CHOLESTEROL OXIDASE MODIFIED
MICROELECTRODES FOR DETECTION OF
CHOLESTEROL IN THE PLASMA MEMBRANE OF
SINGLE CELLS
by
ANANDO DEVADOSS
Submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
Thesis Advisor: Dr. James David Burgess
Department of Chemistry
CASE WESTERN RESERVE UNIVERSITY
January, 2006 CASE WESTERN RESERVE UNIVERSITY
SCHOOL OF GRADUATE STUDIES
We hereby approve the dissertation of
______
candidate for the Ph.D. degree *.
(signed)______(chair of the committee)
______
______
______
______
______
(date) ______
*We also certify that written approval has been obtained for any proprietary material contained therein.
Dedicated to my parents and brother TABLE OF CONTENTS
TABLE OF CONTENTS ...... i
LIST OF FIGURES ...... v
LIST OF SCHEMES ...... xiii
LIST OF ABBREVIATIONS ...... xiv
ACKNOWLEDGEMENTS ...... xv
ABSTRACT ...... xvi
INTRODUCTION...... 1
CHAPTER 1. DETECTION OF CHOLESTEROL THROUGH ELECTRON
TRASFER TO CHOLESTEROL OXIDASE IN ELECTRODE-
SUPPORTED LIPID BILAYER MEMBRANE...... 4
1.1 INTRODUCTION...... 4
1.2 EXPERIMENTAL...... 8
1.3 RESULTS AND DISCUSSION...... 11
1.4 CONCLUSIONS...... 18
APPENDIX 1-1...... 20
APPENDIX 1-2...... 24
APPENDIX 1-3...... 28
APPENDIX 1-4...... 30
APPENDIX 1-5...... 32
1.5 REFERENCES...... 33
i CHAPTER 2. STEADY-STATE OXIDATION OF CHOLESTEROL
CATALYZED BY CHOLESTEROL OXIDASE IN LIPID BILAYER
MEMBRANES ON PLATINUM ELECTRODES...... 40
2.1 INTRODUCTION...... 40
2.2 EXPERIMENTAL...... 44
2.3 RESULTS AND DISCUSSION...... 48
2.4. CONCLUSIONS...... 60
2.5 ACKNOWLEDGEMENTS...... 61
2.6 REFERENCES...... 62
CHAPTER 3.ENZYME MODIFICATION OF PLATINUM
MICROELECTRODES FOR DETECTION OF CHOLESTEROL IN
VESICLE LIPID BILAYER MEMBRANES ...... 66
3.1 INTRODUCTION...... 66
3.2 EXPERIMENTAL SECTION...... 68
3.3 RESULTS AND DISCUSSION...... 71
3.4 CONCLUSIONS...... 86
3.5 ACKNOWLEDGEMENTS...... 86
3.6 REFERENCES...... 87
CHAPTER 4. 89STEADY STATE DETECTION OF CHOLESTEROL
CONTAINED IN THE PLASMA MEMBRANE OF SINGLE CELL
USING LIPID BILAYER MODIFIED MICROELECTRODES
INCORPORATING CHOLESTEROL OXIDASE...... 89
4.1 INTRODUCTION...... 89
ii 4.2 EXPERIMENTAL...... 90
4.3 RESULTS AND DISCUSSION...... 91
4.4 CONCLUSIONS...... 96
4.5 ACKNOWLEDGMENT...... 96
APPENDIX 4-1...... 97
4.6 REFERENCES...... 102
CHAPTER 5. 103COVALENT MODIFICATION OF CHOLESTEROL OXIDASE
TO PLATINUM MICROELECTRODES FOR DETECTION OF
CHOLESTEROL IN THE PLASMA MEMBRANE OF SINGLE
CELLS ...... 103
5.1 INTRODUCTION...... 103
5.2 EXPERIMENTAL...... 104
5.3 RESULTS AND DISCUSSION...... 105
5.4 CONCLUSION...... 110
5.5 ACKNOWLEDGEMENTS...... 111
5.6 REFERENCES...... 112
CHAPTER 6. SCANNING FORCE MICROSCOPY IMAGES OF
CHOLESTEROL OXIDASE IMMOBILIZED IN SUPPORTED
LIPID BILAYER MEMBRANE...... 113
6.1 INTRODUCTION...... 113
6.2 EXPERIMENTAL SECTION...... 115
6.3 RESULTS AND DISCUSSION...... 116
6.4 CONCLUSIONS...... 119
iii 6.5 ACKNOWLEDGEMENTS...... 119
6.6 REFERENCES...... 120
BIBLIOGRAPHY...... 122
iv LIST OF FIGURES
Figure1-1. Idealized structure of the electrode-supported lipid bilayer membrane
containing cholesterol oxidase...... 6
Figure 1-2. Dual chambered electrochemical dialysis cell. The dimensions of the cell are
1 cm x 5 cm x 5 cm. The electrode is mechanically clamped in the sample
chamber and the dialysis membrane is sandwiched between the flow
chamber and the sample chamber...... 9
Figure 1-3. A) Atomic force microscopy images of an indium tin oxide surface (contact
mode). B) A zoomed portion of the scan from A showing a cross-section.
...... 13
Figure 1-4. Cyclic voltammetry of 5mM potassium ferricyanide at A) bare ITO, B) ITO
modified with the thiolipid, and C) ITO after deposition of the outer lipid
leaflet...... 15
Figure 1-5. Amperometric responses for exposure of cholesterol to a cholesterol
oxidase modified electrode and to a lipid bilayer modified electrode
containing no cholesterol oxidase (control). The electrode area is 0.2 cm2.
...... 17
Figure A1-1. Linear sweep voltammograms showing the capacitance change upon
modification of the electrode with thiolipid and after deposition of the
outer lipid leaflet. A) voltammogram at bare ITO electrode B)
voltammogram at the ITO electrode after treating with ethanolic solution
v of thiolipid. C) voltammogram at the ITO electrode after the dialysis
procedure. All scans are taken in 6 mM KNO3 solution at 100 mV/s...... 28
Figure A1-2. Cyclic voltammetry of FAD indicating partitioning of FAD into the lipid
bilayer modified electrode. A) voltammogram at a lipid bilayer modified
ITO electrode. B) voltammogram at a lipid bilayer modified ITO electrode
containing FAD. All scans were taken in 50 mM sodium phosphate buffer
at a scan rate of 10 mV/s...... 30
Figure A1-3. Background subtracted linear sweep voltammograms of hydrogen peroxide
reduction at A) a bare ITO electrode, and B) a lipid bilayer modified ITO.
The solution was 1 mM hydrogen peroxide and 50 mM sodium phosphate
buffer. The scan rate is 2 mV/s...... 32
Figure 2-1. Idealized structure of the electrode-supported lipid bilayer membrane
containing cholesterol oxidase...... 42
Figure 2-2. Reaction sequence for oxidation of cholesterol as catalyzed by cholesterol
oxidase from Pseudomonas sp. Cholesterol oxidase catalyzes the two
electron oxidation of cholesterol through electron transfer to the FAD
prosthetic group of the enzyme, yielding FADH2. FAD is regenerated
through reduction of molecular oxygen to hydrogen peroxide. The enzyme
also catalyzes further oxidation of the intermediate cholest-5-en-3-one to
6β-hydroperoxycholest-4-en-3-one, the major initial product. Over time
the intermediate undergoes spontaneous conversion to more stable
products (Δ indicates spontaneous reaction, R is used to abbreviate the
remaining carbon skeleton that is unaffected by oxidation)...... 43
vi Figure 2-3. Schematic diagram of the thin-layer electrochemical cell as viewed from the
side and top. Dashed arrows represent flow of solution across the working
electrode surface. (A) Bulk platinum sheet working electrode (shown cut-
away in side view). (B) Platinum-plated auxiliary electrode and flow cell
housing. (C) Ag/AgCl/KCl (1 M) reference electrode. (D) Teflon gasket
that defines the shape and thickness of the fluid layer in contact with the
working and auxiliary electrodes...... 44
Figure 2-4. Amperometric response at a cholesterol oxidase-modified electrode to 100
μM cholesterol containing 50mM CD. (A) Trace of response for a
cholesterol oxidase-modified electrode. The downward arrow (↓) indicates
change of flow from cholesterol-free buffer to buffer containing
cholesterol. The upward arrow (↑) indicates change of flow back to
cholesterol-free buffer. (B) Trace of response of the same electrode before
immobilization of cholesterol oxidase (control experiment)...... 51
Figure 2-5. Amperometric response at a cholesterol oxidase-modified electrode for
exposure to aerobic (air saturated) cholesterol solution (A and C) and
anaerobic (nitrogen purged) cholesterol solution (B). The current for
reverting the flow to buffer containing no cholesterol is also shown (D).
The cholesterol concentration is 100 μM containing 50mM CD. The sharp
noise spikes are due to changing the injection valve setting...... 53
Figure 2-6. Amperometric responses at a cholesterol oxidase-modified electrode to
continuous flow exposure of various cholesterol concentrations containing
50mM CD. A downward arrow (↓) indicates change of flow from buffer to
vii cholesterol solution, and an upward arrow (↑) indicates change of flow
back to buffer. (A) 10 μM; (B) 30 μM; (C) 60 μM; (D) 20 μM; (E) 50
μM; (F) 40 μM; (G) 10 μM and (H) 30 μM cholesterol. The trace was
biased by −12.4 nA to zero the initial baseline...... 55
Figure 2-7. (A) Plot of steady-state current vs. cholesterol concentration showing
saturation behavior for an electrode on Day 2 of use. (B) Lineweaver–
Burk plot of data in Figure 2-7A. (C) Plot of steady-state current vs.
cholesterol concentration showing saturation behavior for an electrode on
Day 8 of use. (D) Lineweaver–Burk plot of data in Figure 2-7C. All
experiments were conducted using 50mM CD...... 58
Figure 2-8. Amperometric response at a cholesterol oxidase-modified electrode for
exposure to 100 μM cholesterol in buffered CD (A), buffered LDL
solution (B and C), 100 μM cholesterol (D), and buffered LDL solution
containing 50mM CD (E). The cholesterol concentration (not including
cholesterol esters) of the LDL solutions is 100 μM...... 59
Figure 3-1. Ferrocyanide voltammetry for deposition of the thiolipid/lipid bilayer
membrane containing cholesterol oxidase on a platinum electrode. Cyclic
voltammetry of potassium ferrocyanide at A) a bare platinum
microelectrode, B) after deposition of thiolipid, C) after deposition of
DPPC, and D) after incorporating cholesterol oxidase...... 74
Figure 3-2. Amperometric response obtained for an exposure of 25 μM cholesterol
solution at (A) an oxidase modified electrode (B) the same electrode with
viii thiolipid/lipid membrane prior to immobilization of the oxidase . The up
arrow (↑) indicates the times of cholesterol exposure...... 75
Figure 3-3. Amperometric responses obtained at an oxidase modified electrode for
exposure to cholesterol solution and sequential dilutions. The up arrow
(↑) indicates the time of cholesterol exposure and the down arrows (↓)
indicate the times of buffer dilution...... 77
Figure 3-4. Flow injection data at an oxidase modified electrode for exposure to three
different cholesterol concentrations. The up arrows (↑) indicate the times
of cholesterol injection and the down arrows (↓) indicate the times where
the flow is reverted to buffer...... 78
Figure 3-5. Photographs showing an electrode (A) positioned about 15 μm away from a
giant lipid vesicle and (B) contacting the giant lipid vesicle...... 79
Figure 3-6. Amperometric responses obtained at three different oxidase modified
electrodes for contacting three different giant lipid vesicles prepared with
0.5 cholesterol to phospholipid ratio. The up arrows (↑) indicate the times
of contact...... 80
Figure 3-7. Control experiments for (A) contacting a giant vesicle formed with no
cholesterol with an oxidase modified electrode and (B) for contacting a
giant vesicle with 0.5 cholesterol to phospholipid ratio with a bare
platinum electrode. The up arrows (↑) indicate the times of contact and
the down arrows (↓) indicate the times of withdrawal...... 81
Figure 3-8. Amperometric responses of an oxidase modified electrode for two
consecutive contacts at the same giant vesicle formed with 0.66
ix cholesterol to lipid ratio. The up arrows (↑) indicate the times of contact
and the down arrows (↓) indicate the times of withdrawal...... 82
Figure 3-9. Amperometric responses obtained at an oxidase modified electrode for
contacting three different vesicles containing cholesterol to lipid ratios of
(A) 0.66, (B) 0.5, and (C) 0.33. The arrow (↑) indicates the times of
contact...... 84
Figure 4-1. Photographs showing the electrode: (A) positioned about 5 μm from the
plasma membrane and (B) contacting the plasma membrane...... 91
Figure 4-2. Amperometric data for detection of cellular cholesterol at a microelectrode
(11.5 μm diameter) modified with a lipid bilayer membrane containing
cholesterol oxidase. No contact (Figure 4-1): baseline data; no cholesterol
detection. Adjacent: data for positioning the electrode within about 1 μm
of (or partially touching) the plasma membrane. Cell contact: data for
contacting the oocyte plasma membrane. Arrows approximate the times of
changing electrode position. The buffer is 0.1 M sodium phosphate, pH
6.5. The electrode potential is 800 mV vs NHE...... 92
Figure 4-3. Replicate experiments showing amperometric responses for contacting an
oocyte with a lipid bilayer modified microelectrode containing cholesterol
oxidase. Other conditions as in Figure 4-2...... 93
Figure 4-4. Amperometric responses for contacting an oocyte with a lipid bilayer
modified microelectrode containing cholesterol oxidase. Response (A)
prior to cholesterol depletion, (B) after partial cholesterol depletion, and
(C) after near complete cholesterol depletion. All data collected at the
x same oocyte using the same electrode. Other conditions as in Figure 4-2.
...... 94
Figure A4-1. Giant vesicle captured at the tip of a glass capillary. The picture also shows
small multilamellar vesicles attached to the giant vesicle...... 97
Figure A4-2. Response of an enzyme modified electrode for contacting a vesicle that
contains cholesterol as a membrane constituent (A). Response of an
enzyme modified electrode for contacting a vesicle that contains no
cholesterol (B). Other conditions as in Figure 4-2...... 98
Figure A4-3. Electrode response for contacting an oocyte with a lipid bilayer modified
electrode incubated in enzyme solution for (A) 1 hour and (B) the
response of the same electrode at the same cell after incubated in enzyme
solution for an additional 2 hours. Other conditions as in Figure 4-2...... 99
Figure A4-4. Response of an enzyme modified electrode for contacting an oocyte that
has been depleted of cholesterol (A). Response of the same electrode after
delivery of cholesterol to the cell plasma membrane (B). Other conditions
as in Figure 4-2...... 100
Figure 5-1. Cyclic voltammogram of 5 mM potassium ferrocyanide at A) a bare platinum
electrode with 100 μm diameter B) same electrode after modified with
MUA-11 and C) same electrode after modified with cholesterol oxidase.
Scan rate 50 mVs-1...... 106
Figure 5-2. Flow injection amperometry data for injection of various concentration of
cholesterol dissolved with Triton X-100. The (↑) shows the time when the
xi flow is changed from buffer to cholesterol and (↓) shows the time when
the flow is changed from cholesterol to buffer...... 107
Figure 5-3. Photographs showing cholesterol oxidase modified A) positioned ca.150 μm
away from oocyte plasma membrane B) contacting the plasma membrane
of oocyte...... 108
Figure 5-4. Amperometry data obtained when cholesterol oxidase modified electrode (A)
1) contacting the plasma membrane of the oocyte 2) contacting the plasma
membrane with additional force 3) withdrawn to an applied force closer to
postion 1 4) withdrawn away from the plasma membrane (B) Bare
platinum electrode 1) contacting the plasma membrane 2) contacting the
plasma membrane with additional force...... 109
Figure 6-1. TM-SFM images of lipid bilayer membrane modified mica surface A)
topographical image B) phase contrast image...... 116
Figure 6-2. TM-SFM images of cholesterol oxidase immobilized in lipid bilayer
membrane on mica surface A) topographical image B) phase contrast
image...... 117
xii LIST OF SCHEMES
Scheme 1-1. Reaction sequence at the cholesterol oxidase modified ITO electrode...... 7
Scheme 2-1. Detection scheme at cholesterol oxidase modified platinum electrode...... 52
xiii LIST OF ABBREVIATIONS
cm centimeter
EDC N-(3-Dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride
FIA flow injection analysis hr hour i current
ITO Indium Tin Oxide
M molar min. minute mM millimolar mm millimeter
MUA 11- mercapto undecanoic acid mV millivolt
NHE normal hydrogen electrode
Pt platinum
QCM quartz crystal microbalance s sec
TM-SFM Tapping mode-scanning force microscopy
μM micromolar
μm micrometer
xiv ACKNOWLEDGEMENTS
I would like to thank my Ph.D. advisor Prof. James D. Burgess for his
enthusiastic, inspirational and patient guidance. I remain grateful to Professors Daniel
Scherson and Barry Miller for their helpful discussions and for their constructive criticism during the group meetings. It is a pleasure to acknowledge and thank Prof.
