Citation
Purification, spectral and kinetic characterization, and determination of midpoint potential of short-chain Acyl-coa Dehydrogenase

Material Information

Title:
Purification, spectral and kinetic characterization, and determination of midpoint potential of short-chain Acyl-coa Dehydrogenase
Creator:
Hays, Tiffany
Publication Date:
Language:
English
Physical Description:
xii, 89 leaves : illustrations ; 28 cm

Subjects

Subjects / Keywords:
Dehydrogenases ( lcsh )
Oxidation-reduction reaction ( lcsh )
Dehydrogenases ( fast )
Oxidation-reduction reaction ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 87-89).
Thesis:
Department of Chemistry
Statement of Responsibility:
by Tiffany Hays.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
42611545 ( OCLC )
ocm42611545
Classification:
LD1190.L46 1999m .H39 ( lcc )

Downloads

This item has the following downloads:


Full Text
PURIFICATION, SPECTRAL AND KINETIC CHARACTERIZATION,
AND DETERMINATION OF MIDPOINT POTENTIAL
OF SHORT-CHAIN ACYL-COA DEHYDROGENASE
by
Tiffany Hays
B.S., University of Colorado at Denver, 1997
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Chemistry
1999


This thesis for the Master of Science
degree by
Tiffany Hays
has been approved
by
Donald C. Zapien
Frank E. Frerman
Larry G. Anderson


Hays, Tiffany (M.S., Chemistry)
Purification, Spectral and Kinetic Characterization, and Determination of Midpoint
Potential of Rat Short-Chain Acyl-CoA Dehydrogenase
Thesis directed by Assistant Professor Donald C. Zapien
ABSTRACT
Short-chain acyl-CoA dehydrogenase (SCAD) is a member of a family of
mitochondrial enzymes involved in the p-oxidation of fatty acids. In the reaction
catalyzed by SCAD, two electrons are removed from a fatty acyl-CoA substrate
introducing a trans-a, P double bond. The electrons are passed through the proteins
electroactive center, flavin adenine dinucleotide (FAD), and then are shuttled into a
chain of redox enzymes which feed the electron transport chain where ATP is
generated. As the only source of useful energy for living organisms, oxidation-
reduction reactions such as the SCAD catalysis are of biological significance.
This thesis presents a purification method for rat SCAD that has been
modified from previous protocols. Results show a method including DEAE gel
exclusion chromatography, hydroxylapatite column chromatography, and Sephacryl
S-200 gel filtration chromatography produce a protein in good yield and of increased
purity from previously published methods. Kinetic and mechanistic data also
in


support the purification of a protein which is of comparable quality, if not better, than
those used in previous publication.
An important research question with any redox protein is that of its standard
redox potential (E). After determining the E for a wild-type enzyme,
modification of the potential may show the effect of specific amino acid mutations
near the redox center. The E of SCAD was determined in this work using two
different methods. The widely accepted method utilizing a reduction dye led to a
E= -.182 V. An unexplored method of cyclic voltammetry using a carbon electrode
led to a E= -.240 V. This method is quicker and does not rely on achieving
equilibrium between a dye and the protein for the potential determination. The
difference between the two values leads to questions regarding the exact mechanism
by which electrons flow in the two methods. With both values lower than expected,
questions remain regarding whether the more negative potential is a result of a
physiological role or if it has been artificially lowered. The unexpected results of
this work will lead to further interest in SCADs important biochemical role.
This abstract accurately represents the content o^h^andidate^hesi^I
recommend its publication.
Signed
Donald C. Zapien'
IV


ACKNOWLEDGEMENT
My thanks to my advisors, Frank Frerman and Don Zapien, for their support and
understanding. My thanks to the following: Greg DeGala, Tim Dwyer, Michelle
Muller, Mai Pham, and Denise Salazar for their help and guidance. My thanks to my
committee members Ellen Levy and Larry Anderson for their time and knowledge. I
would finally like to thank all those who have supported me outside of school
without whom this would not be possible; Dave Hays, Brendan Hays, Leslie and Bob
Reichardt, Stephanie and Rob Hays, John and. Cheryl Claybough, and Alisha
Cambria.


CONTENTS
Figures................................................... x
Tables.................................................... xii
CHAPTER
1. INTRODUCTION........................................... 1
Fatty Acid Oxidation................................ 2
Acyl-CoA Dehydrogenases............................. 5
Flavin as the Coenzyme......................... 7
Acyl-CoA Dehydrogenase Mechanism............... 9
Structure of the Acyl-CoA Dehydrogenases...... 15
Oxidation-Reduction Reactions....................... 16
Cyclic Voltammetry............................ 20
Conclusions......................................... 21
2. PURIFICATION, SPECTRAL AND KINETIC
CHARACTERIZATION OF RAT SHORT-CHAIN ACYL-COA
DEHYDROGENASE.......................................... 23
VI


Introduction..............................................
Materials and Methods......................................
Purification of SCAD...............................
Determination of SCAD Enzyme Activity..............
Determination of Extinction Coefficients for SCAD and
Charge-Transfer Complex............................
Steady-State Kinetic Assays..........................
Determination of SCAD Mechanism....................
Results and Discussion....................................
Purification.......................................
Spectral Properties................................
KineticProperties..................................
DETERMINATION OF THE MIDPOINT POTENTIAL OF RAT
SHORT-CHAIN ACYL-COA DEHYDROGENASE USING THE
XANTHINE/XANTHINE OXIDASE REDUCTION
METHOD.......................................................
Introduction..............................................
Materials and Methods.....................................
23
24
25
28
28
30
31
32
32
37
46
48
.48
.49
Determination of Extinction Coefficient for Reduced
SCAD............................................
49


Re-crystallization of Redox Dye...................... 50
Determination of Xanthine Oxidase Enzyme
Activity............................................. 50
Determination of SCAD Midpoint Potential............. 51
Results and Discussion...................................... 53
Determination of Extinction Coefficient for Reduced
SCAD................................................. 53
Determination of Xanthine Oxidase Enzyme
Activity............................................. 55
Determination of SCAD Midpoint
Potential.............................................55
4. DETERMINATION OF THE MIDPOINT POTENTIAL FOR
SHORT CHAIN ACYL-COA DEHYDROGENASE BY
CYCLIC VOLTAMMETRY.................................................64
Introduction.................................................64
Materials and Methods....................................... 65
Design of the Electrochemical Cell....................65
Construction of the Electrochemical
Cell................................................. 67
Cyclic Voltammetry....................................69
Results and Discussion...................................... 72
Determination of Cell Function....................... 72
viii


Determination of Midpoint Potential of
SCAD...................................... 72
Scan Rate Dependence...................... 79
5. GENERAL DISCUSSION.................................82
REFERENCES..................................................87
ix


FIGURES
Figure
1.1 Reactions of the (3-oxidation pathway....................................4
1.2 Flavin adenine dinucleotide (FAD).......................................8
1.3 Oxidation states of flavin..............................................10
1.4 Overall enzyme sequence.................................................11
1.5 Catalytic process by which ACD is bound and regenerated............:....13
1.6 Mechanism of dehydrogenation............................................14
1.7 Structure of the MCAD subunit...........................................17
1.8 Arrangement of the MCAD binding pocket..................................18
2.1 Elution of SCAD from Hydroxylapatite column.............................36
2.2 Absorbance spectrum of purified rat SCAD................................38
2.3 Aerobic reduction of SCAD with butyryl-CoA..............................40
2.4 Anaerobic titration of SCAD using sodium dithionite with product present.... 42
2.5 Anaerobic titration of SCAD using sodium dithionite without product
present................................................................. 43
2.6 Anaerobic titration of electron transfer flavoprotein using SCAD........44
2.7 Double reciprocal (Lineweaver-Burke) plot of steady-state kinetic assay.47
3.1 Photoreduction of SCAD using a deazaflavin catalyst.....................54
3.2 Plot to determine the activity of xanthine oxidase......................56
3.3 Anaerobic reduction f indigo carmine using xanthine/xanthine oxidase....58
3.4 Anaerobic reduction of SCAD using xanthine/xanthine oxidase.............59
3.5 Anaerobic reduction of SCAD and indigo carmine using xanthine/xanthine
oxidase..................................................................61
3.6 Minnaert plot of reduction of SCAD and indigo carmine data..............62
4.1 Electrochemical cell used for determination of midpoint potential of SCAD... 68


4.2 Cyclic voltammogram of potassium ferrocyanide.................73
4.3 Cyclic voltammogram of SCAD...................................76
4.4 Plot to determine the scan rate dependence for SCAD...........80
XI


TABLES
Table
2.1 Purification of rat SCAD


CHAPTER 1
INTRODUCTION
Living organisms perform mechanical work, electrical work, and synthetic
work in order to acquire energy from their environment. The only source of
biologically useful energy is derived from oxidation-reduction reactions. In aerobic
metabolism, animals consume organic molecules and carbohydrates which are
oxidized releasing the necessary energy to perform work. In some bacteria,
anaerobic metabolism generates energy from oxidation of relatively reduced
nitrogen and sulfur compounds yielding products such as sulfate or nitrate. Plants
use photosynthesis where trapped light energy from the sun is converted via
reduction of carbon dioxide and oxidation of water to biological energy in the form
of ATP. Oxidation-reduction reactions, therefore, represent one of the most
important biochemical processes necessary for life.
The common energy currency and final product of oxidative breakdown for
living organisms is the high energy molecule adenosine triphosphate (ATP). ATP
hydrolysis is the essential driving force for many biological reactions which would
otherwise not occur due to positive Gibbs free energy values. ATP is produced as
1


metabolites are broken down and enter the electron-transport chain whose products
are water and ATP. One pathway which supplies the electron-transport chain is the
[3-oxidation of fatty acids. A family of enzymes involved in this pathway extracts
electrons from activated fatty acids using them to eventually synthesize ATP. This
family of enzymes includes short-chain acyl-CoA dehydrogenase (SCAD).
Although some enzymes in this family have been studied in detail, this particular
member seems to have been less emphasized in research. Its operation in such a
vital pathway for sustaining life warrants the further investigation put forth in this
report. The studies done here open the door to the more important work in
understanding how a malfunctioning enzyme may ultimately be handled in a clinical
setting.
Fatty Acid Oxidation
Lipids are oxidized to C02 and H,0 yielding significantly larger amounts of
energy than their counterparts, proteins and carbohydrates. After a complicated
pathway of adsorption into cells, fatty acids must be activated before they are
oxidized. Activation is ATP-dependent and forms the fatty acyl-coenzyme A
through an acylation reaction. The reaction catalyzed by the acyl-CoA synthetases
or thiokinases according to the length of the fatty acid chain follows.
2


Fatty acid + Co A + ATP -> acyl-CoA + AMP + PPi (1.1)
The activation process occurs in the cytosol after which the fatty acyl-CoA must be
transported across the mitochondrial membrane so that oxidation can occur. For
transport, the CoA portion of the fat is exchanged with a carnitine molecule in the
cytosol and then transported by specific carnitine carriers across the membrane. A
new CoA from inside the mitochondria then replaces the carnitine to reform the
fatty acyl-CoA (Voet and Voet, 1990).
Oxidation of the fatty acyl-CoA molecule entails the four reactions depicted
in Figure 1.1. The first reaction, which will be discussed in detail later in this
chapter, is catalyzed by members of the acyl-CoA dehydrogenase family. The
dehydrogenation reaction introduces a trans-a, (3 double bond resulting in the trans-
enoyl-CoA product. The second reaction, catalyzed by the enoyl-CoA hydratase,
hydrates the double bond yielding the hydroxyacyl-CoA product. The third reaction
is an NAD+-linked dehydrogenation at the (3 carbon yielding the ketoacyl-CoA
product. Finally, the bond between the a and P carbons is cleaved in a thiolysis
reaction catalyzed by thiolase. The result of these four reactions is the release of an
acetyl-CoA and the formation of a new fatty acyl-CoA that is now two carbons
shorter which can re-enter the cycle for further breakdown (Seager and Slabaugh,
1987).


H H O
I I II
CH3(CH2) CC C SCoA
H H
Fatty Acyl-CoA
H O
CH3(CH2) C =C C SCoA
H
trans-Enoyl-CoA
H20
enoyl-CoA hydratase
H
I
O
CH3(CHj),,CCH2C SCoA
OH
Hydroxyacyl-CoA
NAD
hydroxyacyl-CoA
/
^dehydrogenase
, NADH + H"
O O
CH3 (CH2)n CCH2 C SCoA
Ketoacyl-CoA
- CoASH
ketoacyl-CoA thiolase
O
II
CH3 (CH2) C SCoA
Fatty acyl-CoA
O
II
CH3cSCoA
Acetyl-CoA
Figure 1.1. Reactions of the (3-oxidation pathway. The fatty acyl-CoA undergoes
a dehydrogenation in the first reaction introducing a trans-a,|3 double bond. The
double bond is then hydrated in the second reaction. A dehydrogenation at the (3
carbon is the third reaction followed by the cleavage of the a,(3 bond in the final
reaction.
4


The first round of a fatty acid through the p-oxidation pathway produces one
acetyl-CoA as a product, one NADH molecule via the third reaction, and one
FADH2 molecule via the first reaction. The acetyl-CoA then enters the citric acid
cycle where an additional FADH2 molecule and NADH molecule are generated.
Therefore, P-oxidation functions as an energy generating pathway with 129 ATP
being formed after the repeated cycles of just one 16 carbon fatty acid molecule
(Voet and Voet, 1990).
Acyl-CoA Dehydrogenases
Enzymes are specialized proteins that function as biological catalysts in
nearly all biochemical reactions occurring in living organisms. Acting under the
identical laws of nature which govern all chemical reactions, enzymes have an
increased capacity to speed reaction rates by lowering activation energy barriers. In
addition, these reactions take place under chemically mild conditions of pH and
temperature, have increased substrate specificity, and are under strict regulatory
control. Without such characteristics, biological systems would be unable to extract
the necessary energy from chemical reactions needed for life.
The acyl-CoA dehydrogenases (ACDs) are one such family of enzymes
which produce energy via the first step of fatty acid oxidation. The members of the
5


family which participate in fatty-acid metabolism are short chain acyl-CoA
dehydrogenase (SCAD), medium chain acyl-CoA dehydrogenase (MCAD), and
long chain acyl-CoA dehydrogenase (LCAD). Though the en2ymes are thought to
have evolved perhaps from the same gene and are similar in many ways, they are
identified and so named for their substrate chain length specificity. Each enzyme
catalyzes the same oxidation-reduction reaction, a, (3-dehydrogenation, of its
specific substrate.
Though many enzymes are able to catalyze reactions such as acid-base
reactions and formation of covalent bonds using only the functional groups of their
side chains, oxidation-reduction reactions require the use of an enzyme cofactor.
Examples of cofactors include metal ions and small organic molecules such as the
heme. When the cofactor is a small organic molecule, it is referred to as a
coenzyme. Such molecules are chemically altered during the catalyzed reaction and
require regeneration via an additional catalytic reaction performed by another
enzyme. Thus, the cycle is completed for the initial enzyme.
6


Flavin as the Coenzyme
The specific coenzyme involved in the reaction catalyzed by the acyl-CoA
dehydrogenases is flavin adenine dinucleotide (FAD) and is shown in Figure 1.2.
Any protein such as the ACDs containing FAD is known as a flavoprotein with a
flavin defined as any compound containing the isoalloxazine ring system. Since
most animals cannot synthesize the isoalloxazine system, it is made available
through the diet in forms such as riboflavin (vitamin B,).
The electroactive portion of FAD is the 7,8-dimethylisoalloxazine. The three
rings of this planar system are designated the xylene, pyrazine, and pyrimidine rings
seen from left to right in Figure 1.2. FAD is substituted at the N(10) position by a
D-ribitol residue derived from the alcohol of the sugar D-ribose. Bound to the
opposite end of the ribityl side chain, the adenosine component of FAD is found.
The functions of the side chain portions of the FAD are confined to the anchoring of
the coenzyme into the active site of the protein with redox activity being confined to
the isoalloxazine system.
Reduction of FAD may take place via a two electron reduction to its fully
reduced or hydroquinone state (FADH2) or via two single electron reductions
whereby the compound exists after half-reduction as the radical semiquinone
(FADH-)- The semiquinone form may itself exist in various protonation states.
7


Figure 1.2. Flavin Adenine Dinucleotide (FAD). The standard numbering of the
isoalloxazine ring system is noted. The isoalloxazine is substituted at N(10) by the
D-ribitol residue which binds the adenosine portion of FAD.
8


Comparing their absorption spectra with the spectrum of fully oxidized flavin
identifies these states. The characteristic spectrum of fully oxidized flavin includes
peaks near 436 nm and 373 nm. The protonation states are known as the anionic red
form so named for the negatively charged species with a red shift seen near 370 nm,
the neutral red species although without carrying a charge also has a red spectral
shift, and the neutral blue form wherein a blue shift is seen in the spectrum near 560
nm (Draper and Ingraham, 1968; Muller et al., 1972). The various oxidation states
of flavin are shown in Figure 1.3. In addition to the spectral changes, reduction of
FAD is evident through the loss of the intense yellow color flavoproteins
characteristically show in their native forms.
Acyl-CoA Dehydrogenase Mechanism
The overall enzyme sequence beginning with the fatty acid substrate of the
ACDs is shown in Figure 1.4. In the mitochondria, the acyl-CoA substrate is
converted to its enoyl-CoA product as the first step in (3-oxidation by the ACDs.
The reduced ACD then passes its electrons onto its physiological electron acceptor
electron transfer flavoprotein (ETF). Subsequently, electrons are passed from ETF
to the membrane bound ETF dehydrogenase and on into the main respiratory chain
via ubiquinone with a final outcome of production of ATP (Thorpe and Kim, 1995).
9


N Y
,nh
0
le-
Jf
Semiquionones partially (le-) reduced flavin
*-ir
-w
Hydroquinones fully (2e-) reduced flavin
Figure 1.3. Oxidation states of flavin. The three oxidation states of flavin are
shown along with the various protonation states.
10


Fatty Acyl-CoA
(substrate)
-ACD-ox 4r
-^ACD-red-
X
->h I h-red
ETF-ox
ETFdh
-ox
N/
Enoyl-CoA
(product)
Figure 1.4. Overall enzyme sequence. Electrons from the reduction of the fatty
acyl-CoA are delivered to the electron transport chain via the acyl-CoA
dehydrogenase (ACD) followed by electron transfer flavoprotein (ETF), followed
by ETF-dehydrogenase (ETFdh) and finally coenzyme Q (CoQ). Oxidized (ox) and
reduced (red) indicate the respective oxidation states of the enzymes.


The accepted catalytic process by which ACD is bound and reduced by
substrate and then regenerated by ETF is illustrated in Figure 1.5. By this method,
the acyl-CoA substrate is bound by oxidized enzyme forming the Michaelis
complex. Electrons are then passed from substrate to enzyme leaving the enoyl
product bound to the fully reduced flavin. Since the binding of the enoyl product in
the active site is extremely tight (Steyn-Parve and Beinert, 1958), its release is only
achieved after the ACD has been reoxidized by ETF. This reoxidation takes place in
two subsequent one electron reductions of ETF. With product binding reduced by
the process, the next substrate compound then easily displaces the product and the
catalysis is repeated.
The mechanism of dehydrogenation of the substrate is depicted in Figure
1.6. As shown, the catalytic amino acid base of the ACD abstracts an a-hydrogen
from the fatty-acyl substrate. This step is favored because the a position has been
activated by the Coenzyme A. As charge is pulled towards the CoA end of the
molecule, the a position takes on a slight positive charge allowing the basic residue
of the protein to abstract the hydrogen. The electrons remaining place a negative
charge on C(3) of the substrate allowing the elimination a P-hydrogen at this
position as a hydride equivalent to N(5) position on FAD (Ghisla and Massey,
1989).
12


REDUCTIVE HALF-REACTION
FAD-ox + substrate FAD-ox substrate ^=^FAD2-redproduct
FAD-ox + product
Figure 1.5. Catalytic process by which ACD is bound and regenerated. ACDs
are reduced in a two-electron reduction by substrate. Product remains bound until
electrons are passed to ETF in two subsequent one-electron reductions. The
regenerated protein then releases its product.
13


R
R
0
SCoA
HB
Figure 1.6. Mechanism of substrate dehydrogenation. Dehydrogenation occurs
via the abstraction of the activated a-hydrogen by the active site glutamate. The
carbanion then initiates the elimination of the [3-hydrogen to the N(5) position of
FAD.
14


The catalytic base responsible for the proton extraction in the
dehydrogenation mechanism has been identified in several species of the ACDs as
a glutamic acid. In short-chain acyl-CoA dehydrogenase, the catalytic glutamate has
been identified as Glu368 while in porcine MCAD it is Glu376 and in long-chain it
can be found at Glu261 (Battaile, et al., 1996). The relative location although
conservatively placed in the binding pocket of the protein plays a role in the
substrate specificity of the enzyme. The exact location of the catalytic residue
makes each protein more suited to act on differing lengths of fatty acid chains.
Structure of the Acyl-CoA Dehydrogenases
All of the members of the acyl-CoA dehydrogenase family which participate
in fatty acid metabolism are homotetramers with each subunit non-covalently
binding one FAD molecule. The quaternary structure and individual subunit size
are consistent from enzyme to enzyme. The native molecular weight for short-chain
is 160,000 while that for both medium and long-chain is 180,000 daltons.
Individual subunit size only varies from short-chains 41,000 to 45,000 daltons for
medium and long-chain (Ikeda et al., 1985).
Since medium-chain acyl-CoA dehydrogenase is the most studied of the
enzyme family, its structure will be presented as a model for the other related
15


enzymes. The structure of the MCAD subunit is shown in Figure 1.7. There are
three domains known as the N and C terminal and the p sheet domains which make
up each subunit of MCAD. Both the N and C terminal domains contain six alpha
helices each with these domains separated by the p sheet domain. The FAD
binding site is within the subunit with the electroactive isoalloxazine ring portion
buried at the interface of the domains. The binding site for the MCAD substrate is
buried among the alpha helices of the N and C terminal domains. When bound, a
substrate molecule will lie length wise in the binding pocket with the isoalloxazine
ring flat against one side and the catalytic amino acid base directly on the other.
Such a sandwich arrangement accommodates the flow of electrons from substrate
to FAD (Thorpe and Kim, 1995). Figure 1.8 depicts this arrangement showing the
MCAD active Glu 376, the substrate, and FAD.
Oxidation-Reduction Reactions
Oxidation-reduction (redox) reactions involve the transfer of electrons from
electron donor to electron acceptor. These reactions may be divided into two half-
reactions or redox couples each representing the reduction of the electron acceptor
or the oxidation of the electron acceptor. A unique property of redox reactions is
that the two half reactions can be physically separated in what is known as an
16


Figure 1.7. Structure of the MCAD subunit. The ribbon diagram of a MCAD
subunit is shown. FAD is shown in black. Picture drawn by Tim Dwyer using the
Rasmol software.


