation parameters that are shown in Table 5.1 are quite reasonable because an
estimate for the parameters was obtained from experimental data as discussed below. The
determination of an appropriate g-value for the simulation was made by using the apparent
g-value of the experimental spectra in Figure 4. la and Figure 4.1b. Because all photolyase
samples that were used were in solution, an isotropic g-value of 2.0028 was used for both
the natural abundance and specifically deuterated samples“. As discussed in Chapters 2
and 3, the large hyperfine couplings in Figure 4.1a and Figure 4.1b have been assigned to
hyperfine coupling to two equivalent nuclei that are the B-methylene protons on
tryptophan’. The coupling is estimated to be 15-16 G by measuring Am; in the figure.

The best fit for the simulated spectra was obtained with a hyperfine coupling of 15.5 G as
shown in Table 5.1. The experimental spectra in Figure 2.83 and Figure 2.8b were used to
assign the smaller hyperfine coupling to a single or-ring proton with a coupling of 5 G5.

This value was used to simulate the natural abundance spectrum in Figure 5.3a, and a

167

Figure 5.3. Comparison of simulated and experimental spectra: (a) shows the natural
abundance spectra and (b) shows the a-ring deuterated spectra. The top spectrum in each
figure is the experimental spectrum and the lower spectrum is the simulated one.

168

a) Natural abundance

b) ot-ring deuterated

 

l l l l 1_ l a l I l

 

3240 3260 3280 3300 3320 3340
Gauss

Figure 5.3

169

Table 5.1 Parameters used to simulate the spectra in Figure 5.3a and Figure 5.3b.

 

 

 

Parameters Natural Abundance Deuterated
g-value 2.0028 2.0028
Frequency (MHz) 9222.26 9222.26
Linewidth (G) 4.1 4.1
Exchange Coupling (G) 7.8 7.8
Ratio 3 2
Precursor State Doublet Doublet
2 B—methylene protons 15.6 15.6
hyperfine coupling (G)

C(2) proton 4.8 0.8
hyperfine coupling (G)

N(1) nitrogen 4.1 4.1

hyperfine coupling

 

 

 

 

170
hyperfine coupling of 0.8 G (one-sixth of 5 G) was used to simulate deuterium labeled

sample. Figure 3.3b shows the transient EPR spectrum for a photolyase sample that
contains l’N-labeled tryptophan’. From a comparison with the natural abundance
spectrum, an estimate of less than 5 G is estimated for hyperfine coupling to the nitrogen
on the tryptophan ring. Table 5.1 shows that the best fit to the experimental spectra was
obtained with a hyperfine coupling of 3.8 G for the tryptophan ring nitrogen. The
exchange interaction is estimated to be 7-8 G in photolyase from a measurement of 2] in
Figure 4.1a”. The simulation used 7 .5 G for the exchange interaction, which is within the
predicted range for it. An estimate of the ratio of the low field exchange split pair can be
made by examining Figure 4. la. The figure shows that the ratio of the emissive to the
absorptive transition in the exchange split pair is at least two. The best fit to the transient
spectra was obtained with a ratio of 3.5 as shown in Table 5.1. The linewidth was
estimated to be 4 G from the half width at half height of the transitions in Figure 4.1a.

The simulation of the natural abundance transient spectrum that is shown in Figure 5.3a
agrees quite well with the experimental spectrum, which is shown directly above. The
three major transitions that show hyperfine coupling of 15 G in a 122:1 ratio in the
experimental spectrum are also present in the simulation as well as the smaller splittings
that arise from a single ring proton. A comparison of the magnetic field where a transition
occurs in each spectrum shows that the simulation shows transitions that are at the same
field as the experimental spectrum. The intensity of each transition does not agree as well

with the experimental spectrum, but it is probably due to a lack of experimental data such

171
as anisotropy in the g and hyperfine tensors, accurate hyperfine couplings, and the relative

orientations of the g and hyperfine axes rather than a limitation of the simulation program.

The simulation of the ring deuterated tryptophan transient spectrum that is shown in
Figure 5.3b is also qualitatively similar to the experimental spectrum that is shown above.
The simulation shows the same large hyperfine couplings as the experimental spectrum,
and the intensity and magnetic field position of each transition are the same in each
spectrum. The simulation does show small splittings that are not present in the
experimental spectrum, but this error is again probably due to a lack of experimental data.

Figure 5.4a shows the integrated absorption spectrum of the spectra in Figure 5.3a for
the natural abundance sample, and Figure 5.4b shows the integrated absorption spectra of
the spectra in Figure 5.3b for the deuterated sample. The simulation in each case agrees
very well with the experimental spectra in the figure. The simulations show the same six
line polarization pattern that alternates between absorption and emission. The splitting
between each exchange split pair and the splitting within the pair are also simulated very
well when compared to the experimental spectra. The net polarization of each exchange
split pair is also preserved in each simulation.

The simulation results in this chapter give strong support for the proposed mechanism
of spin polarization in photolyase that was presented in Chapter 4. The simulation
program is able to reproduce the experimental spectra to a high degree of accuracy, and
the results are quite good for a first attempt at simulating spin polarization in photolyase.
When better experimental data such as anisotropy in the g and hyperfine tensors, more
accurate hyperfine couplings, and the relative orientation of the g and hyperfine axes is

available; the simulation program should be able to better reproduce the transient spectra

172

Figure 5.4. Comparison of simulated and experimental absorption spectra for photolyase.
(a) is the natural abundance spectrum and (b) is the deuterated spectra. The top spectrum
in each figure is the experimental spectrum and the bottom one is the simulated one.

173

a) Natural
abundance

 

 

b) a ring-deuterated

 

 

 

l l L I l J I I l l l l I I

3220 3240 3260 3280 3300 3320 3340 3360

Gauss
Figure 5.4

l 74
in photolyase.

1 75
References

l. Sancar, A. Photolyase: In Advances in Electron T rarrsfer Chemistry; Mariano, P.E.,
Ed.; JAI Press: London, 1992; Vol. 2, pp. 215-272.

2. Kim, S.T.; Malhotra, K. and Sancar, A. In Biological Responses to Ultraviolet
Radiation; Urbach, F., Ed.; Valdenmar: Overland Park, KS, 1992.

3. Heelis, P.F.; Okamura, T. and Sancar, A. Biochemistry 1990, 29, 5694.

4. Essemnacher, C.; Kim, S.T.; Atamian, M.; Babcock, GT. and Sancar, A. J. Am. Chem.
Soc. 1993, 115, 1602.

5. Kim, S.T.; Sancar, A; Essenmacher, C. and Babcock, G.T. Proc. Natl. Acad. Sci. USA
1993, 90, 8023.

6. Brok, M.; Babcock, G.T.; DeGroot, A. and Hoff, A.J. J. Magn. Res. 1986, 70, 368.

7. Hoganson, CW. and Babcock, G.T. Biochemistry 1992, 31, 11874.

8. Norris, J.R.; Morris, A.L.; Thumauer, MC and Tang, J. J. Chem. Phys. 1990, 92,
4239.

9. Snyder, SW. and Thumauer, M.C. Electron Spin Polarization in Photosynthetic
Reaction Centers: The Photosynthetic Reaction Center 1993, Vol.2 ed. Deisenhofer, J.
and Norris, J .R. (Academic Press, SanDiego) p. 285.

10. Jenks, W.S. and T urro, N.J. Rev. Chem. Int. 1990, 13, 237.

11. Adrian, F.J. Rev. Chem. Int. 1986, 7, 173.

12. Kawai, A; Okutsa, T. and Obi, K. J. Phys. Chem. 1991, 95, 9130.

13. Kawai, A. and Obi, K. J. Phys. Chem. 1992, 96, 52.

14. Lhoste, J .M.; Haug, A. and Hemmerich, P. Biochemistry 1966, 5, 3290.

APPENDIX

176

APPENDIX

Listing of program to simulate spin polarized EPR spectra

#include ”polheadh”

char F ileName[MAXPATH];
char F ileNamel[MAXPATH];
char F ileName2 [MAXPATH];
PRINTDLG pd;

BOOL PrevSave = FALSE:
int SimType=l;

float HMin,l-1Max;

float xepr [1024];

float yepr [1024];

float spec [5 12];

int DataNumber;
PTFileWindow ImportFileWindow;
P'I‘Window ParamWindow;
float Freq = 9400;

int NumberGroups = 0;

float gx = 2.0023;

float gy = 2.0023;

float gz = 2.0023;

int StepSize = 15;

int NoDiv = 5;

float Dalpha = 30;

float MemTheta = 0;

float MemPhi = 0;

float Omega = 0;

float Wobble = 5;

float Linewidth = 2;

float CrystalTheta = 0;

float CrystalPhi = 0;

float AHYD [3113118];

float NuclearSpin [8];

int NucleiPerGroup [8];

float Theta[8];

float Phi[8];

float Psi[8];

BOOL AccurGen = TRUE;
BOOL SpecCRPP = FALSE;
float Exchange = 5.0;

float Ratio = 5.0;

BOOL PrecurDou = TRUE;

177

BOOL SimTrue = FALSE;

BOOL ExperLoaded = F ALSE'.

int DrawMode = 0;

HPEN PaintPen;

float Back1=0;

float Tilt=0;

BOOL F ileSaved = TRUE;

PTScroller GlobalScroller,

BOOL Integrate = FALSE;

HDC PolarDC; // Device context for main window

void DrawSpectrum (HDC DrawDC)
{
SetCursor(LoadCursor(NULL,IDC_WAIT));
float dspec[512];
float yint[1024];
if (lIntegrate && ExperLoaded)
{
for (int g=0;g<1024;++g)
{
yint[g] = yeprlgl;
}
}
if (Integrate && ExperLoaded)
{
float Back=0;
for (int f=5;f<55;++l)
Back += yepr[f|;
Back /= 50;
for (int g=0;g<1024;++g)
{
yint[g] = 0;
}
for (int d=l;d<DataNumber;++d)
{
yint[d] = yint[d-l] + ((yepr[d]-Back-Back1-(Tilt*d))*(xepr[DataNumber-1]-xepr[0])/ 1022);
}

}

if (DrawMode != 2)

{

int istarLiend;

float w,a,b,c,br,dl,d2;

float gaussperch = (HMax - HMin) / 511;
w = Linewidth / gaussperch;
a = WW3);
b = 2 * pow(w,2);
c=4*w;
for (int d = 0;d<512;++d)
dspec [d] = 0;
for (int e=0;e<512;-H-e)
{
if (speCIel != 0)
{

istart = e - c;

178
iend = e + c;
if (istart < 0)
istart = 0;
if (iend > 511)
iend = 511;
for (int f= istart;f< iend + l;++l)
{
d1 = f - e;
d2 = POW(dl.2);
hr = -dl / a*exp(-dZ/b);
dspecm += br "‘ speCIe];
}
}
}
}
if (Integrate && SimTrue)
{
for (int a=l;a<512;++a)
{
dspec[a] = dspec[a-l] + dspec[a]"'(1-IMax-HMin)/51 l;
1
}
if (DrawMode == 1)
{
SetMapMode(DrawDC,MM_ANISOTROPIC):
RECT WindowRect;
HWND MainHandle = GetActiveWindowO;
GetClientRect(MainHand1e,&WindowRect);
SetWindowExt(DrawDC,681, 1200);
SetViewportExt(DrawDC,WindowRect.right,-WindowRectbottom);
float dspecmax = dspec[O];
float d8pecmin = dspeCIOI;
float hfl,ds;
for (int g=0;g<512;++g)
{
if(dspeCIgl < dsrrecmin)
dspecmin = dspec[gl;
if(d8pedgl > dspecmax)
} dwecmax = 6896013];
SetWindowOrg(DrawDC,-85, 1000);
MoveTo(DrawDC,0,800);
LineTo(DrawDC,0,0);
LineTo(DrawDC,51 1,0);
float dstemp = (dspec[O]-dspecmin)*800/(dspecmax-dspecmin);
MoveTo(DrawDC,0,dstemp);
for (int h=1;h<512;++h)
{
hfl = h;
ds = (dspec[h]-dspecmin)*800/(dspecmax-dspecmin);
LineTo(DrawDC,hfl,ds);
}
MoveTo(DrawDC.0.0);
LineTo(DrawDC,0,-20);