Jonathan Smith, our collaborator for the development of this project and providing us with macrophages and related biological samples, Prof. Guy Chisolm for providing us with low density lipoproteins and Prof. Jianmin Cui for providing us with oocytes.
I would like to thank my colleagues Simona Palencsar , Dechen Jiang and Danjun
Fang for helping me with the experiments. Also acknowledged are the great help of the
undergraduate students, Michael Bokoch, Jonathan Ipsaro, Andrea Cendrowski and
Michael Honkonen. I would also like to thank of Dr. Cai for helping me with the
microelectrode setup Gayathri Krishnamoorthy in Prof. Cui’s lab for teaching me oocyte
protocol, and Dr. Ionel Stefan for his advice regarding the vesicle fusion method.
Thanks to the financial support from Department of Chemistry, Case Western
Reserve University, The Eveready Battery Company Inc., The Ernest B.Yeager Center
for Electrochemical Sciences, and National Institute of Health.
I am grateful for all the help and support I have received from my friends, seniors
and juniors during my stay in Cleveland.
This thesis is dedicated to my parents and my brother for their love and support.
xv Cholesterol Oxidase Modified Microelectrodes for Detection of Cholesterol in the
Plasma Membrane of Single Cells
ABSTRACT
BY
ANANDO DEVADOSS
Cholesterol oxidase was immobilized on electrode supported lipid bilayer
membranes in an active state. Lipid molecules with a thiol functionality were used to
form a sub-monolayer covalently linked to either indium tin oxide (ITO) or platinum
electrode surfaces. The outer leaflet of the lipid bilayer membrane was formed using a
deoxycholate dialysis method for modification of ITO electrodes. The vesicle fusion
method was used to deposit the outer lipid leaflet on platinum electrode surfaces.
Ferrocyanide/ferricyanide cyclic voltammetry was used to qualitatively monitor the
formation of lipid bilayer at the electrode surfaces. Amperometry was used to detect hydrogen peroxide produced during the enzymatic oxidation of cholesterol at room temperature. Flow injection analysis of solution phase cholesterol at oxidase modified
ITO, conventionally sized platinum electrodes, and platinum microelectrodes are reported. Cholesterol oxidase modified platinum electrodes showed Michaelis-Menten kinetic behavior for oxidation of solution phase cholesterol. The characterization studies using solution phase cholesterol at microelectrodes also suggested that the electrode response is limited by enzyme kinetics. Steady state current responses are obtained when
xvi the enzyme modified platinum microelectrodes are positioned in contact with the lipid membrane of giant vesicles containing cholesterol as a membrane constituent. The electrode responses correlated with the cholesterol content of the vesicle membrane where higher cholesterol content produces larger responses. Data obtained on contacting
Xenopus oocytes demonstrated detection of cholesterol present in the cell plasma membrane. When the membrane cholesterol was decreased by exposure to cyclodextrin solution, decreased amperometric response were observed. For studies at 37 °C, cholesterol oxidase was covalently linked to platinum microelectrodes modified with a sub-monolayer of 11- mercapto undecanoic acid (MUA) using N-(3-
Dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (EDC) as a cross-linker.
Covalently modified oxidase electrodes were used to detect cholesterol in the plasma membrane of oocytes and in the macrophage plasma membrane at 37 °C. Also reported are tapping mode – scanning force microscopy of cholesterol oxidase immobilized in a lipid membrane on mica. Topographic and phase contrast images are consistent with monomers and aggregates of cholesterol oxidase molecules immobilized in the lipid bilayer membrane.
xvii INTRODUCTION
Cholesterol is an amphipathic lipid molecule that is known to play essential roles
in cellular membrane structure and function. The cholesterol content of cellular
membranes is believed to be tightly regulated through vesicular trafficking pathways that
consume energy and alterations in cell plasma membrane cholesterol content are likely
involved in the onset of atherosclerosis. A better understanding of cholesterol
homeostasis can be attained if the cholesterol content of the plasma membrane is
monitored in a live cell under physiological conditions. Developing a direct
electrochemical method to track the cholesterol content of the plasma membrane of cells is a goal of this research group.
Cholesterol oxidase is a membrane associated enzyme that catalyzes the
oxidation of cholesterol by molecular oxygen. Chapter 1 details the initial studies
performed to evaluate the activity of cholesterol oxidase immobilized in a lipid bilayer
membrane on indium tin oxide electrodes. Cholesterol present in solution phase as a
cholesterol-cyclodextrin complex was extracted to the bilayer and detected at the oxidase
modified electrodes. One of the products of enzymatic oxidation of cholesterol, hydrogen
peroxide, was electrochemically reduced for electrochemical detection of cholesterol.
While studies at indium tin oxide showed that electrode supported lipid
bilayer membrane can be used to immobilize cholesterol oxidase in an active state, it
suffered from large background currents due to electrochemical reduction of oxygen. To
avoid the background current due to oxygen reduction, cholesterol oxidase was
immobilized on conventionally sized platinum electrodes modified with lipid bilayer
1 membrane. Hydrogen peroxide produced during enzymatic oxidation of solution phase cholesterol was electrochemically oxidized for detection and oxygen reduction did not occur. Chapter 2 discusses modification of platinum electrodes with cholesterol oxidase and the Michaelis-Menten kinetic behavior exhibited by these oxidase modified electrodes.
To detect cholesterol present in the plasma membrane of single cells, platinum microelectrodes were fabricated. A typical macrophage has a diameter of about
8 μm. Therefore, platinum microelectrodes with a diameter of ca. 10 μm were fabricated and modified with a lipid bilayer membrane containing cholesterol oxidase. Chapter 3 describes the fabrication of cholesterol oxidase modified microelectrodes and detection of cholesterol in the lipid membrane of giant lipid vesicles at room temperature. Giant lipid vesicles were used as model systems for cell plasma membrane. Giant vesicles prepared with higher cholesterol/lipid ratios lead to higher steady state current responses demonstrating the potential application of the microelectrodes as a single cell cholesterol sensor.
Cholesterol oxidase modified microelectrodes discussed in Chapter 3 were used to detect cholesterol present in the Xenopus oocyte (single cell). Chapter 4 discusses the current responses obtained on contacting the cell plasma membrane at room temperature. Most importantly the current responses correlated with the cholesterol content of the cell plasma membrane. The cholesterol content of the plasma membrane was altered by exposing the cells to cyclodextrin.
To achieve greater thermal stability of the electrode architecture so that experiments could be conducted at physiological temperatures, covalent attachment of
2 the enzyme to the platinum electrode was considered. Chapter 5 discusses work for cholesterol oxidase covalently linked to platinum electrodes. Platinum electrodes (100
μm diameter) covalently modified with cholesterol oxidase were used to detect cholesterol in the plasma membrane of oocyte. At physiological temperatures, these electrodes showed activity for detecting cholesterol present in the macrophage plasma membrane.
Exploratory research was conducted in order to understand the structure of lipid membrane formed by vesicle fusion at solid surfaces and the nature of cholesterol oxidase binding to the membrane. Tapping mode-scanning force microscopy (TM-SFM) studies of mica surfaces with cholesterol oxidase modified lipid membranes are presented in chapter 6. Topographic and phase contrast images are consistent with cholesterol oxidase molecules partially immersed in the lipid bilayer membrane.
3 CHAPTER 1.
DETECTION OF CHOLESTEROL THROUGH ELECTRON TRASFER TO
CHOLESTEROL OXIDASE IN ELECTRODE-SUPPORTED LIPID BILAYER
MEMBRANE.
1.1 INTRODUCTION.
Lipid bilayer membranes on solid supports have been the subject of numerous
publications,1-40 especially over the last decade.1-3,5,8,10-20,22,24-38,40 The focus of several
review articles is given in the appendix 1-1.1-10 Lipid bilayer membranes prepared on
various solids by optimized variations of the available deposition chemistries have been
shown to accommodate a variety of proteins and enzymes in controlled orientations and in active conformations. The supported bilayer membranes somewhat mimic the native environment of membrane-associated biomolecules. Another attractive feature is that the
resulting electrode architecture is thin (i.e., 50 Å) so that the incorporated biomolecules are close to both the electrode surface and to the membrane solution interface. The
electrode-supported lipid bilayer membranes prepared here for immobilization of cholesterol oxidase sequester cholesterol from aqueous solution because equilibration of aqueous phase cholesterol into the hydrophobic environment of the lipid bilayer is energetically favored.41,42
Lang et al.6,43 have used thiol functionalized lipid monolayers as the inner
lipid leaflet of bilayers formed on gold. In the surface structure reported here the inner
4 lipid monolayer is similarly fixed to the electrode surface through a thiol functionality at
the lipid head group. It is noted that this electrode modification step is required to
produce membranes that exhibit enzymatic activity (vide infra). The outer lipid leaflet is
deposited using the cholate dialysis procedure reported earlier for immobilizing
cytochrome c oxidase in electrode-supported lipid bilayer membranes.34-36,40 Because the
inner lipid monolayer is bound to the electrode surface via a thiol functionality prior to
dialysis, the resulting bilayer structure is also conceptually similar to the hybrid alkane
thiol/lipid bilayer structures described by Plant6 where vesicle spreading produces the outer lipid leaflet on the hydrophobic alkane thiol monolayer. The outer lipid leaflet of
such hybrid bilayer structures has been shown to be a reasonable model for the surface of
biological lipid bilayers in that membrane associated proteins can be immobilized in an
active state and are able to diffuse laterally in the outer lipid leaflet.6 The cholate dialysis
procedure used here reconstitutes cholesterol oxidase in the outer leaflet of the membrane
as it is formed (i.e., the enzyme and the outer lipid leaflet are deposited simultaneously).
Lipid bilayer membranes are effective collectors of aqueous phase cholesterol.41,42
Lipid bilayer vesicles have been used to stimulate cholesterol efflux from the lipid monolayer of lipoproteins and from the plasma lipid bilayer of cells. A primary mechanism of cholesterol efflux from “donor” lipid membranes (e.g., lipoproteins and cell plasma membranes) to “acceptor” vesicle lipid bilayer membranes is aqueous diffusion. For a solution (or suspension) of cholesterol donor particles, the equilibrium between cholesterol in the lipid membrane and aqueous phase cholesterol lies largely towards membrane resident cholesterol. Addition of vesicles containing no cholesterol to the donor particle solution results in cholesterol efflux from the donor particles as
5 aqueous cholesterol is sequestered by the vesicles.41,42 Based on this background, it is
reasonable to assume that the electrode-supported lipid bilayer membranes prepared here
collect aqueous phase cholesterol from solution.
Cholesterol oxidase is a flavin adenosine dinucleotide (FAD) containing enzyme
(i.e., flavoenzyme) that catalyzes the isomerization and oxidation of cholesterol.
Molecular oxygen is reduced by the FAD moiety generating cholestenone (oxidized cholesterol) and hydrogen peroxide.44 The crystal structures of the oxidase isolated from
Brevibacterium sterolicum and Streptomyces species indicate dimensions of ca. 7.3 nm
×5.3nm×5.1nm and 5.1nm×7.3nm×6.3nm for the monomers, respectively, with the FAD
group buried in the hydrophobic interior of the polypeptide.45,46 The enzyme is widely
used for assessing the cholesterol content of cell culture samples.44 Generally, a
spectrophotometric detection scheme is used where a dye is produced upon enzymatic
generation of hydrogen peroxide. The model for enzymatic oxidation of cholesterol in the plasma membrane of cells has the oxidase inserted in the lipid bilayer through
S- = thiolipid S- = outer lipid S-
Pt S- Cholesterol
S- Oxidase
S- Indium Tin Oxide Tin Indium
S-
Figure1-1. Idealized structure of the electrode-supported lipid bilayer membrane
containing cholesterol oxidase.
6 hydrophobic interactions such that cholesterol movement occurs directly from the plasma
membrane into the enzyme pocket. This model has been proposed by Sampson and
coworkers based on kinetic studies for treatment of vesicles containing cholesterol with
the oxidase enzyme.47-49 Acrylodan (fluorophore) labeled enzyme were produced using site directed mutagenesis to probe the depth of protein insertion into the lipid bilayer
membrane.48 Further support for this model is given by studies involving the treatment of
cells with cholesterol oxidase. Complete lysis of mid-gut epithelial cells50 and
morphological changes in the plasma membrane of smooth red muscle cells51 have been
reported on exposure to the enzyme. Also, the activity of cholesterol oxidase has been
shown to be dependent on the cell type suggesting significant interaction between the
enzyme and the cell surface.52
In the present work, a method is described for forming lipid bilayer membranes
containing cholesterol oxidase on tin-doped indium oxide (ITO) electrodes.53 The
procedure produces cholesterol oxidase modified electrodes that show stable enzymatic activity for days. The model of the electrode architecture (Figure 1-1) has the enzyme partially inserted in the outer leaflet of the bilayer as has been proposed by others for the interaction of the enzyme with the plasma membrane of cells.47 Our proposed mechanistic model for detection of cholesterol at the cholesterol oxidase modified electrode is energetically favorable equilibration of cholesterol into the electrode- supported lipid bilayer membrane from aqueous solution followed by lateral diffusion of
Membrane: Cholesterol + O2 Æ Cholestenone + HOOH + - Electrode: HOOH + 2H +2e Æ 2H2O
Scheme 1-1. Reaction sequence at the cholesterol oxidase modified ITO electrode.
7 cholesterol within the bilayer to enzyme sites. Though no evidence for lateral diffusion
of cholesterol is presented in this report, lateral diffusion of species within the outer leaflet of hybrid bilayer membranes has been shown.54-56 Lateral diffusion coefficient
values for lipids and steroids in monolayers lie in the range of 10-7 to 10-8 cm2/s. The
lateral diffusion in a solid supported lipid bilayer is dependent on the phase of the bilayer, the solid support, and the type of monolayer that is attached to the surface.3,4,56,57 The
outer lipid monolayer in this system is expected to be present in a liquid crystalline state
under the experimental conditions. It is noted that the cholesterol content affects the
rigidity of lipid bilayer membranes.41 Hydrogen peroxide generated by the enzyme is
electrochemically reduced at the ITO electrode surface for detection of cholesterol.
Scheme 1 shows the enzyme catalyzed reaction and the electrode reaction.
The method described here builds on other strategies for preparing electrodes
modified with cholesterol oxidase58-71 by demonstrating the potential advantages
discussed above of using an electrode-supported lipid bilayer membrane as the host for
immobilization of the enzyme. A chronological history and brief descriptions of the studies that have been published on the topic of electrochemical detection of cholesterol
are given in the appendix 1-2. Many groups have reported electrodes that respond
linearly with cholesterol concentration and the lifetimes of the various electrodes range
from hours to months.
1.2 EXPERIMENTAL.
The cyclic voltammetric and amperometric experiments were conducted using a
Bioanalytical Systems CV-50 potentiostat. The atomic force microscope images were
8 acquired using a Molecular Imaging picoscanTM instrument. The ITO (on glass)
(Information Products, Inc.) electrodes were cleaned by successive sonication in aqueous
Alconox for 60 min., in ethanol for 60 min., and in water for 60 min. 0.01 M sodium phosphate aqueous solution adjusted to a pH 6.5 by addition of sodium hydroxide is used as buffer . Cholesterol (Sigma) solutions were prepared by dissolving cholesterol in a
100 mM hydroxypropyl β-cyclodextrin (Cerestar USA, Inc.), 0.01 M sodium phosphate
(Sigma) pH 6.5 solution. The cyclodextrin is used to solubilize cholesterol.72
The inner lipid leaflet was formed by sonicating the electrode in an ethanolic
Figure 1-2. Dual chambered electrochemical dialysis cell. The dimensions of the cell are
1 cm x 5 cm x 5 cm. The electrode is mechanically clamped in the sample
chamber and the dialysis membrane is sandwiched between the flow
chamber and the sample chamber.