Figure 1.8. Arrangement of the MCAD binding pocket. The catalytic Glu 376 of
MCAD is shown along with the crotonyl-CoA substrate (shown in black) and FAD.
The diagram shows the alignment in the binding pocket. Picture drawn by Tim
Dwyer using Rasmol.
18


electrochemical cell. The free energy of the reaction occurring between the half-
cells can then be determined by measuring their voltage difference. In a redox
reaction such as the one below
Aox + Bred -> + Box (1.2)
where B is the reductant and A is the oxidant, the free energy can be written as
follows.
AG = AG + RT In ([A^tBJ / [AJP.J) (1.3)
Under reversible conditions,
AG = w e) (1.4)
where w e) is the electrical work required to move-electrons across an electrical
potential and
wel=nFE (1.5)
where F is the Faraday constant and E is the electrical potential difference.
Substituting equation 1.5 into equation 1.4,
AG = n F E (1.6)
If equations 1.5 and 1.1 are then combined, the Nemst equation is generated
E = E RT/nF In ([A^JPBJ / [Aj[Bred]) (1.7)
Where E is the redox potential, E is the standard redox potential, R is the Rydberg
constant, T is temperature in Kelvin, n is the number of electrons transferred, and F
19


is the Faraday constant. From this equation, it should be noted that a positive value
for E indicates a spontaneous reaction (negative AG). For a redox couple,
E = E (e_ acceptor) E (e. donor) (1-8)
So for the reaction to occur spontaneously (E is positive), the electron acceptor must
have a more positive standard potential than the electron donor. A species with a
more positive potential has a greater affinity for electrons and therefore will accept
charge from any source with a lower affinity.
Cyclic Voltammetry
One method for determining the standard redox potential of a species is by
cyclic voltammetry. Half-cells are arranged in an electrochemical cell, one of which
containing the species of interest and the other containing a suitable reference
solution for which the E is known. Initially, a potential is applied to the cell at
which no reaction occurs at the working electrode. That is, the potential applied is
high enough to sustain the oxidized species in the oxidized form. A potential ramp
is then applied in the negative direction eventually causing the reduction of the
species in question. This reduction provides a measurable current (electron flow)
which increases as the maximum concentration of species is reduced. The current
then begins to decrease creating an anodic peak as the amount of species
20


approaching the electrode is limited by diffusion. At this point, the potential ramp is
reversed and an identical process of oxidation takes place creating a cathodic peak.
The standard redox potential, or midpoint potential, is the average potential between
that of the anodic and cathodic peaks (Zapien, 1998). Other details of a species
kinetics and electrochemical behavior can also be determined from more in-depth
cyclic voltammetry experiments.
Conclusions
Although much is currently known about the acyl-CoA dehydrogenases and
the purification and certain characterizations of the short-chain acyl-CoA
dehydrogenase have been published, it is still one of the least studied members of
the family. The standard redox potential for any mammalian SCAD has yet to be
determined. This value will provide not only information about the native enzyme
but will allow future experiments which could have broader applications in areas
such has how electrons travel through complex systems like proteins.
In order to produce reliable results for a potential, a purification method and
confirmation of its success are necessary. The results of further investigation on that
protein are then valid and accepted. Although current methods for determining the
redox potential of flavoproteins provide reliable and accurate results, these methods
21


may have flaws and new methods which would improve on current techniques are
always desirable. The purpose of this research is to purify an active rat SCAD
protein and characterize its spectral and kinetic properties. Two methods can then
be used to determine its redox potential, a vital characteristic of any redox protein.
22


CHAPTER 2
PURIFICATION, SPECTRAL AND KINETIC
PROPERTIES OF RAT SHORT-CHAIN ACYL-COA
DEHYDROGENASE
Introduction
In-depth study of short-chain acyl-CoA dehydrogenase began when Shaw
and Engel (1984) used ammonium sulfate fractionation and chromatographic
techniques in one of the initial purifications. The enzyme when purified was
present in two forms. A green form contains CoA-persulfide in the active site of the
enzyme. This ligand is absent in the other yellow form. Shaw and Engel (1987)
later characterized the enzymes green form and outlined further purification
methods to remove the ligand contaminant. Later, upon working with the bacterial
counterpart enzyme to SCAD, Fink et al. (1986) discovered other impurities present
after purification and again modified the protocol.
After a purification technique has been employed, spectral and kinetic
properties are determined to characterize a properly separated enzyme. Spectral
properties play an important role in the study of many enzymes, especially flavin
containing dehydrogenases, because of the existence of very characteristic patterns.
Detection of a flavin containing protein and a measure of purity are readily


determined from its spectrum. Other spectral assays provide information on the
mechanism of catalysis. Kinetic parameters are also determined in a complete
analysis of a purified protein. Such parameters allow comparison with related
enzymes and further insight into the proteins physiological role.
Complete characterization, both spectral and kinetic, is necessary when any
modifications are made to a purification protocol. Spectral and kinetic parameters
can then be compared with the available literature to validate the new purification
method. Small changes in previous protocols were made in the purification of the
SCAD used in the subsequent studies of this work. The objective of this chapter is
to present this purification method and the full spectral and kinetic characterization
of the enzyme ensuring the use of an adequately purified protein in further study.
Materials and Methods
Ultraviolet/visible spectroscopy was performed using a Shimadzu UV2401
spectrophotometer unless otherwise noted. All reagents were reagent grade or better
and were obtained from commercial sources. Plasmid containing rat SCAD was the
generous gift of Dr. Jerry Vockley, Mayo Clinic, Rochester, Minnesota.
24


Purification of SCAD
Inoculation with 10 mL of starter culture took place in 10 flasks containing
800 mL of Tryptone-Yeast Phosphate (TYP) media (16 g/L tryptone, 16 g/L yeast
extract, 2.5 g/L Potassium Phosphate, and 5 g/L NaCl, brought to pH 6.8 with HC1),
100 pg/mL ampicillin, and 500 pg/mL IPTG. Cells were grown at 37 C overnight
with vigorous aeration to an optical density at 600 nm of approximately 8.5. Cells
were harvested by centrifugation at 7500 rpm for 10 minutes, resuspended in
phosphate buffered saline and placed in the centrifuge again at 8000 rpm for 20
minutes. The resulting cell pellet was then stored at -70 C. Cells were kept frozen
no longer than 3 days before disruption. To disrupt the cells, the thawed pellet was
resuspended in approximately 100 mL of 10 mM Tris-HCl, pH 7.5 and
homogenized with a Potter-Elvehjem homogenizer. Cells were then passed through
a French pressure cell twice at 1300 psi. The disrupted cells were placed in the
centrifuge for one hour at 100,000 X g to remove cell debris.
After ultracentrifugation, the supernatant was loaded onto a 373 mL DEAE
gel exclusion chromatography column previously equilibrated with 10 mM Tris-
HCl, pH 7.5. The column was then washed with 400 mL of equilibration buffer,
and the wash through was assayed for SCAD activity to assure the protein had
successfully bound to the column. The protein was then eluted from the column
25


with a 2 L gradient of 0-300 mM NaCl in 10 mM Tris-HCl, pH 7.5. Fractions were
collected and assayed for catalytic SCAD activity. Fractions containing peak
activity were pooled and concentrated in an Amicon concentrator using a YM100
membrane.
The concentrated protein solution was brought to 0.1 M Tris-HCl, pH 7.5
and degassed with 10 cycles of alternating argon and vacuum. After addition of 600
mg of sodium dithionite, 10 more degassing cycles were performed. The enzyme
was then allowed to sit at room temperature for several hours before dialyzing
against 50 mM Tris-HCl, pH 7.5 with 10 mM sodium dithionite overnight. A final
dialysis was performed against 10 mM Tris, pH 7.5 to remove dithionite.
The resulting protein was loaded onto a 150 mL hydroxyapatite (HA)
column previously equilibrated with 5 mM potassium phosphate, pH 7.5 buffer.
While the column was washed with equilibration buffer, fractions were collected
and absorbance monitored at 280 nm to detect the elution of any sodium dithionite
not removed by prior dialysis. After an increase and subsequent decrease in
absorbance at 280 nm, SCAD was eluted from the column with a 2 L gradient of 5-
250 mM potassium phosphate, pH 7.5. Fractions were collected and again assayed
for enzyme activity. Fractions containing dehydrogenase with the highest specific
26


activity were pooled. The enzyme could then be frozen at -70 C for several
months.
Pooled fractions from the HA column elution were concentrated to
approximately 5 mL, spun at 7000 rpm for 15 minutes if any precipitate was present,
and loaded onto a Sephacryl S-200 gel filtration column (2.5 cm X 90 cm)
equilibrated with 10 mM Hepes, pH 7.5. Protein was eluted from the column with
equilibration buffer and fractions were collected. Spectra of individual fractions
eluted from the S-200 column were determined and the dehydrogenase was pooled
according to R value. The R value is calculated from the absorbance spectrum of
the protein by dividing the absorbance at 270 nm by the absorbance at 450 nm. This
value ascertains the protein to flavin ratio indicating the relative amounts of all
proteins versus those containing flavin. Non-flavin containing proteins will
contribute absorbance to the 270 nm peak due to aromatic rings without increasing
absorbance near 450 nm where flavin absorbs. Therefore, the lower the R value, the
higher the purity of the flavoprotein. Pools were concentrated and stored in either
20% glycerol at -70 C.
27


Determination of SCAD Enzyme Activity
Enzyme activity was measured spectrophotometrically using a standard dye-
coupled assay (Engel and Massey, 1971; Williamson and Engel, 1984). Protein was
reduced by its physiological substrate, butyryl-CoA, and regenerated using a final
electron accepting dye, dichlorophenolindopbenol, which undergoes a color change
from blue to clear upon reduction. Reaction mixtures contained 50 pM butyryl CoA
as substrate, 38 pM dichlorophenolindophenol (DCPIP) as the terminal electron
acceptor, and 2.5-5.0 pL of SCAD in 20 mM Tris-HCl, pH 8.0. Reactions were
performed at 25 C and were initiated with the addition of phenazine methosulfate
to a final concentration of 0.001 M. The progress of the reaction was monitored at
the absorbance maximum of the dye, 600 nm, for 60 seconds and rates were
calculated from the linear portion of the curve. Activities were then calculated from
the change in absorbance per min using Beers Law and the extinction coefficient
for DCPIP of 21 X 103cmImM1 (Armstrong, 1964).
Determination of Extinction Coefficients
for SCAD and the Charge-Transfer Complex
Determination of Extinction Coefficient for Oxidized SCAD. The
extinction coefficient for oxidized protein at the visible absorption maximum, 450
28


nm, was determined as follows (McKean et al., 1983). The purified protein was
dialyzed against 10 mM sodium phosphate, pH 7.5 for 16 hours and precipitated
protein was removed by centrifugation. A visible absorbance spectrum of the
protein was obtained. FAD was released from the protein by treatment with 1%
SDS and an additional spectrum obtained. The absorbance of free FAD and its
known extinction coefficient in SDS of 12.045 X 103 cm'1 mM'1 (Salazar, 1998)
were used to calculate the extinction coefficient of enzyme bound FAD. The
technique was repeated for accuracy.
Determination of the Extinction Coefficient for the Charge-Transfer
Complex. The extinction coefficient for the charge-transfer complex was
determined by reduction of the protein with its physiological substrate without a
further electron acceptor present. Absorbance at the 450 nm peak of SCAD in 10
mM Hepes, pH 7.5 at 25 C was measured and its concentration calculated using
s450nm,ox = 14.5 X 103 cm'1 mM'1. The substrate, butyryl CoA, was added to a final
concentration of 100 pM and a subsequent spectrum taken. Absorbance at the
charge-transfer peak of 560 nm was determined and the extinction coefficient
calculated using Beers Law and the calculated SCAD concentration. Saturation
29


was assured by further addition of butyryl CoA with no change noted in the
spectrum.
Steady-state Kinetic Assays
Kinetic parameters for SCAD were determined using the standard dye-
coupled assay previously utilized for activity assays. For kinetic assay, the enzyme
is reduced by varying concentrations of substrate and electrons passed to a terminal
electron acceptor where reduction is monitored. Reaction mixtures contained a final
protein concentration of 7 nM 600 pM phenazine ethosulfate, and 40 pM
dichlorophenolindophenol in 20 mM Tris, pH 7.5. Reactions took place at 25 C
and were initiated with the addition of 2-20 pM butyryl CoA as the varied substrate.
Reaction progress was monitored spectrophotometrically at 600 nm for 60 seconds
and duplicate runs were performed. Rates were calculated as changes in absorbance
per minute using only the linear portion of the reduction curve. Kinetic parameters
and their standard errors were determined by plotting a standard reciprocal plot
using the Trinity Softwares Enzyme Kinetics program
30


Determination of SCAD Mechanism
SCAD was initially reduced using sodium dithionite with crotonyl CoA,
SCADs physiological product, present. The titration was performed at 25 C in 10
mM Hepes, pH 7.5. All components of the reaction mixture were made anaerobic
by repeating 10 cycles of argon gas and vacuum. The reaction mixture contained
7.5 pM SCAD and 125 pM crotonyl CoA. An absorbance spectrum was obtained
and the titration initiated using 10 pL of a 1.00 mM sodium dithionite solution in
0.001 M sodium pyrophosphate, pH 9.0. Successive additions of 4 pL of the
dithionite solution were made until full reduction had occurred. Spectra were
obtained after each dithionite addition.
The same method was employed in the reduction of SCAD without a
product present. All conditions were kept constant as well as reaction mixture
component concentrations. The crotonyl CoA was not added. The titration was
begun using 6 pL of the dithionite solution. One successive addition of 3 pL was all
that was required for full reduction. Spectra were measured before the titration and
after the addition of reducing agent.
SCADs reduction of its physiological electron acceptor, ETF, was
monitored spectrophotometrically at 25 C in 10 mM Hepes, pH 7.5. A reaction
mixture containing 9 pM pig-ETF and .6 mM butyryl CoA was made anaerobic by
31


10 alternating cycles of argon and vacuum. The reaction was initiated with the
addition of 250 nM SCAD and spectra were taken every 2 minutes until full
reduction of ETF had occurred.
Results and Discussion
Purification
A typical purification of short-chain acyl-CoA dehydrogenase produced pine
protein in good yield. A summary of a typical purification is listed in Table 2.1.
The initial pellet weight of cells was 45.5 g and led to the final recovery of 127 mg
of SCAD. Table 2.1 shows a decrease in the total number of enzyme units from the
crude extract to the DEAE column fraction pool. This data may seem unlikely as a
continual decrease in units is expected as a purification protocol proceeds. It is
likely, however, that the increase is actually due to removal of enzyme activity
inhibitors. Table 2.1 also shows a decrease in units, as expected, after the
completion of the chromatography using the HA column. This decrease, however,
may be accentuated by the additional removal of protein by the de-greening step
performed between the DEAE and HA columns. The de-greening step is not a
separation step,
32


Table 2.1. Purification of rat SCAD. Enzyme activity was measured with 50 p.M
butyryl-CoA as outlined in the Materials and Methods section. Total protein was
calculated using s450nm = 14.5 X 103 M'1.
Volume (ml) Total Activity (fjmol/ min) Total Protein (mg) mg/ml R value (A 27(/ A 450)
Crude Extract 110 1120 nd nd nd
DEAE eluate 223 2950 nd nd nd
HA eluate (pool 1) 114 390 nd nd nd
HA eluate (pool 2) 250 745 nd nd nd
HA eluate (pool 3) 184 192 nd nd nd
Gel filtration eluate 26 Nd 72.1 2.77 5.50
(pool 2a)
Gel filtration eluate 26 Nd 31.6 1.22 5.65
(pool 2b)
Gel filtration eluate 26 Nd 23.3 .932 5.78
(pools 1 & 3)
nd: not determined


but the harsh reducing conditions used to remove the Co A-persulfide ligand lead to
some denaturation and therefore, loss of enzyme units. It is necessary to remove the
persulfide ligand, however, because it is an enzyme inhibitor. Adjusting the pH
during the addition of sodium dithionite so that the protein is left in a slightly less
harsh environment may minimize loss.
Additional loss of protein was experienced if, as in the purification shown in
Table 2.1, the HA column pools were stored at -70 C for a period of time. The
first pool from the HA column was not frozen and showed no additional
precipitation upon completion of the purification process.. However, when the
second two HA column pools were removed from the freezer and concentrated for
the final chromatography step, there was considerable loss of protein. This
precipitation could be due to instability brought about by removal of the CoA-
persulfide ligand and compounded by the fact that the pools were frozen without
addition of glycerol to aid in stabilization. It would, therefore, be best not to store
protein prior to the completion of the entire purification protocol when glycerol is
added for storage.
There were additional problems with stabilizing the short-chain protein.
Initially, the protein was stored in 20% glycerol at -70 C. When this protein was
thawed for use, there was some precipitation noted. The precipitation problem
34


compounded when there was need for concentration or dialysis of the protein.
Perhaps the protein was denaturing because of the freezing and thawing process, so
in a subsequent purification, the protein was stored in 50% glycerol but only a -20
C. After only a short time, this protein began to denature rapidly as well and so
storage of the protein was returned to -70 C, Certain losses were expected upon
any manipulation of the protein in preparation for its use.
The precipitation problem may be explained by previous reports indicating
that the green form of the protein may actually be more stable than the yellow form
(Shaw and Engel, 1984). Such observations could be explained by the Co-A
persulfide ligand occupying the binding site of the protein therefore giving it
physical and chemical stability. The yellow form without the artificial substrate
may be more affected by solvent interactions or physical deformations of the protein
as would be present in the various conditions mentioned. A solution for the problem
or any further affects besides loss has not been identified.
After the completion of all steps of the purification, the SCAD obtained was
highly pure. The hydroxyapatite column can illustrate the success of the
chromatography steps. This columns elution profile is shown in Figure 2.1. The
profile not only shows the activity of the protein as it comes off the column but also
indicates purity. Purity for the SCAD is established by the A270/A450 or R value. As
35


Fraction Number
Figure 2.1. Elution of SCAD from Hydroxyapatite column. Protein was applied
to the HA column previously equilibrated with 5 mM potassium phosphate, pH 7.5.
The column was washed with equilibration buffer and eluted with a 2 L gradient of
5-250 mM potassium phosphate, pH 7.5. The absorbance at 270 nm (O) and 450
nm () were monitored and the enzyme activity (A) towards butyryl CoA was
determined. The R value (x) was also calculated.
36
R value


shown in the figure, R values were below 7 for the majority of the fractions
containing SCAD activity. After the protein had run through the final polishing step
with the gel filtration column, typical R values of final protein were around 5.5.
These values exceed the purity used in previous publications using short-chain
dehydrogenases including that of rat.
A spectrum of the purified yellow form of rat SCAD is pictured in Figure 2.2
showing the characteristic pattern for a flavoprotein. The peaks seen are at 270 nm,
368 nm, and 450 nm. The large absorption at 270 nm arises from aromatic ring
structures present in proteins as well as the adenine and isoalloxazine moieties of
FAD. Peaks at 368 nm and 450 nm are the result of the bound FAD prosthetic
group. (Mahler, 1954).
Spectral Properties
The extinction coefficient for the oxidized form of SCAD at its absorbance
peak of 450 nm was determined to be 14.5 mM^cm'1 using SDS denaturation.
Initial attempts were made in determining e450nI1, using heat denaturation but were
met with unexplained large errors. The final value calculated is in good agreement
with other reported values ranging from 14.4 mM''cm reported by Williamson and
Engel (1984) for SCAD from M. elsdenii to 15.4 mM''cm for pig kidney general
37


Wavelength (nm)
Figure 2.2. Absorption spectrum of purified rat SCAD. The figure shows the
absorption spectrum of SCAD in 10 mM Hepes, pH 7.5. The protein was purified
as noted in the text.
38


acyl-CoA dehydrogenase (Thorpe et al., 1979). The extinction coefficient for free
FAD also determined at the 450 nm maximum is 11.3 mM^cm'1. Thus, the value
determined for SCAD shows the characteristic increase in flavin absorption upon
protein binding.
The spectrum resulting from the addition of saturating concentrations of
SCADs physiological substrate butyryl CoA is shown in Figure 2.3. Changes in
the spectrum from initial conditions include a blue shift of the 370 nm band of
approximately 20 nm, a disappearance of the flavin maximum at 450 nm, and the
appearance of a peak at 560 nm. The new band at the longer wavelength near 560
nm initially thought to be the result of flavin-semiquinone (Beinert, 1957) formation
corresponds to a charge-transfer complex between the fully reduced (two electron)
flavin and the bound crotonyl-CoA (Massey and Ghisla, 1974; Hall et al., 1979;
Frerman et al., 1980; Thorpe et al., 1981; Schmidt et al., 1981). This complex
provides the thermodynamic stability necessary for the electrons to be transferred
from the dehydrogenase to its electron acceptor, ETF, in two sequential one electron
steps. The bound product also protects the reduced flavin from reaction with
molecular oxygen which would short circuit the energy yielding pathway and
generate hydrogen peroxide, a toxic oxygen species (Crane and Beinert, 1956).
Calculations from Figure 2.3 lead to an extinction coefficient for the charge-transfer
39