179

MoveTo(DrawDC,51 l/3,-20);
LineTo(DrawDC,51 l/3,0);
MoveTo(DrawDC,51 1‘2/3,-20);
LineTo(DrawDC,5 1 1 *2/10);
MoveTo(DrawDC,511,0);
LineTo(DrawDC,51 1,-20):
char FieldText [7];
RECT FormatField;
FormatField.t0p = -25;
FonnatField.bottom = -80_;
FormatFieldJeft = -20;
FormatField.right=80;
gcvt(HMin,5.FieldText);
DrawText(DrawDC,FieldText,-l,&FormatField,DT_TOP | DT_LEF'I‘);
Ofl'setRect(&FonnatField,51 1/3,0);
gcvt((1-Nax-1Mn)/3+I-IMin,5,FieldText);
DrawText(DrawDC,FieldText,-1,&FormatField,DT_TOP | DT_LEFT);
Ofl‘setRect(&FormatField,51 l/3,0);
gcvt((HMax-1~I1VIin)"2/3+I-Mn,5,FieldText);
DrawText(DrawDC,FieldText,-l,&FormatField,DT_TOP | DT_LEFT);
Ofl‘setRect(&FormatField.51 1/3,0);
gcvtflIMax,5,FieldText);
DrawText(DrawDC.FieldText.-l,&FormatField,DT_TOP | DT_LEFT);
Ofl‘setRect(&FormatF ield,-51 1/2 - 50,930);
FormatField.right = 400;
DrawText(DrawDC,”Simulated Spectrum", l8.&FormatField,DT_TOP | DT_LEFT);
}
if (DrawMode == 2)
{
HWND MainHandle = GetActiveWindowO;
RECT WindowRect;
GetClientRectMainHandle,&WindowRect);
SetMapMode(DrawDC,MM_ANISOTROPIC);
SetWindowExt(DrawDC,681, 1200);
SetViewportExt(DrawDC,WindowRect.right,-WindowRect.bottom);
float specmin = yint[O];
float specmax = yint[0];
for (int a=1;a<DataNumber,++a)
{
if (yint[a] < specmin)
specmin = yint[a];
if (yint[a] > specmax)
}specmax = yint[a];
SetWindowOrg(DrawDC,-85, 1000);
MoveTo(DrawDC,0,800);
LineTo(DrawDC,0,0);
LineTo(DrawDC,511,0):
float YConv = 800/(specmax-specmin);
float XConv = 511/(xepr[DataNumber-1] - xepr[0]);
MoveTo(DrawDC,0,(yint[0]-specmin)*YConv);
for (int b=1;b<DataNumber;+—+b)
{
LineTo(DrawDC,(xepr[b]-xepr[0])*XConv,(yint[b]-specmin)*YConv);

180

}
MoveTo(DrawDC,0,0);
LineTo(DrawDC,0,-20);
MoveTo(DrawDC,51 1/3,-20);
LineTo(DrawDC,51 1/3,0);
MoveTo(DrawDC,5 1 1 ‘2/3,-20);
LineTo(DrawDC,51 1*2/3.0);
MoveTo(DrawDC,51 1,0);
LineTo(DrawDC,51 1,-20):
char FieldT ext [7];
RECT FormatField;
FormatField.t0p = -25;
FormatField.bottom = -80;
FonnatField.lefl = -20;
FonnatField.right=80;
gcvt(xepr[0],5,FieldText);
DrawText(DrawDC,FieldText.-l.&FonnatField,DT_TOP | DT_LEF1‘);
Ofi’setRect(&FormatField,51 l/3.0);
gcvt((xepr[DataNumbcr-l ]-xepr[0] )/3+xepr[0] ,5,FieldText);
DrawText(DrawDC,FieldText,-l.&FormatFicld,DT_TOP | DT_LEFI);
OflsetRect(&FormatField,51 l/3,0):
gcvt((xepr[DataNumber-l]-xepr[0])*2/3+xepr[0].5,FieldText);
DrawText(DrawDC,FieldText.-1.&FonnatF ield,DT_TOP | DT_LEFF);
Ofl‘setRect(&FonnatField,5 l 1/3,0);
gcvt(xepr[DataNumber-l],5,FieldText);
DrawText(DrawDC,FieldText,-1.&FormatField,DT_TOP | DT_LEF1’);
Ofl‘setRect(&FormatF ield,-51 1/2 - 50,930);
FormatField.right = 400;
DrawText(DrawDC,'Experimental Spectrum”,2l.&FomtatField,DT_TOP | DT_LEFI‘);
1
if (DrawMode == 3)
{
HWND MainI-Iandle = GetActiveWindowO;
RECT WindowRect;
GetClientRect(MainHandle,&WindowRect);
SetMapMode(DrawDC,1V[M_ANlSOTROPIC);
SetWindowExtCDrawDC,68 1 . 1200);
SetViewportExt(DrawDC,WindowRect.righL-WindowRectbottom);
float specmin = yint[O];
float specmax = yint[O];
for (int a=l;a<DataNumber,++a)
{
if (yint[a] < specmin)
specmin = yint[a];
if (yint[a] > specmax)
specmax = yint[a];
}
float dspecmin = dspec[O];
float dspecmax = dspec[O];
for (int b=1;b<512;++b)
{
if (dspedbl < dsvecmin)
dSPecmin = (1896011)];
if (dweclbl > dSPecmax)

181

}dSp°cmax = (1896611)];
SetWindowOrg(DrawDC,-85, 1000);
MoveTo(DrawDC,0,800);
LineTo(DrawDC,0,0);
LineTo(DrawDC,51 1,0);
float YConv = 800/(specmax-specmin);
float XConv = 511/(xeprlDataNumber-1] - xepr[0]);
MoveTo(DrawDC,0,(yint[0]-specmin)"'YConv);
for (int e=1 ;e<DataNumber;++e)
{
LineTo(DrawDC,(xepr[e]-xepr[0])*XConv,(yint[e]-specmin)*YConv);
}
YConv = 800/(dspecmax-dspecmin);
SelectObject(DrawDC,GetStockObject(BLACK_PEN));
PaintPen = CreatePen(PS_SOLID,1,0x000000FF);
SelectObject(DtawDC,PaintPen);
MoveTo(DrawDC,(HMin-xepr[0])‘XConv,(dspec[0]-dspecmin)“'YConv);
for (int c=1;c<512;++c)
{
LineTo(DrawDC,((HM in+c*(HMax-H]V11n)/5 1 1)-xepr[0])*XConv,(dspec[c]-dspecmin)*YConv);
}
MoveTo(DrawDC.0.0):
LineTo(DrawDC,O,-20):
MoveToGDrawDC,51 1/3,-20);
LineTo(DrawDC,51 1/3,0);
MoveTo(DrawDC,5 1 1*2/3,-20);
LineTo(DrawDC,5 1 l *2/3,0);
MoveTo(DrawDC,51 1,0);
LineTo(DrawDC,51 1,-20);
char FieldText [7];
RECT FormatField;
FormatField.top = -25;
FonnatField.bottom = -80;
FormatFieldJefi = -20;
FonnatField.right=80;
gcvt(xepr[0],5,FieldText);
DrawText(DrawDC,FieldText,-l.&FormatField,DT_TOP | DT_LEFT);
Ofl‘setRect(&FonnatField,51 l/3,0);
gcvt(( xeprlDataN umber-1 ]-xepr[0])/3 +xepr[0],5,FieldText);
DrawText(DrawDC,FieldText.-l .&FormatField.DT_TOP | DT_LEPT);
OflsetRect(&FomatF ield.51 l/3,0);
gcvt((xepr[DataNumber-1 ]-xepr[0])"‘2/ 3 +xepr[0],5,FieldText);
DrawText(DrawDC,FieldText.-1.&FormatF ield,DT_TOP | DT_LEF1‘);
Ofl'setRect(&FormatField,51 1/3,0)'.
gcvt(xepr[DataNumber-l],5,FieldText);
DrawText(DrawDC,FieldText,-l,&FonnatField,DT_TOP | DT_LEFT);
OlfsetRect(&FormatFicld,-511/2 - 50,930);
FonnatField.right = 400;
DrawText(DrawDC,"Overlaid Spectra",16.&FormatField,DT_TOP | DT_LEFT);
}
if (DrawMode == 4)

{
HWND MainHandle = GetActiveWindowO;

182

RECT WindowRect;
GetClientRect(MainHandle,&WindowRect);
SetMapMode(DrawDC,MM_ANISOTROPIC);
SetWindowExt(DrawDC,681, 1200);
SetViewportExt(DrawDC.WindowRect.right,-WindowRect.bottom);
float specmin = yint[O];
float specmax = yint[O];
for (int a=1;a<DataNumber;++a)
{
if (yint[a] < specmin)
specmin = yint[a];
if (yint[a] > specmax)
specmax = yint[a];
}
float dspecmin = dspec[O];
float dspecmax = dspec[O];
for (int b=l;b<512;++b)
{
if (dspec[b] < dspecmin)
dspecmin = dSPGCIb];
if (dSPeCIbl > dspecmax)
)dsmcmax = dSPeclb];
SetWindowOrg(DrawDC,-85, 1000);
MoveTo(DrawDC,0.800);
LineTo(DrawDC,0,0);
LineTo(DrawDC,51 1,0);
float YConv = 800/(specmax-specmin);
float XConv = 511/(xepr[DataNumber-l] - xepr[0]);
MoveTo(DrawDC,0,(yint[O]-specmin)*YConv);
for (int e= l ;e<DataNumber,++e)
{
LineTo(DrawDC,(xepr[e]-xepr[0])*XConv,(yint[e]-specmin)*YConv);
}
YConv = 800/(dspecmax-dspecmin);
MoveTo(DrawDC,0,(dspec[0]-dspecmin)*YConv- 1 200);
for (int c=1 ;c<5 12;++c)
{
LineTo(DrawDC,c,(dspec[c]-dspecmin)*YConv-1200);
}
MoveTo(DrawDC,0,0);
LineTo(DrawDC,0,-20);
MoveTo(DrawDC,51 l/3,-20);
LineTo(DrawDC,51 l/3,0);
MoveTo(DrawDC,5 1 1*2/3,-20):
LineTo(DrawDC,5 1 l *2/3,0);
MoveTo(DrawDC,51 1,0);
LineTo(DrawDC,51 l,-20);
char F ieldText [7];
RECT FormatField;
FonnatField.top = -25;
FonnatField.bottom = -80;
FormatFieldJeft = -20;
FonnatField.right=80;

183

gcvt(xepr[0],5,FieldText);
DrawText(DrawDC,FieldText,-l.&FormatField.DT_TOP | DT_LEFT);
OflsetRect(&FormatF ield,51 1/3,0);
gcvt(( xeprlDataNumber— 1 ]-xepr[0])/3 +xepr[0],5,FieldT ext);
DrawText(DrawDC,FieldTexL-1.&FormatF ield,DT_TOP | DT__LEFT);
OffsetRect(&FormatField,51 1/3,0);
gcvt((xepr[DataNumber-1]-xepr[0])*2/3+xepr[0],5,FieldText);
DrawText(DrawDC.FieldText,-l,&FormatField,DT_TOP | DT_LEFI');
Ofl‘setRect(&FormatF ield,51 1/3,0);
gcvt(xepr[DataNumber- l ],5,FieldText);
DrawText(DrawDC,Fie1dText,-l,&FormatField,DT_TOP | DT_LEFI‘);
Ofl'setRect(&FormatF ield,-51 1/2 - 50,930);
FormatF ield. right = 400;
DrawText(DrawDC,”Expen'mental Spectrum”,2l,&FonnatField,DT_TOP | DT_LEFI‘);
MoveTo(DrawDC,0,-1200);
LineTo(DrawDC,0,-1220);
MoveTo(DrawDC,51 1/3,-1220);
LineTo(DrawDC,51 1/3,-1200);
MoveTo(DrawDC,5 l 1*2/3,-1220);
LineTo(DrawDC,51 1*2/3,-1200);
MoveTo(DrawDC,51 l,-1200);
LineTo(DrawDC,51 1,-1220);
MoveTo(DrawDC,0,-400);
LineTo(DrawDC,0,-1200);
LineTo(DrawDC,51 1,-1200);
FormatField.top = -1225;
FormatField.bottom = ~1280;
FormatFieldJefl = -20;
FonnatField.right=80;
gcvt(H]Vlin,5,FieldText);
DrawText(DrawDC,FieldText,-1,&FormatF ield,DT_TOP | DT_LEF1‘);
OflsetRect(&FormatField,51 l/3.0);
gcvt(1fl\4in+(1flV13x-INin)/3,5,F ieldText);
DrawText(DrawDC,FieldText,-1,&FormatField,DT_TOP | DT_LEF'I’);
Ofl’setRect(&FormatField,51 1/3,0);
gcvt(1-IMin+(Hl\/1ax-HMin)*2/3,5,FieldText);
DrawText(DrawDC,FieldText,-1.&FormatField,DT_TOP | DT_LEFT);
Ofl‘setRect(&FonnatF ield,51 l/3,0);
gcvt(1-INIax,5,FieldText);
DrawText(DrawDC,FieldText,-l,&FormatField,DT_TOP | DT_LEFF);
Ofl'setRect(&FormatF ield,-51 1/2 - 50,930);
FormatFieldJight = 600;
DrawText(DrawDC,"Simulated Spectrum",18,&FormatField.DT_TOP l DT_LEFT):
1
if (DrawMode == 10)
{

HWND MainHandle = GetActiveWindowO;
ofstream exportos(FileName2);
if (lexportos.bad())
{

for (int s=0;s<512;++s)