9 solution of the 1,2-dipalmitoyl-sn-glycero-3-phosphothioethanol (Avanti Polar Lipids
Inc.) for 90 mins at 50o C. The dual chambered electrochemical dialysis cell (Figure 1-2)
and dialysis conditions have been described earlier.35 One exception is that the dialysis
solution contained 0.3 mM DOPE (1,2-dioleoyl-sn-glycero-3-phosphoethanolamine)
(Avanti Polar Lipids) (earlier work35 used a mixture of DOPE and dioleoyl
phosphatidylcholine). In brief, the dual chambered electrochemical dialysis cell is
composed of two separate pieces of Lucite®. One forms the sample chamber including
the wall jet inlet, and reference and auxiliary electrodes. The other piece forms the flow
chamber. The ITO electrode is clamped to the sample chamber and a dialysis membrane
is sandwiched between the two cell pieces separating the sample chamber from the flow
chamber. The sample chamber is filled with deoxycholate (40.2 mM), lipid, and the
cholesterol oxidase enzyme (Wako Pure Chemical Industries, Ltd.). Buffer is passed through the flow chamber to remove deoxycholate from the sample chamber. This procedure drives the formation of the outer lipid leaflet containing the oxidase enzyme on the inner lipid leaflet on the ITO electrode. Dialysis was performed for 12 hours and the sample chamber of the dialysis cell was then flushed with buffer for 12 hours at a flow rate of 10 μl/min to remove remaining deoxycholate, lipids, and enzyme from the sample chamber. Continuous sample flow experiments for exposure of the electrodes to cholesterol were conducted using a six-way valve and two syringe pumps (one containing buffer and the other a buffered cholesterol solution) at a flow rate of 0.450 ml/min.
Cholesterol exposure times that allowed the current responses to approach a steady state value were controlled manually. The electrode potential was 120 mV vs. NHE and the
reference electrode was silver/silver chloride (1 M KCl). At this electrode interface, this
10 potential is sufficient to reduce hydrogen peroxide and it does not result in large oxygen
reduction background currents. All potentials are reported vs. NHE.
1.3 RESULTS AND DISCUSSION.
Contact mode atomic force microscopic (AFM) images of the bare ITO surface
(Figure 1-3) show a topography that closely resembles other reported images of ITO on
glass.73,74 However, it is noted that these structures can vary.73,74 Relatively flat plateaus
with lateral dimensions on the order of tens of nanometers are observed. It is
hypothesized that the bilayer may form on these regions with some two dimensional
continuity. The regions of the electrodes surface containing steep edges may not contain
a complete lipid layer coverage. Hydrogen peroxide produced by the enzyme reaction
may be electrochemically reduced at the potentially bare regions of the electrode surface
(vide infra).
The inner leaflet of the bilayer is attached to the ITO surface presumably through sulfur
metal bonds. Alkane thiols have been shown to form thiolate bonds at ITO surfaces.73,75
However, it is not known if the thiolate bond is with the tin, indium or both metal centers.
The fact that hydrogen peroxide generated by the enzyme is electrochemically reduced at these electrode surfaces (vide infra) suggests that the inner lipid monolayer is not tightly packed or that it contains a substantial defect density (i.e., portions of the ITO surface are not covered by the lipid layer). The voltammetry of potassium ferricyanide before
(Figure 1-4A) and after deposition of the inner lipid leaflet (Figure 1-4B) indicate deposition of the thiol functionalized lipids. The increase in the peak potential separation indicates a slower heterogeneous electron transfer rate. Slowed electron transfer kinetics
11 contribute to the smaller peak currents. It is possible that regions of the electrode surface
are completely blocked from ferricyanide in solution. Voltammograms conducted in 1
mM potassium nitrate indicate an electrode capacitance for the bare indium oxide surface
of ca. 22 μF/cm2 decreasing to 17 μF/cm2 after formation of the inner lipid monolayer
(Appendix 1-3). The lipids may be dispersed evenly over the surface with a relatively
large nearest neighbor distance. Alternately, regions of the surface could be covered by
more tightly packed domains of lipid (i.e., islands of lipid on the plateau regions of the
surface). In either case, the regions of the surface that are modified with the thiol functionalized lipid do evidently provide a hydrophobic template so that the second leaflet of the bilayer containing cholesterol oxidase is deposited during the cholate dialysis step. It is noted that performing the dialysis step on bare ITO electrodes (the ITO is not pretreated with the thiol functionalized lipid) does not produce interfaces that exhibit enzymatic activity upon exposure to cholesterol.
12
Figure 1-3. A) Atomic force microscopy images of an indium tin oxide surface (contact
mode). B) A zoomed portion of the scan from A showing a cross-section.
13 Cholesterol oxidase and presumably the outer leaflet of the bilayer are deposited on the lipid monolayer modified ITO surface during dialysis. It is also possible that the dialysis procedure results in deposition of additional lipids with the polar head groups associated with the hydrophilic ITO surface. This could occur at regions of the electrode surface that are not initially modified with thiolipid. This possibility would be consistent with the model for bilayer deposition on submonolayer coverages of alkanethiol on silver where deoxycholate dialysis was used to reconstitute cyctochrome c oxidase.34-36,40 The
ferricyanide characterization studies (Figure 1-4C) show a further decrease of the
electron transfer rate and possibly increased blocking after dialysis. However, the
procedure apparently does not result in lipid deposition over the entire electrode surface
as indicated by the voltammetric waves shown in Figure 1-4C. The electrode
capacitance is also further decreased by 31 % after dialysis (Appendix 1-3). The
structure of the lipid membrane containing the enzyme is not known. However, the data
could reflect formation of a lipid bilayer structure on regions of the electrode that initially
contained a coverage of the thiol functionalized lipid. The possibility that lipid
multilayer aggregates are formed on regions of the electrode surface cannot be ruled out.
Tapping-mode atomic force microscopy did not reveal the existence of such structures.
This indicates that either the structures do not exist or that the stability of the structures is
insufficient to allow imaging under the experimental conditions (i.e. images resemble that
shown in Figure 1-3). Again, electrochemical reduction of the generated hydrogen
peroxide may occur at regions on the electrode surface that are not modified with the
bilayer. Control experiments for hydrogen peroxide reduction for this system in the
absence of cholesterol also show smaller currents after deposition of the bilayer
14 compared to bare ITO (appendix 1-5). Additional evidence for bilayer deposition is given by voltammetric data that are consistent with partitioning of FAD (hydrophobic probe molecule) into the electrode supported membrane (appendix 1-4).
Amperometry using a wall jet configuration under continuous sample flow
conditions was used to detect cholesterol at the cholesterol oxidase modified electrodes.
Figure 1-5 shows the current measured for changing the flow from buffer to buffer
containing cholesterol and for changing the flow back to buffer. Exposure of three
different cholesterol concentrations to the electrode is shown where the last exposure is a
replicate of the first. The data are consistent with electrochemical reduction of hydrogen
90 A 70
50 B
30 C
10
-10 current (mA) -30 C -50 B
-70 A -90 -280 -80 120 320 520 720 920 potential (mV)
Figure 1-4. Cyclic voltammetry of 5mM potassium ferricyanide at A) bare ITO, B) ITO
modified with the thiolipid, and C) ITO after deposition of the outer lipid
leaflet.
15 peroxide generated by the enzyme upon exposure to cholesterol. The relatively slow
approach to a steady state current is attributed to non-ideal wall jet hydrodynamics for the
cell geometry used76 (an initial experiment using a thin layer electrochemical cell showed
a much faster approach to steady state, e.g., 20 s). The current responses depend on the cholesterol concentration with the higher concentrations yielding larger responses (Figure
1-5). Control experiments for lipid bilayer modified electrodes that contain no
cholesterol oxidase do not show current responses upon exposure to cholesterol (Figure
1-5, control). As expected, cholesterol exposure experiments conducted under reduced
oxygen concentrations (i.e., purged solutions) show ca. 40% smaller current responses.
These data are consistent with cholesterol oxidase being immobilized on the ITO surface
in an active state and with the proposed reactions shown in Scheme 1.
The responses are reproducible (within 10%) for 3-5 days at a given electrode
however, the lifetime of the electrodes has not been fully characterized. The current for
injection of a given cholesterol concentration varies between electrodes. The data shown in Figure 1-5 were collected at the electrode that yielded the largest responses for exposure to cholesterol. A different electrode that yielded the smallest responses observed (data not shown) showed currents that were ca. 30% of those shown in Figure
1-5. The differences in response between electrodes are likely due to variations in the rate of hydrogen peroxide reduction between electrodes and/or disparity in the amount of oxidase immobilized. The applied potential (120 mV) does not reduce hydrogen peroxide at a mass transfer controlled rate at these electrode surfaces (appendix 1-5).
While the limited
16 data presented for the cholesterol concentration dependence do not suffice for evaluation of mass transport and kinetically controlled regimes, the data do suggest that the electrode-supported lipid bilayer effectively sequesters cholesterol that exists in solution as a cyclodextrin-cholesterol complex 72 (see Experimental). These attributes may allow these electrodes to extract cholesterol from the lipid monolayer of lipoproteins (e.g., low density lipoproteins, LDL). At the applied potential, steady state oxygen reduction current (e.g. 4 nA) is observed during the flow experiments. For data presentation the nonzero baseline current due to oxygen reduction was subtracted from the data shown in
Figure 1-5.
The reaction catalyzed by the enzyme is a two electron oxidation of cholesterol and a two electron reduction of molecular oxygen producing hydrogen peroxide which is
100 μM 100 μM 4
50 μM
2 25 μM current (nA) current
0 control
0 500 1000 1500 2000 time (s)
Figure 1-5. Amperometric responses for exposure of cholesterol to a cholesterol
oxidase modified electrode and to a lipid bilayer modified electrode
containing no cholesterol oxidase (control). The electrode area is 0.2 cm2.
17 detected at the electrode. Some of the generated hydrogen peroxide could diffuse out of
the lipid membrane into solution and this may be a function of the electrode preparation.
Detection of cholesterol by electrochemical oxidation of the generated hydrogen peroxide
would eliminate nonzero baseline currents due to oxygen reduction. However, mass
transfer controlled hydrogen peroxide oxidation requires potentials as positive as ca. 1.32
V at ITO electrodes in this media and experiments for detection of cholesterol by
electrochemical oxidation of hydrogen peroxide have so far not yielded reproducible
current responses. Evidence for accumulation of cholestenone (oxidized cholesterol) in
the bilayer inhibiting responses for sequential cholesterol exposures has not been
observed (Figure 1-5). Lipid bilayer membranes containing cholesterol oxidase formed
on platinum electrodes are currently being investigated. Using platinum as the substrate
allows electrochemical oxidation of the generated hydrogen peroxide.77-82 Another approach to overcome kinetic and structural inconsistencies between electrode preparations (described above) is to implement a coulometric detection scheme for quantitative analysis of cholesterol in a known volume of sample.
1.4 CONCLUSIONS.
This method for immobilizing cholesterol oxidase in electrode-supported lipid bilayer membranes allows detection of cholesterol through electrochemical reduction of hydrogen peroxide generated by the enzyme. The data shows that the enzyme remains active upon immobilization in the lipid bilayer. The data also suggest that cholesterol
18 partitions into the lipid bilayer membrane from solution due to hydrophobic interactions between cholesterol and the lipid tails.
19 APPENDIX 1-1
DISCUSSION ON SOLID SUPPORTED LIPID BILAYER MEMBRANES
The area of lipid bilayer membranes on solid supports has been the subject of
numerous publications,1-40 especially over the last decade. 1-3,5,8,10-20,22,24-38,40 The
reviews by Guidelli et al.,1 Sinner et al.,2 Sackmann,3 and Tiede4 discuss a wide range of
lipid bilayers formed on various solid supports, the physical properties of the membranes,
and studies characterizing the interactions of proteins with the membranes. A review by
Plant focuses on hybrid alkane thiol/lipid membranes where a lipid monolayer is anchored to the surface through hydrophobic interactions between the lipid acyl tails and the methyl terminus of a self-assembled monolayer.5,6 Gold, silver, and mercury have
been used as the support and capacitance measurements on these systems are common.
This review details the vesicle fusion method of depositing the lipid monolayer on the
hydrophobic alkane thiol surface. Studies concerning lateral diffusion of proteins
adsorbed to or inserted within the lipid monolayer surface are presented (e.g. Ubiquinone
diffusion in alkanethiol/phospholipid bilayers with a diffusion coefficient of 1.7 x 10-8 cm2 s-1). Reviews by Tien et al. and Ottova et al. discuss the formation of self-assembled
lipid bilayer membranes on freshly cut metal surfaces.7-9 Here hydrophilic interactions between the polar lipid head groups and the hydrophilic metal surface result in deposition of a layer of lipids with the acyl tail groups extending normal to the surface plane. The
outer leaflet of the bilayer forms due to hydrophobic interactions analogous to the alkane
thiol/lipid membranes discussed above. A review by Wagner et al. concentrates on the
formation and characterization of tethered polymer-supported planar lipid bilayer
20 systems.10 In many of these structures the lipid bilayer is supported by the hydration layer
of the polymer support. The reports by Steinem et al.11 and Ding et al.12 present a
comparison of different preparation techniques for forming supported lipid bilayer
membranes. Groves et al. have prepared patterned arrays of fluid membrane corrals on
oxidized silicon substrates by fusion of small unilamellar vesicles.13 Parikh et al. have
formed alkyksiloxane/lipid hybrid membranes at oxidized silicon substrates using
Langmuir-Blodgett techniques.14 Ross et al have used a polymerizable lipid to form a
self-assembled lipid bilayer membrane on silicon dioxide surfaces that were stable in the
presence of surfactants and organic solvents.15 A lipid bilayer with the acyl tails of the
two bilayer leaflets intercalated has been reported by Kanzaki et al. using layered
synthetic mica as the substrate.16 Wu et al. have recently shown that polarization of
glassy carbon electrodes at higher positive potentials (e.g., 1720 mV vs. NHE) results in
the self-assembly of a lipid bilayer membrane under their solution conditions.17 Gao et al. have fabricated self-assembled lipid bilayer membranes on ITO surfaces and have characterized the photoelectric properties of the membranes.18 Hillebrandt et al. have
prepared polymer/lipid composite films on ITO electrodes and this report focuses on the
electrical properties of the films.19 In a recent report Wiegand et al. studied the electrical
properties of lipid bilayer membranes on gold and ITO prepared by both vesicle fusion
and a solvent exchange method.20 Huang reported the dependence of lipid concentration
on the formation of lipid membrane layers on glass using a dialysis method.21 He also studied the binding of H2K glycoproteins to the supported bilayers.
Lipid bilayer membranes prepared on various solids by optimized variations of the available deposition chemistries have been shown to accommodate a variety of
21 proteins and enzymes in controlled orientations and in an active conformations. Salamon
et al. have reviewed studies that involve lipid bilayer membranes containing proteins
probed by surface plasmon resonance spectroscopy.22 Nakanishi reported a freeze-thaw
method to reconstitute a transmembrane protein(human glycophorin A) into supported
lipid bilayer on glass cover slips.23 Salamon et al. reported the electrochemical reduction
of horse heart cytochrome c at a platinum electrode modified with a lipid bilayer
membrane which also accomodated a hydrophobic mediator (vinyl ferrocene).24 In subsequent reports this group has reported direct electrochemistry of spinach plastocyanin, thioredoxins, and glutathione at lipid bilayer modified gold electrodes.
They also reported direct electron transfer of two cytochromes (cytochrome f and cytochrome c oxidase) at ITO electrodes using lipid bilayers as the host for immobilization of the enzymes.25-27 Fischer et al, have formed a hydrid antibody/lipid
membrane on alkylated gold surfaces (alkane thiol) using the Langmuir Blodgett technique.28 Here the binding site of a monoclonal antibody binds the head group(s) of
phospholipids. Bianco et al. have studied the electron transfer reactions between c-type
cytochromes and lipid modified electrodes.29,30 Zhang et al. have used lipid film modified
pyrolytic graphite electrodes to immobilize cytochrome P450 cam and direct electron
transfer between the enzyme and the electrode is reported.31 Pierrat et al. have
incorporated pyruvate oxidase in an hybrid alkane thiol/lipid bilayer on gold electrodes.32
A mediator (ferrocene methanol) was incorporated into the bilayer to establish electrochemical communication between the electrode and the oxidase. This allowed electrochemical detection of pyruvate from solution through enzymatic oxidation.