Figure 23. Aerobic reduction of SCAD with butyryl-CoA. Protein (11.7 pM) in
a 1 ml reaction mixture containing 10 mM Hepes, pH 7.5, was reduced with the
addition of 25 pi of 5.76 mM butyryl- CoA. Spectra were recorded before and after
the addition. Saturation was verified with an additional 10 pi of substrate leading to
no change in the spectrum.
40


band of 3.19 mlvr'cm'1. Other reported values for this e were not found but the
effects of substrate saturation are in agreement with other literature.
The titration of SCAD using sodium dithionite with product is shown in
Figure 2.4. This titration results in a red shift of the 370 run peak of approximately
15-20 nm, the appearance of a shoulder around 480 nm, and a long-wavelength band
near 580 nm. As SCAD is titrated, the 370 nm band initially shows a slight
decrease after which it continually increases until completion. Such spectral
changes are the result of the formation of a stable anion semiquinone. The
semiquinone indicates the flavin existing in its one electron .reduced state. The
presence of product makes this reduction possible as is shown by Figure 2.5. This
figure shows the same conditions with the exception that there is no product present.
Without product, the spectra resulting from sodium dithionite titration are markedly
different. There is full reduction of the flavin as seen in the absence of absorbance
at 450 nm and no formation of any long-wavelength band indicating any
semiquinone formation. Thus, the existence of product allows the one electron
reduction of flavin.
Figure 2.6 shows spectrally how SCAD reduces its physiological electron
acceptor ETF. The characteristic spectrum of ETF is similar to that of SCAD with
the absorbance maxima around 370 nm and 450 nm. Additionally, ETF shows a
41


Figure 2.4. Anaerobic titration of SCAD using sodium dithionite with product
present. SCAD (7.5 pM ) in a 1 ml anaerobic reaction mixture containing 125 pM
crotonyl-CoA in 10 mM Hepes, pH 7.5, was reduced using 4-10 pi additions of 1
mM sodium dithionite in 0.001 M sodium pyrophosphate, pH 9.0. Spectra were
recorded after each successive addition of dithionite until no further change was
detected.
42


O.llOr
Wavelength (nm)
Figure 2.5. Anaerobic titration of SCAD using sodium dithionite without
product present. SCAD (7.5 pM) in 10 mM Hepes, pH 7.5, was made anaerobic
and reduced using 3-6 pi additions of 1 mM sodium dithionite in 0.001 M sodium
pyrophosphate, pH 9.0. Spectra were recorded after each successive addition of
dithionite until no further change was detected.
43


Figure 2.6. Anaerobic titration of Electron Transfer Flavoprotein using SCAD.
ETF (9 JJ.M) in an anaerobic reaction mixture containing 0.6 mM butyryl-CoA in 10
mM Hepes, pH 7.5, was reduced using 250 nM SCAD. Spectra were recorded
before SCAD addition and every 2 minutes thereafter until complete reduction was
noted. Numbering indicates order of spectra.
44


shoulder around 460 nm. As ETF is allowed to be reduced by SCAD, an initial
increase of the 370 nm peak is seen followed by subsequent decreases. The 450 nm
peak continually decreases with time. The initial increase at 370 nm indicates initial
formation of the semiquinone form of ETF followed by the decreases indicating the
formation of the fully reduced hydroquinone. Since the spectra were taken with
time, the formation of semiquinone occurred quickly while the full reduction took
significantly more time.
Upon consideration of the spectral evidence presented, the mechanism used
by this protein can be assumed much like other members of the acyl-CoA
dehydrogenase family already studied. The two electron reduction of FAD occurs
upon availability of substrate forming the charge-transfer complex. The product is
not released until the re-oxidation of the FAD is possible and so remains stable in
the absence of an electron acceptor. When the electron acceptor, ETF, becomes
available, one electron is transferred out of SCAD at a time. The transfer of the
second electron is the slow, rate-limiting process completing the return of SCAD to
its native form.
45


Kinetic Properties
The double reciprocal (Lineweaver-Burke) plot used to determine the steady-
state kinetic properties for rat SCAD is shown in Figure 2.7. The plot shows
excellent linearity with a maximal velocity, Vmax, value of 4.256 s'1 and a value
of 1.248 pM. A calculated value of Vmax/Kn, indicating the enzymes overall
efficiency is 3.41 pM"'sec"'. These kinetic values are similar to those currently
reported in the literature.
46


1/[SCAD] (pM)
Figure 2.7. Double reciprocal (Lineweaver-Burke) plot of steady-state kinetic
assay. The standard dye-coupled reaction mixtures contained 7 nM SCAD, 600 pM
phenozine ethosulfate, and 40 pM DCPIP in 20 mM Tris-HCl, pH 7.5. Reactions
were initiated using 2-20 uM additions of butyryl-CoA as the varied substrate.
Change in absorbance was monitored at 600 nm for 60 seconds. Rates were
calculated in enzyme units per time.
47


CHAPTER 3
DETERMINATION OF THE MIDPOINT POTENTIAL
OF RAT SHORT-CHAIN ACYL-COA
DEHYDROGENASE USING THE
XANTHINE/XANTHINE OXIDASE
REDUCTION METHOD
Introduction
Although the redox potentials of several of the acyl-CoA dehydrogenases
have been explored (Fink et al., 1986; Gustafson et al., 1986; Lenn et al., 1990), the
potential for any mammalian SCAD species remains undetermined. Since the
dehydrogenase family has been characterized in terms of electron donors and
acceptors, the interest in the redox potential of SCAD lies with the effect on this
property of amino acid changes in the vicinity of the cofactor, FAD. After the
potential of the wild-type SCAD is known, comparisons can then be made with the
potentials of mutant proteins of interest. Any changes noted in mutant potentials
can then be directly correlated to the amino acid residue in question. Such
information may play a role in determining the pathway of the electrons as they
flow from substrate to enzyme and finally to electron acceptor.
The most commonly used method for determining the redox potential of
flavoproteins is the xanthine/xanthine oxidase reduction system developed by
Massey (1991). Although this method is limited by its approximation of an
48


equilibrium system, it provides accurate values for potentials without requiring the
specialized equipment of other methods such as potentiometry. The objective of
this chapter is to use the Massey method to determine the redox potential of rat
short-chain acyl-CoA dehydrogenase.
Materials and Methods
Rat SCAD was expressed and purified from E. Coli as described in Chapter
2 of this thesis. Ultraviolet/visible spectroscopy was performed using a Shimadzu
UV2401 spectrophotometer. All reagents were reagent grade or better and were
obtained from commercial sources.
Determination of Extinction Coefficient for
Reduced SCAD
The extinction coefficient for the reduced SCAD was determined
using the method described by Massey and Hemmerich (1978) in which the
flavoprotein is photoreduced using deazaflavin as a catalyst. Reactions were
completed at 4 C in 10 mM potassium phosphate, 10% ethylene glycol, 10 mM
EDTA, pH 7.0. Reaction mixtures were made anaerobic by initial bubbling of
buffers with argon for 15 minutes and subsequent alternating of argon and vacuum
for 20 cycles of final reaction mixtures. Mixtures of 13.6 pM SCAD and 16 nM
49


5-deazaflavin were kept in the dark and allowed to stand for 2 minutes before the
experiment. The reaction was initiated with a 5-second exposure to the light of a
slide projector. A full absorption spectra was then obtained to monitor the
proteins reduction. Light exposure was then repeated and additional spectra
obtained until full reduction of the SCAD had occurred. The extinction coefficient
of the reduced enzyme was then calculated using Beers Law and the previously
calculated value for the oxidized coefficient of 14.5 X 103 cm'1 mM'1 at 450 nm.
Re-crystallization of Redox Dye
Redox dye Indigo Carmine was re-crystallized by dissolution in a minimal
volume of water. Acetone was added until complete precipitation of the dye
occurred. The crystals were then collected by vacuum filtration and allowed to dry
overnight.
Determination of Xanthine Oxidase Enzyme Activity
Enzyme activity was measured spectrophotometrically by monitoring the
reduction of xanthine by xanthine oxidase. Reaction mixtures contained 400 pM
xanthine in 0.1 M potassium phosphate bufffer, pH 8.4. The reaction was initiated
by addition of 5 pi of 0.90 nM xanthine oxidase and absorbance at 296 nm was
50


monitored for 60 seconds. Successive additions up to 25 pi xanthine oxidase were
made with activities calculated as change in absorbance per time. These values
were plotted versus the volume xanthine oxidase added to assure linearity.
Determination of SCAD Midpoint Potential
The midpoint potential of SCAD was determined using the
xanthine/xanthine oxidase reduction method of Massey (1991). With this method, a
redox indicator dye is used with a known potential near that of the protein. The
reduction of both dye and protein are monitored and the apparent potential of the
protein is determined using the Nemst equation. Reduction reaction mixtures of the
indicator dye alone contained 15 pM indicator dye, 1.5 pM benzyl viologen as the
mediator dye included to facilitate equilibrium between indicator dye and protein,
and 250 pM xanthine in deaerated 10 mM potassium phosphate buffer, pH 7.0. A
final concentration of 40 nM xanthine oxidase in buffer was added to the sidearm of
a quartz cuvette and the reaction mixture was made anaerobic by 25 alternating
cycles of argon and vacuum. After allowing the solution to rest for 2 minutes, the
reaction was initiated by addition of the xanthine oxidase and spectra were
monitored every 120 seconds for 30 minutes at 15 C. Isosbestic points were
determined from the accumulated spectra.
51


Reduction reaction mixtures of the protein alone contained 10 pM SCAD,
1.5 pM benzyl viologen, and 250 pM xanthine in deaerated 10 mM potassium
phosphate buffer, pH 7.0. After the mixture was made anaerobic as described
above, the reaction was initiated using 63 nM xanthine oxidase and monitored every
60 seconds at 15 C until complete reduction of protein was observed.
Reduction reaction mixtures to determine the redox potential of the protein
contained 15 pM SCAD, 15 pM indicator dye, 1.5 pM benzyl viologen, and 250
pM xanthine in deaerated 10 mM potassium phosphate buffer, pH 7.0. These
mixtures were made anaerobic and 25-65 nM xanthine oxidase was added to initiate
the reaction. Reduction of protein and dye were monitored every 60 seconds at 15
C until either had reduced completely. Reduction of protein was monitored at dye
isosbestic points near the proteins maximum. The midpoint potential was
determined by plotting the log of the oxidized over reduced concentrations of dye
versus the log of the oxidized over reduced concentrations of the protein. This plot
represented the Nemst equations for dye and protein set equal and the potential
value for the protein was calculated from the y-intercept and the known potential of
the dye.
52


Results and Discussion
Determination of the Extinction Coefficients for
Reduced SCAD
The spectra obtained during the photoreduction of SCAD using a
deazaflavin catalyst are shown in Figure 3.1. The spectra show the complete
quenching of the visible absorbance peak of SCAD at 450 nm indicating the full
reduction of the protein. In initial runs of the experiment, some protein reduction
was observed prior to obtaining spectra. Keeping subsequent reaction mixtures
covered during the preparation time eliminated the problem. In addition, re-
oxidation of the protein was observed as the experiment time progressed due to
oxygen exposure. Addition of the deazaflavin from a sidearm cuvette instead of
using a gas tight syringe kept the reaction mixture anaerobic for the duration of the
reduction. The extinction coefficient calculated for the reduced SCAD species at
468 nm was 2.48 X 103 cm'1 mM'1. The extinction coefficient for the oxidized
species at 468 nm was 12.4 X 103 cm'1 mM'1. Determination of the extinction
coefficients for the reduced and oxidized species at this wavelength was necessary
for later calculation of SCAD concentrations during the midpoint potential
determination experiments.
53


Figure 3.1. Photoreduction of SCAD using a deazaflavin catalyst. Anaerobic
reaction mixtures containing 13.6 pM SCAD and 16 nM deazaflavin in 10 mM
potassium phosphate, 10 mM EDTA, 10% ethylene glycol, pH 7.0 were exposed to
10 second intervals of light at 4 C. Spectra were taken after each light exposure
until full reduction has taken place. Not all spectra are shown.
54


Determination of Xanthine Oxidase Catalytic Activity
The catalytic activity of xanthine oxidase was measured by the reduction of
its electron acceptor xanthine. Xanthine has an absorbance maximum at 296 nm
which increases upon reduction. The plot of xanthine oxidase activity measured in
change in absorbance at 296 nm per 20 seconds versus volume of xanthine oxidase
added is seen in Figure 3.2. The graph shows excellent linearity with increase in
enzyme concentration. Linearity indicates that the enzyme is functioning properly.
The assay was performed to ensure that the xanthine oxidase used for the midpoint
potential determination experiments could adequately catalyze the reduction
reactions.
Determination SCAD Midpoint Potential
The oxidation-reduction midpoint potential of short-chain acyl-CoA
dehydrogenase was determined using the xanthine/xanthine oxidase method of
Massey. In order to obtain accurate values with this method, the indicator dye used
should have a standard potential within 30 mV of the protein in question. A dye
with nearly equivalent potential will closest approach equilibrium conditions. In
looking to the literature for related protein potentials, the expected value of SCADs
potential was approximately -120 mV (Gustafson, 1986). In addition to specific
potentials, indicator dyes should also provide little absorbance interference near the
55


W
u
c
o
o
CD
cn
O
CM
0.014
0.012
0.01
O)
CN
W
SI
<
a)
ra
c
CO
SI
O
0.008
0.006
0.004
0.002
0
I
I
5 10 15 20 25 30
|il Xanthine Oxidase
Figure 3.2. Plot to determine the activity of xanthine oxidase. Xanthine (400
pM) in 0.1 M potassium phosphate. pH 8.4 was reduced using successive additions
of 5 pi of 0.9 nM xanthine oxidase. Activity was measured as a decrease in
absorbance at 296 nm for 60 seconds after each addition. This activity is plotted
versus the volume of xanthine oxidase added and the plot is examined for linearity.
56


proteins maximum and have their absorbance maximum in a region where the
protein does not absorb. In this way, the reduction of protein and dye can be
monitored simultaneously without interference. SCADs maximum absorbance
peak is at 450 nm. An ideal dye would contain an isosbestic point at or near the
proteins absorbance maximum with a potential nearly identical to that of the
protein.
With an expected approximate value for SCADs potential and knowledge of
its spectral properties, initial experiments were run using indigo carmine as the
redox indicator dye. This dye is commonly seen in the literature with other
flavoprotein potential determination experiments. Initial runs of an experiment with
dye alone showed no reduction or incomplete reduction even with increased
concentrations of xanthine oxidase, the reaction initiator. After replacement of the
gas train used to make the reaction mixtures anaerobic in the laboratory and re-
crystallization of the dye, complete reduction of indigo carmine was accomplished
as shown in Figure 3.3. The spectra show an ideal isosbestic point at 468 nm near
SCADs maximum and a dye peak at 620 nm where SCAD is not expected to
absorb. It was then necessary to reduce SCAD alone in order to show its behavior
in the experimental conditions at the proposed wavelengths. The spectra obtained
from this reduction are shown in Figure 3.4. These spectra show a decreasing
absorbance at 468 nm which could be used to monitor SCAD reduction and minimal
57


-0.05 -i
-OT
-S.lS.-i
_______300_____________wi 400 4S0 SOP 5SD 600________________WavgWncJi term
Figure 3.3. Anaerobic reduction of indigo carmine using xanthine/xanthine
oxidase. Anaerobic reaction mixtures contained 15 p.M indigo carmine, 1.5 pM
benzyl viologen, and 250 pM xanthine in 10 mM potassium phosphate, pH 7.0.
Reactions were run at 15 C and initiated with 40 nM xanthine oxidase. Spectra
were monitored.
58


Figure 3.4. Anaerobic reduction of SCAD using xanthine/xanthine oxidase.
Anaerobic reaction mixtures contained 10 p.M SCAD, 1.5 pM benzyl viologen, and
250 pM xanthine in 10 mM potassium phosphate, pH 7.0. Reactions were run at 15
C and initiated with 63 nM xanthine oxidase. Spectra were monitored.
59


absorbance at 620 nm where the reduction of dye could be monitored. Therefore,
with the potential of the indigo carmine (E = -.121 V) within the expected range of
the protein, the experiment was performed with both dye and protein present.
The midpoint potential of SCAD using indigo carmine as indicator dye was
determined to be -. 182 V. Sample spectra from a reduction are shown in Figure
3.5. Figure 3.6 shows the Minneart plot from which the potential was calculated.
This potential value is considerably lower than what had been anticipated and so
was met with skepticism. In preliminary experiments, it was clear that the protein
was reducing considerably slower than the dye indicating its lower potential.
Therefore, numerous trials were made using alternate dyes of differing potentials.
Indigo tetrasulfonic acid (E= -.046 V) was an unlikely alternative due to its higher
potential but was used to rule out the possibility that indigo carmine was perhaps not
interacting with the protein as expected. Results again showed the protein reducing
slower than dye verifying previous conclusions.
Another dye used was phenosafranine (E0,= -.252 V) which has a
considerably lower potential. Results with this dye proved inconclusive as a
correction had to be made for the dyes absorbance at the protein absorbance
wavelength. This correction introduced enough error that results were not in
agreement with previous calculations. Other dyes looked at included anthraquinone
60


400 450 500 550 600 650 700
Wavelength (nm)
Figure 3.5. Anaerobic reduction of SCAD and indigo carmine using
xanthine/xanthine oxidase. Anaerobic reaction mixtures contained 15 p.M SCAD,
15 p.M indigo carmine, 1.5 pM benzyl viologen, and 250 pM xanthine in 10 mM
potassium phosphate, pH 7.0. Reactions were run at 15 C and initiated with 25-65
nM xanthine oxidase. Spectra were monitored. Not all spectra are shown.
61


3 -
0.5 -
---------9--------------:-----------:----------------------------------
.-0.5 0 0.5 1 1.5 2 2.5
log [ox] dye / [red] dye
Figure 3.6. Minneart plot of reduction of SCAD and indigo carmine data.
Concentrations of dye and protein were determined and the Minneart plot was
drawn. A regression line was fit to the data. E0 was calculated from the y-intercept
and slope was calculated.
62


2,6-disulfonate (E= -.184 V), neutral blue (E= -.190 V) and cresyl violet acetate
(E=-. 166 V). The anthraquinone had a large absorbance at the protein peak which
prevented monitoring of protein reduction. Both neutral blue and cresyl violet
acetate did not equilibrate with the protein indicating perhaps these compounds
could not enter the proteins binding site. Many other available dyes in the range of
potentials needed were quinone containing with their absorbance maxima near that
of SCAD preventing their use in the experiment.
The results obtained using indigo carmine were, therefore, used to calculate
the SCAD midpoint potential. The consequences of not finding an ideal dye for the
protein can be seen in the Minnaert plot (Figure 3.6). In an experiment such as this
with protein and dye reducing the same number of electrons, a slope of one is
expected. The slope seen with SCAD varied from this ideal. Such variations may
be directly related to the differences in potentials between the dye and protein. The
potential difference in this experiment was approximately 60 mV, twice the value
suggested by Massey. Under these conditions, the experiment was not conducted
under equilibrium conditions. The rate at which the dye reduced was significantly
faster than the protein. Therefore, variations would be expected and the result
obtained should be seen as a valid approximation of the true midpoint potential for
this protein.
63


CHAPTER 4
DETERMINATION OF THE MIDPOINT POTENTIAL
OF SHORT-CHAIN ACYL CO-A DEHYDROGENASE
BY CYCLIC VOLTAMMETRY
Introduction
Cyclic voltammetry of proteins is a useful method for probing electron
transfer properties. Such properties include the standard rate constant (k), a
measure of electron transfer kinetics, the transfer coefficient (a), a measure of
energy barrier symmetry, and the formal potential for a solution species (E) or
adsorbed species (Eads). The first two properties can only be determined when
electrons are transferred directly from the electrode to the protein in solution.
Voltammetry also offers an advantage over some spectroelectrochemical techniques
used in which the species of interest does not come in contact with the electrode.
With the latter, equilibrium is approached between a dye and the protein, and the
potential is calculated with respect to the dye. In cyclic voltammetry, direct electron
transfer eliminates the uncertainty of the equilibrium approximation.
Although cyclic voltammetry may seem to be an ideal method for the
determination of formal potential, its use with proteins is somewhat problematic.
Many initial experiments with proteins such as cytochrome c gave no currents using
64


cyclic voltammetry. After much work in the area, methods have been developed
and new electrode surfaces have made cyclic voltammetry of proteins feasible.
Many proteins still await electrochemical characterization as with SCAD. The
objective of this chapter is to show how the midpoint potential of short-chain acyl-
CoA dehydrogenase was obtained in order to verify the values previously
determined with the Massey method.
Materials and Methods
Rat SCAD was expressed and purified from E. Coli as described in Chapter
2 of this thesis. Teflon for the electrochemical cell was purchased from Regal
Plastics, Denver, Colorado. Gold wire and foil and silver wire were purchased from
Alfa-Aesar, Mass. Platinum wire was purchased from The Wilkinson Company,
Agoura Hills, California. Pyrolitic graphite was obtained from Union Carbide. All
reagents were purchased from commercial sources and were of reagent grade or
better, and used without further purification.
Design of the Electrochemical Cell
The design of an electrochemical cell for the determination of the SCAD
midpoint potential took several limitations into consideration. The availability of
65


analyte for such a measurement was restricted to extremely small volumes as
purification of large amounts of protein was costly in time and materials. A cell that
accommodates such volumes was not widely available commercially and so was,
therefore, constructed in the laboratory. While minimizing the cell volume was
essential, the cell also had to be sufficient in size to accommodate three-electrodes.
Electrodes available in this laboratory would have required a much larger sample
size than was allowed in these experiments. Therefore, electrodes were also
constructed and sized to function in a small volume cell.
In addition, all other requirements for any working electrochemical cell had
to be met by the cell design. The material from which a cell is made had to be non-
reactive, readily available, and easily machined. Reference and auxiliary electrodes
had to be separated from analyte solution normally accomplished by use of a porous
frit. In order to run experiments under anaerobic conditions, there also had to be an
inlet and an exhaust vent built into the cell to allow a stream of nitrogen to blanket
the solution.
66


Construction of the Electrochemical Cell and Electrodes
The electrochemical cell designed to accommodate 300-500 pL of solution
is pictured in Figure 4.1. The cell was constructed using a % Teflon rod measuring
approximately 3 inches in length as starting material. A hole, 27/64 X 1 Vz\ was
drilled at one end of the rod. The cell was then tapped to accommodate a 14-13
coarse Nylon screw. To insert electrodes, three holes were drilled lengthwise
through the screw. The hole for the reference electrode required a 7/64 bit with the
auxiliary and working electrodes requiring 11/64 and 1/16 bits, respectively. The
hole for the reference electrode was purposely made larger than needed so that it
could also serve as the vent. A final hole was drilled through the side of the rod at a
45 angle away from the cell opening to accommodate the tube fitting for the gas
inlet. The gas inlet was positioned just below the screw threads.
A Ag/AgCl reference was constructed by placing 1/8 diameter X 1/8
Vycor rod into the end of a 3 length of heat shrink tubing. The Vycor served as the
frit. A heat gun was used to shrink the tubing around the Vycor and to shrink the
remaining tubing to allow it to fit through the 764 diameter hole size drilled
through the top of the cell cap. The tubing was then filled with 1M KC1 and a Ag
wire plated with AgCl was placed through the open end. A platinum auxiliary
electrode was constructed in the same manner using the Vycor and heat shrink
67