{
exportos << (HMin + s*(HMax-HMin)/511) << " " << dspec[sl << "\n":
}

184
}

else
MessageBox(MainHandle."File cannot be opened”,"Disk Error",
MB_OK | MB_ICONSTOP );
exportos.close();
}
if (DrawMode == 11)
{
POINT PageSize,Ofl‘setSize;
Escape(DrawDC.GETPHYSPAGESIZENULL,NULL,&PageSize);
Escape(DrawDC,GETPRINTINGOFFSET,NULL,NULL.&OlTsetSize);
SetMapMode(DrawDC.MM_ANISOTROPIC);
SetWindowExt(DrawDC,690, 1200);
SetViewportExt(DrawDC.PageSize.x - 2*Ol’fsetSize.x,-(PageSize.y-2*OflsetSize.y));
SetWindowOrg(DrawDC.-85, 1000);
SetViewportOrg(DrawDC,-Ofl'setSize.x,—Ofl‘setSize.y);
float dspecmax = dspec[O];
float dspecmin = dspec[O];
for (int v=l;v<512;++v)
{
if (dspec[v] < dspecmin)
dsrrecmin = dspwlv]:
if (dSPOCIVI > dsrmmu)
mm = dSPeCIVI;
}

float YConv = 800/(dspecmax-dspecmin);
Escape(DrawDC.STARTDOC.6.”SimDoc",NULL);
MoveTo(DrawDC,0,800);
LineTo(DrawDC,0.0);
LineTo(DrawDC,51 1,0);
LineTo(DrawDC,5 1 l,-20);
MoveTo(DrawDC,51 1/3‘2,0);
LineTo(DrawDC,51 1/3'2,-20);
MoveTo(DrawDC,51 l/3 ,0);
LineTo(DrawDC,51 1/3,-20);
MoveTo(DrawDC,0.(dspec[0]-dspecminYYConv);
for (int y=1;y<512;++y)
{

LineTo(DrawDC,y,(dspec[y]dspecminY'YConv);
}
char FieldPrint [10];
RECT FieldRect;
FieldRectJefi = -20;
FieldRect. right = 80;
FieldRect.top = -25;
FieldRect.bottom = -85;
gcvt(Hl\din,6,FieldPrint);
DrawText(DrawDC,FieldPrint,-l,&FieldRect,DT_TOP | DT_LEFT);
Ofl'setRect(&FieldRect,51 l/3.0)‘.
gcvt((1Nin+(HMax-I—Il\din)/3),6.FieldPrint);
DrawText(DrawDC,FieldPrint.-l.&FieldRect.DT_TOP | DT_LEFI'):
Ofl‘setRect(&F ieldRect.51 1/3,0);
gcvt((HlVlin+(HI\4ax-HMin)*2/3),6.FieldPrint);
DrawText(DrawDC,FieldPrint,-1,&FieldRect,DT_TOP | DT_LEFI‘);

185

Ofl‘setRect(&FieldRect,51 l/3,0);
gcvt(H]\dax,6,FieldPrint);
DrawText(DrawDC,F ieldPrinL-1,&FieldRect,DT_TOP | DT_LEFT ):
Ofl‘setRect(&FieldRect,—51 1/2-10,-40);
DrawText(DrawDC,"Gauss",5,&FieldRect,DT_TOP | DT_LEFT);
F ieldRectJeft = 511/2-70;
FieldRect.right = 511/2 +200;
FieldRect.top = 900;
FieldRect.bottom = 820;
DrawText(DrawDC,"Simulated Spectrum”,l8,&FieldRect,DT_TOP | DT_LEFT);
Escape(DrawDC,NEWFRAME,0.NULL,NULL);
Escape(DrawDC,ENDDOC,0,NULL,NULL);

}

SetCursor(LoadCursor(NULL,IDC_ARROW));

}

// define the application class derived from TApplication
class TPolarization :public TApplication
{

public:

TPolarization (LPSTR AName. HANDLE hlnstance, HANDLE
hPrevInstance. LPSTR lpCmdLine, int nCmdShow) :
TApplication ( AName. hlnstance, hPrevInstance,

lpCmdLine, nCmdShow) {};

virtual void InitMainWindow 0;

};

// Construct a Window class for main window derived from TWindow

class TPolarWindow : public TWindow

{

public: // Data members for TPolarWindow

PTWindow ImportWindow,PolParamWindow;
PTWindow FieldWindow;
HPEN PaintPen;
BOOL helpb;
// Declare functions for the file menu choices
virtual void CMFileNew (RTMessage Msg) = [CM_FIRST + CM_FILENEW];
virtual void CMFileOpen (RTMessage Msg) = [CM_FIRST + CM_FILEOPEN];
void CMFileImport (RTMessage Msg) = [CM_FIRST + CM_IMPORT];
virtual void CMFileSave (RTMessage Msg) = [CM_FIRST + CM_FILESAVE]:
virtual void CMFileSaveAs (RTMessage Msg) = [CM_FIRST + CM_FILESAVEAS];
void CMFileExport (RTMessage Msg) = [CM_FIRST + CM_EXPORT];
void CMPrint (RTMessage Msg) = [CM_FIRST + CM_PRINT];
void CMPrintSetup (RTMessage Msg) = [CM_FIRST + CM_PRINTSETUP]:
virtual void CMExit (RTMessage Msg) = [CM_FIRST + CM_EXIT];
virtual BOOL CanCloseO;
virtual LPSTR GetClassName();
virtual void GetWindowClass(WNDCLASS& AWndClass);

// Declare functions for View menu choices
void CMViewExper (RTMessage Msg) = [CM_FIRST + CM_VIEWEXJPER];
void CMViewSim (RTMessage Msg) = [CM_FIRST + CM_VIEWSIM];
void CMViewBoth (RTMessage Msg) = [CM_FIRST + CM_VIEWBOTH];
void CMViewPaged (RTMessage Msg) = [CM_FIRST + CM_VIEWPAGED];

186

// Declare functions for simulate menu choices
void CMSimulate (RTMessage Msg) = [CM_FIRST + CM_SIMSPEC];
void CMSimParam (RTMessage Msg) = [CM_FIRST + CM_SIMPARAM];
void CMSimPolParam (RTMessage Msg) = [CM_FIRST + CM_SIMPOLPARAM];
void CMSimLineChange (RTMessage Msg) = [CM_FIRST + CM_SIMLINE];

// Declare integrate menu functions
void CMIntegDrawInteg (RTMessage Msg) = [CM_FIRST + CM_DRAWINTEG];
void CMIntegEPR (RTMessage Msg) = [CM_FIRST + CM_INTEGEPR];
void CMIntegBase (RTMessage Msg) = [CM_FIRST + CM_INTEGBASE];
void CMIntegBaseOrg (RTMessage Msg) = [CM_FIRST + CM_INTEGBASEORG];
// Declare various utility functions that are need
void OpenFileO;
virtual void Paint(1-IDC PDC,PAINTSTRUCT& PS);
void F ileSave();
void CNflielplndex(RTMessage Msg) = [CM_FIRST + 901];
void CMCommands(RTMessage Msg) = [CM_FIRST + 903];
void CMHelpOnHelp(RTMessage Msg) = [CM_FIRST + 902];
// Constructor for TPolarWindow
TPolarWindow (PTWindowsObject AParent, LPSTR ATitle);
~TPolarWindow()

{
if (helpb)
WinHelp(HWindow,"simhelp.hlp".HELP_QUIT,0);
}

};

TPolarWindowzzTPolarWindow (PT WindowsObject AParent, LPSTR ATitle) :
TWindow (AParent. ATitle)

{
helpb = FALSE;
Attr.X = 10;
Attr.Y = 10;
Attr.W = 620;
Attr.H = 460;
Attr. Style |= WS_VSCROLL;
Scroller = new TScroller(this,1.23.2, l9):
AssignMenu("MENU_l");
GlobalScroller = Scroller;
PolarDC = GetDC(HWindow);
long int begin = GetCurrentTimeO;
BITMAP bm;
HBITMAP OldBit;
HDC BitmapDC = CreateCompatibleDC(PolarDC)z
HBITMAP IntroBitmap = LoadBitmap(GetApplication()->h1nstance,”BITMAP_1");
GetObjectantroBitmap.sizeof(BITMAP).&bm);
OldBit = SelectObject(BitmapDC.IntroBitmap):
BitBlt(PolarDC,160,80,bm.bmWidth,bm.bmHeight.BitmapDC,0.0.SRCCOPY);
start:
long int end = GetCurrentTimeO;
if ((end-begin) < 4000 || (end-begin) < 0)
goto start;
SelectObject(BitmapDC.OldBit):

187

DeleteObject(OldBit);
DeleteObject(IntroBitmap);
InvalidateRect(HWindow,NULL,TRUE);
DeleteDC(BitmapDC);
ReleaseDC(HWindow.PolarDC):

}

”Define class to create the parameter window

class TParamWindowzpublic TWindow

{

public:

char FreqText [10];

char NumberGroupsText [3];

char ngext [10];

char gyText[10];

char ngext[10];

char StepSizeText[5];

char NoDivText [5];

char DalphaText[ 10];

char MemThetaTextl 10];

char MemPhiTextl 10];

char OmegaText[ 10];

char WobbleText[ 10];

char LinewidthText[10];

char CrystalThetaText[10];

char CrystalPhiTextl 10];

PTComboBox SimTypeCombo;

P'I'Window GroupsWindow;

PTEdit LinewidthEdiLNoDivEdiLDalphaEdit.WobbleEdiLOmegaEdit;
PTEdit CrystalThetaEdit.CrystalPhiEdiLMemThetaEdit,MemPhiEdit;
PTButton OKButtomCanButton:

PTListBox AccurListBox;

PTEdit Frqudit.NoGroupsEdit.ngdiLgyEdit,ngdit,StepSizeEdit;
virtual void SetupWindowO;

void HandleCancelButton (RTMessage Msg) = [ID_F1RST + 28];
void HandleOKButton (RTMessage Msg) = [ID_F1RST + 27];
TParamWindow(PTWindowsObject AParent, LPSTR ATitle);

};
// define a class to enter hyperfine values for each group
class TNumberGroupWindowzpublic TWindow

{

public:

clmr AxText[8],AyText[8],AzText[8]'.

char PhiText[8],ThetaText[8],PsiText[8]:

char NoNucleiTextl8];

char NuclearSpinTextl8];

char GroupNumber [2];

PTEdit AxEdit [8];

PTEdit AyEdit [8];

PTEdit AzEdit [8];

PTEdit PhiEdit [8];

PTEdit PsiEdit [8];

PTEdit Theth [8];

PTEdit NoNucleiEdit [8]:

188

PTEdit NuclearSpinEdit [8];

PTButton OKButtomCancelButton;

void HandleButtonOK(RTMessage Msg) = [ID_F1RST + 94];

void HandleButtonCancel (RTMessage Msg) = [ID_F1RST + 95];
TNumberGroupWindow(PTWindowsObject AParent,LPSTR ATitle);
};

/l Define class to query the user to choose I-I field values for simulation
class TFieldWindowzpublic TWindow
{
public:
char HMAX [10],HMIN [10]:
float ammt;
float atemp;
float gmax;
float gmin;
void SimulateSpectrum ();
void goniomf(float Thetal.float Phi1,float Alpha,float Hohm,float &X1,float &Y 1 .
float &Zl);
void HandleYesButtonField (RTMessage Msg) = [ID_F1RST + 150];
void HandleNoButtonField (RTMessage Msg) = [ID_F1RST + 151];
TFieldWindow(PTWindowsObject AParenLLPSTR ATitle);
};
// Define a class to create an import window
class 'I‘ImportWindowzpublic TWindow
{
public:
char delimiter,
int F ileBeginRead,FileEndRead;
P'I‘GroupBox DelimGroupBox; // Creates a group of radio buttons
PTGroupBox LineGroupBox; // another group
PTRadioButton ImportRadio;
PTRadioButton ImportRadiol;
P'I'RadioButton ImportRadiOZ;
PTRadioButton ImportRadio3;
PTRadioButton ImportRadio4;
P'I‘Static ImportStatic;
PTButton ImportButton;
PTButton ImportButton 1;
virtual void SetupWindowO;
void HandleButtonOKImport (RTMessage Msg) = [ID_F1RST + 5];
void HandleButtonCanImport (RTMessage Msg) = [ID_F1RST + 6];
void HandleRadioOtherlmport (RTMessage Msg) = [ID_F1RST + 4];
void HandleRadioOtherLinesImport (RTMessage Msg) = [ID_F1RST + 8];
TlmportWindow(PTWindowsObject AParent, LPSTR ATitle):
TWindow(AParent.ATitle)
{
FileBeginRead = 1;
FileEndRead = -1;
DisableAutoCreateO;
Attr. Style |= WS_POPUP | WS_CAPTION;
Attr.X = 20;
Attr.Y = 20;
Attr.W = 600;

189

Attr.H = 210;
// Creation of the groupbox
DelimGroupBox = new TGroupBox(this. l."Delimiter in File”,
10,10,185,120);
// Creation of radiobuttons
ImportRadio = new TRadioButton(this.2,”whitespace”,
30,30,100,30.DelimGroupBox);
ImportRadio->Attr. Style |= WS_TABSTOP;
ImportRadiol = new TRadioButton(this.3,”comma”.
30,60, 100,30,DelimGroupBox);
ImportRadioZ = new TRadioButton(this,4,"other",
30,90,100,30,DelimGroupBox);
// creation of another groupbox
LineGroupBox = new TGroupBox(this.9,"Lines to Input”,
196, 10,250, 120);
// Creation of radiobuttons for LineGroupbox
ImportRadio3 = new TRadioButton(this.7,"All',
215,40,100,30,LineGroupBox);
ImportRadio3->Attr. Style |= WS_TABSTOP;
ImportRadio4 = new TRadioButton(this.8,"Other",
215,80, 100,30,LineGroupBox);
// Create static text in the window
ImportStatic = new TStatic(this,-1,”View of the file”,
0,140,580,30,17);
ImportStatic->Attr. Style = ImportStatic->Attr.Style & ~SS_LEFT l SS_CENTER;
// MAKE A PUSH BUTTON FOR USER TO MAKE A CHOICE
ImportButton = new TButton(this.5,”OK",500,30.70,30,TRUE);
ImportButton->Attr. Style |= WS_TABSTOP | BS_DEFPUSHBU'ITON;
ImportButton] = new TButton(this.6."Cancel",500,70,70,30,TRUE);
ImportButtonl->Attr. Style |= WS_TABSTOP;
EnableKBHandlerO; // Enables the tab key
1
1;
// Make a window for the polarization parameters
class TPolParamWindowzpublic TWindow
{
public:
PTGroupBox PolGroupBox,PolGrouplBox;
PTEdit PolEdiLPolEditl;
PTRadioButton PolCheckBox,PolChecleox;
PTRadioButton PolCheckZBox,PolCheck3Box;
PTButton PolButtomPolButtonl;
char CIDEPRatio [9];
char EXCHANGE [9];
virtual void SetupWindowO:
void HandleOKPolButton (RTMessage Msg) = [ID_F1RST + 168];
void HandleCanPolButton (RTMessage Msg) = [ID_F1RST + 169];
TPolParamWindowfl’I'WindowsObject AParenLLPSTR ATitle);

}.