Naumann et al. have incorporated H+ -ATPase into supported lipid bilayer separated from
22 the gold electrode by a peptide spacer.33 They studied the coupling of proton translocation through the incorporated ATPase using square wave volatmmetry and double potential-pulse chronoamperometry. The Hawkridge group has conducted voltammetric studies of cytochrome c oxidase immobilized in lipid bilayer membranes on gold electrodes using cholate dialysis.34,35,40 Amperometry of enzyme mediated electron
transfer from cytochrome c (the enzymes native redox partner) in solution to the
electrode has been used to follow transitions of the oxidase between two distinct kinetic
states (resting and pulsed). Scanning force microscopy images support the proposed structural model having the oxidase partially inserted in the lipid bilayer membrane mimicking its orientation within the inner mitochondrial membrane.36 Edmiston et al have discussed the molecular orientation distributions in cytochrome c films immobilized on phospholipids bilayers.37,38 They have immobilized cytochrome c on pyridyldisulfide-
capped bilayers and on an streptavidin layer on a planar supported phopholipid bilayer
formed on fused quartz slides, and on silica-titania surfaces. Dolfi et al. deposited
bacteriorhodopsin on hybrid alkanethiol/lipid bilayers on mercury in their recent study.39
The electrode was used to study the kinetics of light driven proton transport by the protein. Electrochemical characterizations of the alkanethiol/phospholipid bilayers formed on mercury have been reported by Buoninsegni et al.83
23 APPENDIX 1-2
DISCUSSION ON CHOLESTEROL OXIDASE MODIFIED ELECTRODE
SYSTEMS
A cholesterol assay based on an electrochemical detection scheme was reported in
1977 by Satoh et al. Cholesterol oxidase was immobilized in a collagen matrix on a
platinum electrode.58 Because oxygen is consumed by the enzyme only in the presence
of cholesterol, it was possible to monitor the decrease in oxygen reduction current at the
electrode upon exposure to cholesterol. The electrochemical response was linear up to
0.2 mM. Another group demonstrated that this system responded to cholesterol for
months.59 Subsequent work involved enzyme immobilization in a octyl-agarose gel
trapped in a glass reactor.60 Hydrogen peroxide produced during the enzymatic reaction
was electrochemically oxidized at the platinum electrode (down stream) in a solution
flow experiment. This system also showed a linear response vs. cholesterol concentration
o (2.6 - 10.4 mM) and the electrodes were stable for one month when stored at 4 C. In
1983, Wollenberger et al. covalently immobilized cholesterol oxidase on 2- hydroxyethylmethacrylate gel particles using glutaraldehyde.61 A suspension of the
particles was fixed to the platinum electrode using a membrane (silk cloth) and the assay
relied on the oxidation of generated hydrogen peroxide. In this system, detection involved
electrochemical reduction of ferricyanide generated through oxidation of ferrocyanide by
hydrogen peroxide. This group also showed that horseradish peroxidase could be co-
immobilized for catalytic hydrogen peroxide reduction through direct electron transfer
from the electrode to the peroxidase enzyme. With this detection scheme a linear
24 response vs. cholesterol concentration (0.4-12 mM ) was reported and the electrode was
stable for 10 days. In 1985 Masoom et al reported the preparation of a column of
controlled porosity glass containing immobilized cholesterol oxidase.62 Hydrogen
peroxide produced by the enzymatic reaction was amperometrically detected at an
oxidizing potential downstream of the column. The difference in activity of the enzyme for various sterols was demonstrated using this strategy. In 1991 Kajiya et al. incorporated the enzyme and a mediator (ferrocenecarboxylate) in a polypyrrole film on a platinum electrode.63 The electrode was stable for 20 measurements and showed a linear
response up to 0.05 mM. In 1993, Motonaka et al. prepared a carbon microelectrode
modified with cholesterol oxidase using a porous carbon composite with a Nafion™
coating.64 The enzyme was adsorbed on the surface of the Nafion and an osmium
bipyridyl complex was loaded in the carbon composite for mediation. Linearity of the
electrochemical response vs cholesterol concentration (0.02 - 1.4 mM) was demonstrated
and a lifetime for the electrode of 2-3 months was reported. Covalent immobilization of
cholesterol oxidase on gold electrodes was achieved in 1999 by Nakaminami et al.
Glutaraldehyde was used to cross-link enzyme molecules forming a thin film on a self-
assembled monolayer of 2-aminoethanethiolate.65 The amperometric response was
obtained with thionin present as a mediator in the membrane (and in solution). The
stability of the electrode ranged from 40 to 50 hrs, and was dependent on the storage conditions. Kumar et al. have co-immobilized cholesterol oxidase and horseradish peroxidase in a tetraethyl orthosilicate derived sol-gel film on an indium tin oxide electrode.66 Cyclic voltammetric studies for the oxidation of hydrogen peroxide (at 970 mVvs NHE) with varying cholesterol concentration were described and a calibration
25 curve for cholesterol concentrations ranging 2-10 mM is given. The electrode was stable
for about 8 weeks at 25o C and 12 weeks at 4-5o C. In 2001, Venkatajalabathy et al.
reported the co-immobilization of cholesterol oxidase and a peroxidase enzyme.71 A
layer-by-layer film of cholesterol oxidase and poly (styrenesulfonate) was prepared on a
monolayer of peroxidase covalently-immobilized on gold. A self-assembled monolayer
of alkanethiolates (3-mercaptopropionic acid, 2-aminoethanethiol) was used to link the
peroxidase to the gold electrode. Direct electron transfer to the peroxidase enzyme from
the gold electrode was used in the detection scheme. Ropers et al. immobilized cholesterol oxidase on gold and glassy carbon electrodes using liquid crystal matrices composed of monoolein and a fluorinated surfactant (11-perflourohexyl-9-thio-3,6- dioxaundecanol).67 Electrochemical oxidation of hydrogen peroxide during cyclic
voltammetry experiments was used to demonstrate a linear response for cholesterol
concentrations (0.01 - 5 mM). Bongiovanni et al. immobilized cholesterol oxidase and
horseradish peroxidase on pyrolytic graphite electrodes using a tributylmethyl
phosphonium chloride polymer.68 Direct electron transfer from the electrode to the
peroxidase enzyme allowed amperometric detection of hydrogen peroxide. This electrode
was stable for more than a week. A quasi-linear relationship between the current response
and cholesterol concentration (0.04 - 0.27 mM) was obtained. Ram et al. have reported a cholesterol oxidase modified electrode prepared by a layer-by-layer technique using the
polyelectroytes, poly(styrene sulfonate) and poly(ethylene imine) deposited on quartz,
ITO and platinum surfaces.69 Electrochemical oxidation of hydrogen peroxide was used
in the detection scheme. Linear responses for cholesterol (0.05 to 1 mM) were reported.
Vidal et al. have prepared a cholesterol biosensor with two sequentially
26 electrosynthesized polymer layers on platinized platinum electrodes.70 The polymer bilayer contains an inner polypyrrole layer hosting cholesterol oxidase and an outer poly(o-phenylenediamine) layer. The polymer bilayer was used to decrease the electrochemical response due to the interferences (e.g. ascorbic acid) present in blood.
The electrode was shown to have a lifetime of one month and a linear range of 0.012-0.35 mM.
27 APPENDIX 1-3
CAPACITANCE DATA FOR FORMATION OF THE BILAYER LIPID
MEMBRANE
The voltammetric studies were conducted using a Bioanalytical systems CV-50 potentiostat. All the potentials are reported with respect to NHE. The geometric area of the electrode is 0.38 cm2. The capacitance values were calculated from the cathodic scans (920 - 720 mV vs NHE), using the relation,84 C = i/νA Where C = capacitance in F/
Cm2, i = current in Amperes, A = Area in cm2 and ν = scan rate in V/s.
0
-0.2
C
A) -0.4 μ B
current ( current -0.6
-0.8 A
-1 720 770 820 870 920 potential (mV)
Figure A1-1. Linear sweep voltammograms showing the capacitance change upon
modification of the electrode with thiolipid and after deposition of the
outer lipid leaflet. A) voltammogram at bare ITO electrode B)
voltammogram at the ITO electrode after treating with ethanolic solution
of thiolipid. C) voltammogram at the ITO electrode after the dialysis
procedure. All scans are taken in 6 mM KNO3 solution at 100 mV/s.
28 The capacitance data were estimated from the linear sweep voltammogram
(Figure A1-1) taken in 6 mM KNO3 solution. The electrode capacitances (averaged from
data taken at 740 mV, 780 mV and 820 mV) are 21.4 μF/cm2 for the bare ITO electrode,
14 μF/cm2 after modification with thiolipid, and 9.7 μF/cm2 after deposition of the outer
lipid leaflet. Lower electrode capacitance has been reported for a lipid bilayer formed on
a thiolipid inner leaflet on gold (eg. 0.93 μF/cm2).43 However, our percentage decrease
(ca. 31%) observed for deposition of the outer leaflet is consistent with other reports (34
% decrease for deposition of a POPC outer leaflet on a thiolipid inner layer on gold, 37% for a DOPC outer leaflet on a tetradecanethiol inner layer on mercury).83 The actual surface area is not known with certainty and our absolute capacitance values cannot be compared with those obtained for other electrodesupported lipid bilayer membranes.
The decrease in capacitance after treatment of the electrode with thiolipid and after dialysis qualitatively indicates deposition of lipids on the electrode.
29 APPENDIX 1-4
ADSORPTION OF FAD IN THE LIPID BILAYER MEMBRANE
The data for these experiments were collected at an ITO electrode modified with a lipid bilayer membrane containing no cholesterol oxidase. Figure A1-2 shows the voltammograms taken at the bilayer modified indium tin oxide electrode and at the same electrode after being immersed in a FAD solution for 2 hrs removed from the solution and extensively rinsed. The increase in the reduction current (negative of -300 mV) after
2
0
-2 A A)
μ -4
-6 current ( -8 B
-10
-12 -500 -300 -100 100 300 potential (mV)
Figure A1-2. Cyclic voltammetry of FAD indicating partitioning of FAD into the lipid
bilayer modified electrode. A) voltammogram at a lipid bilayer modified
ITO electrode. B) voltammogram at a lipid bilayer modified ITO electrode
containing FAD. All scans were taken in 50 mM sodium phosphate buffer
at a scan rate of 10 mV/s.
30 partitioning of FAD into the lipid bilayer membrane is attributed to a one or two electron
reduction of FAD.85 Because FAD is largely hydrophobic, the data suggest that there is a
hydrophobic region (i.e. the core of the lipid bilayer membrane) on the electrode surface
where FAD is stabilized. This result is in line with other studies showing partitioning of
hydrophobic probe molecules (e.g. vinyl ferrocene, C60) into supported lipid bilayer membranes.9,18,24
31 APPENDIX 1-5
HYDROGEN PEROXIDE REDUCTION AT THE LIPID BILAYER MODIFIED
ELECTRODE
A comparison of electrochemical reduction of 1 mM hydrogen peroxide (scan rate, 2 mV/s) between the bilayer modified electrode and the bare electrode (Figure A1-3) shows that the reactions are not mass-transfer limited at 120 mV vs NHE. The bilayer modified electrode shows smaller current at 120 mV compared (ca. 71%) to the bare
electrode.
1
0
-1 B A)
μ -2
-3 current ( -4 A -5
-6 -100 -50 0 50 100 150 200 250 potential (mV)
Figure A1-3. Background subtracted linear sweep voltammograms of hydrogen peroxide
reduction at A) a bare ITO electrode, and B) a lipid bilayer modified ITO.
The solution was 1 mM hydrogen peroxide and 50 mM sodium phosphate
buffer. The scan rate is 2 mV/s.
32 1.5 REFERENCES.
(1) Guidelli, R.; Aloisi, G.; Becucci, L.; Dolfi, A.; Moncelli, M. R.; Tadini Buoninsegni, F. "Bioelectrochemistry at metal|water interfaces." J. Electroanal. Chem. 2001, 504, 1-28
(2) Sinner, E.-K.; Knoll, W. "Functional tethered membranes." Curr Opin Chem Biol 2001, 5, 705-711
(3) Sackmann, E. "Supported membranes: scientific and practical applications." Science (Washington, D. C.) 1996, 271, 43-8
(4) Tiede, D. M. "Incorporation of membrane proteins into interfacial films: model membranes for electrical and structural characterization." Biochim Biophys Acta 1985, 811, 357-79
(5) Plant, A. L. "Supported Hybrid Bilayer Membranes as Rugged Cell Membrane Mimics." Langmuir 1999, 15, 5128-5135
(6) Plant, A. L. "Self-assembled phospholipid/alkanethiol biomimetic bilayers on gold." Langmuir 1993, 9, 2764-7
(7) Tien, H. T.; Salamon, Z. "Formation of self-assembled lipid bilayers on solid substrates." Bioelectrochemistry and Bioenergetics 1989, 22, 211-18
(8) Ottova, A. L.; Ti Tien, H. "Self-assembled bilayer lipid membranes: from mimicking biomembranes to practical applications." Bioelectrochemistry and Bioenergetics 1997, 42, 141-152
(9) Tien, H. T.; Ottova, A. L. "Supported planar lipid bilayers (s-BLMs) as electrochemical biosensors." Electrochim. Acta 1998, 43, 3587-3610
(10) Wagner, M. L.; Tamm, L. K. "Tethered polymer-supported planar lipid bilayers for reconstitution of integral membrane proteins: silane-polyethyleneglycol-lipid as a cushion and covalent linker." Biophys J 2000, 79, 1400-1414
(11) Steinem, C.; Janshoff, A.; Ulrich, W.-P.; Sieber, M.; Galla, H.-J. "Impedance analysis of supported lipid bilayer membranes: a scrutiny of different preparation techniques." Biochim Biophys Acta 1996, 1279, 169-80
(12) Ding, L.; Li, J.; Dong, S.; Wang, E. "Supported phospholipid membranes: comparison among different deposition methods for a phospholipid monolayer." J. Electroanal. Chem. 1996, 416, 105-112
(13) Groves, J. T.; Ulman, N.; Boxer, S. G. "Micropatterning fluid lipid bilayers on solid supports." Science (Washington, D. C.) 1997, 275, 651-653
33 (14) Parikh, A. N.; Beers, J. D.; Shreve, A. P.; Swanson, B. I. "Infrared Spectroscopic Characterization of Lipid-Alkylsiloxane Hybrid Bilayer Membranes at Oxide Substrates." Langmuir 1999, 15, 5369-5381
(15) Ross, E. E.; Bondurant, B.; Spratt, T.; Conboy, J. C.; O'Brien, D. F.; Saavedra, S. S. "Formation of Self-Assembled, Air-Stable Lipid Bilayer Membranes on Solid Supports." Langmuir 2001, 17, 2305-2307
(16) Kanzaki, Y.; Hayashi, M.; Minami, C.; Inoue, Y.; Kogure, M.; Watanabe, Y.; Tanaka, T. "Intercalation and bilayer formation of phospholipids in layered synthetic mica. 2. Solvent effect of the intercalation reaction of natural and reduced-type phosphatidylcholines." Langmuir 1997, 13, 3674-3680
(17) Wu, Z.; Wang, B.; Cheng, Z.; Yang, X.; Dong, S.; Wang, E. "A facile approach to immobilize protein for biosensor: self-assembled supported bilayer lipid membranes on glassy carbon electrode." Biosens Bioelectron 2001, 16, 47-52
(18) Gao, H.; Luo, G. A.; Feng, J.; Ottova, A. L.; Ti Tien, H. "Fabrication and photoelectric properties of self-assembled bilayer lipid membranes on conducting glass." J. Photochem. Photobiol., B 2000, 59, 87-91
(19) Hillebrandt, H.; Wiegand, G.; Tanaka, M.; Sackmann, E. "High Electric Resistance Polymer/Lipid Composite Films on Indium-Tin-Oxide Electrodes." Langmuir 1999, 15, 8451-8459
(20) Wiegand, G.; Arribas-Layton, N.; Hillebrandt, H.; Sackmann, E.; Wagner, P. "Electrical properties of supported lipid bilayer membranes." J. Phys. Chem. B 2002, 106, 4245-4254
(21) Huang, L. "Incorporation of acylated antibody into planar lipid multilayers: characterization and cell binding." Biochemistry 1985, 24, 29-34
(22) Salamon, Z.; Macleod, H. A.; Tollin, G. "Surface plasmon resonance spectroscopy as a tool for investigating the biochemical and biophysical properties of membrane protein systems. II: Applications to biological systems." Biochim Biophys Acta 1997, 1331, 131-152
(23) Nakanishi, M. "Rapid reconstitution of a transmembrane protein into supported planar lipid membranes." FEBS Lett 1984, 176, 385-8
(24) Salamon, Z.; Tollin, G. "Reduction of cytochrome c at a lipid bilayer modified electrode." Bioelectrochemistry and Bioenergetics 1991, 25, 447-54
(25) Salamon, Z.; Gleason, F. K.; Tollin, G. "Direct electrochemistry of thioredoxins and glutathione at a lipid bilayer-modified electrode." Arch Biochem Biophys 1992, 299, 193-8