3/4'
1/2"
Cell-
side view
Nylon screw-
side view
Figure 4.1. Electrochemical cell used for determination of Em of SCAD. The
cell was constructed of Teflon and designed to accommodate 300-500 jj.1 of solution.
A nylon screw was used to contain the samples. Top and side views are shown.
Drilling sizes for electrodes and vent are also as shown.
68


tubing. The diameter of the tubing above the frit was not additionally decreased so
that solution could easily be removed and re-filled. The carbon working electrode
was formed by sanding a piece of carbon from a larger block into a rod shape of
approximately 1/8 diameter and V in length. This piece was then placed in a 1
length of heat shrink tubing and secured at one end of the tubing with a heat gun
leaving approximately 1/8 of the carbon exposed. Epoxy was then used to seal the
edge where the carbon met the tubing. A small drop of mercury was placed in the
open end of the tubing and a gold wire was used as the conductor. The gold
electrode used was constructed using a gold flag approximately V square. Three
holes were drilled near one end of the flag through which a gold wire was threaded.
Cyclic Voltammetry
The electrochemical cell was first cleaned by treatment with chromic acid for
at least 10 minutes and rinsed thoroughly using pyrolitically distilled water (PDW)
(Conway, 1973). Other components of the cell including the screw top and fittings
were soaked in base bath (1 L 95% ethanol, 100 g KOH, 100 ml H20) for 10
minutes and also rinsed with PDW. Reference and auxiliary electrodes were rinsed
with PDW before cell assembly. The gold electrode was soaked in warm chromic
acid for 10 minutes and rinsed. It was then placed in an H-cell in 1 M sulfuric acid
69


and cleaned electrochemically by cycling between -.35 V and 1.5 V for ten minutes.
If a clean scan could not be obtained by electrochemical cleaning, the electrode was
heated to incandescence in a natural gas/air flame. After obtaining a clean scan in
sulfuric acid, the electrode was rinsed and ready for use in the micro-cell. The
working carbon electrode was activated by polishing with alumina on a glass plate
and then rinsed well with PDW. The auxiliary electrode was filled with electrolyte
and all electrodes were inserted through their respective holes in the screw cap.
Electrode wires were secured in place using Teflon tape. After placing 300-500 pL
of sample into the cell, the cap was screwed in place and nitrogen line attached.
Nitrogen gas was blanketed over the solution for 10 minutes prior to
scanning. All scans were run at room temperature using a Cypress Model Omni 90
potentiostat. Data were recorded using a BioAnalytical Systems Model RXY
Recorder. The scan range used for the protein potential determinations was 0.000 V
to-0.600 V.
Determination of the Midpoint Potential. Prior to any electrochemical
experiment, SCAD was dialyzed against 10 mM Hepes, pH 7.0 overnight. After
dialysis, protein was concentrated in an Amicon Centricon 10 concentrator to 60
pM. The midpoint potential was determined using cyclic voltammetry as described
70


above. Blank scans were run using buffer only. The scan rate used was 100 mV/s.
Midpoint potentials were calculated by hand using the average of the anodic and
cathodic peak potentials and are reported versus the standard hydrogen electrode
(SHE).
Scan Rate Dependence. Scan rate dependence was determined using cyclic
voltammetry and 60 pM SCAD in 10 mM Hepes, pH 7.0 as described above. Scans
were run at rates of 25, 50, 100, and 200 mV/s. The working electrode was cleaned
prior to each scan to remove any buildup of adsorbed species. The peak current was
measured by hand with subtraction of background for each scan rate. Plots were
then made of peak height versus scan rate and peak height versus the square root of
scan rate. The plots were then examined for linearity to determine whether the
species was in the dissolved or adsorbed state. Linearity of the peak height versus
scan rate plot would indicate a dissolved species while linearity of the peak height
versus the square root of scan rate would indicate a species has adsorbed onto the
electrode.
71


Results and Discussion
Determination of Cell Function
The cyclic voltammogram for Potassium Ferrocyanide is shown in Figure
4.2. This scan clearly shows the sharp peaks expected for an ideal compound. The
midpoint potential calculated from the scan is 0.580 V vs Ag/AgCl. This value is in
agreement with reported values of 0.583 V vs Ag/AgCl (Rieger, 1994). Obtaining a
scan for a well characterized analyte of known midpoint potential was necessary in
confirming the proper functioning of the cell and electrodes constructed for the
SCAD experiments. With an accurate determination of the midpoint potential of
potassium ferrocyanide, the cell could then be used for unknown potential
determinations with Confidence-
Determination of the Midpoint Potential of SCAD
There is no universal material used for the working electrode which produces
notable peaks for every species. Therefore, when, working with an
electrochemically uncharacterized compound, the first obstacle is to identify a
material from which a reasonable scan can be obtained. The process not only entails
looking to the literature for previous reports on similar species but also trial and
72


1 /pA
Figure 4.2. Cyclic voltammogram of potassium ferrocyanide. The scan of
7 X lO"4 M potassium ferrocyanide was performed in 1 M KC1 on a clean gold
electrode. The scan rate used was 100 mV/s.
73


error. For the determination of SCADs midpoint potential, many electrode surfaces
were tried with unsuccessful results.
A most recent report on SCADs electron acceptor ETF showed it to adsorb
and give a well-defined scan on idium oxide electrodes (Salazar, 1998). Since this
proteins electroactive center is also flavin, it seemed reasonable to attempt using
indium oxide as the working electrode material with SCAD. Initial scans of SCAD
on idium oxide led to what appeared to be small peaks. After increasing the
sensitivity of the experiment by using a differential pulse technique, the peaks were
deemed to be due to buffer and not reactivity of the protein.
Upon referring to further literature reports, attempts were made at using a
gold electrode surface (Barker and Hill, 1988; Martin et ah, 1997; Rivera et al.,
1994). The scan on bare gold showed no peaks for SCAD. Another technique used
widely with proteins is to modify a gold surface with compounds which render that
surface charged. The charged surface can then attract an oppositely charged region
of the protein. Proper alignment of the proteins electroactive center with the
electrode will then allow electron transfer. Such electrode modification was
attempted using 3-mercaptopropionic acid (MPA) or cysteimine leaving a negative
or positive charge, respectively, on the surface. Neither attempt successfully
produced a scan for SCAD. In addition to surface modification, other promoters
74


may be used in an analyte solution. These promoters may be ions which would
assist in the diffusion of protein towards the electrode surface (Rivera, 1994).
Magnesium chloride was added to a solution of SCAD and tries with both modified
surfaces still showed no peaks. A final attempt with gold was made using a poly-
lysine promoter as reported in the literature (Rivera, 1994). The lysine with its
charged polar side chain could then attract oppositely charged side chains within the
protein structure. Since SCAD is known to be acidic with a pi of approximately 5,
this also seemed a reasonable method. However, no scan was obtained using the
poly-lysine alone or in conjunction with the MgCl2 promoter. This environment
seemed especially harsh as denaturation of protein was noted.
Finally, an attempt was made using a carbon electrode without a promoter
(Barber, 1997). The scan obtained is shown in Figure 4.3. This scan shows clear
peaks from which a midpoint potential was calculated to be -.240 V. Since the
potential of other members of the dehydrogenase family are considerably higher,
this determination was met with skepticism. Upon running another blank after the
initial scan was obtained, it appeared that the peaks were still present. Further
cleaning of the electrode led to a decrease in these peaks. To alleviate any concern
that the peaks were not due to SCAD reactivity, a new electrode was constructed.
75


-.8 -.6 -.4 -.2 0 .2
E / V (Ag/AgCI)
Figure 4.3. Cyclic voltammogram of SCAD. The scan was performed using 60
pM SCAD in 10 mM Hepes, pH 7.0 on a carbon electrode. The scan rate used was
100 mV/s.
76


After a blank was run with no appearance of peaks, the SCAD was scanned again.
Peaks were again visible and a midpoint potential of -.260 V was calculated
indicating that the proper electrode surface had been found for the determination of
SCAD potential.
The shape of a voltammogram gives other indications of the nature of the
SCAD system. The large background or vertical separation of the peaks seen is due
to the high sensitivity with which the scan was made. Background in cyclic
voltammetry, although partially caused by residual currents from minute quantities
of contaminants in the system, arises due to the existence of capacitive current.
Capacitive current is due to the double-layer charging at an electrode surface upon
application of a potential. Although a transient portion of this current disappears
quickly, a linear portion exists throughout a cyclic voltammetry scan. This
capacitive current produces the vertical separation above and below zero on the
voltammogram.
As sensitivity is increased on the instrumentation, the capacitive current
response is also increased. In the scan shown for SCAD, a further increase in
sensitivity would have pushed the background capacitive current and, therefore,
current peaks off the scale. The protein used was concentrated to a point just high
enough to see reliable peaks over background and so was the lowest concentration
77


possible for obtaining the scan. Even though the protein is retained after cyclic
voltammetry is performed, significant amounts of protein are lost in the dialysis and
concentration steps required in preparation. Therefore, the least amount of protein
necessary for a well-defined scan was used to avoid waste. It should also be noted
that the relatively large protein concentration would not have been possible if it
werent for the small volume cell constructed for use in the experiment.
The separation of peaks on a voltammogram is due to the necessary
application of an overpotential to convert oxidized species to reduced and vice
versa. Overpotential is a measure of the polarization or departure of electrode
potential from the Nemstian value. Polarization effects are due significantly to
electron transfer kinetics. The peak separation for SCAD is 145 mV. This value is
greater than that of 29.5 mV which would be expected of a completely reversible 2-
elecron transferring system. Such a quasi-reversible system indicates the existence
of an activation barrier which must be overcome by application of potential. This
characteristic was quantified by the kinetic studies presented in Chapter 2 of this
report and are again shown by the voltammogram.
The overall slant of the SCAD voltammogram is due to the ohmic potential
or iR drop of the electrochemical cell. That is, a certain driving force (potential) is
required to overcome the resistance of ions in solution towards movement to the
78


electrodes. The net effect in the case of SCAD is to increase the amount of potential
required to see Faradaic current. The iR drop can be minimized in a cell by
increasing salt concentrations, increasing permeability of frits, using mobile ions,
and decreasing the distance between the electrodes. For the micro-cell, the electrode
distance was very minimal as was required by design. The salt concentration was
not increased further due to rapid denaturation of the protein at high salt
concentrations. Other factors such as the permeability of the frit were determined
by the availability of materials. Overall, the existence of the ohmic potential caused
by the requirements of the experiment did not impede obtaining reliable data.
Scan Rate Dependence
The next step in ensuring that the midpoint potential obtained by cyclic
voltammetry could be compared to that of the spectroscopically obtained value
discussed in Chapter 3 was to determine the scan rate dependence of the peak
current height. The plot of peak current versus square root of the scan rate is shown
in Figure 4.4. The plot is linear with a correlation coefficient of 0.997. A plot of
peak current versus scan rate dependence shows less linearity. The equation
representing peak current for an adsorbed species is
79


Figure 4.4. Plot to determine the scan rate dependence for SCAD. Cyclic
voltammogram scans were performed using 60 pM SCAD in 10 mM Hepes, pH 7.0
at various scan rates. The electrode was cleaned between scans. Peak currents were
plotted versus the square root of the scan rate. Scan rates used were 25, 50, 100, and
200 mV/s. Linearity indicates a dissolved species scan.
80


(4.1)
in = n2 F2 A T* v 0
RT (1 + 0)2
where ip is peak current, n is number of electrons transferred, A is electrode surface
area, T* and 0 are measures of surface concentrations, R is the gas constant, T is
temperature in Kelvin, and v is the scan rate. This equation shows peak current, ip,
directly proportional to scan rate, v. On the other hand, the equation representing
the peak current for a dissolved species is
ip = (2.69 X 105) n3/2 A D1/2 v1/2 C* (4.2)
where ip is peak current, n is number of electrons transferred, A is the electrode area,
D is the diffusion constant for the species, v is the scan rate, and C* is the bulk
species concentration. In this equation, peak current, ip, is directly proportional to
the square root of the scan rate, v1/2. Since all other factors in both equations are
constant for an experiment, a linear plot for either ip vs. v or ip vs. vm indicates the
state of the electroactive species. For SCAD, the plot indicates that the scan
obtained was due dissolved SCAD and not SCAD which might have adsorbed to the
electrode surface. Since adsorption could shift the calculated midpoint potential, it
is necessary to rule out this possibility as an explanation for the difference in values
calculated from the two different methods used in this thesis.
81


CHAPTER 5
GENERAL DISCUSSION
The results put forth in this thesis provide an alternate method of purifying
rat short-chain acyl-CoA dehydrogenase from E. Coli. The data shows the protocol
including DEAE gel exclusion chromatography, hydroxylapatite column
chromatography, and S-200 Sephacryl gel filtration chromatography produces
protein in good yield and of a higher purity than what has currently been published.
The mechanistic data support the previously proposed mechanism for other
dehydrogenases. SCAD is reduced after the removal of two-electrons from its
substrate forming the charge-transfer complex. The semiquinone species is formed
although to a lesser extent than some of the dehydrogenases. This suggests the
dehydrogenases have different primary structures near the FAD. These amino acids
stabilize the semiquionone and so the amount formed depends on the particular side
chains present. SCAD then reduces its electron acceptor, ETF, in two one-electron
reduction steps. The charge-transfer complex makes this possible by
thermodynamic stabilization of the intermediate. Kinetic data found in this work
was in line with previous reports and is comparable to the other dehydrogenases as
well.
82


The main goal of this research was to determine the midpoint potential of rat
SCAD. With the initial characterization complete, the expectation was the protein
would behave much like its family members with a potential near -.100 V. Upon
initiating the potential experiments, however, it was clear that the similarities may
not apply to midpoint potentials. The first experiments using the xanthine/xanthine
oxidase reduction method of Massey clearly showed a protein with a potential more
negative than expected. The dyes which were readily available and normally used
for such determinations were not the best candidates for this protein. Other dyes
and methods may provide a more accurate determination.
Although numerous dyes were used in attempts to more closely approach
equilibrium with the Massey method, little success was met. Dyes in the potential
range needed, around -.200 V, either had high spectral interference or seemed not to
equilibrate with the protein. The dye, indigo carmine, showed reproducible results
and is one which has been commonly used with other flavoproteins. A midpoint
potential was calculated to be -.180 V. As discussed previously, this dyes potential
(E0 = -.121 V) is out of the recommended range for use in the experiment.
However, with the complications that exist with this protein, it provides the best
approximation for SCADs potential using this method.
The method of determining midpoint potentials using cyclic voltammetry
has been used recently with proteins like SCAD. After determining proper
83


conditions, this method provides a quick determination through direct electron
transfer between the working electrode and the protein. Thus, the equilibration or
interference with a redox dye is not an issue. A cell was constructed for this work
so that protein volume was minimized. Working electrodes were tried including
indium oxide and gold as well as modifiers with gold. A carbon electrode was
found to provide an excellent signal with SCAD. The potential calculated from this
method was -.240 V, a more negative and somewhat different value than the Massey
method.
One question regarding the data is why these two methods provide different
values for midpoint potential. A possibility, which has already been discussed, is
that the Massey method, without an ideal dye, has provided an approximation.
Thus, a different dye may produce a value more similar to that determined by cyclic
voltammetry. Another possibility is that the two methods inherently provide
different results based on their mechanisms. The Massey method makes use of the
redox mediator, benzyl viologen. Electrons are passed through this easily reducible
and highly diffusable compound which then reduces both dye and protein in
equilibrium. Therefore, benzyl viologen is entering the protein binding pocket and
reducing FAD directly.
Using cyclic voltammetry, the protein is reduced via an applied potential at
an electrode. The protein approaches the electrode by diffusion and if the force
84


(potential) is strong enough, electrons are transferred from the electrode to the
protein. The exact path of the electrons in this case is unknown. Electrons may
have to travel through a large amount of relative space to reach the FAD buried
inside the protein. Such a difference in the mechanism between the two methods
may result in different apparent potentials for SCAD. The two methods used in this
research, therefore, provide values for SCADs midpoint potential which may both
be valid for the given technique. Since the exact mechanisms are not known
definitively, neither can be discounted.
The midpoint potential for SCAD, no matter how obtained, is lower than
expected based on previous determinations of related proteins. Thus, the other
question left from this research is why the potential would be more negative.
Possibilities include that the potential is low due to a physiological role. A lower
potential indicates that the reaction catalyzed would happen more slowly due to a
higher AG than if the potential were higher. Therefore, the reaction may not
proceed as easily without high concentrations of substrate pushing the equilibrium
forward. In this way, the protein may play a regulatory role within the oxidation
pathway.
Another possibility is that the potential measured here has been artificially
lowered by some treatment of the protein. As was discussed previously, the protein
seemed fairly unstable and likely to precipitate out of solution. The proposed
85


reasoning for the instability was the removal of the CoA-persulfide ligand during
purification. Perhaps this removal caused a structural change within the protein
leading to an apparent low potential. Thus, the proteins potential could more
closely resemble those of the other dehydrogenases.
Without further study, it is impossible to say which of these hypotheses, if
any, are true. As is often the case, this report has left many questions unanswered.
With the unexpected results, however, interest is spurred to look at rat SCAD more
closely. As with many enzyme families, particular members are studied more than
others and it is assumed that the data will be consistent across the family. The data
in this report disproves this theory and shows that unexpected results are valuable.
In addition, future research may give insight into the potentials determined here.
Perhaps purification or potential determination methods need to be modified or
perhaps SCAD possesses an important, yet unknown physiological role.
86


REFERENCES
Armstrong, J. McD (1964)Biochim. Biophys. Acta., 86,194-197.
Barber, M.J., Trimboli, A.J., Nomikos, S., and Smith, E.T. (1997) Arch. Biochem.
Biophys., 345, 88-96.
Barker, P.D., and Hill, H.A. (1988) Prog. Clin. Biol. Res., 274,419-433.
Battaile, K.P., Mohsen, A.A. and Vockley, J. (1996) Biochemistry., 35, 15356-
15363.
Beinert, H. (1957) J. Biol. Chem., 225,465-478.
Conway, B.E., Angerstein-Kozlowska, H., Sharp, W.B.A. and Criddle, E.E. (1973)
Anal. Chem., 45, 1331.
Crane, F.L. and Beinert, H. (1956) J. Biol. Chem., 218, 717-731.
Draper, R.D. and Ingraham, L.L. (1968) Arch. Biochem. Biophys., 125, 802-808.
Engel, P.C. and Massey, V. (1971) Biochem. J., 125, 879-887.
Fink, C.W., Stankovich, M.T. and Soltysik, S. (1986) Biochemistry, 25, 6637-6643.
Frerman, F.E., Miziorko, H.M. and Beckman, J.D. (1980) J. Biol. Chem., 255,
11192-11198.
Ghisla, S. and Massey, V. (1989) Eur. J. Biochem., 181,1-17.
Gustafson, W.G., Feinberg, B.A. and McFarland, J.T. (1986) J. Biol. Chem., 261,
7733-7741.
Hall, C.L., Lambet, D.L. and Kamin, H. (1979) J. Biol. Chem., 254,2023-2031.
Ikeda, Y., Okamura-Ikeda, K. and Tanaka, K. (1985) J. Biol. Chem., 260(2), 1311-
1325.
87


Lerm, N.D., Stankovich, M.T. and Liu, H. (1990) Biochemistry, 29,3709-3715.
Martin, T.D., Monheit, S.A., Niichel, R.J., Peterson, S.C., Campbell, C.H. and
Zapien, D.C. (1997) J. Electroanal. Chem., 420,279-290.
Massey, V. (1991) In Curti, B., Ronchi, S. and Zanetti, G. (eds) Flavins and
Flavoproteins 1990. Walter de Gruyter and Co., New York, pp. 59-66.
Massey, V. and Hemmerich, P. (1978) Biochemistry,. 17(1), 9-16.
Massey, V. and Ghisla, S. {\97 A) Ann. N. Y. Acad. Sci., 227,446-465.
McKean, M.D., Beckman, J.D. and Frerman, F.E. (1983) J. Biol. Chem., 258, 1866-
1870.
Muller, F., Brustlein, M., Hemmerich, P., Massey, V. and Walker, W.H. (1972) Eur
JBiochem., 25, 573-580.
Rieger, P.H. (1994) Electrochemistry Second Edition. Chapman & Hall, New York,
p. 440.
Rivera, M., Wells, M.A. and Walker, F.A. (1994) Biochemistry, 33,2161-2170.
Salazar, D. (1998) Ph.D. dissertation, University of Colorado Health Sciences
Center.
Schmidt, J., Reinsch, J. and McFarland, J.T. (1981) J. Biol. Chem., 256,11667-
11670.
Seager, S.L. and Slabaugh, M.R. (1987) Chemistry for Today General, Organic,
and Biochemistry. West Publishing Company, St. Paul, Minnesota, pp. 750-754.
Shaw, L., and Engel, P.C. (1984) J. Biochem., 218, 511-520.
Shaw, L., and Engel, P.C. (1987) Biochim. Biophys. Acta., 919,171-174.
Steyn-Parve, E.P. and Beinert, H. (1958) J. Biol. Chem., 233, 843-852.
88


Full Text

PAGE 1

PURIFICATION, SPECTRAL AND KINETIC CHARACTERIZATION, AND DETERMINATION OF MIDPOINT POTENTIAL OF SHORT-CHAIN ACYL-COA DEHYDROGENASE by Tiffany Hays B.S., University of Colorado at Denver, 1997 A thesis submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Master of Science Chemistry 1999

PAGE 2

This thesis for the Master of Science degree by Tiffany Hays has been approved by Donald C. Zapien 7 Frank E. Frerman Date

PAGE 3

Hays, Tiffany (M.S., Chemistry) Purification, Spectral and Kinetic Characterization, and Determination of Midpoint Potential of Rat Short-Chain Acyl-CoA Dehydrogenase Thesis directed by Assistant Professor Donald C. Zapien ABSTRACT Short-chain acyl-CoA dehydrogenase (SCAD) is a member of a family of mitochondrial enzymes involved in the of fatty_ acids. In the reaction catalyzed by SCAD, two electrons are removed from a fatty acyl-CoA substrate introducing a trans-a, double bond. The electrons are passed through the protein's electroactive center, flavin adenine dinucleotide (FAD), and then are shuttled into a chain of redox enzymes which feed the electron transport chain where ATP is generated. As the only source of useful energy for living organisms, oxidationreduction reactions such as the SCAD catalysis are of biological significance. This thesis presents a purification method for rat SCAD that has been modified from previous protocols. Results show a method including DEAE gel exclusion chromatography, hydroxylapatite column chromatography, and Sephacryl S-200 gel filtration chromatography produce a protein in good yield and of increased purity from previously published methods. Kinetic and mechanistic data also lll

PAGE 4

support the purification of a protein which is of comparable quality, if not better, than those used in previous publication. An important research question with any redox protein is that of its standard redox potential (E0'). After determining the E0 for a wild-type enzyme, of the potential may show the effect of specific amino acid mutations near the redox center. The E0 of SCAD was determined in this work using two different methods. The widely accepted method utilizing a reduction dye led to a E0'= -.182 V. An unexplored method of cyclic voltammetry using a carbon electrode led to a E0'= -.240 V. This method is quicker and does rely on achieving equilibrium between a dye and the protein for the potential determination. The difference between the two values leads to questions regarding the exact mechanism by which electrons flow in the two methods. With both values lower than expected, questions remain regarding whether the more negative potential is a result of a physiological role or if it has been artificially lowered. The unexpected results of this work will lead to further interest in SCAD's important biochemical role. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. Signed Donald C. Zap"en iv

PAGE 5

ACKNOWLEDGEMENT My thanks to my advisors, Frank Frerman and Don Zapien, for their support and understanding. My thanks to the following: Greg DeGala, Tim Dwyer, Michelle Muller, Mai Pham, and Denise Salazar for their help and guidance. My thanks to my committee members Ellen Levy and Larry Anderson for their time and knowledge. I would finally like to thank all those who have supported me outside of school without whom this would not be possible; Dave Hays, Brendan Hays, Leslie and Bob Reichardt, Stephanie and Rob Hays, John and. Cheryl Claybough, and Alisha Cambria.