TPolParamWindow::TPolParamWindow (PTWindowsObject AParent. LPSTR ATitle):
TWindow(AParent,ATitle)

{

DisableAutoCreateO;

190

Attr.Style |= WS_POPUP l WS_CAPTION;

Attr.X = 50;

Attr.Y = 120;

Attr.W = 540;

Attr.H = 240;

gcvt(Exchange,5,EXCHANGE);

gcvt(Ratio,5,CIDEPRatio);

PolGroupBox = new TGroupBox(this. 160,”Spectrum Type”,10.10,125,70);
PolCheckBox = new TRadioButton(this.161,"Nonnal",40,30.80,20,PolGroupBox);
PolChecleox = new TRadioButton(this,162,”CRPP”,40,50,80,20,PolGroupBox);
PolGrouplBox = new TGroupBox(this.163,"Precursor",145.10, 125.70);
PolCheckZBox = new TRadioButton(this.164,"Doub1et",l75,30,80.20.PolGrouplBox);
PolCheck3Box = new TRadioButton(this.165,"Quartet",175.50,80,20.PolGrouplBox);
new TStatic(this,-l,"Ratio of peaks with largest coupling".20,100,150,40,36);
PolEdit = new TEdit(this.166.CIDEPRatio.30,150.70.25.6,FALSE);
PolEdit->Attr.Style |= WS_TABSTOP;

new TStatic(this.-l,"Exchange value(Gauss)",305,10,125,502l);

PolEdit] = new TEdit(this,167,EXCHANGE,305,55,100,25,6,FALSE);
PolEdit1->Attr. Style |= WS_TABSTOP;

PolButton = new TButton(this,168,”OK",450,10,70,30,TRUE);
PolButton->Attr.Style |= BS_DEFPUSHBU'1TON | WS_TABSTOP;

PolButtonI = new TButton(this, l69,"Cancel',450.50,70,30,TRUE);
PolButtonl->Attr. Style |= WS_TABSTOP;

}

void TPolParamWindow: :SetupWindow()

{
TWindow::SetupWindow();
if (lSpecCRPP)
PolCheckBox->Check0;
else
PolChecleox->Check();
if (PrecurDou)
PolCheckZBox->Check();
else
PolCheck3Box->Check();

}

// function that handles ok response for polarization parameters
void TPolParamWindow: :HandleOKPolButton (RTMessage)
{
if (PolCheckBox->GetCheck0 = BF_CHECKED)
SpecCRPP = FALSE;
if (PolChecleox->GetCheck() = BF_CHECKED)
SpecCRPP = TRUE;
if (PolCheck280x->GetCheck() == BF_CHECKED)
PrecurDou = TRUE;
if (PolCheck3Box->GetCheck() == BF_CHECKED)
PrecurDou = FALSE;
PolEdit->GetLine(CIDEPRatio,6. 1);
Ratio = atof(CIDEPRatio);
PolEdit l ->GetLine(EXCHANGE.6. 1 )1
Exchange = atoflEXCHANGE):
CloseWindowO;

191
}

// Function to handle a cancel response in the polarization parameter window
void TPolParamWindow: :HandleCanPolButton (RTMessage)

{
CloseWindowO;

}

TFieldWindow::TFieldWindow (PTWindowsObject AParenLLPSTR ATitle):
TWindow(AParent,ATitle)
{
Attr.Style |= WS_POPUP | WS_CAPTION:
Attr.X = 100;
Attr.W = 400;
Attr.Y = 100;
Attr.H = 240;
// Function to calculate hmax and hmin for simulation
8m=gx;
gnun=gx;
if (gmax < gy)
gmax=gy;
if (gmax < 31)
gmax=gz;
if (gmin > gy)
min = gy;
if (gmin > 32)
min = gz;
HMax = Freq‘0.7l4484/gmin;
HMin = Freq‘0.714484/gmax;
amax = 0;
if (NumberGroups > 0)
{
for (int i=0;i<NumberGroups:++i)
{
for (int j=0;i<3;-H-j)
{
atemp = AHyp Lilljlli];
if (atemp < 0)
atemp = -atemp;
if (atemp > amax)
amax = atemp;
}
HMin -= amax " NucleiPerGroup [i] " NuclearSpin [i];
HMax += amax * NucleiPerGroup [i] * NuclearSpin [i];
amax = 0;
1
}
HMax = HMax + 5 "' Linewidth;
HMin = HMin - 5 "‘ Linewidth:
gethNaXfiJNAX);
gcvtflIMinfiJ-IMIN);
new TStatic(this,-l,"The Calculated Value for Hmin is ",10.10.290,20.
33);
new TStatic(this,-1,1-IMIN.300.10.60.20.5izeof HMIN);
new TStatic(this,-1,"The Calculated Value for Hmax is ",10,50.290.20,

192

33);

new TStatic(this,-1,HMAX,300,60,60,20,sizeof HMAX);

new TStatic(this,-l,"Do you wish to use these values?",30,100,290,20,
32);

new TButton(this,150,”Yes", 150,140,70,30,FALSE);

new TButton(this, 151,”No",250,140,70,30,FALSE);

}

TParamWindowzzTParamWindow (PTWindowsObject AParent,LPSTR ATitle):
TWindow(AParent,ATitle)
{
DisableAutoCreateO;
Attr. Style |= WS_POPUP l WS_CAPTION;
Attr.X=l;
Attr.Y=l;
Attr.W=639;
Attr.H=479;
gcvt(Freq.6,FreqText);
gcvt(NumberGroups,2,NumberGroupsText);
gcvt(gx.6.ngext);
gcvt(gy..6.gyText);
gcvt(g2.6,ngext);
itoa(StepSize,StepSizeText, 10);
itoa(NoDiv,NoDivText, 10);
gcvt(Dalpha,6,DalphaText);
gcvt(MemTheta,6,MemThetaText);
gcvt(MemPhi,6.MemPhiText);
gcvt(Omega,6,0megaText);
gcvt(Wobble,6,WobbleText);
gcvt(Linewidth*2,6,LinewidthText);
gcvt(CrystalTheta,6,CrystalThetaText);
gcvt(CrystalPhi,6,CrystalPhiText);
new TStatic (this,-1,"Sim. Type”,50,1,90,20,9);
SimTypeCombo = new TComboBox(this,10,50,22,90.95,
CBS_DROPDOWNLIST, 10);
SimTypeCombo->Attr. Style |= WS_TABSTOP;
new TStatic (this,-I,”Freq. (M—Iz)".50,174,90,20,11);
Frqudit = new TEdit(this,] l,FreqText,50,196,90,25,9,FALSE);
Frqudit->Attr. Style |= WS_TABSTOP;
new TStatic (this,-l,"Accuracy",190, l,90,20,8);
AccurListBox = new TListBox(this,12,190,22,90,35);
AccurListBox->Attr. Style |= WS_TABSTOP;
new TStatic (this,-1,"Linewidth", 190, 174,90,20,9);
LinewidthEdit = new TEdit(this, l3,LinewidthText, 190, l96.90,25,9,FALSE);
LincwidthEdit->Attr.Style |= WS_TABSTOP;
new TStatic(this,-1,”gx-value”,50,87,90,20,8);
new TStatic(this,-l,"gy-value", 190.87 ,90,20,8);
new TStatic(this,-1,"gz-value”,330,87,90,20,8);
new TStatic(this,-l,"# of groups”,330,1.90.20,11);
ngdit = new TEdit(this,l4,ngext,50,109,90,25,9,FALSE);
ngdit->Attr.Style |= WS_TABSTOP;
gyEdit = new TEdit(this,15,gyText,190,109,90.25,9,FALSE);
gyEdit->Attr.Sty1e |= WS_TABSTOP;
ngdit = new TEdit(this, 16,ngext,330, 109.90.25.9,FALSE);

193

ngdit->Attr. Style |= WS_TABSTOP;

NoGroupsEdit = new TEdit(this, l7.NumberGroupsText,330,22,90,25.2,FALSE);
NoGroupsEdit->Attr. Style |= WS_TABSTOP;

new TStatic(t11is,-l,"Step Size”,330,174.90.20,9);

new TStatic(this,-l,"# div.”,470,174,90.20,6);

StepSizeEdit = new TEdit(this,l8.StepSizeText,330,196.90,25,4,FALSE);
StepSizeEdit->Attr. Style |= WS_TABSTOP;

NoDivEdit = new TEdit(this,19.NoDivText.470, l96,90,25,5,FALSE);
NoDivEdit->Attr. Style |= WS_TABSTOP;

new TStatic(this,-l,"da1pha",50.261.90.20.6);

new TStatic(t1us,-l,"omega', 190,26 1,90,20,5);

new TStatic(this,-1,"wobble",330.26l,90,20,6);

DalphaEdit = new TEdit(this,20,DalphaText,50,283,90,25,9,FALSE);
DalphaEdit->Attr.Style |= WS_TABSTOP;

OmegaEdit = new TEdit(this,2 l,OmegaText,190,283,90,25,9,FALSE);
OmegaEdit->Attr. Style |= WS_TABSTOP;

WobbleEdit = new TEdit(this,22,WobbleText,330,283,90,25,9,FALSE);
WobbleEdit->Attr. Style |= WS_TABSTOP;

new TStatic(this,-l,"Mem. Theta”,50,348,90,20,10);

new TStatic(this,-1,”Mem. Phi”,l90,348,90.20,8);

new TStatic(this,-1,"Cry. Theta”,330,348.90.20, 10);

new TStatic(this,-1,"Cry. Phi",470,348,90,20,8);

MemThetaEdit = new TEdit(this.23,MemThetaText,50,370,90,25,9,FALSE);
MemThetaEdit->Attr. Style |= WS_TABSTOP;

MemPhiEdit = new TEdit(this,24,MemPhiText,190,370,90,25,9,FALSE);
MemPhiEdit->Attr. Style |= WS_TABSTOP;

CrystalThetaEdit = new TEdit(this,25,CrystalThetaText,330,370,90,25,9,FALSE);
CrystalThetaEdit->Attr. Style |= WS_TABSTOP;

CrystalPhiEdit = new TEdit(this,26,CrystalPhiText,470,370,90,25,9,FALSE);
CrystalPhiEdit->Attr. Style |= WS_TABSTOP;

OKButton = new TButton(this,27,”OK”,500,30,70,30,TRUE);

CanButton = new TButton(this,28,"Cancel",500,70,70,30,TRUE);
EnableKBHandlerO;

1
void TParamWindow: :SetupWindowO

{
TWindow::SetupWindow();
SimTypeCombo->AddString("Powder”)'.
SimTypeCombo->AddString(”Oriented”);
SimTypeCombo->AddString(”Crystal");
SimTypeCombo—>AddStn'ng("Solution"):

if (SimType = 1)
SimTypeCombo->SetSelIndex(2);
if (SimType == 2)
SimTypeCombo—>SetSellndex( l );
if (SimType == 3)
SimTypeCombo->SetSelIndex(0);
if (SimType = 4)

SimTypeCombo—>SetSelIndex(3);
AccurListBox->AddString(”General");
AccurListBox->AddString("Approx.");
if (AccurGen)
AccurListBox->SetSelIndex( 1);

else

194

AccurListBox->SetSelIndex(0);
1
// Handles the ok choice in the parameter window
void TParamWindow: :HandleOKButton (RTMessage )
{
Frqudit->GetLine(FreqText,7, 1);
Freq = atof(FreqText);
if (Freq < 0)
Freq = -Freq;
if (Freq = 0)
Freq= 9400,
if (SimTypeCombo->GetSelIndex()== 0)
SimType= 3;
if (SimTypeCombo->GetSelIndex() = l)
SimType = 2;
if (SimTypeCombo->GetSelIndex() = 2)
SimType = 1;
if (SimTypeCombo—>GetSelIndex() = 3)
SimType = 4;
if (AccurListBox->GetSelIndex() = 0)
AccurGen = FALSE;
else
AccurGen = TRUE;
NoGroupsEdit—>GetLinemumberGroupsText, 3, 1);
NumberGroups= atof (NumberGroupsText);
if (NumberGroups > 8)
NumberGroups= 8;
if (NumberGroups < 0)
NumberGroups = 0;
ngdit->GetLine(ngext,7, 1);
gx = atof(ngext);
if (gx < 0)
3" = -gx;
if (gx = 0)
gx = 2.0023;
gyEdit->GetLine(gyText,7, l);
gy = atof(gyText);
if (gy < 0)
83' = -gy;
if (gy = 0)
gy = 2.0023;
ngdit->GetLine(ngext,7, 1);
gz = atof(ngext);
if (32 < 0)
81 = '82;
if (g2 = 0)
gz = 2.0023;
StepSizeEdit->GetLine(StepSizeText.5, 1);
StepSize = atoi(StepSizeText);
if (StepSize < 0)
StepSize = -StepSize;
if (StepSize < 1 || StepSize > 180)
StepSize = 15;
NoDivEdit->GetLine(NoDivText, 5, l);