34 (26) Salamon, Z.; Tollin, G. "Direct electrochemistry of spinach plastocyanin at a lipid bilayer-modified electrode: cyclic voltammetry as a probe of membrane-protein interactions." Arch Biochem Biophys 1992, 294, 382-7
(27) Salamon, Z.; Hazzard, J. T.; Tollin, G. "Direct measurement of cyclic current- voltage responses of integral membrane proteins at a self-assembled lipid-bilayer- modified electrode: cytochrome f and cytochrome c oxidase." Proc Natl Acad Sci U S A 1993, 90, 6420-3
(28) Fischer, B.; Heyn, S. P.; Egger, M.; Gaub, H. E. "Antigen binding to a pattern of lipid-anchored antibody binding sites measured by surface plasmon microscopy." Langmuir 1993, 9, 136-40
(29) Bianco, P.; Haladjian, J. "Electrochemistry of cytochrome c3 at a lipid-modified graphite electrode." J. Electroanal. Chem. 1994, 367, 79-84
(30) Bianco, P.; Haladjian, J. "Control of the electron transfer reactions between c-type cytochromes and lipid-modified electrodes." Electrochim. Acta 1994, 39, 975-82
(31) Zhang, Z.; Nassar, A.-E. F.; Lu, Z.; Schenkman, J. B.; Rusling, J. F. "Direct electron injection from electrodes to cytochrome P450cam in biomembrane-like films." J. Chem. Soc., Faraday Trans. 1997, 93, 1769-1774
(32) Pierrat, O.; Lechat, N.; Bourdillon, C.; Laval, J.-M. "Electrochemical and surface plasmon resonance characterization of the step-by-step self-assembly of a biomimetic structure onto an electrode surface." Langmuir 1997, 13, 4112-4118
(33) Naumann, R.; Jonczyk, A.; Hampel, C.; Ringsdorf, H.; Knoll, W.; Bunjes, N.; Graeber, P. "Coupling of proton translocation through ATPase incorporated into supported lipid bilayers to an electrochemical process." Bioelectrochemistry and Bioenergetics 1997, 42, 241-247
(34) Cullison, J. K.; Hawkridge, F. M.; Nakashima, N.; Yoshikawa, S. "A Study of Cytochrome c Oxidase in Lipid Bilayer Membranes on Electrode Surfaces." Langmuir 1994, 10, 877-82
(35) Burgess, J. D.; Rhoten, M. C.; Hawkridge, F. M. "Cytochrome c Oxidase Immobilized in Stable Supported Lipid Bilayer Membranes." Langmuir 1998, 14, 2467-2475
(36) Burgess, J. D.; Jones, V. W.; Porter, M. D.; Rhoten, M. C.; Hawkridge, F. M. "Scanning force microscopy images of cytochrome c oxidase immobilized in an electrode-supported lipid bilayer membrane." Langmuir 1998, 14, 6628-6631
(37) Edmiston, P. L.; Saavedra, S. S. "Molecular Orientation Distributions in Protein Films. 4. A Multilayer Composed of Yeast Cytochrome c Bound through an Intermediate Streptavidin Layer to a Planar Supported Phospholipid Bilayer." J Am Chem Soc 1998, 120, 1665-1671
35 (38) Edmiston, P. L.; Saavedra, S. S. "Molecular orientation distributions in protein films: III. Yeast cytochrome c immobilized on pyridyl disulfide-capped phospholipid bilayers." Biophys J 1998, 74, 999-1006
(39) Dolfi, A.; Tadini-Buoninsegni, F.; Moncelli, M. R.; Guidelli, R. "Photocurrents Generated by Bacteriorhodopsin Adsorbed on Thiol/Lipid Bilayers Supported by Mercury." Langmuir 2002, 18, 6345-6355
(40) Rhoten, M. C.; Burgess, J. D.; Hawkridge, F. M. "The reaction of cytochrome c from different species with cytochrome c oxidase immobilized in an electrode supported lipid bilayer membrane." J. Electroanal. Chem. 2002, 534, 143-150
(41) Phillips, M. C.; Johnson, W. J.; Rothblat, G. H. "Mechanisms and consequences of cellular cholesterol exchange and transfer." Biochim Biophys Acta 1987, 906, 223-76
(42) Rothblat, G. H.; De la Llera-Moya, M.; Atger, V.; Kellner-Weibel, G.; Williams, D. L.; Phillips, M. C. "Cell cholesterol efflux: integration of old and new observations provides new insights." J Lipid Res 1999, 40, 781-796
(43) Lang, H.; Duschl, C.; Vogel, H. "A new class of thiolipids for the attachment of lipid bilayers on gold surfaces." Langmuir 1994, 10, 197-210
(44) MacLachlan, J.; Wotherspoon, A. T. L.; Ansell, R. O.; Brooks, C. J. W. "Cholesterol oxidase: sources, physical properties and analytical applications." J Steroid Biochem Mol Biol 2000, 72, 169-195
(45) Vrielink, A.; Lloyd, L. F.; Blow, D. M. "Crystal structure of cholesterol oxidase from Brevibacterium sterolicum refined at 1.8 .ANG. resolution." J Mol Biol 1991, 219, 533-54
(46) Yue, Q. K.; Kass, I. J.; Sampson, N. S.; Vrielink, A. "Crystal structure determination of cholesterol oxidase from Streptomyces and structural characterization of key active site mutants." Biochemistry 1999, 38, 4277-4286
(47) Sampson, N. S.; Kass, I. J.; Ghoshroy, K. B. "Assessment of the Role of an W Loop of Cholesterol Oxidase: A Truncated Loop Mutant Has Altered Substrate Specificity." Biochemistry 1998, 37, 5770-5778
(48) Chen, X.; Wolfgang, D. E.; Sampson, N. S. "Use of the Parallax-Quench Method to Determine the Position of the Active-Site Loop of Cholesterol Oxidase in Lipid Bilayers." Biochemistry 2000, 39, 13383-13389
(49) Ghoshroy, K. B.; Zhu, W.; Sampson, N. S. "Investigation of membrane disruption in the reaction catalyzed by cholesterol oxidase." Biochemistry 1997, 36, 6133- 6140
36 (50) Greenplate, J. T.; Duck, N. B.; Pershing, J. C.; Purcell, J. P. "Cholesterol oxidase: An oostatic and larvicidal agent active against the cotton boll weevil, Anthonomus grandis." Entomol Exp Appl 1995, 74, 253-8
(51) Liu, K.; Maddaford, T. G.; Ramjiawan, B.; Kutryk, M. J. B.; Pierce, G. N. "Effects of cholesterol oxidase on cultured vascular smooth muscle cells." Mol Cell Biochem 1991, 108, 39-48
(52) Crockett, E. L.; Hazel, J. R. "Sensitive assay for cholesterol in biological membranes reveals membrane-specific differences in kinetics of cholesterol oxidase." J Exp Zool 1995, 271, 190-5
(53) Devadoss, A.; Burgess, J. D. "Detection of cholesterol through electron transfer to cholesterol oxidase in electrode-supported lipid bilayer membranes." Langmuir 2002, 18, 9617-9621
(54) Fahey, P. F.; Koppel, D. E.; Barak, L. S.; Wolf, D. E.; Elson, E. L.; Webb, W. W. "Lateral diffusion in planar lipid bilayers." Science (Washington, DC, U. S.) 1977, 195, 305-6
(55) Stroeve, P.; Miller, I. "Lateral diffusion of cholesterol in monolayers." Biochim Biophys Acta 1975, 401, 157-67
(56) Gyoervary, E.; Wetzer, B.; Sleytr, U. B.; Sinner, A.; Offenhaeusser, A.; Knoll, W. "Lateral Diffusion of Lipids in Silane-, Dextran-, and S-Layer-Supported Mono- and Bilayers." Langmuir 1999, 15, 1337-1347
(57) Lewis, R. N. A. H.; Sykes, B. D.; McElhaney, R. N. "Thermotropic phase behavior of model membranes composed of phosphatidylcholines containing cis- monounsaturated acyl chain homologs of oleic acid: differential scanning calorimetric and phosphorus-31 NMR spectroscopic studies." Biochemistry 1988, 27, 880-7
(58) Satoh, I.; Karube, I.; Suzuki, S. "Enzyme electrode for free cholesterol." Biotechnol Bioeng 1977, 19, 1095-9
(59) Bertrand, C.; Coulet, P. R.; Gautheron, D. C. "Enzyme electrode with collagen- immobilized cholesterol oxidase for the microdetermination of free cholesterol." Anal Lett 1979, 12, 1477-88
(60) Karube, I.; Hara, K.; Matsuoka, H.; Suzuki, S. "Amperometric determination of total cholesterol in serum with use of immobilized cholesterol esterase and cholesterol oxidase." Anal Chim Acta 1982, 139, 127-32
(61) Wollenberger, U.; Kuehn, M.; Scheller, F.; Deppmeyer, V.; Jaenchen, M. "Amperometric enzyme sequence electrodes for cholesterol." Bioelectrochemistry and Bioenergetics 1983, 11, 307-17
37 (62) Masoom, M.; Townshend, A. "Determination of cholesterol by flow injection analysis with immobilized cholesterol oxidase." Anal Chim Acta 1985, 174, 293-7
(63) Kajiya, Y.; Tsuda, R.; Yoneyama, H. "Conferment of cholesterol sensitivity on polypyrrole films by immobilization of cholesterol oxidase and ferrocenecarboxylate ions." Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 1991, 301, 155-64
(64) Motonaka, J.; Faulkner, L. R. "Determination of cholesterol and cholesterol ester with novel enzyme microsensors." Anal Chem 1993, 65, 3258-61
(65) Nakaminami, T.; Ito, S.-i.; Kuwabata, S.; Yoneyama, H. "Amperometric Determination of Total Cholesterol at Gold Electrodes Covalently Modified with Cholesterol Oxidase and Cholesterol Esterase with Use of Thionin as an Electron Mediator." Anal Chem 1999, 71, 1068-1076
(66) Kumar, A.; Malhotra, R.; Malhotra, B. D.; Grover, S. K. "Co-immobilization of cholesterol oxidase and horseradish peroxidase in a sol-gel film." Anal Chim Acta 2000, 414, 43-50
(67) Ropers, M.-H.; Bilewicz, R.; Stebe, M.-J.; Hamidi, A.; Miclo, A.; Rogalska, E. "Fluorinated and hydrogenated cubic phases as matrices for immobilisation of cholesterol oxidase on electrodes." Phys. Chem. Chem. Phys. 2001, 3, 240-245
(68) Bongiovanni, C.; Ferri, T.; Poscia, A.; Varalli, M.; Santucci, R.; Desideri, A. "An electrochemical multienzymatic biosensor for determination of cholesterol." Bioelectrochemistry 2001, 54, 17-22
(69) Ram, M. K.; Bertoncello, P.; Ding, H.; Paddeu, S.; Nicolini, C. "Cholesterol biosensors prepared by layer-by-layer technique." Biosens Bioelectron 2001, 16, 849-856
(70) Vidal, J.-C.; Garcia-Ruiz, E.; Castillo, J.-R. "Design of a multilayer cholesterol amperometric biosensor for preparation and use in flow systems." Electroanalysis 2001, 13, 229-235
(71) Vengatajalabathy Gobi, K.; Mizutani, F. "Layer-by-layer construction of an active multilayer enzyme electrode applicable for direct amperometric determination of cholesterol." Sens. Actuators, B 2001, B80, 272-277
(72) Haynes, M. P.; Phillips, M. C.; Rothblat, G. H. "Efflux of Cholesterol from Different Cellular Pools." Biochemistry 2000, 39, 4508-4517
(73) Yan, C.; Zharnikov, M.; Goelzhaeuser, A.; Grunze, M. "Preparation and Characterization of Self-Assembled Monolayers on Indium Tin Oxide." Langmuir 2000, 16, 6208-6215
38 (74) Donley, C.; Dunphy, D.; Paine, D.; Carter, C.; Nebesny, K.; Lee, P.; Alloway, D.; Armstrong, N. R. "Characterization of Indium-Tin Oxide Interfaces Using x-ray Photoelectron Spectroscopy and Redox Processes of a Chemisorbed Probe Molecule: Effect of Surface Pretreatment Conditions." Langmuir 2002, 18, 450- 457
(75) Kondo, T.; Takechi, M.; Sato, Y.; Uosaki, K. "Adsorption behavior of functionalized ferrocenylalkane thiols and disulfide onto Au and ITO and electrochemical properties of modified electrodes: effects of acyl and alkyl groups attached to the ferrocene ring." J. Electroanal. Chem. 1995, 381, 203-9
(76) Elbicki, J. M.; Morgan, D. M.; Weber, S. G. "Theoretical and practical limitations on the optimization of amperometric detectors." Anal Chem 1984, 56, 978-85
(77) Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part 1. An adsorption-controlled mechanism." Electrochim. Acta 1997, 43, 579-588
(78) Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part II: effect of potential." Electrochim. Acta 1998, 43, 2015-2024
(79) Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part IV: phosphate buffer dependence." Electrochim. Acta 1999, 44, 4573-4582
(80) Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part III: effect of temperature." Electrochim. Acta 1999, 44, 2455-2462
(81) Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part V: inhibition by chloride." Electrochim. Acta 2000, 45, 3573-3579
(82) Bokoch, M. P.; Devadoss, A.; Palencsar, M. S.; Burgess, J. D. "Steady-state oxidation of cholesterol catalyzed by cholesterol oxidase in lipid bilayer membranes on platinum electrodes." Analytica Chimica Acta 2004, 519, 47-55
(83) Buoninsegni, F. T.; Herrero, R.; Moncelli, M. R. "Alkanethiol monolayers and alkanethiol|phospholipid bilayers supported by mercury: an electrochemical characterization." J. Electroanal. Chem. 1998, 452, 33-42
(84) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 1980.
(85) Niemz, A.; Imbriglio, J.; Rotello, V. M. "Model Systems for Flavoenzyme Activity: One- and Two-Electron Reduction of Flavins in Aprotic Hydrophobic Environments." J Am Chem Soc 1997, 119, 887-892
39 CHAPTER 2.
STEADY-STATE OXIDATION OF CHOLESTEROL CATALYZED BY
CHOLESTEROL OXIDASE IN LIPID BILAYER MEMBRANES ON PLATINUM
ELECTRODES
2.1 INTRODUCTION.
In the earlier chapter we have demonstrated that cholesterol oxidase can be
immobilized in a functional conformation by lipid bilayer membranes on ITO.1 In this initial chapter, turnover of cholesterol was monitored by electrochemical reduction of hydrogen peroxide generated by the enzyme. This detection scheme suffered from nonzero baseline currents due to reduction of molecular oxygen at the electrode surface.
This chapter describes the use of platinum electrodes that allow hydrogen peroxide to be electrochemically oxidized, thus avoiding a large background current from oxygen reduction. This step also regenerates molecular oxygen, which oxidizes cholesterol in the enzyme-catalyzed reaction. Other research groups have demonstrated electrochemical oxidation of hydrogen peroxide at platinum as a detection scheme for reactions catalyzed by cholesterol oxidase2-9, as well as other oxidase enzymes.10-12 This chapter reports the use of the supported lipid bilayer membrane to sequester cholesterol from solution and immobilize cholesterol oxidase near the electrode surface. This electrode architecture exhibits steady-state responses allowing characterization of enzyme activity under
discrete levels of lipid bilayer membrane cholesterol content.
The idealized model of the electrode architecture with cholesterol oxidase partially inserted in the thiolipid/lipid bilayer membrane is shown in Figure 2-1. The
40 enzyme immobilization scheme is supported by structural studies of cholesterol oxidase,
as well as reports indicating that the enzyme associates with lipid bilayer vesicles and cell
plasma membranes during oxidation of membrane-resident cholesterol.13-18
It is proposed that cholesterol partitions into the electrode-supported lipid bilayer
membrane from solution. This is consistent with literature describing the use of lipid
bilayer vesicles as solution phase cholesterol acceptors in cellular cholesterol efflux
experiments.19,20 Based on these studies, it is reasonable to assume that cholesterol
partitions into the electrode-supported lipid bilayer membrane from solution. Lateral
diffusion of cholesterol (diffusion coefficient of 10−8 cm2 s−1 at 20 ◦C in a bilayer of 6:4
dipalmitoylphosphatidylcholine:cholesterol)21-23, and possibly of the enzyme within the
membrane24, allows binding between cholesterol and the enzyme. Enzymatic oxidation of
cholesterol proceeds with the two electron reduction of the cholesterol oxidase flavin
adenine dinucleotide (FAD) prosthetic group to FADH2. Most cholesterol oxidases
catalyze both the oxidation of cholesterol to an intermediate (cholest-5-en-3-one) and the
subsequent isomerization to cholest-4-en-3-one 25. However, the cholesterol oxidase
produced by Pseudomonas sp. (the enzyme used in the work reported here) yields
different products. As shown in Figure 2-2, the Pseudomonas enzyme produces 6β- hydroperoxycholest-4-en-3-one as the major initial product of cholesterol oxidation
(Figure 2-2) 26,27. This intermediate spontaneously reacts to form two stable products,
cholest-4-ene-3,6-dione and 6 β-hydroxycholest-4-en-3-one. FAD, the redox active
center of cholesterol oxidase, is regenerated through reduction of molecular oxygen to
hydrogen peroxide.
41 Various detergents and surfactants have been used to solubilize cholesterol during
kinetic experiments, and all have been shown to affect the activity of cholesterol
oxidase.28 A number of Michaelis–Menten constants (Km) have been reported for cholesterol oxidase in solution, ranging from 3.0 to 800 μM.7,14,29-34 The values for Km
vary depending on the genera of bacterial enzyme used, the method used to measure
reaction rate (including electrochemical methods 7,34), pH, ionic strength, and the
concentration and type of detergent used for the studies. Sampson et al. 14 have reported
initial rates for the cholesterol oxidase-catalyzed turnover of cholesterol contained in
vesicle membranes. The rate of cholesterol oxidation in various cell plasma membranes
upon exposure to cholesterol oxidase has also been investigated.35 However, there is
apparently no literature that reports Km for cholesterol oxidase associated with a lipid
bilayer or cell plasma membrane (the enzymes native environment).