PAGE 6

CONTENTS Figures.............................................................................. x Tables............................................................................... xu CHAPTER 1. INTRODUCTION............................................................. 1 Fatty Acid Oxidation...................................................... 2 Acyl-CoA Dehydrogenases ............................................... 5 Flavin as the Coenzyme.......................................... 7 Acyl-CoA Dehydrogenase Mechanism........................ 9 Structure of the Acyl-CoA Dehydrogenases .................. 15 Oxidation-Reduction Reactions.......................................... 16 Cyclic Voltammetry..................... .. . . . . . . . . . . . 20 Conclusions ................................................................. 21 2. PURIFICATION, SPECTRAL AND KINETIC CHARACTERIZATION OF RAT SHORT-CHAIN ACYL-COA DEHYDROGENASE......................................................... 23 VI

PAGE 7

Introduction ................................................................. 23 Materials and Methods ................................................... 24 Purification of SCAD........................................... 25 Determination of SCAD Enzyme Activity .................... 28 Determination ofExtinction Coefficients for SCAD and ChargeTransfer Complex ....................................... 28 Steady-State Kinetic Assays ..................................... 30 Determination of SCAD Mechanism .......................... 31 Results and Discussion ................................................... 32 Purification........................................................ 32 Spectral Properties ................................................ 37 KineticProperties ................................................. 46 3. DETERMINATION OF THE MIDPOINT POTENTIAL OF RAT SHORT-CHAIN ACYL-COA DEHYDROGENASE USING THE XANTHINE/XANTHINE OXIDASE REDUCTION METHOD ...................................................................... 48 Introduction ................................................................. 48 Materials and Methods .................................................... 49 Determination of Extinction Coefficient for Reduced SCAD ............................................................... 49 vii

PAGE 8

Re-crystallization of Redox Dye ............................... 50 Determination of Xanthine Oxidase Enzyme Activity ............................................................ 50 Determination of SCAD Midpoint Potential................. 51 Results and Discussion ................................................... 53 Determination of Extinction Coefficient for Reduced SCAD ............................................................. 53 Determination of Xanthine Oxidase Enzyme Activity ............................................................ 55 Determination of SCAD Midpoint Potential ............................................................ 55 4. DETERMINATION OF THE MIDPOINT POTENTIAL FOR SHORT CHAIN ACYL-COA DEHYDROGENASE BY CYCLIC VOLTAMMETRY ................................................. 64 Introduction ................................................................. 64 Materials and Methods ................................................... 65 Design of the Electrochemical Cell ............................ 65 Construction of the Electrochemical Cell ................................................................. 67 Cyclic Voltammetry .............................................. 69 Results and Discussion ................................................... 72 Determination of Cell Function................................ 72 viii

PAGE 9

Determination of Midpoint Potential of SCAD .............................................................. 72 Scan Rate Dependence.................................................. 79 5. GENERAL DISCUSSION ..................................................... 82 REFERENCES ............................................................................... 87 ix

PAGE 10

FIGURES Figure 1.1 Reactions of the pathway .................................................. 4 1.2 Flavin adenine dinucleotide (FAD) ..................................................... 8 1.3 Oxidation states of flavin ................................................................ I 0 1.4 Overall enzyme sequence ............................................................... II 1.5 Catalytic process by which ACD is bound and regenerated ................. : ...... 13 1.6 Mechanism of dehydrogenation ........................................................ 14 1. 7 Structure of the MCAD subunit ........................................................ 17 1.8 Arrangement ofthe MCAD binding pocket .............. ............................ 18 2.1 Elution of SCAD from Hydroxylapatite column .................................... 36 2.2 Absorbance spectrum of purified rat SCAD ......................................... 38 2.3 Aerobic reduction of SCAD with butyryl-CoA ..................................... 40 2.4 Anaerobic titration of SCAD using sodium dithionite with product present .... 42 2.5 Anaerobic titration of SCAD using sodium dithionite without product present .................................................................................... 43 2.6 Anaerobic titration of electron transfer flavoprotein using SCAD ................ 44 2. 7 Double reciprocal (LineweaverBurke) plot of steady -state kinetic assay ....... 4 7 3.1 Photoreduction of SCAD using a deazaflavin catalyst .............................. 54 3.2 Plot to determine the activity of xanthine oxidase ................................... 56 3.3 Anaerobic reduction findigo carmine using xanthine/xanthine oxidase ......... 58 3.4 Anaerobic reduction of SCAD using xanthine/xanthine oxidase .................. 59 3.5 Anaerobic reduction of SCAD and indigo carmine using xanthine/xanthine oxidase ..................................................................................... 61 3.6 Minnaert plot of reduction of SCAD and indigo carmine data .................... 62 4.1 Electrochemical cell used for determination of midpoint potential of SCAD ... 68 X

PAGE 11

4.2 Cyclic voltammogram of potassium ferrocyanide .................................. 73 4.3 Cyclic voltammogram ofSCAD ....................................................... 76 4.4 Plot to determine the scan rate dependence for SCAD .............................. 80 xi

PAGE 12

TABLES Table 2.1 Purification of rat SCAD ............................................................ 33 xii

PAGE 13

CHAPTER 1 INTRODUCTION Living organisms perform mechanical work, electrical work, and synthetic work in order to acquire energy from their environment. The only source of biologically useful energy is derived from oxidation-reduction reactions. In aerobic metabolism, animals consume organic molecules and carbohydrates which are oxidized releasing the necessary energy to perform work. In some bacteria, anaerobic metabolism generates energy from oxidation of relatively reduced nitrogen and sulfur compounds yielding products such as sulfate or nitrate. Plants use photosynthesis where trapped light energy from the sun is converted via reduction of carbon dioxide and oxidation of water to biological energy in the form of ATP. Oxidation-reduction reactions, therefore, represent one ofthe most important biochemical processes necessary for life. The common energy currency and fmal product of oxidative breakdown for living organisms is the high energy molecule adenosine triphosphate (ATP). ATP hydrolysis is the essential driving force for many biological reactions which would otherwise not occur due to positive Gibbs free energy values. A TP is produced as 1

PAGE 14

metabolites are broken down and enter the electron-transport chain whose products are water and ATP. One pathway which supplies the electron-transport chain is the !)-oxidation of fatty acids. A family of enzymes involved in this pathway extracts electrons from activated fatty acids using them to eventually synthesize A TP. This family of enzymes includes short-chain acyl-CoA dehydrogenase (SCAD). Although some enzymes in this family have been studied in detail, this particular member seems to have been less emphasized in research. Its operation in such a vital pathway for sustaining life warrants the further investigation put forth in this report. The studies done here open the door to the more important work in understanding how a malfunctioning enzyme may ultimately be handled in a clinical setting. Fatty Acid Oxidation Lipids are oxidized to C02 and H20 yielding significantly larger amounts of energy than their counterparts, proteins and carbohydrates. After a complicated pathway of adsorption into cells, fatty acids must be activated before they are oxidized. Activation is ATP-dependent and forms the fatty acyl-coenzyme A through an acylation reaction. The reaction catalyzed by the acyl-CoA synthetases or thiokinases according to the length of the fatty acid chain follows. 2

PAGE 15

Fatty acid+ CoA + ATP -7 acyl-CoA +AMP+ PPi (1.1) The activation process occurs in the cytosol after which the fatty acyl-CoA must be transported across the mitochondrial membrane so that oxidation can occur. For transport, the Co A portion of the fat is exchanged with a carnitine molecule in the cytosol and then transported by specific camitine carriers across the membrane. A new CoA from inside the mitochondria then replaces the camitine to reform the fatty acyl-CoA (Voet and Voet, 1990). Oxidation of the fatty acyl-CoA molecule entails the four reactions depicted in Figure 1.1. The first reaction, which will be discussed in detail later in this chapter, is catalyzed by members ofthe acyl-CoA dehydrogenase family. The dehydrogenation reaction introduces a trans-a, double bond resulting in the trans enoyl-CoA product. The second reaction, catalyzed by the enoyl-CoA hydratase, hydrates the double bond yielding the hydroxyacyl-CoA product. The third reaction is an NAD+-linked dehydrogenation at carbon yielding the ketoacyl-CoA product. Finally, the bond between the a carbons is cleaved in a thiolysis reaction catalyzed by thiolase. The result of these four reactions is the release of an acetyl-CoA and the formation of a new fatty acyl-CoA that is now two carbons shorter which can re-enter the cycle for further breakdown (Seager and Slabaugh, 1987). 3

PAGE 16

H H 0 I I II CH3-(CH2>,.C11C.;CSCoA I I H H Fatty Acyl-CoA -CoA \ 1 ydrogenase } ADH2_./ H 0 I II CH3-
PAGE 17

The first round of a fatty acid through the 13-oxidation pathway produces one acetyl-CoA as a product, one NADH molecule via the third reaction, and one F ADH2 molecule via the first reaction. The acetyl-CoA then enters the citric acid cycle where an additional F ADH2 molecule and NADH molecule are generated. Therefore, 13-oxidation functions as an energy generating pathway with 129 ATP being formed after the repeated cycles of just one 16 carbon fatty acid molecule (V oet and V oet, 1990). Acyl-CoA Dehydrogenases Enzymes are specialized proteins that function as biological catalysts in nearly all biochemical reactions occurring in living organisms. Acting under the identical laws of nature which govern all chemical reactions, enzymes have an increased capacity to speed reaction rates by lowering activation energy barriers. In addition, these reactions take place under chemically mild conditions of pH and temperature, have increased substrate specificity, and are under strict regulatory control. Without such characteristics, biological systems would be unable to extract the necessary energy from chemical reactions needed for life. The acyl-CoA dehydrogenases (ACD's) are one such family of enzymes which produce energy via the first step of fatty acid oxidation. The members of the 5

PAGE 18

family which participate in fatty-acid metabolism are short chain acyl-CoA dehydrogenase (SCAD), medium chain acyl-CoA dehydrogenase (MCAD), and long chain acyl-CoA dehydrogenase (LCAD). Though the enzymes are thought to have evolved perhaps from the same gene and are similar in many ways, they are identified and so named for their substrate chain length specificity. Each enzyme catalyzes the same oxidation-reduction reaction, a, 13-dehydrogenation, of its specific substrate. Though many enzymes are able to catalyze reactions such as acid-base reactions and formation of covalent bonds using only the functional groups of their side chains, oxidation-reduction reactions require the use of an enzyme cofactor. Examples of cofactors include metal ions and small organic molecules such as the heme. When the cofactor is a small organic molecule, it is referred to as a coenzyme. Such molecules are chemically altered during the catalyzed reaction and require regeneration via an additional catalytic reaction performed by another enzyme. Thus, the cycle is completed for the initial enzyme. 6

PAGE 19

Flavin as the Coenzyme The specific coenzyme involved in the reaction catalyzed by the acyl-CoA dehydrogenases is flavin adenine dinucleotide (FAD) and is shown in Figure 1.2. Any protein such as the ACD's containing FAD is known as a flavoprotein with a flavin defined as any compound containing the isoalloxazine ring system. Since most animals cannot synthesize the isoalloxazine system, it is made available through the diet in forms such as riboflavin (vitamin B2). The electroactive portion of FAD is the 7,8-dimethylisoalloxazine. The three rings of this planar system are designated the xylene, pyrazine, and pyrimidine rings seen from left to right in Figure 1.2. FAD is substituted at the N(l 0) position by a D-ribitol residue derived from the alcohol of the sugar D-ribose. Bound to the opposite end of the ribityl side chain, the adenosine component ofF AD is found. The functions of the side chain portions of the FAD are confined to the anchoring of the coenzyme into the active site of the protein with redox activity being confmed to the isoalloxazine system. Reduction ofF AD may take place via a two electron reduction to its fully reduced or hydroquinone state (F ADH2 ) or via two single electron reductions whereby the compound exists after half-reduction as the radical semiquinone (F ADH). The semiquinone form may itself exist in various protonation states. 7

PAGE 20

o-Ribitol N;:_N> 0 0 cH I 2 I_ I_ 0 H-C-OH 0 0 H H H-6-oH H H I HO OH H-C-OH I CH2 I N N 0 5 .?l WN ......_ Isoalloxazine N n H 0 Adenosine Figure 1.2. Flavin Adenine Dinucleotide (FAD). The standard numbering of the isoalloxazine ring system is noted. The isoalloxazine is substituted at N( 1 0) by the D-ribitol residue which binds the adenosine portion ofF AD. 8

PAGE 21

Comparing their absorption spectra with the spectrum of fully oxidized flavin identifies these states. The characteristic spectrum of fully oxidized flavin includes peaks near 436 nm and 373 nm. The protonation states are known as the anionic red form so named for the negatively charged species with a red shift seen near 370 nm, the neutral red species although without carrying a charge also has a red spectral shift, and the neutral blue form wherein a blue shift is seen in the spectrum near 560 nm (Draper and Ingraham, 1968; Muller et al., 1972). The various oxidation states of flavin are shown in Figure 1.3. In addition to the spectral changes, reduction of FAD is evident through the loss of the intense yellow color flavoproteins characteristically show in their native forms. Acyl-CoA Dehydrogenase Mechanism The overall enzyme sequence beginning with the fatty acid substrate of the ACD's is shown in Figure 1.4. In the mitochondria, the acyl-CoA substrate is converted to its enoyl-CoA product as the first step in !3-oxidation by the ACD's. The reduced ACD then passes its electrons onto its physiological electron acceptor electron transfer flavoprotein (ETF). Subsequently, electrons are passed from ETF to the membrane bound ETF dehydrogenase and on into the main respiratory chain via ubiquinone with a final outcome of production of ATP (Thorpe and Kim, 1995). 9

PAGE 22

leJr Semiquiononespartially (le-) reduced flavin I I H H+ ){N(yo __:.. ._...-N I NH 0 0 Anionic red Neutral red H Jl Hydroquinones fully (2e-) reduced flavin Anionic H+ IJ H :r::(1: H 0 Neutral I ){N('fo NH N H a Neutral blue Figure 1.3. Oxidation states of flavin. The three oxidation states of flavin are shown along with the various protonation states 10

PAGE 23

Fatty Acyi-CoA (substrate) ETFdh c.--Aco-ox coa-red ....._ ETF-ox __.) ETFdh __.) CoO-ox --ox Enoyi-CoA (product) Figure 1.4. Overall enzyme sequence. Electrons from the reduction of the fatty acyl-CoA are delivered to the electron transport chain via the acyl-CoA dehydrogenase (ACD) followed by electron transfer flavoprotein (ETF), followed by ETF-dehydrogenase (ETFdh) and finally coenzyme Q (CoQ). Oxidized (ox) and reduced (red) indicate the respective oxidation states of the enzymes. II

PAGE 24

The accepted catalytic process by which ACD is bound and reduced by substrate and then regenerated by ETF is illustrated in Figure 1.5. By this method, the acyl-CoA substrate is bound by oxidized enzyme forming the Michaelis complex. Electrons are then passed from substrate to enzyme leaving the enoyl product bound to the fully reduced flavin. Since the binding of the enoyl product in the active site is extremely tight (Steyn-Parve and Beinert, 1958), its release is only achieved after the ACD has been reoxidized by ETF. This reoxidation takes place in two subsequent one electron reductions of ETF. With product binding reduced by the process, the next substrate compound then easily displaces the product and the catalysis is repeated. The mechanism of dehydrogenation of the substrate is depicted in Figure 1.6. As shown, the catalytic amino acid base of the ACD abstracts an a-hydrogen from the fatty-acyl substrate. This step is favored because the a position has been activated by the Coenzyme A. As charge is pulled towards the Co A end of the molecule, the a position takes on a slight positive charge allowing the basic residue of the protein to abstract the hydrogen. The electrons remaining place a negative charge on C(3) of the substrate allowing the elimination a at this position as a hydride equivalent to N(5) position on FAD (Ghisla and Massey, 1989). 12

PAGE 25

REDUCTIVE HALF-REACTION FAD-ox + substrate FAD-ox substrate 2;!ed product OXIDATIVE HALF-REACTION FAD -ox + product Figure 1.5. Catalytic process by which ACD is bound and regenerated. ACD's are reduced in a two-electron reduction by substrate. Product remains bound until electrons are passed to ETF in tv, o subsequent one-electron reductions The regenerated protein then releases its product. 13

PAGE 26

R R I I N N N N /"yo -yo NH NH N H 0 0 SCoA SCoA H t.aHB-Figure 1.6. Mechanism of substrate dehydrogenation. Dehydrogenation occurs via the abstraction of the activated a-hydrogen by the active site glutamate. The carbanion then initiates the elimination of the P-hydrogen to the N(5) position of FAD. 14

PAGE 27

The catalytic base responsible for the proton extraction in the dehydrogenation mechanism has been identified in several species of the ACD's as a glutamic acid. In short-chain acyl-CoA dehydrogenase, the catalytic glutamate has been identified as Glu368 while in porcine MCAD it is Glu376 and in long-chain it can be found at Glu261 (Battaile, et al., 1996). The relative location although conservatively placed in the binding pocket of the protein plays a role in the substrate specificity of the enzyme. The exact location of the catalytic residue makes each protein more suited to act on differing lengths of fatty acid chains. Structure of the Acyl-CoA Dehydrogenases All of the members of the acyl-CoA dehydrogenase family which participate in fatty acid metabolism are homotetramers with each subunit non-covalently binding one FAD molecule. The quaternary structure and individual subunit size are consistent from enzyme to enzyme. The native molecular weight for short-chain is 160,000 while that for both medium and long-chain is 180,000 daltons. Individual subunit size only varies from short-chain's 41,000 to 45,000 daltons for medium and long-chain (Ikeda et al., 1985). Since medium-chain acyl-CoA dehydrogenase is the most studied of the enzyme family, its structure will be presented as a model for the other related 15

PAGE 28

enzymes. The structure of the MCAD subunit is shown in Figure 1. 7. There are three domains known as the N and C terminal and the sheet domains which make up each subunit of MCAD. Both theN and C terminal domains contain six alpha helices each with these domains separated by the sheet domain. The FAD binding site is within the subunit with the electroactive isoalloxazine ring portion buried at the interface of the domains. The binding site for the MCAD substrate is buried among the alpha helices of the N and C terminal domains. When bound, a substrate molecule will lie length wise in the binding pocket with the isoalloxazine ring flat against one side and the catalytic amino acid base directly on the other. Such a "sandwich" arrangement accommodates the flow of electrons from substrate to FAD (Thorpe and Kim, 1995). Figure 1.8 depicts this arrangement showing the MCAD active Glu 376, the substrate, and FAD. Oxidation-Reduction Reactions Oxidation-reduction (redox) reactions involve the transfer of electrons from electron donor to electron acceptor. These reactions may be divided into two half reactions or redox couples each representing the reduction of the electron acceptor or the oxidation of the electron acceptor. A unique property of redox reactions is that the two half reactions can be physically separated in what is known as an 16

PAGE 29

__ /' _Jo_ ... / ... .... -. \-. -Figure 1.7. Structure of the MCAD subunit. The ribbon diagram of a MCAD subunit is shown. FAD is shown in black. Picture drawn by Tim Dwyer using the Rasmol software. 17

PAGE 30

Figure 1.8. Arrangement of the MCAD binding pocket. The catalytic Glu 376 of MCAD is shown along with the crotonyl-CoA substrate (shown in black) and FAD. The diagram shows the aligrunent in the binding pocket. Picture drawn by Tim Dwyer using Rasmol. 18

PAGE 31

electrochemical cell. The free energy of the reaction occurring between the half cells can then be determined by measuring their voltage difference. In a redox reaction such as the one below (1.2) where B is the reductant and A is the oxidant, the free energy can be written as follows. (1.3) Under reversible conditions, Ll.G =-W el (1.4) where w el is the electrical work required to move electrons across an electrical potential and we1=nFE where F is the Faraday constant and E is the electrical potential difference. Substituting equation 1.5 into equation 1.4, Ll.G = n FE (1.5) (1.6) If equations 1.5 and 1.1 are then combined, the Nemst equation is generated (1.7) Where E is the redox potential, E0 is the standard redox potential, R is the Rydberg constant, T is temperature in Kelvin, n is the number of electrons transferred, and F 19