195

NoDiv = atci(NoDivText);
if (NoDiv < 0)
NoDiv = -NoDiv;
if (NoDiv == 0)
NoDiv = 5;
DalphaEdit->GetLine(DalphaText,6, 1);
Dalpha = atof(DalphaText);
if (Dalpha = 0)
Dalpha = 30;
MemThetaEdit->GetLine(MemThetaText,6, l );
MemTheta = atof(MemThetaText);
MemPhiEdit->GetLine(MemPhiText,6, 1 );
MemPhi = atof(MemPhiText);
OmegaEdit->GetLine(OmegaText,6, 1);
Omega = atof(OmegaText);
WobbleEdit->GetLine(WobbleText,6, 1);
Wobble = atof(WobbleText);
CrystalThetaEdit->GetLine(CrystalThetaText,6, l );
CrystalTheta = atof(CrystalThetaText);
CrystalPhiEdit->GetLine(CrystalPhiText,6, l);
CrystalPhi = atof(CrystalPhiText); ’
LinewidthEdit->GetLine(LinewidthText,6, 1 );
Linewidth = atof(LinewidthText);
Linewidth /= 2;
if (Linewidth < 0.05)
Linewidth = 0.05;
if (NumberGroups = 0)
CloseWindowo;
if (NumberGroups > 0)
{
GroupsWindow = new TNumberGroupWindow(this,"Group Parameters");
GetApplication()->MakeWindow(GroupsWindow);
1
1
void TParamWindow::HandleCancelButton(RTMessage )
{
CloseWindow();
1
”Constructor for TNumberGroupWindow
TNumberGroupWindow::TNumberGroupWindow(PTWindowsObject AParent,
LPSTR ATitle):TWindow(AParent.ATitle)
{
Attr. Style |= WS_POPUP | WS_CAPT ION;
Attr.X = 5;
Attr.Y = 10;
Attr.W = 635;
Attr.H = 469;
new TStatic(this,-l,"Group",10,80,70,20,5);
new TStatic(this,-1,"Ax",80.80,70,20,2);
new TStatic(t1us,-l,"Ay",15l,80,70,20,2);
new TStatic(this,-l,"Az”,222,80,70,20,2);
new TStatic(this,-1,”Phi",293,80,70,20,3);
new TStatic(this,-l,"Theta",364,80,70,20,5);
new TStatic(this,-1,"Psi',435,80,70,20,3);

196

new TStatic(this,-l,"# Nuc”,506,80,70,20,5);

new TStatic(this,-1,"Spin”,577,80,50,20,4);

for (int i =0;i<NumberGroups;+-+i)

{
itoa(i+ l ,GroupNumber, 10);
new TStatic(this,-1,GroupNumber,10,100+i*25,70,25,sizeof GroupNumber);
gcvt(AHyplol [011i],7.AxText);
gcvt(AHypl lllllli],7,AyText);
gcvt(AHypl2][2][il.7.AzText);
gcvt(Phi[i],6,PhiText);
gcvt(I‘hetali],6,'I'hetaText);
gcvt(Psi[i],6,PsiText);
itoa(NucleiPerGroup[i],NoNucleiText,4);
gcvt(NuclearSpin[i],4,NuclearSpinText);
AxEdit[i] = new TEdit(this,30+i,AxText,80,100+i*25,70,25,7,FALSE);
AxEdit[i]->Attr.Style |= WS_TABSTOP;
AyEdit[i] = new TEdit(this,38+i,AyText,151,100+i*25,70,25,7,FALSE);
AyEdit[i]->Attr.Style |= WS_TABSTOP;
AzEdit[i] = new TEdit(this,46+i,AzText,222,100+i*25,70,25,7,FALSE);
AzEdit[i]->Attr. Style |= WS_TABSTOP;
PhiEditlil= new 'I'Edit(this,54+i,PhiText,293,100+i*25,70,25,7,FALSE);
PhiEditli]->Attr. Style |= WS_TABSTOP;
ThetaEditli] = new TEdit(this,62+i,ThetaText,364,100+i*25,70,25,7,FALSE);
ThetaEditli]->Attr. Style |= WS_TABSTOP;
PsiEdit[i] = new TEdit(this,70+i,PsiText,435,100+i*25,70,25,7,FALSE);
PsiEdit[i]->Attr. Style |= WS_TABSTOP;
NoNucleiEdit[i] = new TEdit(this,78+i,NoNucleiText,506,100+i*25,70,25,7,FALSE);
NoNucleiEdit[i]->Attr. Style |= WS_TABSTOP;
NuclearSpinEditlil=new TEdit(this,86+i,Nuc1earSpinText,577,100+i*25,50,25,7.FALSE);
NuclearSpinEditli]->Attr. Style |= WS_TABSTOP;

1
OKButton = new TButton(this,94,"OK",500,10,70,30,TRUE);
CancelButton = new TButton(this,95,"Cancel",500,45,70,30,TRUE);
EnableKBHandlerO;

1
void TNumberGroupWindow: :HandleButtonOK(RTMessage)

{

for (int i=0;i<NumberGroups;++i)

{
AxEdit[i]->GetLine(AxText,8, I );
AI-lyp[0][0][i] = atof(AxText);
AyEdit[i]->GetLine(AyText,8, l);
AHyp[ l ] [ l][i] = atof(AyText);
AzEdit[i]->GetLine(AzText,8, l);
AHyp[2][2][i] = atof(AzText);
PhiEdit[i]->GetLine(PhiText,8, 1);
Phi[i] = atof(PhiText);
PsiEdit[i]->GetLine(PsiText,8, l);
Psi[i]= atof(PsiText);
ThetaEdit[i]—>GetLine('I'hetaText,8, l );
Theta[i]=atof(ThetaText);
NoNucleiEdit[i]->GetLine(NoNucleiText,8, l);
NucleiPerGroup[i]=atoi(NoNucleiText);
NuclearSpinEdit[i]->GetLine(NuclearSpinText,8, 1 );

197

NuclearSpin[i]=atof(NuclearSpinText):
1
CloseWindow();
ParamWindow->CloseWindow();
1
void TNumberGroupWindow::HandleButtonCance1(RTMessage)
{
CloseWindow();
ParamWindow->CloseWindow();

1

// Actual setup of the controls in the import window
void TImportWindow: :SetupWindow()

{

TWindow::SetupWindow();
ImportRadio->SetCheck(BF_CI-IECKED);
lmportRadio3->SetCheck(BF_CHECKED);

1

// Make response function for 0k command in the import window
void TImportWindow: :HandleButtonOKImport (RTMessage)
{
SetCursor(LoadCursor(NULL,1DC_WAIT));
// Get the data from the file
ifstream is(FileName);
if (lis.bad())
{
if (ImportRadioZ->GetCheck()=BF_CHECKED)
{
char xeprchar [10],yeprchar [10];
if (FileEndRead=-1 && FileBeginRead 1= 1)
{
for (int d= l ;d<FileBeginRead;4-+d)
is.ignore(80,'\n');
int c=0;
while (lis.eof())
{
is. getline(xeprchar, 10,de1imiter);
is. getline(yeprchar, 10);
xeprlc] = atoflxeprchar);
yepr[c] = atof(yeprchar);
++c;
1
DataNumber = c;
1
else if (FileBeginRead 1= 1 || FileEndRead != -l)
{
for (int e=l;e<FileBeginRead;+1-e)
is.ignore(80,'\n');
for(int f=0;f < (F ileEndRead - FileBeginRead + l);++l)
{
is. getline(xeprchar, 10,delimiter);
is. getline(yeprchar, 10);
xepr[f] = atof(xeprchar);

198

yepr[f] = atof(yeprchar);
1
DataNumber = FileEndRead - FileBeginRead + 1'.

else

{
int g=0;
while (l(is.eof()))
{
is. getline(xeprchar, 10,delimiter);
is. getline(yeprchar, 10);
xepr[g] = atof(xeprchar);
yepr[g] = atof(yeprchar);
Hg;
1
DataNumber = g;
1
1
else if (ImportRadiol->GetCheck()=BF_CHECKED)
{
char xeprchar [10],yeprchar [10];
if (F ileEndRead=-1 && FileBeginRead 1= 1)

{
for (int d=l;d<FileBeginRead;++d)
is.ignore(80,’\n');
int c=0;
while (lis.eof())
{
is. getline(xeprchar, 10,',');
is. getline(yeprchar,]O);
xepr[c] = atof(xeprchar);
yepr[c] = atof(yeprchar);
++c;
1
DataNumber = c;
1
else if (FileBeginRead 1= 1 || FileEndRead 1= -1)
{
for (int e=l;e<FileBeginRead;+-le)
is.ignore(80,’\n');
for(int f=0;f < (FileEndRead - FileBeginRead + l);++l)
{
is. getline(xeprchar, 10,',');
is. getline(yeprchar, 10);
xepr[f] = atof(xeprchar);
yepr[f] = atoflyeprchar);
1
DataNumber = F ileEndRead - F ileBeginRead + 1;

else
{
int g=0;
while (l(is.eof()))
{

199

is. gefline(xeprchar,10,',' ;
is. getline(yeprchar, 10);
xepr[g] = atof(xeprchar);
yepr[g] = atof(yeprchar);
Hg;
1
DataNumber = g:

1

else
if (FileEndRead-=4 && FileBeginRead l= l)

{

for (int d=1;d<F ileBeginRead;++d)
is.ignore(80,'\n');
int c=0;
while (lis.eof())
{
is >> xepr[c] >> yepr[c];
-H-c;

1
DataNumber = c-l;

1

else if (FileBeginRead 1= 1 || FileEndRead l= -l)

l
for (int e=1;e<FileBeginRead;+—+e)
is.ignore(80,’\n');
for(int f=0;f < (FileEndRead - FileBeginRead + l);++t)
{

is >> xepr[fl >> yepr[fl;
1
DataNumber = FileEndRead - FileBeginRead + 1;

else
{
int g=0;
while (lis.eof())
{
is >> xepr[gl >> yepr[g];
'H'g;
1
DataNumber = g-l;
1
}
is.close();
1mportFileWindow->CloseWindow();
CloseWindow();
ExperLoaded = TRUE;
HMENU MainMenu = GetMenu(Parent->HWindow);
EnableMenuItem(MainMenu,CM_V1EWEXPER,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM__IMPORT,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGBASE. MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_INTEGBASEORG, MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_INTEGEPR,MF_BYCOMMAND | MF_GRAYED);

200

EnableMenuItem(MainMenu.CM_DRAWINTEG.MF_BYCOMMAND | MF_ENABLED);
if (SimTrue)
{
EnableMenuItem(MainMenu,CM_VIEWBOTH.MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_VIEWPAGED, MF_BYCOMMAND | MF_ENABLED);

1
1
else
{
is.close();
HWND ImporthdHandle = GetActiveWindowO;
MessageBox(ImporthdHandle,”Cannot Open F ile","File Error",IDOK):
1
SetCursor(LoadCursor(NULL,IDC_ARROW));

1

void TImportWindow::HandleButtonCanImport (RTMessage )
{
ImportFileWindow->CloseWindow();
CloseWindow();
1
// fimction that prompts user to enter a delimiter character
void TImportWindow: :HandleRadioOtherImport (RTMessage )
{
char delim [5] = "";
// Displays a dialog box for the user
if (GetApplication()->ExecDialog(new T1nputDialog(this,"Delimiter Character”
,"Enter the delimiter character in the file:",delim,sizeof
delim» == IDOK)
{
sscanf(delim,”°/oc",&delimiter);
1

1
// function that prompts the user to enter file lines to read

void TImportWindow::HandleRadioOtherLinesImport (RTMessage )
{
// Displays the input boxes
char begin [10] = ”1";
char end [10] = "end";
if (GetApplication()->ExecDialog(new TInputDialog(this,”Begin Line",
"Enter the line to begin reading',begin.sizeof begin» = IDOK)
FileBeginRead = atoi(begin);
if (GetApplimtionO->ExecDialog(new TInputDialog(this,"End Linc",
”Enter the line to end reading”,end,sizeof end)) = IDOK)
{
if (lstrcmp(end,"end"))
FileEndRead = -1;
else
FileEndRead = atoi(end);
1

1
void TFieldWindow::goniomf(float Thetal,float Phi l,float Alpha,

float Hohm,float &X1,float &Y1,float &Zl)
{

201

float const convert = 0.017453292;

X1 = (cos(I-Iohm‘convert) * sin(Thetal‘convert) "‘ cos(Phil*convert));

X1 = (XI + sin(Hohm‘convert) * sin(Alpha'convert) * sin(Phil*convert));