S- = thiolipid S- = outer lipid S-
Pt S- Cholesterol
S- Oxidase
S-
S-
Figure 2-1. Idealized structure of the electrode-supported lipid bilayer membrane
containing cholesterol oxidase.
42 Continuous solution flow experiments allow the cholesterol content of the reported initial rates for the cholesterol oxidase-catalyzed turnover of cholesterol contained in vesicle membranes. The rate of cholesterol oxidation in various cell plasma membranes upon exposure to cholesterol oxidase has also been investigated.35 However, there is apparently no literature that reports Km for cholesterol oxidase associated with a lipid bilayer or cell plasma membrane (the enzymes native electrode-supported lipid bilayer membrane to be held constant during steady-state enzymatic turnover of cholesterol. The anodic current observed at the electrode provides a measure of the steady-state rate of enzymatic turnover under discrete cholesterol concentrations. This is
Figure 2-2. Reaction sequence for oxidation of cholesterol as catalyzed by cholesterol
oxidase from Pseudomonas sp. Cholesterol oxidase catalyzes the two
electron oxidation of cholesterol through electron transfer to the FAD
prosthetic group of the enzyme, yielding FADH2. FAD is regenerated
through reduction of molecular oxygen to hydrogen peroxide. The enzyme
also catalyzes further oxidation of the intermediate cholest-5-en-3-one to
6β-hydroperoxycholest-4-en-3-one, the major initial product. Over time the
intermediate undergoes spontaneous conversion to more stable products (Δ
indicates spontaneous reaction, R is used to abbreviate the remaining carbon
skeleton that is unaffected by oxidation).
43 possible because the current responses are reasonably stable (e.g., over 2 h) during continuous flow of cholesterol solution. The cholesterol concentration dependence of steady-state current is analyzed with respect to Michaelis–Menten kinetic behavior. Data are presented for exposure of low density lipoprotein solution to the electrode.
2.2 EXPERIMENTAL.
Electrochemical cell. Amperometric data shown in Figure 2-4, 2-6–2-8 were collected in
Figure 2-3. Schematic diagram of the thin-layer electrochemical cell as viewed from the
side and top. Dashed arrows represent flow of solution across the working
electrode surface. (A) Bulk platinum sheet working electrode (shown cut-
away in side view). (B) Platinum-plated auxiliary electrode and flow cell
housing. (C) Ag/AgCl/KCl (1 M) reference electrode. (D) Teflon gasket that
defines the shape and thickness of the fluid layer in contact with the working
and auxiliary electrodes.
44 a thin-layer type electrochemical flow cell. The cell (Figure 2-3) was constructed using a
modified liquid chromatography flow-injection apparatus (BAS, Model CC-5E). The
working electrode (0.56 cm2) is a bulk platinum sheet mounted in the flow-injection
housing (Figure 2-3A). The region of the cell housing that is adjacent to the working electrode is plated with platinum and serves as the auxiliary electrode (Figure 2-3B). The reference electrode (Ag/AgCl/KCl (1 M), Figure 2-3C) is threaded into a cavity
downstream of the working electrode. A Teflon gasket (30 μm thick) sandwiched between the platinum surfaces (Figure 2-3D) defines the volume of the fluid layer in contact with the working electrode (ca. 17 μl). Liquid enters and exits the cell through two openings in the auxiliary electrode (Figure 2-3, dashed arrows).
The platinum working electrode was polished to a mirror finish using 1 μm
diamond grit followed by rinsing, and sonicating for 10–15 min first in ethanol and then
in water. The polished platinum was heated to red hot in a hydrogen flame and quenched
in deionized water to remove adsorbed organic material from the electrode surface.
Injection lines and cell components were flushed with ethanol and water before each set
of experiments. Data shown in Figure 2-5 for nitrogen purged solution were collected
using a wall jet flow cell described elsewhere 36 which allowed a higher solution flow
rate (0.5 ml/min). The working electrode (0.2 cm2) was bulk platinum and the electrode
cleaning procedure was that described above.
Lipid membrane preparation. Chloroform stock solutions of 1,2-dipalmitoyl-sn- glycero-3-phosphothioethanol(thiolipid) and 1,2-dipalmitoylsn-glycero-3-phosphocholine
(DPPC) were obtained from Avanti Polar Lipids. In each case the stock was dried under
45 nitrogen to evaporate the chloroform solvent. Lipids were dissolved in 100% ethanol and
again dried under nitrogen to remove any trace chloroform. Thiolipid was dissolved in
ethanol to a final concentration of 5.5 mM. This relatively high thiolipid concentration
was used to minimize the time required to modify the electrode with a monolayer of
thiolipid. DPPC lipid vesicles were prepared according to a procedure reported by
Plant.24 In brief, dried DPPC (2 μmoles) was dissolved in isopropanol (50 μl) and this
solution was diluted with buffer (50mM sodium phosphate, pH 6.5) to yield to a final
DPPC concentration of 2 mM. The lipid membrane was formed on the platinum working
electrodes by injecting lipid solution directly into the flow cell. The cell was filled with
thiolipid (5.5 mM) and incubated for 2 h to form a self-assembled monolayer of thiolipid
on the platinum working electrode (i.e., the inner lipid leaflet of the electrode-supported
lipid bilayer). The cell was subsequently flushed with buffer. The DPPC vesicle solution
(2mM DPPC) was injected into the cell and incubated for 2 h to deposit the outer leaflet
of the lipid bilayer membrane by vesicle spreading.24 Electrode capacitance was
calculated from cyclic voltammograms recorded in buffer (E = +420 to +620mV versus
NHE, ν = 10 mV/s).
Cholesterol solution preparation. Solutions of hydroxypropyl-cyclodextrin (Cargill) in
sodium phosphate buffer (50 mM, pH 6.5) were used to prepare all cholesterol solutions.
Cyclodextrin (CD) is a large amphipathic molecule (molecular weight ca. 1400–1600
g/mol; 1600 g/mol was assumed in preparing CD solutions) that increases the solubility
of cholesterol in aqueous solution.37 A 20mM ethanolic stock solution of 5-cholesten-3-ol
(cholesterol) was prepared from powdered reagent (Sigma). The cholesterol stock was
46 then dried under a stream of nitrogen and re-dissolved in 50mM CD to a cholesterol
concentration of 100 μM. For higher cholesterol concentrations, a 2mM cholesterol
solution was first prepared in 100mM CD. This stock was then diluted to 1mM
cholesterol and 50mM CD by adding an equal volume of buffer. Lower cholesterol
concentrations were prepared by serial dilution from 100 μM and 1mM stocks. Buffered
CD (50 mM) was used for all cholesterol dilutions. Human low density lipoprotein (LDL)
was obtained from the Cleveland Clinic Foundation protein bank through collaboration with Dr. Guy Chisolm. LDL (6020 μg/ml total cholesterol) was prepared in buffer and in buffered CD.
Amperometry. Two syringe pumps (Harvard Apparatus) were connected to the flow cell
through a six-way valve to allow the flow to be alternated between buffer and cholesterol
solution. One pump contained a given concentration of cholesterol in buffer. The other pump contained only buffer (no cholesterol). The flow rate for both pumps was calibrated to a true rate of 88 μl/min, at which rate the volume of solution in the cell is exchanged
over about 5 s. The working electrode was held at +822mV versus NHE. This potential is
sufficient to oxidize H2O2 at a mass transfer controlled rate at platinum (data not shown).
Before each cholesterol exposure a baseline current was established by flowing
cholesterol-free buffer through the cell. After a stable baseline was observed the flow was
changed to cholesterol solution for a minimum of 12–15 min and then changed back to
buffer. Cholesterol exposures were performed before (control) and after immobilization
of cholesterol oxidase in the electrode-supported lipid membrane. Electrode responses
exhibited a rise to a steady-state plateau current for each cholesterol concentration. The
47 reported numerical values for steady-state current were obtained by averaging data points
in the plateau regions. The range of data averaged was consistent for all exposures and
was defined as beginning at the first visible maximum(on a given plateau) and ending at
the time when flow was changed back to buffer. Current traces were biased to set the
lowest observed baseline current to zero for analysis (except for Figures 2-4,2-5 and 2-8
which show raw data).
Cholesterol oxidase immobilization. Pseudomonas sp. cholesterol oxidase (activity of
4.8–5.3 units/mg) was obtained from Wako. A 4.0 mg/ml solution of enzyme was
prepared in 50mM phosphate buffer, pH 6.5. The cholesterol oxidase solution was injected into the flow cell and incubated overnight in order to immobilize the enzyme in
the lipid bilayer membrane.
2.3 RESULTS AND DISCUSSION.
Capacitance data indicate deposition of lipids on the platinum electrode surface.
Formation of the thiolipid monolayer (inner lipid leaflet) results in a 90% decrease in
capacitance from the bare platinum value. This decrease in capacitance is similar to that
reported for deposition of a thiolipid monolayer on gold.38 Alkanethiol monolayers
formed on gold electrodes also result in comparable decreases in electrode capacitance .39
The outer lipid leaflet is deposited by a vesicle fusion method.24 Formation of the
DPPC outer leaflet on the monolayer of thiolipid results in a further decrease (21%) in
electrode capacitance. These data are comparable to decreases in capacitance that have
48 been reported by other groups for deposition of outer lipid leaflets on covalently attached
thiol monolayers on metal electrodes (ca. 40% decrease for deposition of a DPPC outer
leaflet on a thiolipid monolayer on gold 38). The capacitance data for deposition of both thiolipid and DPPC qualitatively indicate formation of a thiolipid/lipid bilayer membrane on the platinum electrode surface. It is noted that the structure of the membrane is not known and a lipid multi-bilayer structure could exist on regions of the electrode surface.
Acoustic impedance data collected at platinum quartz crystal microbalance electrodes indicate a mass increase roughly equal to that expected for deposition of a complete lipid bilayer membrane (data not shown). However, voltammetric studies at lipid
membranemodified platinum electrodes do not show complete blocking of ferricyanide
reduction indicating the existence of defects in the membrane. The electrode surface may
have regions containing a lipid multi-bilayer structure, regions of lipid bilayer, and
regions of exposed bare platinum (e.g., membrane defects). In earlier work reporting immobilization of cholesterol oxidase in lipid bilayer membranes on tin-doped indium
oxide electrodes1, we hypothesized that unmodified regions of the electrode surface
provided sites for electrochemical reduction of hydrogen peroxide generated by the
enzyme-catalyzed reaction. The same speculation is offered here for electrochemical
oxidation of hydrogen peroxide at the platinum electrode surface.
Electrodes have been prepared by exposing cholesterol oxidase to platinum
electrodes modified with only a monolayer of thiolipid (i.e., no outer lipid leaflet). These
electrodes did not show steady-state current for cholesterol exposures and the responses
were relatively small. Additionally, electrodes were prepared using no thiolipid. In this
case, bare platinum electrodes were treated with the vesicle solution for deposition of
49 lipids and subsequently with cholesterol oxidase. Similarly, such electrodes did not show steady-state current for cholesterol exposure. Initial characterization studies for vesicle fusion on bare platinum quartz crystal microbalance electrodes suggest that a lipid multilayer is deposited. The data also suggest that initial treatment of electrodes with thiolipid allows subsequent deposition of a single outer lipid leaflet through vesicle fusion. Both steps are required to produce enzyme modified electrodes that exhibit steady-state current for exposure to cholesterol solution. The outer lipid layer provides polar lipid head groups at the membrane/solution interface that likely contribute to binding of the enzyme to the electrode-supported lipid bilayer membrane. Why the thiolipid modification step is required is not yet understood.
Anodic current is observed upon exposure of cholesterol to the cholesterol
oxidase modified electrodes under continuous solution flow conditions. Figure 2-4, trace
A shows the current observed for changing the flow from buffer to a 100 μM cholesterol solution (t = 7 min) and for changing the flow back to buffer (t = 125 min). A current
plateau (at 1350 nA) is observed during the time window (ca. 2 h) when cholesterol is exposed to the electrode (Figure 2-4, trace A). The average current density for the plateau is 2.41 μA/cm2 and the source of the low frequency noise in the current plateau is not
known. The steady-state currents observed at all other electrode preparations under identical conditions are within ±25% of that shown in Figure 2-4, trace A. Control exposures of cholesterol to electrodes modified with a lipid bilayer membrane containing no cholesterol oxidase show no response (Figure 2-4, trace B). This control experiment was conducted at the same electrode used in Figure 2-4, trace A, prior to immobilization of the enzyme. It is noted that cholesterol detection limit is less than 1 μM for these
50 electrodes. Clearly resolvable steady-state current plateaus are observed for exposures to
0.5 μM concentrations of cholesterol.
The current measured under cholesterol exposure is attributed to the electrochemical oxidation of hydrogen peroxide produced upon enzyme-catalyzed oxidation of cholesterol by molecular oxygen. Scheme 2-1 shows the balanced chemical
1500 A
1000
current (nA) 500
B 0 0 60 120 time (min)
Figure 2-4. Amperometric response at a cholesterol oxidase-modified electrode to 100
μM cholesterol containing 50mM CD. (A) Trace of response for a
cholesterol oxidase-modified electrode. The downward arrow (↓) indicates
change of flow from cholesterol-free buffer to buffer containing cholesterol.
The upward arrow (↑) indicates change of flow back to cholesterol-free
buffer. (B) Trace of response of the same electrode before immobilization of
cholesterol oxidase (control experiment).
51 equations for the oxidation of cholesterol in the membrane and for electrochemical oxidation of the generated hydrogen peroxide at the electrode surface. Note that electrochemical oxidation of hydrogen peroxide at the electrode surface regenerates molecular oxygen.
Experiments conducted under near anaerobic conditions (nitrogen purged solutions) show smaller responses (ca. 50%) compared with those observed using air saturated solutions. Figure 2-5 shows the current for changing the flow from aerobic buffer to an aerobic cholesterol solution (A: 230 s) and for changing flow from aerobic cholesterol solution to anaerobic cholesterol solution (B: 900 s). The current for changing the flow back to aerobic cholesterol solution (C: 1800 s) and for changing the flow to buffer (D: 2450 s) is also shown. Exposure of anaerobic versus aerobic buffer (no cholesterol) to the cholesterol oxidase modified electrodes has no detectable effect on the baseline current (data not shown). The decrease in response for anaerobic cholesterol solution supports the electron transfer mechanism given in Scheme 2-1 above where oxygen is a reactant. A smaller response is expected because the concentration of oxygen
(one of the reactants) in purged solution is significantly decreased.
Membrane: Cholesterol + O2 Cholestenone + HOOH
Electrode: HOOH O + 2e- +2H+ 2
Scheme 2-1. Detection scheme at cholesterol oxidase modified platinum electrode.
52 The nonzero current observed for the nitrogen purged solution is likely due to
residual oxygen (and thus electrochemical hydrogen peroxide oxidation). However, a second possibility is that the current observed for the nitrogen purged solutions reflects a contribution from direct electron transfer from the enzyme to the electrode. The idealized structural model (Figure 2-1) suggests a separation distance between the FAD active site of the enzyme and the electrode surface of 30–40 Å. It is noted that electron transfer
80
70
60 B C 50 D 40
30 current (nA) 20 A 10
0 0 500 1000 1500 2000 2500 3000 3500 time (s)
Figure 2-5. Amperometric response at a cholesterol oxidase-modified electrode for
exposure to aerobic (air saturated) cholesterol solution (A and C) and
anaerobic (nitrogen purged) cholesterol solution (B). The current for
reverting the flow to buffer containing no cholesterol is also shown (D). The
cholesterol concentration is 100 μM containing 50mM CD. The sharp noise
spikes are due to changing the injection valve setting.
53 distances of up to 30Å have been reported for zinc/ruthenium modified myoglobins.40 For a monolayer of enzyme on the electrode surface, the current measured under anaerobic conditions corresponds to an enzyme turnover rate of ca. 1 electron per second. Given this relatively low enzyme turnover rate (i.e., 1 electron per second), the current for nitrogen purged solutions could reflect slow direct electron transfer from the enzyme to the electrode via tunneling. It is also possible that some enzyme molecules are immobilized closer to the electrode surface than depicted by the idealized structural model (Figure 2-1). However, direct electron transfer from the enzyme to the electrode is speculative and additional experiments are underway to test this hypothesis.
It is interesting to note that, upon changing the flow to anaerobic solution, the current approaches steady-state slower than for changing the flow back to aerobic solution (ca. 15 min anaerobic for versus ca. 8 min for aerobic). This could reflect depletion of oxygen in the electrode-supported lipid bilayer membrane upon changing the flow to anaerobic solution. Representative amperometric data for a series of cholesterol exposures (under aerobic conditions) in the range of 10–60 μM is shown in Figure 2-6.