PAGE 32

is the Faraday constant. From this equation, it should be noted that a positive value forE indicates a spontaneous reaction (negative For a redox couple, E = Eo (e acceptor) -Eo (e donor) (1.8) So for the reaction to occur spontaneously (E is positive), the electron acceptor must have a more positive standard potential than the electron donor. A species with a more positive potential has a greater affinity for electrons and therefore will accept charge from any source with a lower affinity. Cyclic V oltarnmetry One method for determining the standard redox potential of a species is by cyclic voltammetry. Half-cells are arranged in an electrochemical cell, one of which containing the species of interest and the other containing a suitable reference solution for which the E0 is known. Initially, a potential is applied to the cell at which no reaction occurs at the working electrode. That is, the potential applied is high enough to sustain the oxidized species in the oxidized form. A potential ramp is then applied in the negative direction eventually causing the reduction of the species in question. This reduction provides a measurable current (electron flow) which increases as the maximum concentration of species is reduced. The current then begins to decrease creating an anodic peak as the amount of species 20

PAGE 33

approaching the electrode is limited by diffusion. At this point, the potential ramp is reversed and an identical process of oxidation takes place creating a cathodic peak. The standard redox potential, or midpoint potential, is the average potential between that of the anodic and cathodic peaks (Zapien, 1998). Other details of a species kinetics and electrochemical behavior can also be determined from more in-depth cyclic voltammetry experiments. Conclusions Although much is currently known about the acyl-CoA dehydrogenases and the purification and certain characterizations of the short-chain acyl-CoA dehydrogenase have been published, it is still one of the least studied members of the family. The standard redox potential for any mammalian SCAD has yet to be determined. This value will provide not only information about the native enzyme but will allow future experiments which could have broader applications in areas such has how electrons travel through complex systems like proteins. In order to produce reliable results for a potential, a purification method and confirmation of its success are necessary. The results of further investigation on that protein are then valid and accepted. Although current methods for determining the redox potential of flavoproteins provide reliable and accurate results, these methods 21

PAGE 34

may have flaws and new methods which would improve on current techniques are always desirable. The purpose of this research is to purify an active rat SCAD protein and characterize its spectral and kinetic properties. Two methods can then be used to determine its redox potential, a vital characteristic of any redox protein. 22

PAGE 35

CHAPTER2 PURIFICATION, SPECTRAL AND KINETIC PROPERTIES OF RAT SHORT-CHAIN ACYL-COA DEHYDROGENASE Introduction In-depth study of short-chain acyl-CoA dehydrogenase began when Shaw and Engel (1984) used ammonium sulfate fractionation and chromatographic techniques in one of the initial purifications. The enzyme when purified was present in two forms. A green form contains CoA-persulfide in the active site of the enzyme. This ligand is absent in the other yellow form. Shaw and Engel (1987) later characterized the enzyme's green form and outlined further purification to remove the ligand contaminant. Later, upon working with the bacterial counterpart enzyme to SCAD, Fink et al. (1986) discovered other impurities present after purification and again modified the protocol. After a purification technique has been employed, spectral and kinetic properties are determined to characterize a properly separated enzyme. Spectral properties play an important role in the study of many enzymes, especially flavin containing dehydrogenases, because of the existence of very characteristic patterns. Detection of a flavin containing protein and a measure of purity are readily

PAGE 36

determined from its spectrum. Other spectral assays provide information on the mechanism of catalysis. Kinetic parameters are also determined in a complete analysis of a purified protein. Such parameters allow comparison with related enzymes and further insight into the protein's physiological role. Complete characterization, both spectral and kinetic, is necessary when any modifications are made to a purification protocol. Spectral and kinetic parameters can then be compared with the available literature to validate the new purification method. Small changes in previous protocols were made in the purification of the SCAD used in the subsequent studies of this work. The objective of this chapter is to present this purification method and the full spectral and kinetic characterization of the enzyme ensuring the use of an adequately purified protein in further study. Materials and Methods Ultraviolet/visible spectroscopy was performed using a Shimadzu UV2401 spectrophotometer unless otherwise noted. All reagents were reagent grade or better and were obtained from commercial sources. Plasmid containing rat SCAD was the generous gift of Dr. Jerry Vockley, Mayo Clinic, Rochester, Minnesota. 24

PAGE 37

Purification of SCAD Inoculation with 1 0 mL of starter culture took place in 1 0 flasks containing 800 mL ofTryptone-Yeast Phosphate (TYP) media (16 giL tryptone, 16 giL yeast extract, 2.5 giL Potassium Phosphate, and 5 giL NaCl, brought to pH 6.8 with HCl), 100 J.LglmL ampicillin, and 500 J.Lg/mL IPTG. Cells were grown at 37 C overnight with vigorous aeration to an optical density at 600 run of approximately 8.5. Cells were harvested by centrifugation at 7500 rpm for 10 minutes, resuspended in phosphate buffered saline and placed in the centrifuge again at 8000 rpm for 20 minutes. The resulting cell pellet was then stored at -70 C. Cells were kept frozen no longer than 3 days before disruption. To disrupt the cells, the thawed pellet was resuspended in approximately 100 mL of 10 mM Tris-HCl, pH 7.5 and homogenized with a Potter-Elvehjem homogenizer. Cells were then passed through a French pressure cell twice at 1300 psi. The disrupted cells were placed in the centrifuge for one hour at 1 00,000 X g to remove cell debris. After ultracentrifugation, the supernatant was loaded onto a 373 mL DEAE gel exclusion chromatography column previously equilibrated with 10 mM Tris HCl, pH 7.5. The column was then washed with 400 mL of equilibration buffer, and the wash through was assayed for SCAD activity to assure the protein had successfully bound to the column. The protein was then eluted from the column 25

PAGE 38

with a 2 L gradient of0-300 mM NaCl in 10 mM Tris-HCl, pH 7.5. Fractions were collected and assayed for catalytic SCAD activity. Fractions containing peak activity were pooled and concentrated in an Amicon concentrator using a YM1 00 membrane. The concentrated protein solution was brought to 0.1 M Tris-HCl, pH 7.5 and degassed with 10 cycles of alternating argon and vacuum. After addition of 600 mg of sodium dithionite, 1 0 more degassing cycles were performed. The enzyme was then allowed to sit at room temperature for several hours before dialyzing against 50 mM Tris-HCl, pH 7.5 with 10 mM sodium dithionite overnight. A final dialysis was performed against 10 mM Tris, pH 7.5 to remove dithionite. The resulting protein was loaded onto a 150 mL hydroxyapatite (HA) column previously equilibrated with 5 mM potassium phosphate, pH 7.5 buffer. While the column was washed with equilibration buffer, fractions were collected and absorbance monitored at 280 nm to detect the elution of any sodium dithionite not removed by prior dialysis. After an increase and subsequent decrease in absorbance at 280 nm, SCAD was eluted from the column with a 2 L gradient of 5250 mM potassium phosphate, pH 7.5. Fractions were collected and again assayed for enzyme activity. Fractions containing dehydrogenase with the highest specific 26

PAGE 39

activity were pooled. The enzyme could then be frozen at -70 oc for several months. Pooled fractions from the HA column elution were concentrated to approximately 5 mL, spun at 7000 rpm for 15 minutes if any precipitate was present, and loaded onto a Sephacryl S-200 gel filtration column (2.5 em X 90 em) equilibrated with 10 mM Hepes, pH 7.5. Protein was eluted from the column with equilibration buffer and fractions were collected. Spectra of individual fractions eluted from the S-200 column were determined and the dehydrogenase was pooled according to R value. The R value is calculated from the absorbance spectrum of the protein by dividing the absorbance at 270 nm by the absorbance at 450 nm. This value ascertains the protein to flavin ratio indicating the relative amounts of all proteins versus those containing flavin. Non-flavin containing proteins will contribute absorbance to the 270 nm peak due to aromatic rings without increasing absorbance near 450 nm where flavin absorbs. Therefore, the lower the R value, the higher the purity of the flavoprotein. Pools were concentrated and stored in either 20% glycerol at -70 C. 27

PAGE 40

Determination of SCAD Enzyme Activity Enzyme activity was measured spectrophotometrically using a standard dyecoupled assay (Engel and Massey, 1971; Williamson and Engel, 1984). Protein was reduced by its physiological substrate, butyryl-CoA, and regenerated using a final electron accepting dye, dichlorophenolindophenol, which undergoes a color change from blue to clear upon reduction. Reaction mixtures contained 50 f.LM butyryl CoA as substrate, 38 J.1M dichlorophenolindophenol (DCPIP) as the terminal electron acceptor, and 2.5-5.0 J.lL of SCAD in 20 mM Tris-HCl, pH 8.0. Reactions were performed at 25 C and were initiated with the addition of phenazine methosulfate to a final concentration ofO.OOl M. The progress of the reaction was monitored at the absorbance maximum of the dye, 600 nm, for 60 seconds and rates were calculated from the linear portion of the curve. Activities were then calculated from the change in absorbance per min using Beer's Law and the extinction coefficient forDCPIPof21 X 103cm-1mM-1 (Armstrong, 1964). Determination of Extinction Coefficients for SCAD and the ChargeTransfer Complex Determination ofExtinction Coefficient for Oxidized SCAD. The extinction coefficient for oxidized protein at the visible absorption maximum, 450 28

PAGE 41

nm, was determined as follows (McKean et al., 1983). The purified protein was dialyzed against 1 0 mM sodium phosphate, pH 7.5 for 16 hours and precipitated protein was removed by centrifugation. A visible absorbance spectrum of the protein was obtained. FAD was released from the protein by treatment with 1% SDS and an additional spectrum obtained. The absorbance of free FAD and its known extinction coefficient in SDS of 12.045 X 103 cm-1 mM-1 (Salazar, 1998) were used to calculate the extinction coefficient of enzyme bound FAD. The technique was repeated for accuracy. Determination ofthe Extinction Coefficient for the Charge-Transfer Complex. The extinction coefficient for the charge-transfer complex was determined by reduction of the protein with its physiological substrate without a further electron acceptor present. Absorbance at the 450 nm peak of SCAD in 10 mM Hepes, pH 7.5 at 25 C was measured and its concentration calculated using E4sonm,ox == 14.5 X 103 cm-1 mM-1 The substrate, butyryl CoA, was added to a final concentration of 1 00 !J.M and a subsequent spectrum taken. Absorbance at the charge-transfer peak of 560 nm was determined and the extinction coefficient calculated using Beer's Law and the calculated SCAD concentration. Saturation 29

PAGE 42

was assured by further addition of butyryl Co A with no change noted in the spectrum. Steady-state Kinetic Assays Kinetic parameters for SCAD were determined using the standard dye coupled assay previously utilized for activity assays. For kinetic assay, the enzyme is reduced by varying concentrations of substrate and electrons passed to a terminal electron acceptor where reduction is monitored. Reaction mixtures contained a final protein concentration of 7 nM 600 JlM phenazine ethosulfate, and 40 J.1M dichlorophenolindophenol in 20 mM Tris, pH 7.5. Reactions took place at 25 C and were initiated with the addition of2-20 }lM butyryl CoA as the varied substrate. Reaction progress was monitored spectrophotometrically at 600 nm for 60 seconds and duplicate runs were performed. Rates were calculated as changes in absorbance per minute using only the linear portion of the reduction curve. Kinetic parameters and their standard errors were determined by plotting a standard reciprocal plot using the Trinity Software's Enzyme Kinetics program 30

PAGE 43

Determination of SCAD Mechanism SCAD was initially reduced using sodium dithionite with crotonyl CoA, SCAD's physiological product, present. The titration was performed at 25 C in 10 mM Hepes, pH 7.5. All components of the reaction mixture were made anaerobic by repeating 1 0 cycles of argon gas and vacuum. The reaction mixture contained 7.5 J.LM SCAD and 125 J.LM crotonyl CoA. An absorbance spectrum was obtained and the titration initiated using 10 J.LL of a 1.00 mM sodium dithionite solution in 0.001 M sodium pyrophosphate, pH 9.0. Successive additions of 4 J.LL of the dithionite solution were made until full reduction had occurred. Spectra were obtained after each dithionite addition. The same method was employed in the reduction of SCAD without a product present. All conditions were kept constant as well as reaction mixture component concentrations. The crotonyl CoA was not added. The titration was begun using 6 !J.L of the dithionite solution. One successive addition of 3 !J.L was all that was required for full reduction. Spectra were measured before the titration and after the addition of reducing agent. SCAD's reduction of its physiological electron acceptor, ETF, was monitored spectrophotometrically at 25 oc in 10 mM Hepes, pH 7.5. A reaction mixture containing 9 !J.M pig-ETF and .6 mM butyryl CoA was made anaerobic by 31

PAGE 44

1 0 alternating cycles of argon and vacuum. The reaction was initiated with the addition of 250 nM SCAD and spectra were taken every 2 minutes until full reduction of ETF had occurred. Results and Discussion Purification A typical purification of short-chain acyl-CoA dehydrogenase produced pure protein in good yield. A summary of a typical purification is listed in Table 2.1. The initial pellet weight of cells was 45.5 g and led to the final recovery of 127 mg of SCAD. Table 2.1 shows a decrease in the total number of enzyme units from the crude extract to the DEAE column fraction pool. This data may seem unlikely as a continual decrease in units is expected as a purification protocol proceeds. It is likely, however, that the increase is actually due to removal of enzyme activity inhibitors. Table 2.1 also shows a decrease in units, as expected, after the completion of the chromatography using the HA column. This decrease, however, may be accentuated by the additional removal of protein by the de-greening step performed between the DEAE and HA columns. The de-greening step is not a separation step, 32

PAGE 45

Table 2.1. Purification of rat SCAD. Enzyme activity was measured with 50 11M butyryl-CoA as outlined in the Materials and Methods section. Total protein was calculated using s45onm = 14.5 X 103 M-1 Volume Total Total mg/ml R value (ml) Activity( J.llrlOl! Protein (A27of min) (mg) A450) Crude Extract 110 1120 nd nd nd DEAE eluate 2'Y' _.J 2950 nd nd nd HA eluate (pool 1) 114 390 nd nd nd HA eluate (pool 2) 250 745 nd nd nd HA eluate (pool 3) 184 192 nd nd nd Gel filtration eluate 26 Nd 72_1 2.77 5.50 (pool2a) Gel filtration eluate 26 Nd 31.6 1.22 5.65 (pool2b) Gel filtration eluate 26 Nd ?'"' .... _.J .J _932 5.78 (pools 1 & 3) nd: not determined 33

PAGE 46

but the harsh reducing conditions used to remove the CoA-persulfide ligand lead to some denaturation and therefore, loss of enzyme units. It is necessary to remove the persulfide ligand, however, because it is an enzyme inhibitor. Adjusting the pH during the addition of sodium dithionite so that the protein is left in a slightly less harsh environment may minimize loss. Additional loss of protein was experienced if, as in the purification shown in Table 2.1, the HA column pools were stored at -70 C for a period oftime. The first pool from the HA column was not frozen and showed no additional precipitation upon completion of the purification process .. However, when the second two HA column pools were removed from the freezer and concentrated for the fmal chromatography step, there was considerable loss of protein. This precipitation could be due to instability brought about by removal of the CoA persulfide ligand and compounded by the fact that the pools were frozen without addition of glycerol to aid in stabilization. It would, therefore, be best not to store protein prior to the completion of the entire purification protocol when glycerol is added for storage. There were additional problems with stabilizing the short-chain protein. Initially, the protein was stored in 20% glycerol at -70 C. When this protein was thawed for use, there was some precipitation noted. The precipitation problem 34

PAGE 47

compoWlded when there was need for concentration or dialysis of the protein. Perhaps the protein was denaturing because of the freezing and thawing process, so in a subsequent purification, the protein was stored in 50% glycerol but only a -20 C. After only a short time, this protein began to denature rapidly as well and so storage of the protein was returned to -70 oc_ Certain losses were expected upon any manipulation of the protein in preparation for its use. The precipitation problem may be explained by previous reports indicating that the green form of the protein may actually be more stable than the yellow form (Shaw and Engel, 1984). Such observations could be explained by the Co-A persul:fide ligand occupying the binding site of the protein therefore giving it physical and chemical stability. The yellow form without the artificial substrate may be more affected by solvent interactions or physical deformations of the protein as would be present in the various conditions mentioned. A solution for the problem or any further affects besides loss has not been identified. After the completion of all steps of the purification, the SCAD obtained was highly pure. The hydroxyapatite column can illustrate the success of the chromatography steps. This column's elution profile is shown in Figure 2.1. The profile not only shows the activity ofthe protein as it comes off the column but also indicates purity. Purity for the SCAD is established by the A27r/ A450 or R value. As 35

PAGE 48

1.2 6.4 L -c::: I e 6.2 1 0 E ;.. \ ::1. I -r >-0 8 \ 6 :;o -< :; Ill ;:; I 0 \ c:: <: i I II CD I Q) 5.8 -E 0.6 I \ )> I N >-...... I /I I 0 N -c:: I \ )> w \! / \ .. I Ul I / r 0 0.4 '-;I( (/) =: : c::: ::J Q) L 0 I c::: 0.2 .-. ".;::: 5.4 ca ..o--I ..c -:; / I / I .... ., 0 :---X J (/) .,. ...... t.="-..c ---m _;;I( I <: 0 """!""". . . 5.2 30 35 40 45 50 55 60 Fraction Number Figure 2.1. Elution of SCAD from Hydroxyapatite column. Protein was applied to the HA column previously equilibrated with 5 mM potassium phosphate, pH 7.5 The colwnn was washed with equilibration buffer and eluted with a 2 L gradient of 5-250 mM potassium phosphate, pH 7.5. The absorbance at 270 run (0) and 450 nm (D) were monitored and the enzyme activity (.6.) towards butyryl CoA was determined. The R value (x) was also calculated. 36

PAGE 49

shown in the figure, R values were below 7 for the majority of the fractions containing SCAD activity. After the protein had run through the final polishing step with the gel filtration column, typical R values of final protein were around 5.5. These values exceed the purity used in previous publications using short-chain dehydrogenases including that of rat. A spectrum ofthe purified yellow form of rat SCAD is pictured in Figure 2.2 showing the characteristic pattern for a flavoprotein. The peaks seen are at 270 nm, 368 nm, and 450 nm. The large absorption at 270 nm arises from aromatic ring structures present in proteins as well as the adenine and isoalloxazine moieties of FAD. Peaks at 368 nm and 450 nm are the result of the bound FAD prosthetic group. (Mahler, 1954). Spectral Properties The extinction coefficient for the oxidized form of SCAD at its absorbance peak of 450 nm was determined to be 14.5 mM"1cm1 using SDS denaturation. Initial attempts were made in determining E4sOnm using heat denaturation but were met with unexplained large errors. The final value calculated is in good agreement with other reported values ranging from 14.4 mM"1cm1 reported by Williamson and Engel (1984) for SCAD from M elsdenii to 15.4 mM"1cm1 for pig kidney general 37

PAGE 50

l.SO u 1.00 y c Ill -e 0 ... ..c < 240.0 300.0 400.0 Wavelength (nm) 500.0 600.0 Figure 2.2. Absorption spectrum of purified rat SCAD. The figure shows the absorption spectrum of SCAD in 10 mM Hepes, pH 7.5. The protein was purified as noted in the text. 38

PAGE 51

acyl-CoA dehydrogenase (Thorpe et al., 1979). The extinction coefficient for free FAD also detennined at the 450 run maximum is 11.3 mM"1cm'. Thus, the value detennined for SCAD shows the characteristic increase in flavin absorption upon protein binding. The spectrum resulting from the addition of saturating concentrations of SCAD's physiological substrate butyryl CoA is shown in Figure 2.3. Changes in the spectrum from initial conditions include a blue shift of the 370 run band of approximately 20 run, a disappearance of the flavin maximum at 450 run, and the appearance of a peak at 560 run. The new band at the longer wavelength near 560 run initially thought to be the result of flavin-semiquinone (Beinert, 1957) fonnation corresponds to a charge-transfer complex between the fully reduced (two electron) flavin and the bound crotonyl-CoA (Massey and Ghisla, 1974; Hallet al., 1979; Frennan et al., 1980; Thorpe et al., 1981; Schmidt et al., 1981 ). This complex provides the thennodynamic stability necessary for the electrons to be transferred from the dehydrogenase to its electron acceptor, ETF, in two sequential one electron steps. The bound product also protects the reduced flavin from reaction with molecular oxygen which would short circuit the energy yielding pathway and generate hydrogen peroxide, a toxic oxygen species (Crane and Beinert, 1956). Calculations from Figure 2.3 lead to an extinction coefficient for the charge-transfer 39

PAGE 52

u u c < -e 0 "' .&l < 0.17L-----.------r-----.-------r-----, 0.15 0.10 \ I \ I \ \ \ --400.0 500.0 600.0 700.0 800.0 Wavelength (nm) Figure 23. Aerobic reduction of SCAD with butyryl-CoA. Protein ( 11.7 J..LM) in a 1 ml reaction mixture containing 10 mM Hepes, pH 7.5, was reduced with the addition of25 111 of5.76 mM butyrylCoA. Spectra were recorded before and after the addition. Saturation was verified with an additional 1 0 111 of substrate leading to no change in the spectrum. 40

PAGE 53

band of 3.19 mM"1cm1 Other reported values for this E were not found but the effects of substrate saturation are in agreement with other literature. The titration of SCAD using sodium dithionite with product is shown in Figure 2.4. This titration results in a red shift of the 3 70 nm peak of approximately 15-20 nm, the appearance of a shoulder around 480 nm, and a long-wavelength band near 580 nm. As SCAD is titrated, the 370 nrn band initially shows a slight decrease after which it continually increases until completion. Such spectral changes are the result of the formation of a stable anion semiquinone. The semiquinone indicates the flavin existing in its one electron.reduced state. The presence of product makes this reduction possible as is shown by Figure 2.5. This figure shows the same conditions with the exception that there is no product present. Without product, the spectra resulting from sodium dithionite titration are markedly different. There is full reduction of the flavin as seen in the absence of absorbance at 450 nrn and no formation of any long-wavelength band indicating any semiquinone formation. Thus, the existence of product allows the one electron reduction of flavin. Figure 2.6 shows spectrally how SCAD reduces its physiological electron acceptor ETF. The characteristic spectrum of ETF is similar to that of SCAD with the absorbance maxima around 370 nm and 450 nm. Additionally, ETF shows a 41