X1 = (XI - sin(Hohm"'convert)‘cos(Alpha*convert)'cosCI‘hetal*convert)*
cos(Phi 1 *convert));

Y1 = (cos(Hohm‘convert) "' sin(ThetaPconvert) * sin(Phi1*convert)):

Y1 = (Y 1 - sin(Hohm‘convert) * sin(Alpha‘convert) * cos(Phil‘convert»;

Y1 = (Y 1 - sin(Hohm*convert)*cos(Alpha*convert)*cos(Thetal*convert)*
sin(Phi 1 *convert»;

Z1 = (cos(Hohm*convert) * cos(Theta1*convert));

Z] = (21 + sin(Hohm*convert) "' cos(Alpha*convert) "' sin(l‘hetal*convert)):

1

// Definition of TFieldWindow functions
void TFieldWindow: :HandleYesButtonField (RTMessage )
{
CloseWindow();
DrawMode = 0;
SimulateSpectrumO;
if (SimTrue)
{
FileSaved = FALSE;
Integrate = FALSE;
DrawMode=I;
GlobalScroller->ScrollTo( 1,0);
InvalidateRect(Parent->HWindow,NULL,TRUE);
UpdateWindow(Parent->HWindow);
HMENU MainMenu = GetMenu(Parent->HWindow);
EnableMenuItem(MainMenu,CM_SIMLINE,MF_BYCOIVIMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_VIEWSIM.MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILESAVE,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILESAVEAS,MF_BYCOI\/IMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_EXPORT,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_PRINT,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILEOPEN,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGEPR,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_DRAWINTEG.MF_BYCOMMAND | MF_ENABLED);
if (ExperLoaded)
{
EnableMenuItem(MainMenu,CM_VIEWBOTH,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_VIEWPAGED. MF_BYCOMMAND | MF_ENABLED);
1
DrawMenuBar(Parent->HWindow);

1

else

{
FileSaved = TRUE;
DrawMode = 0;

Global Scroller->ScrollTo( 1 ,0);

HMENU MainMenu = GetMenu(Parent->HWindow);
EnableMenuItem(MainMenu,CM_SIMLINE,MF_BYCOMMAND | MF_GRAYED);
EnableMenultem(MainMenu,CM_VIEWSIM,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_FILESAVE,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItemMainMenu,CM_FlLESAVEAS,MF_BYCOMMAND | MF_GRAYED);

202

EnableMenuItem(MainMenu,CM_EXPORT,MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_PRINT,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWBOT1-I.MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWPAGED. MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_FILEOPEN.MF_BYCOMMAND | MF_ENABLED);

DrawMenuBar(Parent->HWindow);

1

1

void TFieldWindow::HandleNoButtonField (RTMessage )
{
gcvt(HMin,6,HMTN);
gcvt(HMax,6,HMAX);
if (GetApplication()->ExecDialog(new TInputDialog(this,"Minimum H",
”Input a new minimum H”,HMIN,sizeof HMIN)) = IDOK)
HMin = atof(l—IMIN);
if (GetApplication()->ExecDialog(new TInputDialog(this_."Maximum H".
”Input a new maximum H”,I—MAX,sizeof1-IMAX)) == IDOK)
HMax = atof(HMAX);
CloseWindow();
DrawMode = 0;
SimulateSpectrumO;
if (SimTrue)
{
F ileSaved = FALSE;
Integrate = FALSE;
DrawMode = 1;
Global Scroller->ScrollTo( 1,0);
InvalidateRect(Parent->HWindow,NULL,TRUE);
UpdateWindow(Parent->HWindow);
HMENU MainMenu = GetMenu(Parent->HWindow);
EnableMenuItem(MainMenu,CM_SIMLINE,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_V1EWSIM,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILESAVE,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILESAVEAS,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_EXPORT.MF_BYCOMMAND l MF_ENABLED);
EnableMenuItem(MainMenu,CM_PRINT,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILEOPEN,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGEPR,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_DRAWINTEG,MF_BYCOMMAND | MF_ENABLED);
if (ExperLoaded)
{
EnableMenuItem(MainMenu,CM_VIEWBOTH,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_VIEWPAGED, MF_BYCOMMAND | MF_ENABLED);
1
DrawMenuBar(Parent->HWindow);

1
else
{

FileSaved = TRUE;

DrawMode = 0;

Global Scroller->Scrol lTo( 1,0);

HMENU MainMenu = GetMenu(Parent->HWindow);
EnableMenuItem(MainMenu,CM_SIMLINE,MF_BYCOMMAND | MF_GRAYED);

203

EnableMenultem(MainMenu,CM_VIEWSIM,MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_FILESAVE.MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_FILESAVEAS,MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_EXPORT,MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_PRINT.MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWBOTH.MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWPAGED, MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_FILEOPEN,MF_BYCOMMAND | MF_ENABLED);

DrawMenuBar(Parent->HWindow);

1

1

// DEFINITION OF FILE MENU FUNCTIONS BEGINS HERE

// this command erases the screen and clears the experimental and
// simulated spectrum from memory afier prompting the user to
// save their work
void TPolarWindowzzCMFileNew (RTMessage )
{

BOOL Clear = TRUE:

if (lFileSaved)

{
if (MessageBoxflIWindow,”Do you wish to save?”,"File not Saved”,

MB_YESNO | MB_ICONQUESTION) == IDYES)
Clear = FALSE;

1

if (Clear)

{
DrawMode = 0;
Scroller->ScroIITo( 1,0);
InvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(I-IWindow);
Freq = 9400;
SimType = l;
NumberGroups = 0;
gx = 2.0023;
gy = 2.0023;
gz = 2.0023;
StepSize = 15;
NoDiv = 5;
MemTheta = 0;
MemPhi = 0;
CrystalPhi = 0;
CrystalTheta = 0;
Dalpha = 30;
Omega = 0;
Wobble = 5;
Linewidth = 2;
Integrate = FALSE;
Backl = 0;
Tilt = 0;
AccurGen = TRUE;
PrevSave = FALSE;
F ileSaved = TRUE;

204

PrecurDou = TRUE;
SpecCRPP = FALSE;
Ratio = 5;
Exchange = 5;
for (int a=0;a<1024;++a)
{
xepr[a] = O;
yepr[al = 0;
1
for (int b=0;b<512;-H-b)
813660)] = 0;
SimTrue = FALSE;
ExperLoaded = FALSE;
for (int c=0;c<8;++c)
{
NucleiPerGroup[c] = 0;
NuclearSpinlc] = 0;
Theta[c] = 0;
Phi[c] = 0;
Psi[c] = 0;
for (int d=0;d<3;++d)
{
for (int e=0;e<3;-H-e)
{
AHYPlelldIICI = 0;
1
1
1
HMENU MainMenu = GetMenu(HWindow);
EnableMenuItem(MainMenu,CM_FILESAVE,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_FILESAVEAS,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_IMPORT,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_EXPORT,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_FILEOPEN.MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_PRINT,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWEXPER,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWSIM,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWBOTH.MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWPAGED,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_VIEWEXPER. N07_BYCOIVIIVIAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGBASE. MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGBASEORG, MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGEPR.MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_DRAWINTEG,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_SIMLINE,MF_BYCOMMAND | MF_GRAYED);
DrawMenuBar(HWindow);

// this command opens a previously saved simulated spectrum along
// with the experimental spectrum if one was saved
void TPolarWindow: :CMFiIeOpen (RTMessage )

{
SetCursor(LoadCursor(NULL,IDC_WAIT));

205

if (GetApplication()—>ExecDialog(new TFileDialog(this,SD_FILEOPEN,
strcpy(FileNamel,”*.sim"))) == IDOK)
{
ifstream ifs(FileName1);
if (Iifs.bad())
{
char VersionTest [30];
ifs. getline(VersionTesL30);
if (strcmp(VersionTest,"Sim Pol Program Version 1.0") = 0)
{
ifs >> HMax >> HMin;
ifs >> SimType >> Freq >> NumberGroups;
ifs >> gx >> gy >> g2;
ifs >> StepSize >> NoDiv >> Dalpha;
ifs >> MemTheta >> MemPhi >> Omega;
ifs >> Wobble >> Linewidth >> CrystalTheta >> CrystalPhi;
for (int a = 0;a<8;++a)
{
ifs >> NucleiPerGroup[a] >> NuclearSpinla];
1
for (int b=0;b<8;-Hb)
{
ifs >> Theta[b] >> Phi[b] >> Psi[b];
1
for (int c=0;c<8;-Hc)
{
ifs >> AHYPIOIIOIIC] >> AHYPIIIIIHC] >> AHYDIZHZHC];
1
for (int d=0;d<512;+t~d)
{
ifs >> SpeCId];
1
int Speccrrp;
int Precurdou;
ifs >> Speccrrp >> Precurdou >> Exchange >> Ratio;
if (Speccrrp = 1)
SpecCRPP = TRUE;
else
SpecCRPP = FALSE;
if (Precurdou == 2)
PrecurDou = TRUE;
else
PrecurDou = FALSE;
DrawMode = l;
Scroller->ScrollTo( 1,0);
InvalidateRectfl-IWindow,NULL.TRUE);
PrevSave = TRUE;
F ileSaved = TRUE;
SimTrue = TRUE;
HMENU MainMenu = GetMenu(HWindow);
EnableMenuItem(MainMenu,CM_SIMLINE,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_VIEWSIM.MF_BYCOlVIMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILESAVE,MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_FILESAVEAS,MF_BYCOMMAND | MF_ENABLED);

206

EnableMenuItem(MainMenu,CM_EXPORT,MF_BYCOMMAND | MF_ENABLED);

EnableMenuItem(MainMenu,CM_PRINT.MF_BYCOMMAND | MF_ENABLED);

EnableMenuItem(MainMenu,CM_FILEOPEN.MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_VIEWEXPER. MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_INTEGEPR.MF_BYCOMMAND | MF_GRAYED);

EnableMenuItem(MainMenu,CM_DRAWINTEG,MF_BYCOMMAND | MF_ENABLED);

if (ExperLoaded)

{

EnabIeMenuItem(MainMenu,CM_VIEWBOTH.MF_BYCOMMAND | MF_ENABLED);

EnableMenuItem(MainMenu,CM_VIEWPAGED. MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_VIEWEXPER, MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_INTEGBASE, MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_INTEGBASEORG, MF_BYCOMMAND | MF_ENABLED);

1

DrawMenuBarGIWindow);

else
{
MessageBoxGIWindow,"File must be a *.sim file","Incorrect File Format",
MB_OK | MB_ICONSTOP);

else
{
MessageBoxG-IWindow,”File cannot be opened”, ”File Error”,
MB_OK | MB_ICONSTOP);
1
1
SetCursor(InadCursor(NULL,IDC_ARROW));
1
// command to import an experimental spectrum to compare with a
// simulated spectrum that will be created
void TPolarWindow::CMFileImport (RTMessage )
{
// create a file selection box
if ( GetApplication()->ExecDialog(new TFileDialog(this,
SD_FILEOPEN, strcpy(FileName,”*.DAT"))) = IDOK )
{
// Actual creation of the Import window

ImportWindow = new TImportWindow(this,"Import File");

ImportFiIeWindow = new TFileWindow(this,"FileWindow",
F ileName);

// Import file window attributes

ImportFiIeWindow->Attr.X=20;
ImportFiIeWindow->Attr.Y=230;
ImportFiIeWindow->Attr.W=600;
ImportFiIeWindow->Attr.H=240;

// Makes the import window interface elements
GetApplication()->MakeWindow(ImportFileWindow);
GetApplication()->MakeWindow(ImportWindow):

1
1

// command that saves a simulated spectrum that has been

207

// previously saved
void TPolarWindow::CMFileSave (RTMessage )
{
if (lPrevSave)
{
if (GetApplication()->ExecDialog(new TFileDialog(this.SD_FILESAVE.
strcpy(FileNamel,"".sim”))) == IDOK)
FileSave();
1
else
F ileSave();

// command that saves a spectrum that has not been saved
// previously or to save under a different name
void TPolarWindow::CMFileSaveAs (RTMessage )
{
if (GetApplication()->ExecDialog(new TFileDialog(this.SD_FILESAVE.
strcpy(FileNamel,”*.sim"))) == [DOK)
FileSave();
1

// exports the simulated spectrum in an x.y ASCII format text file
void TPolarWindow::CMFileExport (RTMessage )

if (GetApplication()->ExecDialog(new TFileDialog(this.SD_FILESAVE,
strcpy(Fi1eName2,"*.dat”)))== IDOK)

{
int tempDrawMode = DrawMode;
DrawMode = 10;
PolarDC = GetDC(HWindow);
DrawSpectrum(PolarDC);
ReleaseDC(HWindow,PolarDC);
DrawMode = tempDrawMode;
1

1

// command to print a simulation and experimantal spectrum
void TPolarWindow::CMPrint (RTMessage )
{
memset(&pd,0,sizeof(PRINTDLG));
pd.lStructSize = sizeof (PRINTDLG);
pd. hwndOwner = HWindow;
pd.Flags = PD_RETURNDCIPD_1-IIDEPRIN'ITOFILE | PD_NOPAGENUMS | PD_NOSELECTION |
PD_ALLPAGES | PD_SHOWHELP;
if (PrintDlg(&pd) != 0)
{
int tempDrawMode = DrawMode;
DrawMode = 11;
DrawSpectrum(pd.hDC);
DrawMode = tempDrawMode;
if (pd.hDevMode l= NULL)
GlobalFree(pd.hDevMode);
if (pd.hDevNames l= NULL)