The replicate cholesterol exposures (Figure 2-5, responses A and G: 10 μM; Figure 2-5, responses B and H: 30 μM) indicated a slight increase in steady-state current (6–10 nA beyond the baseline drift) over the course of the experiment. Other electrodes have shown a decrease in response of similar magnitude over comparable time. Several factors may contribute to the limiting currents observed. The activity and amount of immobilized enzyme, along with the collection efficiency of hydrogen peroxide at the
54 electrode surface, are major factors determining the steady-state currents observed in the
linear cholesterol concentration range. It is not known if the rates of cholesterol
partitioning into the membrane, and of lateral diffusion within the membrane, also
significantly affect steady-state current.
Enzyme kinetics. The response of the cholesterol oxidase modified electrodes shows
saturation behavior under higher cholesterol concentrations. The steady-state currents
C
300 E
F
200 B B
D current (nA) 100 A A
0 0 60 120 180 240 time (min)
Figure 2-6. Amperometric responses at a cholesterol oxidase-modified electrode to
continuous flow exposure of various cholesterol concentrations containing
50mM CD. A downward arrow (↓) indicates change of flow from buffer to
cholesterol solution, and an upward arrow (↑) indicates change of flow back
to buffer. (A) 10 μM; (B) 30 μM; (C) 60 μM; (D) 20 μM; (E) 50 μM; (F) 40
μM; (G) 10 μM and (H) 30 μM cholesterol. The trace was biased by −12.4
nA to zero the initial baseline.
55 observed for higher cholesterol concentrations are shown for two different cholesterol
oxidase-modified electrodes (up to 600 μM cholesterol for a freshly prepared electrode shown in Figure 2-7A and 1000 μM for an aged electrode shown in Figure 2-7C). The corresponding Lineweaver–Burk (double-reciprocal) plots are also shown (Figure 2-7B
and D, respectively). For both electrodes the plot of steady-state current versus
cholesterol concentration is hyperbolic. At cholesterol concentrations in the range of
200–1000 μM the response deviates from linear and asymptotically approaches a limiting
value. This behavior is consistent with the Michaelis–Menten mechanism for enzyme-
catalyzed reactions in which reversible binding of enzyme and substrate results in either
product formation or dissociation. In this model, the rate of reaction becomes
independent of substrate concentration (zero order) when all enzyme active sites are
occupied by substrate. The extrapolated limiting current corresponds to the maximum
velocity (Vm) that a reaction may attain under given conditions. The Michaelis constant
(Km) is the substrate concentration that yields a reaction rate that is equal to 1/2 Vm. It is
also an apparent dissociation constant for the enzyme–substrate complex.
Figure 2-7A and B are for an electrode on Day 2 of study (fresh electrode) and are
representative of the activity (Vm is 1820 nA and Km is 90μM) observed during the first few days for all electrodes. Figure 2-7C and D show data for an electrode on Day 8 of study (aged electrode) and these data, compared to those in Figure 2-7A and B, reflect the initial decrease in response that is observed at the cholesterol oxidase-modified electrodes
(Vm is 1400 nA and Km is 370μM). That is, during the first week of study, Km increases and Vm decreases. After aging, the electrodes exhibit Km values that are in reasonable agreement with the value reported for a cross-linked cholesterol oxidase film on gold
56 (520 μM).34 The observed decrease in Vm (from 1820 to 1400 nA over 6 days) suggests a
loss of enzyme molecules from the electrode surface and/or a decrease in activity for
immobilized enzymes. The former is expected given that cholesterol oxidase has a
measurable dissociation constant (KD) of 57 ± 21μM for binding to mixed
phospholipid/cholesterol vesicles.14 The possibility that loss of electrode response
(decreased Vm) reflects a decrease in enzyme activity is supported by the observed
increase in Km (e.g., from 90 μM for a fresh electrode on Day 2, to 370 μM for an aged electrode on Day 8), which suggests a decrease in activity for the enzyme molecules that remain immobilized on the electrode surface (it is noted that Km is independent of enzyme concentration). The dissociation rate of enzyme from the membrane must become slow after extended exposure to buffer, given that electrodes have shown responses for up to 60 days. The electrodes retain about 5% of the original response after this time. Re-exposing aged electrodes to enzyme solution does not restore any electrode activity for amperometric cholesterol detection. The decrease in electrode response could be coupled to changes that occur in the structure of the lipid bilayer membrane (resulting in irreversible desorption of enzyme) and the conformation of the enzyme (the enzyme may denature over time). That is, the integrity of the lipid bilayer membrane structure upon electrode ageing is not known. It may be possible to minimize the loss of electrode activity that is observed over days by optimizing the storage conditions (e.g., storing the electrodes at 4 ◦C).
Lipoprotein experiments. Experiments have been conducted to evaluate the ability of
the electrodes to detect cholesterol contained in LDL solutions. Figure 2-8 shows the
electrode response to 100 μM cholesterol in buffered CD (trace A), replicate exposures to
57 buffered LDL solution (traces B and C), 100 μM cholesterol in buffered CD (trace D; replicate of A), and LDL in buffered CD (trace E). The responses for exposure to buffered LDL reflect detection of aqueous phase cholesterol (i.e., cholesterol not contained in the lipid monolayer sheath of LDL). The concentration of cholesterol (not
2000 A 6 C -3 4 ) * 10 10
1000 -1
current (nA) current 2 1/i (nA
0 0 0 500 1000 -0.02 0 0.02 0.04 0.06 0.08 0.1
2000 B 6 D -3 4 ) * 10
1000 -1
current (nA) (nA) 2 1/i (nA
0 0 0 500 1000 -0.005 0.005 0.015 [cholesterol] (μM) 1/[cholesterol] (μM-1)
Figure 2-7. (A) Plot of steady-state current vs. cholesterol concentration showing
saturation behavior for an electrode on Day 2 of use. (B) Lineweaver–Burk
plot of data in Figure 2-7A. (C) Plot of steady-state current vs. cholesterol
concentration showing saturation behavior for an electrode on Day 8 of use.
(D) Lineweaver–Burk plot of data in Figure 2-7C. All experiments were
conducted using 50mM CD.
58 including cholesterol esters) in the buffered LDL solution is 100 μM (including aqueous phase and LDL resident cholesterol). The smaller response for the LDL solution (traces B and C) compared to 100 μM cholesterol in buffered CD (trace A) is consistent with the
180 150 A
120 E 90 D 60 Current (nA) B 30 C 0 0 100 200 300 400 Time (min)
Figure 2-8. Amperometric response at a cholesterol oxidase-modified electrode for
exposure to 100 μM cholesterol in buffered CD (A), buffered LDL solution
(B and C), 100 μM cholesterol (D), and buffered LDL solution containing
50mM CD (E). The cholesterol concentration (not including cholesterol
esters) of the LDL solutions is 100 μM. notion that LDL resident cholesterol is not directly detected at the electrode. The equilibrium that exists between cholesterol contained in the lipid monolayer sheath of
LDL and the aqueous phase lies largely towards LDL resident cholesterol. That is, much of the cholesterol in the LDL solution is contained within the LDL particles and this LDL
59 resident cholesterol does not directly partition into the electrode-supported lipid bilayer membrane.
Electrode response to 100 μM cholesterol in buffered CD was measured before
(trace A) and after (trace D) exposure of the electrode to LDL solution. These data
indicate that exposure of the electrode to LDL results in a substantial decrease in
electrode activity (ca. 50% for this experiment). The decrease in electrode activity may
be due to adsorption of LDL on the electrode surface. It is also noted that the second LDL
response (trace C) is slightly smaller than the first LDL response (trace B) indicating a
decrease in electrode activity. Data are also shown for electrode response to LDL in
buffered CD (trace E). The increased response (ca. six times larger) compared to the
response measured for the same LDL concentration in buffer containing no CD (traces B
and C) reflects a higher aqueous phase cholesterol concentration due to the increased
solubility of cholesterol in the aqueous phase of the buffered CD solution.
2.4. CONCLUSIONS.
Platinum electrodes modified with thiolipid/lipid bilayer membranes have been
prepared to immobilize cholesterol oxidase and to sequester cholesterol from solution. In
this system the close proximity of cholesterol oxidase to the electrode surface facilitates
electrochemical detection of hydrogen peroxide generated by the enzyme. Flow exposure
of aerobic cholesterol solution to the cholesterol oxidase modified electrodes indicates
that the enzyme resides in a conformation capable of oxidizing cholesterol. The data
indicate that the reaction follows the Michaelis–Menten kinetic model and that the
60 enzymatic activity decreases over days. Importantly, this work provides a measure of the
Km for cholesterol oxidase associated with a supported lipid bilayer membrane that may be a better model for the natural environment of the enzyme, as compared with methods that involve detergents or covalent immobilization of the enzyme. Detection of cholesterol in lipoprotein solution suggests that this electrode architecture may be useful in clinical sensing applications.
2.5 ACKNOWLEDGEMENTS.
A.D. acknowledges a student fellowship provided by Eveready Battery Company
Inc. and the Ernest B. Yeager Center for Electrochemical Sciences at Case Western
Reserve University. Helpful discussion with Professors Barry Miller, Daniel A. Scherson, and Morris Burke are acknowledged. Revisions and experiments done by Jonathan Ipsaro are acknowledged, along with experiments conducted by Latifa Odom and Carrie
McDonough.
61 2.6 REFERENCES.
(1) Devadoss, A.; Burgess, J. D. "Detection of cholesterol through electron transfer to cholesterol oxidase in electrode-supported lipid bilayer membranes." Langmuir 2002, 18, 9617-9621
(2) Karube, I.; Hara, K.; Matsuoka, H.; Suzuki, S. "Amperometric determination of total cholesterol in serum with use of immobilized cholesterol esterase and cholesterol oxidase." Anal Chim Acta 1982, 139, 127-32
(3) Wollenberger, U.; Kuehn, M.; Scheller, F.; Deppmeyer, V.; Jaenchen, M. "Amperometric enzyme sequence electrodes for cholesterol." Bioelectrochemistry and Bioenergetics 1983, 11, 307-17
(4) Masoom, M.; Townshend, A. "Determination of cholesterol by flow injection analysis with immobilized cholesterol oxidase." Anal Chim Acta 1985, 174, 293-7
(5) Hin, B. F. Y. Y.; Lowe, C. R. "Amperometric response of polypyrrole entrapped bienzyme films." Sens. Actuators, B 1992, B7, 339-42
(6) Motonaka, J.; Faulkner, L. R. "Determination of cholesterol and cholesterol ester with novel enzyme microsensors." Anal Chem 1993, 65, 3258-61
(7) Wang, H.; Mu, S. "Bioelectrochemical characteristics of cholesterol oxidase immobilized in a polyaniline film." Sens. Actuators, B 1999, B56, 22-30
(8) Vidal, J.-C.; Garcia-Ruiz, E.; Castillo, J.-R. "Design of a multilayer cholesterol amperometric biosensor for preparation and use in flow systems." Electroanalysis 2001, 13, 229-235
(9) Ram, M. K.; Bertoncello, P.; Ding, H.; Paddeu, S.; Nicolini, C. "Cholesterol biosensors prepared by layer-by-layer technique." Biosens Bioelectron 2001, 16, 849-856
(10) Abe, T.; Lau, Y. Y.; Ewing, A. G. "Characterization of glucose microsensors for intracellular measurements." Anal Chem 1992, 64, 2160-3
(11) Meyerhoff, J. B.; Ewing, M. A.; Ewing, A. G. "Ultrasmall enzyme electrodes with response time less than 100 milliseconds." Electroanalysis 1999, 11, 308-312
(12) Matsumoto, N.; Chen, X.; Wilson, G. S. "Fundamental Studies of Glucose Oxidase Deposition on a Pt Electrode." Anal Chem 2002, 74, 362-367
(13) Li, J.; Vrielink, A.; Brick, P.; Blow, D. M. "Crystal structure of cholesterol oxidase complexed with a steroid substrate: Implications for flavin adenine dinucleotide dependent alcohol oxidases." Biochemistry 1993, 32, 11507-15
62 (14) Sampson, N. S.; Kass, I. J.; Ghoshroy, K. B. "Assessment of the Role of an W Loop of Cholesterol Oxidase: A Truncated Loop Mutant Has Altered Substrate Specificity." Biochemistry 1998, 37, 5770-5778
(15) Ghoshroy, K. B.; Zhu, W.; Sampson, N. S. "Investigation of membrane disruption in the reaction catalyzed by cholesterol oxidase." Biochemistry 1997, 36, 6133- 6140
(16) Chen, X.; Wolfgang, D. E.; Sampson, N. S. "Use of the Parallax-Quench Method to Determine the Position of the Active-Site Loop of Cholesterol Oxidase in Lipid Bilayers." Biochemistry 2000, 39, 13383-13389
(17) Atrat, P. G.; Wagner, B.; Wagner, M.; Schumann, G. "Localization of the cholesterol oxidase in Rhodococcus erythropolis IMET 7185 studied by immunoelectron microscopy." J Steroid Biochem Mol Biol 1992, 42, 193-200
(18) Lange, Y. "Tracking cell cholesterol with cholesterol oxidase." J Lipid Res 1992, 33, 315-21
(19) Phillips, M. C.; Johnson, W. J.; Rothblat, G. H. "Mechanisms and consequences of cellular cholesterol exchange and transfer." Biochim Biophys Acta 1987, 906, 223-76
(20) Rothblat, G. H.; De la Llera-Moya, M.; Atger, V.; Kellner-Weibel, G.; Williams, D. L.; Phillips, M. C. "Cell cholesterol efflux: integration of old and new observations provides new insights." J Lipid Res 1999, 40, 781-796
(21) Fahey, P. F.; Koppel, D. E.; Barak, L. S.; Wolf, D. E.; Elson, E. L.; Webb, W. W. "Lateral diffusion in planar lipid bilayers." Science (Washington, DC, U. S.) 1977, 195, 305-6
(22) Stroeve, P.; Miller, I. "Lateral diffusion of cholesterol in monolayers." Biochim Biophys Acta 1975, 401, 157-67
(23) Gyoervary, E.; Wetzer, B.; Sleytr, U. B.; Sinner, A.; Offenhaeusser, A.; Knoll, W. "Lateral Diffusion of Lipids in Silane-, Dextran-, and S-Layer-Supported Mono- and Bilayers." Langmuir 1999, 15, 1337-1347
(24) Plant, A. L. "Supported Hybrid Bilayer Membranes as Rugged Cell Membrane Mimics." Langmuir 1999, 15, 5128-5135
(25) MacLachlan, J.; Wotherspoon, A. T. L.; Ansell, R. O.; Brooks, C. J. W. "Cholesterol oxidase: sources, physical properties and analytical applications." J. Steroid Biochem. Mol. Biol. 2000, 72, 169-195
(26) Doukyu, N.; Aono, R. "Two moles of O2 consumption and one mole of H2O2 formation during cholesterol peroxidation with cholesterol oxidase from Pseudomonas sp. strain ST-200." Biochem. j. 1999, 341, 621-7.
63 (27) Teng, J. I.; Smith, L. L. "Sterol peroxidation by Pseudomonas fluorescens cholesterol oxidase." Steroids 1996, 61, 627-33.