PAGE 54

u u c co -e 0 ., .&:1 < 0.12 0.10 I I I I I ! I I ,: I I ,. I : I ,. I \ I \ .. .-: f : o.os I I 1 I ' I I I I I I I I I I I I I ' .. .... . ----.. ...:_ ___ .......... ;;; __ ;_ 320.0 400.0 soo.o 600.0 700.0 800.0 Wavelength (mn) Figure 2.4. Anaerobic titration of SCAD using sodium dithionite with product present. SCAD (7.5 J.lM) in a 1 ml anaerobic reaction mixture containing 125 J.lM crotonyl-CoA in 10 mM Hepes, pH 7.5, was reduced using 4-10 !J.l additions of 1 mM sodium dithionite in 0.001 M sodium pyrophosphate, pH 9.0. Spectra were recorded after each successive addition of dithionite until no further change was detected. 42

PAGE 55

o.no,---......,----....,......-.......... _--.-----.-----, u u c CQ O.lO -e Q 0.05 Cl.l .c < 400.0 600.0 700.0 aoo. o Wavelength (nm) Figure 2.5. Anaerobic titration of SCAD using sodium dithionite without product present. SCAD (7 5 j..l.M) in 10 mM Hepes, pH 7.5, was made anaerobic and reduced using 3-6 Ill additions of 1 mM sodium d.ithionite in 0.001 M sodium pyrophosphate, pH 9.0. Spectra were recorded after each successive addition of dithionite until no further change was detected. 43

PAGE 56

u u = ca -e 0 "' .&:l < o.o6=zr---;--:-----r----------.----------. 0.06 0.04 0.02 500.0 600.0 Wavelength (mn) Figure 2.6. Anaerobic titration of Electron Transfer Flavoprotein using SCAD. ETF (9 in an anaerobic reaction mixture containing 0.6 mM butyryl-CoA in 10 mM Hepes, pH 7 5, was reduced using 250 nM SCAD. Spectra were recorded before SCAD addition and every 2 minutes thereafter until complete reduction was noted. Numbering indicates order of spectra. 44

PAGE 57

shoulder around 460 nm. As ETF is allowed to be reduced by SCAD, an initial increase of the 370 nm peak is seen followed by subsequent decreases. The 450 nm peak continually decreases with time. The initial increase at 370 nm indicates initial formation of the semiquinone form of ETF followed by the decreases indicating the formation of the fully reduced hydroquinone. Since the spectra were taken with time, the formation of semiquinone occurred quickly while the full reduction took significantly more time. Upon consideration of the spectral evidence presented, the mechanism used by this protein can be assumed much like other members of the acyl-CoA dehydrogenase family already studied. The two electron reduction ofF AD occurs upon availability of substrate forming the charge-transfer complex. The product is not released until the re-oxidation of the FAD is possible and so remains stable in the absence of an electron acceptor. When the electron acceptor, ETF, becomes avai1able, one electron is transferred out of SCAD at a time. The transfer of the second electron is the slow, rate-limiting process completing the return of SCAD to its native form. 45

PAGE 58

Kinetic Properties The double reciprocal (Lineweaver-Burke) plot used to determine the steady state kinetic properties for rat SCAD is shown in Figure 2.7. The plot shows excellent linearity with a maximal velocity, V max> value of 4.256 s-1 and value of 1.248 J.LM. A calculated value ofV indicating the enzyme's overall efficiency is 3.41 J.LM-1 sec-1 These kinetic values are similar to those currently reported in the literature. 46

PAGE 59

0.45 0.4 0.35 0.3 ";' 0.25 "' Q) 0.2 ca ... ""' 0.15 0.1 0.05 0 0 0.1 0.2 0.3 1/[SCAD] (J.LM) 0.4 0.5 Figure 2.7. Double reciprocal (Lineweaver-Burke) plot of steady-state kinetic assay. The standard dye-coupled reaction mixtures contained 7 nM SCAD, 600 ).lM phenozine ethosulfate, and 40 ).lM DCPIP in 20 mM Tris-HCI, pH 7.5. Reactions were initiated using 2-20 ).lM additions ofbutyryl-CoA as the varied substrate. Change in absorbance was monitored at 600 run for 60 seconds. Rates were calculated in enzyme units per time. 47

PAGE 60

CHAPTER3 DETERMINATION OF THE MIDPOINT POTENTIAL OF RAT SHORT-CHAIN ACYL-COA DEHYDROGENASE USING THE XANTHINE/XANTHINE OXIDASE REDUCTION METHOD Introduction Although the redox potentials of several of the acyl-CoA dehydrogenases have been explored (Fink et al., 1986; Gustafson et al., 1986; Lenn et al., 1990), the potential for any mammalian SCAD species remains undetermined. Since the dehydrogenase family has been characterized in terms of electron donors and acceptors, the interest in the redox potential of SCAD lies with the effect on this property of amino acid changes in the vicinity of the cofactor, FAD. After the potential of the wild-type SCAD is known, comparisons can then be made with the potentials of mutant proteins of interest. Any changes noted in mutant potentials can then be directly correlated to the amino acid residue in question. Such information may play a role in determining the pathway of the electrons as they flow from substrate to enzyme and finally to electron acceptor. The most commonly used method for determining the redox potential of flavoproteins is the xanthine/xanthine oxidase reduction system developed by Massey (1991). Although this method is limited by its approximation of an 48

PAGE 61

equilibrium system, it provides accurate values for potentials without requiring the specialized equipment of other methods such as potentiometry. The objective of this chapter is to use the Massey method to determine the redox potential of rat short-chain acyl-CoA dehydrogenase. Materials and Methods Rat SCAD was expressed and purified from E. Coli as described in Chapter 2 of this thesis. Ultraviolet/visible spectroscopy was performed using a Shimadzu UV2401 spectrophotometer. All reagents were reagent grade or better and were obtained from commercial sources. Determination of Extinction Coefficient for Reduced SCAD The extinction coefficient for the reduced SCAD was determined using the method described by Massey and Hemmerich (1978) in which the flavoprotein is photoreduced using deazaflavin as a catalyst. Reactions were completed at 4 C in 10 mM potassium phosphate, 10% ethylene glycol, 1 0 mM EDTA, pH 7.0. Reaction mixtures were made anaerobic by initial bubbling of buffers with argon for 15 minutes and subsequent alternating of argon and vacuum for 20 cycles of final reaction mixtures. Mixtures of 13.6 J.LM SCAD and 16 nM 49

PAGE 62

5-deazaflavin were kept in the dark and allowed to stand for 2 minutes before the experiment. The reaction was initiated with a 5-second exposure to the light of a slide projector. A full absorption spectra was then obtained to monitor the protein's reduction. Light exposure was then repeated and additional spectra obtained until full reduction of the SCAD had occurred. The extinction coefficient of the reduced enzyme was then calculated using Beer's Law and the previously calculated value for the oxidized coefficient of 14.5 X 103 cm-1 mM-1 at 450 nm. Re-crystallization of Redox Dye Redox dye Indigo Carmine was re-crystallized by dissolution in a minimal volume of water. Acetone was added until complete precipitation of the dye occurred. The crystals were then collected by vacuum filtration and allowed to dry overnight. Determination ofXanthine Oxidase Enzyme Activity Enzyme activity was measured spectrophotometrically by monitoring the reduction of xanthine by xanthine oxidase. Reaction mixtures contained 400 J..I.M xanthine in 0.1 M potassium phosphate bufffer, pH 8.4. The reaction was initiated by addition of 5 J.ll of 0.90 nM xanthine oxidase and absorbance at 296 nm was 50

PAGE 63

monitored for 60 seconds. Successive additions up to 25 111 xanthine oxidase were made with activities calculated as change in absorbance per time. These values were plotted versus the volume xanthine oxidase added to assure linearity. Determination of SCAD Midpoint Potential The midpoint potential of SCAD was determined using the xanthine/xanthine oxidase reduction method of Massey (1991). With this method, a redox indicator dye is used with a known potential near that of the protein. The reduction of both dye and protein are monitored and the apparent potential of the protein is determined using the Nemst equation. Reduction reaction mixtures of the indicator dye alone contained 15 11M indicator dye, 1.5 11M benzyl viologen as the mediator dye included to facilitate equilibrium between indicator dye and protein, and 250 11M xanthine in deaerated 10 mM potassium phosphate buffer, pH 7.0. A final concentration of 40 nM xanthine oxidase in buffer was added to the sidearm of a quartz cuvette and the reaction mixture was made anaerobic by 25 alternating cycles of argon and vacuum. After allowing the solution to rest for 2 minutes, the reaction was initiated by addition of the xanthine oxidase and spectra were monitored every 120 seconds for 30 minutes at 15 C. Isosbestic points were determined from the accumulated spectra. 51

PAGE 64

Reduction reaction mixtures of the protein alone contained 10 )lM SCAD, 1.5 )lM benzyl viologen, and 250 )lM xanthine in deaerated 10 m.M potassium phosphate buffer, pH 7.0. After the mixture was made anaerobic as described above, the reaction was initiated using 63 nM xanthine oxidase and monitored every 60 seconds at 15 oc until complete reduction of protein was observed. Reduction reaction mixtures to detennine the redox potential of the protein contained 15 )lM SCAD, 15 )lM indicator dye, 1.5 )lM benzyl viologen, and 250 )lM xanthine in deaerated 10 m.M potassium phosphate buffer, pH 7.0. These mixtures were made anaerobic and 25-65 nM xanthine oxidase was added to initiate the reaction. Reduction of protein and dye were monitored every 60 seconds at 15 oc until either had reduced completely. Reduction of protein was monitored at dye isosbestic points near the protein's maximum. The midpoint potential was determined by plotting the log of the oxidized over reduced concentrations of dye versus the log of the oxidized over reduced concentrations of the protein. This plot represented the Nemst equations for dye and protein set equal and the potential value for the protein was calculated from the y-intercept and the known potential of the dye. 52

PAGE 65

Results and Discussion Determination of the Extinction Coefficients for Reduced SCAD The spectra obtained during the photoreduction of SCAD using a deazaflavin catalyst are shown in Figure 3 .1. The spectra show the complete quenching ofthe visible absorbance peak of SCAD at 450 run indicating the full reduction of the protein. In initial runs of the experiment, some protein reduction was observed prior to obtaining spectra. Keeping subsequent reaction mixtures covered during the preparation time eliminated the problem. In addition, reoxidation of the protein was observed as the experiment time progressed due to oxygen exposure. Addition of the deazaflavin from a sidearm cuvette instead of using a gas tight syringe kept the reaction mixture anaerobic for the duration of the reduction. The extinction coefficient calculated for the reduced SCAD species at 468 run was 2.48 X I 03 cm-1 mM-1 The extinction coefficient for the oxidized species at 468 nm was 12.4 X 103 cm-1 mM-1 Determination of the extinction coefficients for the reduced and oxidized species at this wavelength was necessary for later calculation of SCAD concentrations during the midpoint potential determination experiments. 53

PAGE 66

u u c ca 0.1501-.c 0.10 .. 0 "' .c < 0.0501-600.0 800.0 Wavelength (nm) Figure 3.1. Photoreduction of SCAD using a deazaflavin catalyst. Anaerobic reaction mixtures containing 13.6 J.l.M SCAD and 16 ru\1 deazaflavin in 10 ffi.l\1 potassium phosphate, 10 mM EDTA, 10% ethylene glycol, pH 7.0 were exposed to I 0 second intervals oflight at 4 C. Spectra were taken after each light exposure until full reduction has taken place. Not all spectra are shown. 54

PAGE 67

Determination of Xanthine Oxidase Catalytic Activity The catalytic activity of xanthine oxidase was measured by the reduction of its electron acceptor xanthine. Xanthine has an absorbance maximum at 296 nm which increases upon reduction. The plot of xanthine oxidase activity measured in change in absorbance at 296 nm per 20 seconds versus volume of xanthine oxidase added is seen in Figure 3 .2. The graph shows excellent linearity with increase in enzyme concentration. Linearity indicates that the enzyme is functioning properly. The assay was performed to ensure that the xanthine oxidase used for the midpoint potential determination experiments could adequately catalyze the reduction reactions. Determination SCAD Midpoint Potential The oxidation-reduction midpoint potential of short-chain acyl-CoA dehydrogenase was determined using the xanthine/xanthine oxidase method of Massey. In order to obtain accurate values with this method, the indicator dye used should have a standard potential within 30 m V of the protein in question. A dye with nearly equivalent potential will closest approach equilibrium conditions. In looking to the literature for related protein potentials, the expected value of SCAD's potential was approximately -120 mV (Gustafson, 1986). In addition to specific potentials, indicator dyes should also provide little absorbance interference near the 55

PAGE 68

0.014 i I [ 0.012 f r.n "C c: I 0 u 0.01 r Q,) l r.n 0 N co Cl 0.008 N r.n I I ..0 <( c: Q,) l 0) 0.006 c: ro ..c: (.) i r-0.004 \..-t r I I , ' I . I .. 0.002 0 5 10 15 20 25 30 111 Xanthine Oxidase Figure 3.2. Plot to determine the activity of xanthine oxidase. Xanthine ( 400 ).lM) in 0.1 M potassium phosphate, pH 8.4 was reduced using successive additions of 5 ).ll of 0.9 nM xanthine oxidase. Activity was measured as a decrease in absorbance at 296 run for 60 seconds after each addition. This activity is plotted versus the volume of xanthine oxidase added and the plot is examined for linearity. 56

PAGE 69

protein's maximum and have their absorbance maximum in a region where the protein does not absorb. In this way, the reduction of protein and dye can be monitored simultaneously without interference. SCAD's maximum absorbance peak is at 450 nm. An ideal dye would contain an isosbestic point at or near the protein's absorbance maximum with a potential nearly identical to that of the protein. With an expected approximate value for SCAD's potential and knowledge of its spectral properties, initial experiments were run using indigo carmine as the redox indicator dye. This dye is commonly seen in the literature with other flavoprotein potential determination experiments. Initial runs of an experiment with dye alone showed no reduction or incomplete reduction even with increased concentrations of xanthine oxidase, the reaction initiator. After replacement of the gas train used to make the reaction mixtures anaerobic in the laboratory and re crystallization of the dye, complete reduction of indigo cannine was accomplished as shown in Figure 3.3. The spectra show an ideal isosbestic point at 468 nm near SCAD's maximum and a dye peak at 620 nm where SCAD is not expected to absorb. It was then necessary to reduce SCAD alone in order to show its behavior in the experimental conditions at the proposed wavelengths. The spectra obtained from this reduction are shown in Figure 3.4. These spectra show a decreasing absorbance at 468 nm which could be used to monitor SCAD reduction and minimal 57

PAGE 70

0.2-; C.15 i5' j:!. C. I :g :i .. 0 05 -i o-' 550 Figure 3.3. Anaerobic reduction of indigo carmine using xanthine/xanthine oxidase. Anaerobic reaction mixtures contained 15 J..LM indigo carmine, 1.5 J..LM benzyl viologen, and 250 J..LM xanthine in 10 mM potassium phosphate, pH 7.0. Reactions were run at 15 C and initiated with 40 nM xanthine oxidase. Spectra were monitored. 58

PAGE 71

-----------i 1 J j I I I 1-'"' 1:; lg i: I I 1 I I i I 02 Figure 3.4. Anaerobic reduction of SCAD. using xanthine/xanthine oxidase. Anaerobic reaction mixtures contained 10 }J.M SCAD, 1.5 }J.M benzyl viologen, and 250 1-1M xanthine in 10 mM potassium phosphate, pH 7.0. Reactions were run at 15 oc and initiated with 63 nM xanthine oxidase. Spectra were monitored. 59

PAGE 72

absorbance at 620 run where the reduction of dye could be monitored. Therefore, with the potential of the indigo carmine (E0'= -.121 V) within the expected range of the protein, the experiment was performed with both dye and protein present. The midpoint potential of SCAD using indigo carmine as indicator dye was determined to be -.182 V. Sample spectra from a reduction are shown in Figure 3.5. Figure 3.6 shows the Minneart plot from which the potential was calculated. This potential value is considerably lower than what had been anticipated and so was met with skepticism. In preliminary experiments, it was clear that the protein was reducing considerably slower than the dye indicating its lower potential. Therefore, numerous trials were made using alternate dyes of differing potentials. Indigo tetrasulfonic acid (E0'= -.046 V) was an unlikely alternative due to its higher potential but was used to rule out the possibility that indigo carmine was perhaps not interacting with the protein as expected. Results again showed the protein reducing slower than dye verifying previous conclusions. Another dye used was phenosafranine (E0'= -.252 V) which has a considerably lower potential. Results with this dye proved inconclusive as a correction had to be made for the dye's absorbance at the protein absorbance wavelength. This correction introduced enough error that results were not in agreement with previous calculations. Other dyes looked at included anthraquinone 60

PAGE 73

0.15
PAGE 74

3 -2.5 -c: 'a) 0 2 ... Q. ..... "0 I ... c: 1.5 'i 0 ... c. y = o.643x + 1.248c ';(' 0 R2 = 0.989 Cl ..2 0.5.-0.5 0 0.5 1 1.5 2 2.5 log [ox] dye I [red] dye Figure 3.6. Minneart plot of reduction of SCAD and indigo carmine data. Concentrations of dye and protein were determined and the Minneart plot was drawn. A regression line was fit to the data. E0 was calculated from they-intercept and slope was calculated. 62

PAGE 75

2,6-disulfonate (E0'= -.184 V), neutral blue (E0'= -.190 V) and cresyl violet acetate (E0'=-.166 V). The anthraquinone had a large absorbance at the protein peak which prevented monitoring of protein reduction. Both neutral blue and cresyl violet acetate did not equilibrate with the protein indicating perhaps these compounds could not enter the protein's binding site. Many other available dyes in the range of potentials needed were quinone containing with their absorbance maxima near that of SCAD preventing their use in the experiment. The results obtained using indigo carmine were, therefore, used to calculate the SCAD midpoint potential. The consequences of not finding an ideal dye for the protein can be seen in the Minnaert plot (Figure 3.6). In an experiment such as this with protein and dye reducing the same number of electrons, a slope of one is expected. The slope seen with SCAD varied from this ideal. Such variations may be directly related to the differences in potentials between the dye and protein. The potential difference in this experiment was approximately 60 mV, twice the value suggested by Massey. Under these conditions, the experiment was not conducted under equilibrium conditions. The rate at which the dye reduced was significantly faster than the protein. Therefore, variations would be expected and the result obtained should be seen as a valid approximation of the true midpoint potential for this protein. 63

PAGE 76

CHAPTER4 DETERMINATION OF THE MIDPOINT POTENTIAL OF SHORT -CHAIN ACYL CO-A DEHYDROGENASE BY CYCLIC VOLTAMMETRY Introduction Cyclic voltammetry of proteins is a useful method for probing electron transfer properties. Such properties include the standard rate constant (k0), a measure of electron transfer kinetics, the transfer coefficient (a), a measure of energy barrier symmetry, and the formal potential for a solution species (Eo') or adsorbed species (Eo'ads). The first two properties can only be determined when electrons are transferred directly from the electrode to the protein in solution. Voltammetry also offers an advantage over some spectroelectrochemical techniques used in which the species of interest does not come in contact with the electrode. With the latter, equilibrium is approached between a dye and the protein, and the potential is calculated with respect to the dye. In cyclic voltammetry, direct electron transfer eliminates the uncertainty of the equilibrium approximation. Although cyclic voltammetry may seem to be an ideal method for the determination of formal potential, its use with proteins is somewhat problematic. Many initial experiments with proteins such as cytochrome c gave no currents using 64

PAGE 77

cyclic voltammetry. After much work in the area, methods have been developed and new electrode surfaces have made cyclic voltammetry of proteins feasible. Many proteins still await electrochemical characterization as with SCAD. The objective of this chapter is to show how the midpoint potential of short-chain acyl CoA dehydrogenase was obtained in order to verify the values previously determined with the Massey method. Materials and Methods Rat SCAD was expressed and purified from E. Coli as described in Chapter 2 of this thesis. Teflon for the electrochemical cell was purchased from Regal Plastics, Denver, Colorado. Gold wire and foil and silver wire were purchased from Alfa-Aesar, Mass. Platinum wire was purchased from The Wilkinson Company, Agoura Hills, California. Pyrolitic graphite was obtained from Union Carbide. All reagents were purchased from commercial sources and were of reagent grade or better, and used without further purification. Design of the Electrochemical Cell The design of an electrochemical cell for the determination of the SCAD midpoint potential took several limitations into consideration. The availability of 65

PAGE 78

analyte for such a measurement was restricted to extremely small volumes as purification of large amounts of protein was costly in time and materials. A cell that accommodates such volumes was not widely available commercially and so was, therefore, constructed in the laboratory. While minimizing the cell volume was essential, the cell also had to be sufficient in size to accommodate three-electrodes. Electrodes available in this laboratory would have required a much larger sample size than was allowed in these experiments. Therefore, electrodes were also constructed and sized to function in a small volume cell. In addition, all other requirements for any working electrochemical cell had to be met by the cell design. The material from which a cell is made had to be non reactive, readily available, and easily machined. Reference and auxiliary electrodes had to be separated from analyte solution normally accomplished by use of a porous frit. In order to run experiments under anaerobic conditions, there also had to be an inlet and an exhaust vent built into the cell to allow a stream of nitrogen to blanket the solution. 66