208

GlobalFree(pd.hDevNames);
1

else
MessageBoxGIWindow,"Unable to print!”,"Printer Error”,MB_OI( | MB_ICONINFORMATION);

1

// configures a printer to print
void TPolarWindow: :CMPrintSetup (RTMessage )
{
memset(&pd,0,sizeof(PRINTDLG));
pd.lStructSize = sizecf(PRINTDLG);
pd.hwndOwner = HWindow;
pd.Flags = PD_PRINTSETUP | PD_SHOWHELP;
if (PrintDlg(&pd) l= 0)
{
if (pd.hDevMode l= NULL)
GlobalFree(pd.hDevMode);
if (pd. hDevNames != NULL)
GlobalFree(pd.hDevNames);
1
else
MessageBoxfl-IWindow,"Printer is not initializedl”,”Printer Initialization Error”,
MB_OK | MB_ICONEXCLAMATION);

1
// exits the program after checking that all data is saved
void TPolarWindow: :CMExit (RTMessage )
{
CloseWindow();
1
// determines if the main window can be closed
BOOL TPolarWindow: :CanCloseO
{
if (FileSaved)
return TRUE;
else if (MessageBoxa-IWindow,”Do you wish to save?”,"File not saved”,
MB_YESNO | MB_ICONQUESTION) = IDNO)
return TRUE;
else
retum FALSE;

1

// THIS IS THE END OF THE FILE MENU FUNCTIONS
// VIEW MENU FUNCTIONS BEGIN HERE

// Shows the experimental spectrum on the screen
void TPolarWindow: :CMViewExper (RTMessage )
{
DrawMode=2;
Scroller->ScrollTo( 1,0);
InvalidateRectGIWindow,NULL,TRUE);
UpdateWindow(HWindow);
1

209

// Shows the simulated spectrum on the screen
void TPolarWindow::CMViewSim (RTMessage )
{
DrawMode=l;
Scroller->ScrollTo( 1,0);
lnvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(HWindow);
1

// Shows both the simulated and experimental spectrum on the screen

void TPolarWindow::CMViewBoth (RTMessage )
{

DrawMode=3;

Scroller->ScrollTo( 1,0);

lnvalidateRect(HWindow,NULL,TRUE);

UpdateWindow(HWindow);
1

// Shows both spectra but in difi‘erent windows
void TPolarWindow: :CMViewPaged (RTMessage )
{
DrawMode=4;
Scroller->ScrollTo( 1 ,0);
lnvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(HWindow);
1

// END OF THE VIEW MENU FUNCTIONS
// BEGIN THE SIMULATION MENU FUNCTIONS

void TPolarWindow::CMSimParam (RTMessage )

{
ParamWindow = new TParamWindow(this."Simulation Parameters");
GetApplication()->MakeWindow(ParamWindow);

1

// Does the actual simulation of the EPR spectrum

void TPolarWindow: :CMSimulate(RTMessage)

{
FieldWindow = new TFieldWindow(this,"Calculated Maximum and Minimum H values”);
GetAppIication()->MakeWindow(FieldWindow);
1
// query the user for the polarization parameters
void TPolarWindow: :CMSimPolParam(RTMessage)
{
PolParamWindow = new TPolParamWindowahis,"Polarization Parmeters"):
GetAppIication()->MakeWindow(PolParamWindow);
1
// Redraw the spectrum with new linewidth parameter
void TPolarWindow: :CMSimLineChange(RTMessage)
{
char LineText [10];
float LineOld = Linewidth;
gcvt(2*Linewidth,5,LineText);

210

if (GetApplication()->ExecDialog(new TInputDialog(this,"Linewidth",
”Input 8 new linewidth”,LineText,sizeof LineText» == IDOK)
{
Linewidth = atof(LineText)/2;
HMax = HMax - 5*LineOld + 5*Linewidth;
HMin = HMin + 5*LineOld - 5*Linewidth;
FileSaved = FALSE;
1
Scroller->ScrollTo( l ,0);
lnvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(HWindow);
1

// Begin integrate menu functions

void TPolarWindow: :CMIntegDrawInteg (RTMessage)

{
Integrate = TRUE;
HMENU MainMenu = GetMenu(HWindow);
EnableMenuItem(MainMenu,CM_DRAWINTEG,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGEPRMF_BYCOWAND | MF_ENABLED);
Scroller-> ScrollTo( 1,0);
lnvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(HWindow);

1
void TPolarWindow: :CMlntegEPR (RTMessage)
{
Integrate = FALSE;
HMENU MainMenu = GetMenu(HWindow);
EnableMenuItem(MainMenu,CM_DRAWINTEG.MF_BYCOMMAND | MF_ENABLED);
EnableMenuItem(MainMenu,CM_INTEGEPR,MF_BYCOMMAND | MF_GRAYED);
Scroller->ScrollTo( 1,0);
InvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(HWindow);
1
void TPolarWindow: :CMIntegBase(RTMessage)
{
char Background [5];
gcvt(Backl,3,Background);
char TiltText [10];
gcvt (Tilt,6,TiltText);
if (GetApplication()->ExecDialo'g(new T1nputDialog(this,”Background",
”Enter the amount to raise or lower the Background:",Background,
sizeof Background» == IDOK)
{
Integrate = TRUE;
Backl = atof(Background);
1
if (GetApplication()->ExecDialog(new 'I'InputDialog(this."Tilt".
”Enter the amount in degrees to tilt the baseline:”,TiltText,
sizeof TiltText» = IDOK)
{
Integrate = TRUE;
Tilt = atof(TiltText);
if (Tilt > 20)

211

Tilt = 20;
Tilt = (xepr[DataNumber]-xepr[0])‘tan(TiIt*0. 174532);
Tilt /= (DataNumber - l);
1
if (Integrate)
{
HMENU 382X HMENU MainMenu = GetMenu(HWindow);
EnableMenuItem(MainMenu,CM_DRAWINTEG.MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGEPR,MF_BYCOMMAND | MF_ENABLED);
Scroller->ScrollTo( 1,0);
InvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(HWindow);
1
1
void TPolarWindow: :CMIntegBaseOrg (RTMessage)
{
Tilt = 0;
Backl = 0;
Integrate = TRUE;
HMENU MainMenu = GetMenu(HWindow);
EnableMenultem(MainMenu,CM_DRAWINTEG,MF_BYCOMMAND | MF_GRAYED);
EnableMenuItem(MainMenu,CM_INTEGEPRMF_BYCOMMAND | MF_ENABLED);
Scroller->ScrollTo( 1,0);
InvalidateRect(HWindow,NULL,TRUE);
UpdateWindow(HWindow);
1

ll Begin other utility functions that are used for program

// function that actually saves the file

void TPolarWindow::FileSaveo

{

SetCursor(IoadCursor(NULL,IDC_WAIT)):

ofstream of(FileNamel);

if ( 101‘.bad())

{
of << ”Sim Pol Program Version 1.0\n";
of << HMax << " " << HMin << "\n”;
of << SimType << " " << Freq << " " << NumberGroups << ”\n":
of<<gx<<""<<gy<<""<<gz<<"\n";
of << StepSize << " " << NoDiv << " " << Dalpha << "\n";
of << MemTheta << " " << MemPhi << " " << Omega << ”\n”:
of << Wobble << " " << Linewidth << " " << CrystalTheta << " " << CrystalPhi << ”\n";
for (int a = 0;a<8;++a)
{

of << NucleiPerGroup[a] << " " << NuclearSpin[a] << "\n”;
1
for (int b=0;b<8;++b)
{
of << Theta[b] << " " << Phi[b] << " " << Psi[b] << "\n";

1
for (int c=0;c<8;-H-c)
{
0f << AHyp[0][0][c] << " " << AHyplllllllC] << " " << AHYPIZHZHCI << "\n":

212

1
for (int d=0;d<512;++d)
{
of << specld] << "\n";
1
if (SpecCRPP)
of << ”1";
else
of << ”0";
if (PrecurDou)
0f<< n to << «2»;
else
of<< N I << ”4”,,
of << " " << Exchange << " " << Ratio:
of.close();
FileSaved = TRUE;
1
SetCursor(LoadCursor(NULL,IDC_ARROW));
1
// this function redraws the experimental spectrum
void TPolarWindow: :Paint(HDC PDC,PAINTSTRUCT& )
{
PaintPen = CreatePen(PS_SOLID,1,0x00FF0000);
SelectObject(PDC,PaintPen);
if (ExperLoaded || SimTrue)
{
if (DrawMode 1= 0 || DrawMode > 9)
DrawSpectrum(PDC);
1
DeleteObject(PaintPen);
1

// Function that registers the window class
LPSTR TPolarWindow: :GetClassNameO

{

return ”PolarWindow”;

1

// function that displays the help index for Polar Sim
void TPolarWindow: :CMHelplndex (RTMessage)

{
WinHelp(HWindow,"simhelp.hlp",HELP_INDEX,0);
helpb = TRUE;

1

void TPolarWindow: :CM]-IelpOnHelp(RTMessage)

{

1

// function that displays help on commands in polar Sim

void TPolarWindow: :CMCommands(RTMessage)

{
WinHelp(HWindow,”simhelp.hlp”,I-lELP_CONTEXT,500);
helpb = TRUE;

WinHelp(HWindow,"simhelp.hlp",HELP_HELPONHELP,0):

213
1

// Function to change window characteristics
void TPolarWindow::GetWindowClass(WNDCLASS& AWndClass)
1
TWindow: :GetWindowClass(AWndClass);
AWndClass.hIcon = Loadlcon(GetApplication()->hlnstance,"ICON_l ”);
1
// Function that does the actual simulation of the spectrum
void TFieldWindow: :SimulateSpectrumO
1
SetCursor(LoadCursor(NULL,IDC_WAIT));
float hhfp[3],hhfm[3];
float hefl‘p[3],hefl’m[3];
float maghefl‘p=0;
float magheffm=0;
float CouplingMax=0;
float htheta=0;
const float factor = 658.2109;
float Wobble1=0,weight=0;
float gefl'=0,h=0,bigsplit=0,smasplit=0.bigintens=0,smaintens=0,x=0;
int hl=0,center=0;
float const convert = 0.017453292;
float xl=0,yl=0,zl=0;
float ahl [3][3];
float dc[3],dcg[3];
float ahrot[3][3][8];
float thetatemp=0,phitemp=0,a=0;
float tx[3][3];
float strmos=0,endmos=0,hohm=0,steep=0,alpha=0;
int nflag=0,oddint=0;
float cph=0,sph=0,cth=0,sth=0,cps=0,sps=0;
float t[3][3];
float gaussperch=0;
float r[512],m[512];
gaussperch = (HMax - HMin)/511;
HWND MainHandle = GetActiveWindowO;
PolarDC = GetDC(MainHandle);
MessageBox(MainHandle,"Press F4 to abort”,"To cancel”.MB_OI( | MB_ICONINFORMATION);
if (SpecCRPP)
1
float couplingrnax;
float coupling = 0;
if (NumberGroups > 0)
1
for (int a=0;a<NumberGroups:++a)
1
coupling = 0;
for (int b=0;b<3;++b)
1
couplingmax = AHyp[b][b][a] * NucleiPerGroup[a] * NuclearSpin[a];
if (couplingmax > coupling)
coupling = couplingmax;
1

214

CouplingMax += coupling;
1
1
CouplingMax /= gaussperch;
1
for (int aa=0;aa< 3;++aa)
1
dc[aa]=0;
dcg[aa]=0;
hhfplaal=0;
hhfm[aa]=0;
hefl’plaa]=0;
hefl‘mlaa]=0;
for (int bb=0;bb<3;++bb)
1
ahl[aa][bb] = 0.0;
tx [aa][bb] = 0.0;
t [aa][bb] = 0.0;
for (int cc=0;cc<8;-H-cc)
1

1
1
1
for (int u = 0;u<512;++u)
1
r [u] = 0.0;
rn [u] = 0.0;
spec [u] = 0.0;
1
if (SimType == 1)
1
thetatemp = 0;
phitemp = 0;
1
if (SimType = 2)
1
if (Wobble == 0)
1
Wobblel = 2;
Wobble = I;

ahrot[aa] [bb] [cc] = 0.0;

else
1
if (Wobble < 45 )
Wobble] = 2 "' Wobble;
else
Wobble] = 90;
1
if (Omega < 0)
Omega = -Omega;
if (Omega = 0)
1
if ( (int)(NoDiv/Z) == NoDiv/2)