PAGE 79

Construction of the Electrochemical Cell and Electrodes The electrochemical cell designed to accommodate 300-500 f.LL of solution is pictured in Figure 4.1. The cell was constructed using a%" Teflon rod measuring approximately 3 inches in length as starting material. A hole, 27/64" X 1 Yz", was drilled at one end of the rod. The cell was then tapped to accommodate a Yz" -13 coarse Nylon screw. To insert electrodes, three holes were drilled lengthwise through the screw. The hole for the reference electrode required a 7/64" bit with the auxiliary and working electrodes requiring 11164" and 1116" bits, respectively. The hole for the reference electrode was purposely made larger than needed so that it could also serve as the vent. A final hole was drilled through the side of the rod at a 45 angle away from the cell opening to accommodate the tube fitting for the gas inlet. Tlie gas inlet was positioned just below the screw threads. A Ag/AgCl reference was constructed by placing 1/8" diameter X 1/8" Vycor rod into the end of a 3" length of heat shrink tubing. The Vycor served as the frit. A heat gun was used to shrink the tubing around the Vycor and to shrink the remaining tubing to allow it to fit through the 764" diameter hole size drilled through the top of the cell cap. The tubing was then filled with 1M KCl and a Ag wire plated with AgCl was placed through the open end. A platinum auxiliary electrode was constructed in the same manner using the Vycor and heat shrink 67

PAGE 80

3/4" .. ... 1/2" .. .. Cellside view 1116" 112" 11/64" I -_.. _.. -... ... 1/2" I Nylon screw side view Nylon screw Top view Figure 4.1. Electrochemical cell used for determination of Em of SCAD. The cell was constructed of Teflon and designed to accommodate 300-500 of solution. A nylon screw was used to contain the samples. Top and side views are shown. Drilling sizes for electrodes and vent are also as shown. 68

PAGE 81

tubing. The diameter of the tubing above the frit was not additionally decreased so that solution could easily be removed and re-filled. The carbon working electrode was formed by sanding a piece of carbon from a larger block into a rod shape of approximately 118" diameter and%" in length. This piece was then placed in a 1" length of heat shrink tubing and secured at one end of the tubing with a heat gun leaving approximately 1/8" of the carbon exposed. Epoxy was then used to seal the edge where the carbon met the tubing. A small drop of mercury was placed in the open end of the tubing and a gold wire was used as the conductor. The gold electrode used was constructed using a gold flag approximately W' square. Three holes were drilled near one end of the flag through which a gold wire was threaded. Cyclic Voltammetry The electrochemical cell was first cleaned by treatment with chromic acid for at least 10 minutes and rinsed thoroughly using pyrolitically distilled water (PDW) (Conway, 1973). Other components of the cell including the screw top and fittings were soaked in base bath (1 L 95% ethanol, 100 g KOH, 100 ml H20) for 10 minutes and also rinsed with PDW. Reference and auxiliary electrodes were rinsed with PDW before cell assembly. The gold electrode was soaked in warm chromic acid for 10 minutes and rinsed. It was then placed in an H-cell in 1 M sulfuric acid 69

PAGE 82

and cleaned electrochemically by cycling between -.35 V and 1.5 V for ten minutes. If a clean scan could not be obtained by electrochemical cleaning, the electrode was heated to incandescence in a natural gas/air flame. After obtaining a clean scan in sulfuric acid, the electrode was rinsed and ready for use in the micro-cell. The working carbon electrode was activated by polishing with alumina on a glass plate and then rinsed well with PDW. The auxiliary electrode was filled with electrolyte and all electrodes were inserted through their respective holes in the screw cap. Electrode wires were secured in place using Teflon tape. After placing 300-500 !J.L of sample into the cell, the cap was screwed in place and nitrogen line attached. Nitrogen gas was blanketed over the solution for 10 minutes prior to scanning. All scans were run at room temperature using a Cypress Model Omni 90 potentiostat. Data were recorded using a BioAnalytical Systems Model RXY Recorder. The scan range used for the protein potential determinations was 0.000 V to-0.600 V. Determination of the Midpoint Potential. Prior to any electrochemical experiment, SCAD was dialyzed against 10 rnM Hepes, pH 7.0 overnighL After dialysis, protein was concentrated in an Ami con Centricon I 0 concentrator to 60 J.1M. The midpoint potential was determined using cyclic voltammetry as described 70

PAGE 83

above. Blank scans were run using buffer only. The scan rate used was 1 00 m V /s. Midpoint potentials were calculated by hand using the average of the anodic and cathodic peak potentials and are reported versus the standard hydrogen electrode (SHE). Scan Rate Dependence. Scan rate dependence was determined using cyclic voltammetry and 60 J.LM SCAD in 10 mM Hepes, pH 7.0 as described above. Scans were run at rates of25, 50, 100, and 200 mV/s. The working electrode was cleaned prior to each scan to remove any buildup of adsorbed species. The peak current was measured by hand with subtraction of background for each scan rate. Plots were then made of peak height versus scan rate and peak height versus the square root of scan rate. The plots were then examined for linearity to determine whether the species was in the dissolved or adsorbed state. Linearity of the peak height versus scan rate plot would indicate a dissolved species while linearity of the peak height versus the square root of scan rate would indicate a species has adsorbed onto the electrode. 71

PAGE 84

Results and Discussion Determination of Cell Function The cyclic voltammogram for Potassium Ferrocyanide is shown in Figure 4.2. 1bis scan clearly shows the sharp peaks expected for an ideal compound. The midpoint potential calculated from the scan is 0.580 V vs Ag/AgCl. This value is in agreement with reported values of0.583 V vs Ag/AgCl (Rieger, 1994). Obtaining a scan for a well characterized analyte of known midpoint potential was necessary in confirming the proper functioning of the cell and electrodes constructed for the SCAD experiments. With an accurate determination of the midpoint potential of potassium ferrocyanide, the cell could then be used for unknown potential determinations with confidence. Determination of the Midpoint Potential of SCAD There is no universal material used for the working electrode which produces notable peaks for every species. Therefore, when. working with an electrochemically uncharacterized compound, the first obstacle is to identify a material from which a reasonable scan can be obtained. The process not only entails looking to the literature for previous reports on similar species but also trial and 72

PAGE 85

<( ..... 0 .2 .4 .6 E I V (vs. Ag/ AgCI) Figure 4.2. Cyclic voltammogram of potassium ferrocyanide. The scan of 7 X 1 0"" M potassium ferrocyanide was performed in 1 M KCl on a clean gold electrode. The scan rate used was 1 00 m V /s. 73

PAGE 86

error. For the determination of SCAD's midpoint potential, many electrode surfaces were tried with unsuccessful results. A most recent report on SCAD's electron acceptor ETF showed it to adsorb and give a well-defined scan on idium oxide electrodes (Salazar, 1998). Since this protein's electroactive center is also flavin, it seemed reasonable to attempt using indium oxide as the working electrode material with SCAD. Initial scans of SCAD on idium oxide led to what appeared to be small peaks. After increasing the sensitivity of the experiment by using a differential pulse technique, the peaks were deemed to be due to buffer and not reactivity of the protein. Upon referring to further literature reports, attempts were made at using a gold electrode surface (Barker and Hill, 1988; Martinet al., 1997; Rivera et al., 1994). The scan on bare gold showed no peaks for SCAD. Another technique used widely with proteins is to modify a gold surface with compounds which render that surface charged. The charged surface can then attract an oppositely charged region of the protein. Proper alignment ofthe protein's electroactive center with the electrode will then allow electron transfer. Such electrode modification was attempted using 3-mercaptopropionic acid (MP A) or cysteimine leaving a negative or positive charge, respectively, on the surface. Neither attempt successfully produced a scan for SCAD. In addition to surface modification, other promoters 74

PAGE 87

may be used in an analyte solution. These promoters may be ions which would assist in the diffusion of protein towards the electrode surface (Rivera, 1994). Magnesium chloride was added to a solution of SCAD and tries with both modified surfaces still showed no peaks. A final attempt with gold was made using a poly lysine promoter as reported in the literature (Rivera, 1994). The lysine with its charged polar side chain could then attract oppositely charged side chains within the protein structure. Since SCAD is known to be acidic with a pi of approximately 5, this also seemed a reasonable method. However, no scan was obtained using the poly-lysine alone or in conjunction with the MgC12 promoter. This environment seemed especially harsh as denaturation of protein was noted. Finally, an attempt was made using a carbon electrode without a promoter (Barber, 1997). The scan obtained is shown in Figure 4.3. This scan shows clear peaks from which a midpoint potential was calculated to be -.240 V. Since the potential of other members of the dehydrogenase family are considerably higher, this determination was met with skepticism. Upon running another blank after the initial scan was obtained, it appeared that the peaks were still present. Further cleaning of the electrode led to a decrease in these peaks. To alleviate any concern that the peaks were not due to SCAD reactivity, a new electrode was constructed. 75

PAGE 88

.8 .6 .4 -.2 0 .2 E IV (Ag/AgC!) Figure 4.3. Cyclic voltammogram of SCAD. The scan was performed using 60 1-1M SCAD in 10 rnM Hepes, pH 7.0 on a carbon electrode. The scan rate used was 100 mV/s. 76

PAGE 89

After a blank was run with no appearance of peaks, the SCAD was scanned again. Peaks were again visible and a midpoint potential of -.260 V was calculated indicating that the proper electrode surface had been found for the determination of SCAD potential. The shape of a voltammogram gives Qther indications of the nature of the SCAD system. The large background or vertical separation of the peaks seen is due to the high sensitivity with which the scan was made. Background in cyclic voltammetry, although partially caused by residual currents from minute quantities of contaminants in the system, arises due to the existence of capacitive current. Capacitive current is due to the double-layer charging at an electrode surface upon application of a potential. Although a transient portion of this current disappears quickly, a linear portion exists throughout a cyclic voltammetry scan. This capacitive current produces the vertical separation above and below zero on the voltammogram. As sensitivity is increased on the instrumentation, the capacitive current response is also increased. In the scan shown for SCAD, a further increase in sensitivity would have pushed the background capacitive current and, therefore, current peaks off the scale. The protein used was concentrated to a point just high enough to see reliable peaks over background and so was the lowest concentration 77

PAGE 90

possible for obtaining the scan. Even though the protein is retained after cyclic voltammetry is performed, significant amounts of protein are lost in the dialysis and concentration steps required in preparation. Therefore, the least amount of protein necessary for a well-defmed scan was used to avoid waste. It should also be noted that the relatively large protein concentration would not have been possible if it weren't for the small volume cell constructed for use in the experiment. The separation of peaks on a voltammogram is due to the necessary application of an overpotential to convert oxidized species to reduced and vice versa. Overpotential is a measure of the polarization or departure of electrode potential from the Nernstian value. Polarization effects are due significantly to electron transfer kinetics. The peak separation for SCAD is 145 m V. This value is greater than that of 29.5 m V which would be expected of a completely reversible 2elecron transferring system. Such a quasi-reversible system indicates the existence of an activation barrier which must be overcome by application of potential. This characteristic was quantified by the kinetic studies presented in Chapter 2 of this report and are again shown by the voltammogram. The overall slant of the SCAD voltammogram is due to the ohmic potential or iR drop of the electrochemical cell. That is, a certain driving force (potential) is required to overcome the resistance of ions in solution towards movement to the 78

PAGE 91

electrodes. The net effect in the case of SCAD is to increase the amount of potential required to see Faradaic current. TheiR drop can be minimized in a cell by increasing salt concentrations, increasing permeability of frits, using mobile ions, and decreasing the distance between the electrodes. For the micro-cell, the electrode distance was very minimal as was required by design. The salt concentration was not increased further due to rapid denaturation of the protein at high salt concentrations. Other factors such as the permeability of the frit were determined by the availability of materials. Overall, the existence of the ohmic potential caused by the requirements of the experiment did not impede obtaining reliable data. Scan Rate Dependence The next step in ensuring that the midpoint potential obtained by cyclic voltarnmetry could be compared to that of the spectroscopically obtained value discussed in Chapter 3 was to determine the scan rate dependence of the peak current height. The plot of peak current versus square root of the scan rate is shown in Figure 4.4. The plot is linear with a correlation coefficient of0.997. A plot of peak current versus scan rate dependence shows less linearity. The equation representing peak current for an adsorbed species is 79

PAGE 92

14 12 10 N Ul > .. 8-CD 1: 1!1 u U'J 6-4-2 0 2 4 6 8 10 12 Peak current (IJ.A) Figure 4.4. Plot to determine the scan rate dependence for SCAD. Cyclic voltammogram scans were performed using 60 J..I.M SCAD in 10 mM Hepes, pH 7.0 at various scan rates. The electrode was cleaned between scans. Peak currents were plotted versus the square root of the scan rate. Scan rates used were 25, 50, 100, and 200 m V /s. Linearity indicates a dissolved species scan. 80

PAGE 93

i = n2 F2 A r* v 8 p R T(l +8)2 (4.1) where iP is peak current, n is number of electrons transferred, A is electrode surface area, r* and 8 are measures of surface concentrations, R is the gas constant, T is temperature in Kelvin, and v is the scan rate. This equation shows peak current, iP, directly proportional to scan rate, v. On the other hand, the equation representing the peak current for a dissolved species is (4.2) where iP is peak current, n is number of electrons transferred, A is the electrode area, Dis the diffusion constant for the species, vis the scan rate, and C* is the bulk species concentration. In this equation, peak current, iP, is directly proportional to the square root of the scan rate, v112 Since all other factors in both equations are constant for an experiment, a linear plot for either iP vs. v or iP vs. v112 indicates the state of the electroactive species. For SCAD, the plot indicates that the scan obtained was due dissolved SCAD and not SCAD which might have adsorbed to the electrode surface. Since adsorption could shift the calculated midpoint potential, it is necessary to rule out this possibility as an explanation for the difference in values calculated from the two different methods used in this thesis. 81

PAGE 94

CHAPTERS GENERAL DISCUSSION The results put forth in this thesis provide an alternate method of purifying rat short-chain acyl-CoA dehydrogenase from E. Coli. The data shows the protocol including DEAE gel exclusion chromatography, hydroxylapatite column chromatography, and S-200 Sephacryl gel filtration chromatography produces protein in good yield and of a higher purity than what has currently been published. The mechanistic data support the previously proposed mechanism for other dehydrogenases. SCAD is reduced after the removal of two-electrons from its substrate forming the charge-transfer complex. The semiquinone species is formed although to a lesser extent than some of the dehydrogenases. This suggests the dehydrogenases have different primary structures near the FAD. These amino acids stabilize the semiquionone and so the amount formed depends on the particular side chains present. SCAD then reduces its electron acceptor, ETF, in two one-electron reduction steps. The charge-transfer complex makes this possible by thermodynamic stabilization of the intermediate. Kinetic data found in this work was in line with previous reports and is comparable to the other dehydrogenases as well. 82

PAGE 95

The main goal of this research was to determine the midpoint potential of rat SCAD. With the initial characterization complete, the expectation was the protein would behave much like its family members with a potential near -.100 V. Upon initiating the potential experiments, however, it was clear that the similarities may not apply to midpoint potentials. The first experiments using the xanthine/xanthine oxidase reduction method of Massey clearly showed a protein with a potential more negative than expected. The dyes which were readily available and normally used for such determinations were not the best candidates for this protein. Other dyes and methods may provide a more accurate determination. Although numerous dyes were used in attempts to more closely approach equilibrium with the Massey method, little success was met. Dyes in the potential range needed, around -.200 V, either had high spectral interference or seemed not to equilibrate with the protein. The dye, indigo carmine, showed reproducible results and is one which has been commonly used with other flavoproteins. A midpoint potential was calculated to be -.180 V. As discussed previously, this dye's potential (E0 = -.121 V) is out of the recommended range for use in the experiment. However, with the complications that exist with this protein, it provides the best approximation for SCAD's potential using this method. The method of determining midpoint potentials using cyclic voltammetry has been used recently with proteins like SCAD. After determining proper 83

PAGE 96

conditions, this method provides a quick determination through direct electron transfer between the working electrode and the protein. Thus, the equilibration or interference with a redox dye is not an issue. A cell was constructed for this work so that protein volume was minimized. Working electrodes were tried including indium oxide and gold as well as modifiers with gold. A carbon electrode was found to provide an excellent signal with SCAD. The potential calculated from this method was -.240 V, a more negative and somewhat different value than the Massey method. One question regarding the data is why these two methods provide different values for midpoint potential. A possibility, which has already been discussed, is that the Massey method, without an ideal dye, has provided an approximation. Thus, a different dye may produce a value more similar to that determined by cyclic voltammetry. Another possibility is that the two methods inherently provide different results based on their mechanisms. The Massey method makes use of the redox mediator, benzyl viologen. Electrons are passed through this easily reducible and highly diffusable compound which then reduces both dye and protein in equilibrium. Therefore, benzyl viologen is entering the protein binding pocket and reducing FAD directly. Using cyclic voltammetry, the protein is reduced via an applied potential at an electrode. The protein approaches the electrode by diffusion and ifthe force 84

PAGE 97

(potential) is strong enough, electrons are transferred from the electrode to the protein. The exact path of the electrons in this case is unknown. Electrons may have to travel through a large amount of relative space to reach the FAD buried inside the protein. Such a difference in the mechanism between the two methods may result in different apparent potentials for SCAD. The two methods used in this research, therefore, provide values for SCAD's midpoint potential which may both be valid for the given technique. Since the exact mechanisms are not known definitively, neither can be discounted. The midpoint potential for SCAD, no matter how obtained, is lower than expected based on previous determinations of related proteins. Thus, the other question left from this research is why the potential would be more negative. Possibilities include that the potential is low due to a physiological role. A lower potential indicates that the reaction catalyzed would happen more slowly due to a higher than if the potential were higher. Therefore, the reaction may not proceed as easily without high concentrations of substrate pushing the equilibrium forward. In this way, the protein may play a regulatory role within the oxidation pathway. Another possibility is that the potential measured here has been artificially lowered by some treatment of the protein. As was discussed previously, the protein seemed fairly unstable and likely to precipitate out of solution. The proposed 85

PAGE 98

reasoning for the instability was the removal of the CoA-persulfide ligand during purification. Perhaps this removal caused a structural change within the protein leading to an apparent low potential. Thus, the protein's potential could more closely resemble those of the other dehydrogenases. Without further study, it is impossible to say which of these hypotheses, if any, are true. As is often the case, this report has left many questions unanswered. With the unexpected results, however, interest is spurred to look at rat SCAD more closely. As with many enzyme families, particular members are studied more than others and it is assumed that the data will be consistent across the family. The data in this report disproves this theory and shows that unexpected results are valuable. In addition, future research may give insight into the potentials determined here. Perhaps purification or potential determination methods need to be modified or perhaps SCAD possesses an important, yet unknown physiological role. 86

PAGE 99

REFERENCES Armstrong, J. McD (1964) Biochim. Biophys. Acta., 86, 194-197. Barber, M.J., Trimboli, A.J., Nomikos, S., and Smith, E.T. (1997) Arch. Biochem. Biophys., 345, 88-96. Barker, P.D., and Hill, H.A. (1988) Prog. Clin. Bioi. Res., 274,419-433. Battaile, K.P., Mohsen, A.A. and Vockley, J. (1996) Biochemistry., 35, 1535615363. Beinert, H. (1957) J Bioi. Chern., 225,465-478. Conway, B.E., Angerstein-Kozlowska, H., Sharp, W.B.A. and Criddle, E.E. (1973) Anal. Chern., 45, 1331. Crane, F.L. and Beinert, H. (1956) J Bioi. Chern., 218, 717-731. Draper, R.D. and Ingraham, L.L. (1968) Arch. Biochem. Biophys., 125, 802-808. Engel, P.C. and Massey, V. (1971) Biochem. J., 125, 879-887. Fink, C.W., Stankovich, M.T. and Soltysik, S. (1986) Biochemistry, 25, 6637-6643. Frerman, F.E., Miziorko, H.M. and Beckman, J.D. (1980) J Bioi. Chern., 255, 11192-11198. Ghisla, S. and Massey, V. (1989) Eur. J Biochem., 181, 1-17. Gustafson, W.G., Feinberg, B.A. and McFarland, J.T. (1986) J Bioi. Chern., 261, 7733-7741. Hall, C.L., Lambet, D.L. and Kamin, H. (1979) J Bioi. Chern., 254, 2023-2031. Ikeda, Y., OkamuraIkeda, K. and Tanaka, K. ( 1985) J Bioi. Chern., 260(2), 13111325. 87

PAGE 100

Lenn, N.D., Stankovich, M.T. and Liu, H. (1990) Biochemistry, 29,3709-3715. Martin, T.D., Monheit, S.A., Niichel, R.J., Peterson, S.C., Campbell, C.H. and Zapien, D.C. (1997) J. Electroanal. Chern., 420, 279-290. Massey, V. (1991) In Curti, B., Ronchi, S. and Zanetti, G. (eds) Flavins and Flavoproteins 1990. Walter de Gruyter and Co., New York, pp. 59-66. Massey, V. and Hemmerich, P. (1978) Biochemistry,. 17(1), 9-16. Massey, V. and Ghisla, S. (1974) Ann. N Y Acad. Sci., 227,446-465. McKean, M.D., Beckman, J.D. and Frerman, F.E. (1983) J. Bioi. Chern., 258, 18661870. Muller, F., Brustlein, M., Hemmerich, P., Massey, V. and Walker, W.H. (1972) Eur J Biochem., 25, 573-580. Rieger, P.H. (1994) Electrochemistry Second Edition. Chapman & Hall, New York, p.440. Rivera, M., Wells, M.A. and Walker, F.A. (1994) Biochemistry, 33,2161-2170. Salazar, D. (1998) Ph.D. dissertation, University of Colorado Health Sciences Center. Schmidt, J., Reinsch, J. and McFarland, J.T. (1981) J. Bioi. Chern., 256, 1166711670. Seager, S.L. and Slabaugh, M.R. (1987) Chemistry for Today General, Organic, and Biochemistry. West Publishing Company, St. Paul, Minnesota, pp. 750-754. Shaw, L., and Engel, P.C. (1984) J. Biochem., 218, 511-520. Shaw, L., and Engel, P.C. (1987) Biochim. Biophys. Acta., 919, 171-174. Steyn-Parve, E.P. and Beinert, H. (1958) J. Biol. Chern., 233, 843-852. 88

PAGE 101

Thorpe, C. and Kim, J.-J.P. (1995) FASEB J., 9, 718-725. Thorpe, C., Matthews, R.G. and Williams, C.H. (1979) Biochemistry., 18, 331-337. Thorpe, C., Ciardelli, T.L., Stewart, C.J. and T. Wieland (1981) Eur. J Biochem., 118, 279-282. Voet, D. and Voet, J. (1990) Biochemistry. John Wiley & Sons, New York, 621625. Williamson, G. and Engel, P.C. (1984) J Biochem., 218, 521-529. Zapien, D.C. (1998) Lecture. University of Colorado at Denver. 89