215

NoDiv += 1;
1
strmos = Omega - Wobblel;
endmos = Omega + Wobblel;

if (strmos >= 0)
1
nflag = 0;
steep = (cos(strmos*convert) - cos(endmos‘convert))/ NoDiv;
1
else
1
nflag = 1;
if (Omega == 0)
steep = (2 - cos(strmos‘convert) - cos(endmos‘convert» / NoDiv;
else
1

if (Omega >= fabs(strmos))
strmos = -Omega;
oddint = 1 + 2 "' (NoDiv/2);
steep = 2 / oddint * (l - cos(strmos*convert));
NoDiv = (2 - cos(endmos‘convert) - cos(stnnos‘convert))/steep;
1
1
hohm = strmos;
alpha = 0.0;
1
if (NumberGroups > 0)
1
for (int b=0;b<NumberGroups;++b)
1
for(int c=0;c<3;++c)
1
ahl[CIlCl = AHyplcllcllb];
1
cph = cos(Phi [b]*convert);
sph = sin(Phi[b]*convert);
cth = cos(Theta[b]*convert);
sth = sin(‘l‘heta[b]*convert);
cps = cos(Psi[b]*convert);
sps = sin(Psilbl‘convert);
t[0][01=cp8‘cth‘cph-sp8*sph;
t [0][l] =cps * cth * sph + sps "' cph;
t [01121 = 4398 * sth;
111110] =-sps"cth*cph-cp8* Sph;
t[l][1]=-Sp6"cth‘ Spin-reps *cph;
t [1112] = Sps * sth;
t [2110] = sth '" Cph;
112111] = sth * Sph;
t [2112] = cth;
for (int x=0;x<3;++x)
1
for (int d=0;d<3;++d)

1
a = 0.0;

1

216

for (int e=0;e<3;++e)
1
a = a + t[x][e] "' ahl[clld];
1
tx[x][d] = a;
1
1
for (int f=0;f<3;+—+f)
1
for (int g=0;g<3;++g)
1
a = 0.0;
for (int h=0;h<3;++h)
1
a = a + 0110111] * 11811111;
1
ahrot[fl [glib] = a;
1
1
1
top:
if (SimType == 1)
1

dc[0] = cos(phitemp‘convert) * sin(thetatemp‘convert);
dc[l] = sin(phitemp‘convert) * sin(thetatemp‘convert);
dc[2] = cos(thetatemp‘convert);

1

if (SimType = 2)

1
weight = exp((-log(2) * pow((hohm - Omega).2)) / pow(Wobble.2)) / Wobble;
goniomf(MemTheta,MemPhi,alpha,hohm,x 1,y1,zl);
dc[0] = x1;
dell] = yl;
dc[2] = 21;

1

if (SimType = 3)

1
dc[O] = cos(CrystalPhi’convert) * sin(CrystalTheta‘convert);
dc[l] = sin(CrystalPhi*convert) "' sin(CrystalTheta*convert);
dc[2] = cos(CrystalTheta‘convert);

1

if (SimType = 4)

1
geff = sqrt((pOW(gx,2) + P0W(gy,2) + pOW(gz,2)) / 3);
h = 0.714484 * Freq / gefl‘;
hl = (h-I-IMin) / gaussperch;

else

1
893 = Sqfl(P0W((8X " dc[0]).2) + P0W((8.V*d°lll)-2) + P0W((gZ*dC[2])..2));
h = 0.714484 "' Freq / gefl‘;
hl = (h - HMin) / gaussperch;

1

217

if(hl>=0&&h1<511)
1
if (SimType = l)
1
r[hl] = sin(thetatemp‘convert) "' gel'f;
if (thetatemp = 0)
r[hl] = 0.784 * geff;

1
if (SimType = 2)
r[hl] = weight " geff;
if (SimType == 3)
r[hl] = 100;
if (SimType = 4)
r[hl] = 100;

1

else

MessageBox(GetActiveWindow(),"Transition is not within the spectral limits”.
”Magnetic Field Limit Error",MB_ICONSTOP | MB_OK);
if (NumberGroups > 0)
1
if (SimType < 4)
1
dcglol = gx ‘ dc{01 /geff;
dcg[1] = gy "' dc[l] lgefl’;
dc15121 = 81 " dCI2] / 863‘;
1
for (int k=0;k<NumberGroups;+-+k)
1
if (NucleiPerGroup[k] > 0)
1
if (SimType < 4)
1
a = 0;
if (NuclearSpin[k] > 0.6)
AccurGen = FALSE;
if (lAccurGen)
1
for (int l=0;l<3;++l)
1
for (int m=0;m<3;~Hm)
1
for (int n=0;n<3;++n)
1
a = a + dcg[I]*ahrot[l][n][k]*ahrot[n][m][k]*dcg[m];
1
1
1
bigsplit = sqrt(a) / gaussperch;
bigintens = l;
smasplit = 0;
smaintens = 0;

else

218

for (int ll=0;ll<3;++ll)

1
for (int mm=0;mm<3;++mm)
1

a = a + dcglmm]*ahrot[mm][ll][k];

1
hhfp[ll] = a*(-0.5)*factor;
hhfm[ll] = a*0.5"'factor;
a = 0;

1

maghefl‘p = 0;

maghefl'm = 0;
for (int p=0;p<3;++p)

1
heffplpl = h " delpl + hhfplp];
hefl‘m[p] = h * dc[p] + hhfm[p];
mghcfl'P = maghefl’P + POW(hefl'p[p],2);
maghefl‘m = magheflrn + pow(heffm[p].2);
1
masheffp = sqrt(magheffp);
magheffm = sqrt(magheflm);
bigsplit = (magheffp + magheffm) / factor / gaussperch;
smasplit = (magheffp - maghefl’m) I factor / gaussperch;
htheta = heffp[0]*hefl‘m[0] + hefip[l]*hefi'm[1] + heffp[2]*hefl‘m[2];
if (maghefl'p*magheffm > 0)
1
htheta /= (maghefl‘p*magheffin);
if (htheta > I)
htheta = 0;
else if (htheta < -l)
htheta = 180;
else if (htheta >= -1 && htheta <= 1)
htheta = 57 .29608‘acos(htheta);
1
else
htheta = 0;
bigintens = pow(sin(htheta*convert/2).2);
smaintens = pow(cos(htheta*convert/2),2);
1
1
if (SimType == 4)
1
smaintens = 0;
smasplit = 0;
bigintens = l;
bigsplit = 0;
for (int l=0;l<3;++l)
1
bigsplit += AHyp[1][l][k] / 3:
1
bigsplit /= gaussperch;
1
for (int l=0;l<NucleiPerGroup[k];++l)

1

219

for(int m=0;m<5 12;++m)
1
if (rlml > 0)
1
center = m;
if (NuclearSpin[k] < l)
1
float newintens = r[center] * bigintens;
int newpos = center - (bigsplit / 2);
if(newpos >= 0 && newpos <= 511)
m[newpos] += newintens:
newpos = center + (bigsplit / 2);
if(newpos >= 0 && newpos <= 511)
m[newpos] += newintens;
if (smaintens > 0)
1
newintens = r[center] * smaintens;
newpos = center -(smasp1it/2);
if(newpos>=0&&newpos<512)
m[newpos] += newintens;
newpos = center + (smasplit/2);
if(newpos >= 0 && newpos <512)
m[newpos] += newintens;

else
1
float intens = r[center];
int newpos;
for (int g=-2*NuclearSpin[k];g<2*NuclearSpin[k]+1 ; g +=2)
1
newpos = center + g/2 "' bigsplit;
if(newpos >= 0 && newpos < 512)
m[newpos] += intens;
1
1
1

for (int j=0;j<5 12;++j)
1
rLil = m[i];
mlil = 0;
1
1
1
1
1
MSG Stopmessage;
BOOL SimStop = FALSE;
if (PeekMessage(&Stopmessage,Main1-Iandle,WM_KEYDOWN,WM_KEYDOWN,PM_NOYIELD |
PM_REMOVE))
1
if (Stopmessage.wParam == VK_F4)
SimStop = TRUE;

220
1

if (SpecCRPP)
1
float tempratio=0;
int count = Exchange/gaussperch;
for(int k=0;k<512;-H~k)
1
if (IN > 0)
1
if((hl - k > 0) && PrecurDou)
1
if (Exchange 1= 0)
1
tempratio = (Ratio - 1)‘(hl-k)/(CouplingMax)+l;
if ((k-count/2 >= 0) && (k-count/Z <512))
m[k-count/2] = -l/(tempratio + l)‘r[k]:
if ((k+count/2 >= 0) && (k+count/2 < 512))
m[k+count/2] = tempratio/(tempratio + l)*r[k];

1
1
else if ((hl-k > 0) && lPrecurDou)
1
if (Exchange 1= 0)
1

tempratio = (Ratio - l)*(hl-k)/(CouplingMax)+l;

if ((k-count/2 >= 0) && (k-count/2 <512))
m[k-count/Z] = l/(tempratio + l)*r[k];

if ((k+count/2 >= 0) && (k+count/2 < 512))
m[k+count/2] = -tempratio/(tempratio + 1)*r[k];

1

else

mlk] = -r[k];

1
else if ((hl - k < 0) && PrecurDou)

1
if (Exchange != 0)
1
tempratio = (Ratio - l)*(k-hl)/(CouplingMax)+l;
if ((k-count/2 >= 0) && (k-count/Z <512))
m[k-count/2] = -tempratio/(tempratio + l)*r[k]'.
if ((k+count/2 >= 0) && (k+count/2 < 512))
m[k+count/2] = l/(tempratio + l)"'r[k];
1

else
} mm = - r[k];
else if ( (hl - k < 0) && lPrecurDou)

1
if (Exchange l= 0)

1
tempratio = (Ratio - 1)*(k-hl)/(CouplingMax)+ 1:
if ((k-count/2 >= 0) && (k-count/2 <512))
m[k-count/Z] = tempratio/(tempratio + l)*r[k];

221

if ((k+count/2 >= 0) && (k+count/2 < 512))
m[k+count/2] = -l/(tempratio + l)*r[k];
1
1
elseif(hl -k=0)
1
if (Exchange l=0)
1
if (PrecurDou)
1
if ((k-count/2 >= 0) && (k-count/2 <512))
m[k-count/Z] = -0.5*r[k];
if ((k+count/2 >= 0) && (k+count/2 < 512))
m[k+count/2] = 0.5*r[k];
1
else
1
if ((k-count/2 >= 0) && (k-count/Z <512))
m[k-count/Z] = 0.5*r[k];
if ((k+count/2 >= 0) && (k+count/2 < 512))
m[k+count/2] = -0.5"'r[k];

for (int j=0;j<512;++j)
1
if (lSpecCRPP)
1
59°00] += lrlj];
r[i] = 0;
1
else
1
59°00] +=mlil;
r[i] = 0;
m0] = 0;
1
1
if (SimType = 1)
1
char ThetaText[5];
char PhiText[5];
RECT Change;
Changelefl = 210;
Changetop = 280;
Changeright = 350;
Changebottom = 330;
gcvt(thetatemp,4,'I’hetaText);
gcvt(phitemp,4,PhiText);
if (SimStop)
goto bottom;

222

TextOut(PolarDC,100,150,"Ca1cu1ating the stick spectrum".30);
TextOut(PolarDC,200,250,"Theta is”,8);
TextOut(PolarDC,400,250,”Phi is",6);
SetTextColor(PolarDC,0x00FFFFFF);
DrawText(PolarDC,ThetaText,-1,&Change,DT_TOP);
Ofl‘setRect(&Change,200,0);
DrawText(PolarDC,PhiText,-1,&Change,DT_TOP);
gcvt(thetatemp+StepSize,4,ThetaText);
gcvt(phitemp+StepSize,4,PhiText);
SetTextColor(PolarDC,Ox00000000);
Ofl‘setRect(&Change,-200,0);
DrawText(PolarDC,ThetaText,-l,&Change,DT_TOP);
Ofl’setRect(&Change,200,0);
DrawText(PolarDC,PhiText,-l,&Change,DT_TOP);
if (thetatemp = 0)
1

thetatemp = StepSize;

phitemp = 0;

goto top;
1
phitemp += StepSize;
if (phitemp < 180)

goto top;
phitemp = 0;
thetatemp += StepSize;
if (thetatemp < 180)

goto top;
1
if (SimType == 2)
alpha += Dalpha;
if(alpha < 360)
goto top;
alpha = 0;
if (hohm >= endmos * 1.03)
goto bottom;
if (nflag = 0)
1
x = cos(hohm‘convert) - steep;
hohm = acos(x);
1
else
1
x = cos(hohm*convert) + steep;
if (x<1)
1
hohm = -acos(x);
1
else
1
nflag = 0;
hohm = acos(2-x);
1

223

if (hohm < endmos*l.03)
goto top;
1
bottom:
lnvalidateRect(MainHandle,NULL,TRUE);
SetCursor(LoadCursor(NULL,IDC_ARROW»;
if (lSimStop)
1
MessageBox(MainHandle,”Valid Stick Spectrum exists".”Status”,MB_OK);
MessageBox(MainHandle,”Drawing the EPR spectmm”,”Status”,MB_OK);
SimTrue = TRUE;

else
1
MessageBox(MainHand1e,"A valid spectrum does not exist"."Status”.
MB_OK | MB_ICONSTOP);
SimTrue = FALSE;
for (int a=0;a<512;++a)
1
SPeCla] = 0;
1
1
ReleaseDC(MainHandle,PolarDC);
1

// Construct TPolarization's MainWindow Data member
void TPolarization :: InitMainWindowO
1

1

MainWindow = new TPolarWindow(NULL, "Polarization Simulation Program”);

// Main program function for windows programs
int PASCAL WinMain (HANDLE hlnstance, HANDLE hPrevInstance.
LPSTR lpCmdLine, int nCmdShow)
1
//Create an object of application class TPolarization
TPolarization Polarization (”Polarization”, hlnstance,
hPrevInstance, lpCmdLine, nCmdShow);
Polarization.Run();
return Polarization. Status;

1

"‘111111111111111“