(28) Pollegioni, L.; Gadda, G.; Ambrosius, D.; Ghisla, S.; Pilone, M. S. "Cholesterol oxidase from Streptomyces hygroscopicus and Brevibacterium sterolicum: effect of surfactants and organic solvents on activity." Biotech. Appl. Biochem. 1999, 30, 27-33
(29) Yin, Y.; Sampson, N. S.; Vrielink, A.; Lario, P. I. "The presence of a hydrogen bond between asparagine 485 and the p system of FAD modulates the redox potential in the reaction catalyzed by cholesterol oxidase." Biochemistry 2001, 40, 13779-13787
(30) Pollegioni, L.; Wels, G.; Pilone, M. S.; Ghisla, S. "Kinetic mechanisms of cholesterol oxidase from Streptomyces hygroscopicus and Brevibacterium sterolicum." Eur J Biochem 1999, 264, 140-151
(31) Cheillan, F.; Lafont, H.; Termine, E.; Fernandez, F.; Sauve, P.; Lesgards, G. "Molecular characteristics of the cholesterol oxidase and factors influencing its activity." Biochim Biophys Acta 1989, 999, 233-8
(32) Aleksandrovsky, Y. A.; Titov, V. N. "Influence of the structural organization of cholesterol aggregates in water-organic media on cholesterol oxidation catalyzed by cholesterol oxidase." J. Chem. Biochem. Kinetics 1991, 1, 337-45
(33) Lee, S. Y.; Rhee, H. I.; Tae, W. C.; Shin, J. C.; Park, B. K. "Purification and characterization of cholesterol oxidase from Pseudomonas sp. and taxonomic study of the strain." Appl Microbiol Biotechnol 1989, 31, 542-6
(34) Nakaminami, T.; Ito, S.-i.; Kuwabata, S.; Yoneyama, H. "Amperometric Determination of Total Cholesterol at Gold Electrodes Covalently Modified with Cholesterol Oxidase and Cholesterol Esterase with Use of Thionin as an Electron Mediator." Anal Chem 1999, 71, 1068-1076
(35) Crockett, E. L.; Hazel, J. R. "Sensitive assay for cholesterol in biological membranes reveals membrane-specific differences in kinetics of cholesterol oxidase." J Exp Zool 1995, 271, 190-5
(36) Burgess, J. D.; Rhoten, M. C.; Hawkridge, F. M. "Cytochrome c Oxidase Immobilized in Stable Supported Lipid Bilayer Membranes." Langmuir 1998, 14, 2467-2475
(37) Haynes, M. P.; Phillips, M. C.; Rothblat, G. H. "Efflux of Cholesterol from Different Cellular Pools." Biochemistry 2000, 39, 4508-4517
(38) Lang, H.; Duschl, C.; Vogel, H. "A new class of thiolipids for the attachment of lipid bilayers on gold surfaces." Langmuir 1994, 10, 197-210
64 (39) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. "Spontaneously organized molecular assemblies. 4. Structural characterization of n-alkyl thiol monolayers on gold by optical ellipsometry, infrared spectroscopy, and electrochemistry." J Am Chem Soc 1987, 109, 3559-68
(40) Axup, A. W.; Albin, M.; Mayo, S. L.; Crutchley, R. J.; Gray, H. B. "Distance dependence of photoinduced long-range electron transfer in zinc/ruthenium- modified myoglobins." J Am Chem Soc 1988, 110, 435-9
65 CHAPTER 3.
ENZYME MODIFICATION OF PLATINUM MICROELECTRODES FOR
DETECTION OF CHOLESTEROL IN VESICLE LIPID BILAYER
MEMBRANES
3.1 INTRODUCTION.
This group reported detection of cholesterol contained in the plasma membrane of
a single oocyte using microelectrodes modified with lipid bilayer membrane containing
cholesterol oxidase (chapter 4).1 Cholesterol is an amphipathic lipid molecule that is
known to play essential roles in cellular membrane structure and function.2 The
cholesterol content of cellular membranes is believed to be tightly regulated through
vesicular trafficking pathways that consume energy and alterations in cell plasma
membrane cholesterol content are likely involved in the onset of atherosclerosis.3 The
results presented in this article are focused on electrode fabrication and characterization
including detection of cholesterol contained in vesicle lipid bilayer membranes that serve
as simple model system for the cell plasma membrane.
The electrode supported thiolipid/lipid bilayer membrane is formed by initially
modifying a bare platinum electrode surface with a sub-monolayer of thiolipid.4 A vesicle fusion method is used to deposit a lipid bilayer structure on the electrode where the thiolipid sub-monolayer constitutes a portion of the inner lipid leaflet anchoring the bilayer to the electrode. Cholesterol oxidase has a natural affinity for interaction with
66 lipid bilayer membranes and it spontaneously inserts into the electrode-supported lipid
bilayer membrane from solution.
The structural model for interaction between cholesterol oxidase and the
electrode-supported lipid bilayer membrane4,5 is that proposed for natural interaction of
the bacterial enzyme with the plasma membrane of mammalian cells.6 Cholesterol
oxidase partially inserts in the outer lipid leaflet of the cell plasma membrane undergoing
a conformational change that exposes hydrophobic residues and opens a path for
movement of cholesterol into the active site pocket directly from the plasma membrane.
Based on this background, cholesterol is believed to partition into the electrode-supported
lipid bilayer membrane prior to enzymatic oxidation. That is, the electrode-supported
lipid bilayer membrane functions as an acceptor of cholesterol efflux from a donor
membrane (i.e., vesicle lipid bilayer membrane or cell plasma membrane).
Detection of cholesterol contained in the lipid bilayer membrane of giant vesicles involves positioning the electrode-supported lipid bilayer membrane in contact with the vesicle membrane. A thin aqueous layer between the electrode-supported lipid bilayer membrane and the vesicle membrane is predicted due to lipid head group solvation.
Contacting the vesicle positions the electrode-supported lipid bilayer membrane directly adjacent to the vesicle membrane. Because cholesterol has a finite solubility in the
aqueous phase7 the electrode-supported lipid bilayer membrane containing cholesterol
oxidase consumes aqueous phase cholesterol that is directly adjacent to, and in
equilibrium with, the vesicle membrane. This consumption of aqueous phase cholesterol
at the vesicle membrane surface causes further cholesterol efflux from the vesicle
membrane so that cholesterol mass transport occurs from the high concentration in the
67 vesicle membrane to the lower concentration of the electrode-supported lipid membrane.
This mass transfer model is based on earlier work by Rothblat and co-workers8 that demonstrates the use of lipid vesicles as an aqueous phase acceptor of cellular cholesterol efflux.
The ferrocyanide characterization studies show that the platinum electrodes modified with a lipid bilayer membrane containing cholesterol oxidase have about 15% of the platinum electrode surface unmodified and exposed to solution. The unmodified regions of the platinum electrode may provide sites for facile electro-oxidation of the enzymatically generated hydrogen peroxide. Experiments for exposure of the electrodes to cholesterol solution show steady state responses that are believed to be limited by the rate of enzyme turnover. The current responses obtained at vesicles indicate that electrode response correlates with the cholesterol content of the vesicle membrane where increased cholesterol content produces larger responses.
3.2 EXPERIMENTAL SECTION.
Microelectrode fabrication. Platinum microelectrodes were either purchased from
Cypress systems (10 μm diameter) or fabricated in-house (11.5 µm diameter wire,
Goodfellow Corp.) using a modified procedure reported to construct microelectrodes with carbon fiber.9 Glass capillaries (Kimax – 51 ®, Kimble products) are pulled in a hydrogen flame and platinum wire is inserted. One end of the glass capillary containing the platinum wire is placed inside a heated platinum coil, and the other end is connected to a weight (e.g., 10 g) so that the glass is pulled to create a thin insulating layer on the
68 platinum wire. The capillary microelectrodes are polished using a beveling machine
(WPI, Inc.) to produce a disk electrode. Conductive epoxy (Chemtronics®, USA) was used to make electrical contact between the platinum wire and stainless steel lead
(syringe needle) at the open end of the capillary. The outer diameter of the glass platinum electrodes usually lies in the range of 15-20 μm. It is noted that the overall dimension of the electrode needs to be small so that cells and vesicles can be contacted.
Instrumentation. Voltammetric and amperometric measurements for solution phase cholesterol detection were conducted using a three-electrode cell and a potentiostat (CV-
50, BAS) coupled with a pre-amplifier (PA-1, BAS). Amperometric measurements at giant vesicles were conducted using a two-electrode cell and a voltammeter-amperometer
(CHEM-CLAMP, Dagan corp.). The 3 pole Bessel filter in the voltammeter- amperometer was set to 100 Hz. The filtered response was further processed using a noise rejecting voltmeter (Model 7310 DSP, Signal Recovery Inc.) to digitally filter 60
Hz noise. An Ag/AgCl (1 M KCl) reference electrode was used for all experiments and the applied potential was 820 mV vs NHE.
Electrode modification procedure. The platinum microelectrodes are immersed in a 40
μM ethanolic solution of 1,2 dipalmitoyl-sn-glycero-3-phosphothioethanol (Avanti Polar
Lipids Inc.) (referred to as thiolipid) for 3 hrs to form a covalently attached sub- monolayer. The microelectrode modified with a thiolipid sub-monolayer is immersed in a 1 mM DPPC (1,2 dipalmitoyl-sn-glycero-3-phosphocholine, Avanti Polar Lipids Inc.) vesicle solution (sodium phosphate buffer, 50 mM, pH 7.4,) for 15 minutes to form a
69 bilayer structure on the electrode. The lipid bilayer modified electrode is immersed in 2.5 mg/ml cholesterol oxidase (Wako Pure Chemical Industries, Ltd., 3.6 units/mg) solution for at least 18 hrs to immobilize the enzyme in the lipid bilayer membrane.
Quiet solution experiments. Quiet solution experiments are conducted in a glass cell.
The solutions were introduced into the cell using a six-way (Rheodyne Model 5020,
Supelco Corp.) valve and two syringes, controlled manually. The electrode is first placed
in the buffer solution (0.05 M sodium phosphate, pH 6.5, 50 mM cyclodextrin Cargill
Inc.) until a stable background current is established. A buffered cholesterol solution is injected into the cell. The injected volume is removed, re-injected, and removed again for complete mixing.
Solution flow experiments. Flow injection experiments were conducted in a Lucite®
cell constructed in-house using a six-way valve (Rheodyne Model 5020, Supelco Corp.)
and two syringe pumps (Harvard Apparatus) to direct buffer cyclodextrin or buffered
cyclodextrin containing cholesterol over the electrode. Injection times were controlled
manually.
Giant vesicle experiments. The procedure for electroformation of giant lipid vesicles is
adopted from a method published by Angelova and co-workers.10 L-α phosphatidylcholine from soybean (1mg/ml) and cholesterol (both from Sigma) were dissolved in chloroform/methanol (9:1) to form a lipid mixture. Molar ratios of cholesterol to lipid were 0.33, 0.5, and 0.66 to prepare vesicles with various cholesterol
70 content. A 2 μl aliquot of the lipid mixture is allowed to dry on 250 μm diameter
platinum wires in a vacuum chamber. The platinum wire containing the dried lipid film
is placed on a clean glass slide. The dried lipid film is rehydrated using Tris-HCl buffer
(2 mM, pH 6.5), under an applied ac voltage (10 Hz) with an initial peak to peak
amplitude of 0.2 V that is increased to 1 V over ca. 15 min. The vesicle growth on platinum is monitored under an inverted microscope. About 50% of the experiments
produce vesicles of 40-100 μm diameter. The vesicle bilayer membrane is expected to be
10 unilamellar and in the Lα state where the lipids are present in a loosely packed
disordered state.11 The microelectrode is initially positioned ca.20 μm from the vesicle
membrane and is repositioned for contact where slight vesicle deformation is observed.
It is noted that the experiment must be isolated from mechanical vibrations to maintain
contact between the electrode and the vesicle.
3.3 RESULTS AND DISCUSSION.
Figure 3-1 (trace A) shows the cyclic voltammogram of ferrocyanide at a bare
platinum microelectrode. The steady state current is slightly higher than that expected for
a disk electrode embedded in an infinite insulating plane (e.g., 5.5 nA for reaction of 5
mM ferrocyanide at an 11.5 μm diameter electrode). The increased limiting current is
expected due to the thin glass capillary and diffusion from the back of the electrode
plane.12,13 The limiting current value of 6-7.5 nA is consistent with an outer diameter of the glass capillary of 15-20 μm. This value agrees well with the size of the capillary under optical magnification.
71 Figure 3-1 (trace B) is the cyclic voltammogram of ferrocyanide after the
platinum microelectrode was reacted with thiolipid. The decrease in the limiting current
(ca. 10%) suggests the formation of a sub-monolayer of thiolipid on the electrode surface.
The distribution of thiolipid molecules on the electrode surface is not known and the structure of the thiolipid sub-monolayer could change upon subsequent deposition of
DPPC. The dependence of electrode response on the degree of thiolipid coverage has not
been rigorously characterized. However, it is noted that a near complete monolayer
coverage of thiolipid does not allow construction of functional oxidase modified
electrodes. High coverages of thiolipid likely block facile electro-oxidation of the
hydrogen peroxide that is generated by the enzyme.
Figure 3-1 (trace C) shows ferrocyanide voltammetric data after deposition of
DPPC on the thiolipid sub-monolayer modified platinum electrode. The further decrease
in limiting current suggests deposition of DPPC molecules on the electrode. The positive
shift of the half wave potential after deposition of DPPC suggests a decrease in electron
transfer rate. This is consistent with regions of the electrode containing a lipid coverage that allows reaction of ferrocyanide at an increased distance of closest approach compared to bare platinum. The smaller limiting current measured after DPPC deposition indicates that approximately 50% of the platinum surface is blocked from reaction with ferrocyanide. These microelectrode data are similar to those from ferrocyanide studies conducted for constructing this architecture on conventionally sized platinum electrodes.4
Quartz crystal microbalance (QCM) data obtained at conventionally sized (0.2
cm2) platinum electrodes suggest that some regions of the electrode are covered with a
72 lipid multi-bilayer structure (a multilamellar coating).4 Ferrocyanide characterization studies conducted at the QCM platinum electrodes modified with the thiolipid/lipid bilayer membrane show incomplete blocking indicating that a fraction of the electrode remains unmodified. However, deposition of the thiolipid/lipid bilayer membrane shows a mass increase that is roughly that expected for a complete bilayer membrane (ca. 99 ng, foot print of DPPC assumed to be 60 Å2) that covers the entire electrode surface. Taken together, these data suggest that regions of the electrode are modified with a multilamellar structure. It is noted that atomic force microscopy (AFM) images for vesicle fusion on mica show multilamellar structures (chapter 6). It is hypothesized that the platinum microelectrodes are modified with a lipid bilayer structure that consists, in part, of multilammellar lipid islands as well as defects where bare platinum remains exposed to solution.
Exposing the lipid bilayer modified electrode to enzyme solution further decreases the limiting current for reaction of ferrocyanide (Figure 3-1, trace D). The data are consistent with 15-20% of the platinum electrode remaining unmodified and exposed to solution. One possible explanation for this result is adsorption of cholesterol oxidase at defect sites (bare platinum not modified with a lipid bilayer). While some adsorption of enzyme to bare platinum likely occurs, another possible reason for the increased blocking of ferrocyanide is that a larger fraction of the electrode surface becomes coated with a lipid bilayer structure upon exposure to enzyme. It is noted that direct adsorption of cholesterol oxidase on bare platinum does not produce electrodes that exhibit enzymatic activity. This result indicates that interaction between the enzyme and the electrode-supported lipid bilayer membrane is required for retention of enzymatic
73 activity. It is hypothesized that exposure of the lipid modified electrode to enzyme solution results in destabilization of multilamellar lipid islands and an increase of the electrode surface area that is coated with a lipid bilayer membrane. This hypothesis is supported by AFM studies that show disruption of multilamellar lipid structures on mica upon exposure to cholesterol oxidase solution.
The AFM study for immobilization of the enzyme in lipid bilayer membranes on mica suggest that cholesterol oxidase is immobilized as monomers and aggregates that
Figure 3-1. Ferrocyanide voltammetry for deposition of the thiolipid/lipid bilayer
membrane containing cholesterol oxidase on a platinum electrode. Cyclic
voltammetry of potassium ferrocyanide at A) a bare platinum
microelectrode, B) after deposition of thiolipid, C) after deposition of
DPPC, and D) after incorporating cholesterol oxidase.
74 are 50-100 nm in diameter. It is understood that the lipid bilayer structures formed on mica may not be relevant to those formed on platinum electrodes initially modified with a sub-monolayer of thiolipid. Nevertheless, an electrode architecture is proposed where the surface is covered predominately with a lipid bilayer membrane containing islands of immobilized cholesterol oxidase.
Figure 3-2, trace A, is the amperometric response observed at a oxidase modified electrode upon exposure to cholesterol solution where a quiet buffer solution is spiked with an aliquot of buffered cholesterol and allowed to become quiet. The control experiment (Figure 3-2, trace B) was conducted at the same microelectrode prior to
Figure 3-2. Amperometric response obtained for an exposure of 25 μM cholesterol
solution at (A) an oxidase modified electrode (B) the same electrode with
thiolipid/lipid membrane prior to immobilization of the oxidase . The up
arrow (↑) indicates the times of cholesterol exposure.
75 immobilization of cholesterol oxidase. Scheme 2-1 shows the membrane reaction for
enzymatic oxidation of cholesterol by molecular oxygen and generation of hydrogen
peroxide (labeled Membrane), and the electrode reaction for electrochemical oxidation of
hydrogen peroxide regenerating molecular oxygen (Electrode). It is hypothesized that
unmodified regions of the electrode surface (i.e., 15-20%) provide sites where hydrogen
peroxide predominantly reacts. In an earlier report from this group it was demonstrated
that the limiting current observed at conventionally sized cholesterol oxidase modified
electrodes is significantly diminished under anaerobic conditions. Diminished responses
are also observed for the microelectrodes under anaerobic conditions and these data
suggest that Scheme 2-1 is a dominant mechanism involved in the passage of anodic
current.
A number of factors may contribute to the limiting current observed (e.g., mass
transfer of cholesterol to the electrode, the rate of cholesterol partitioning into the
membrane, the lateral diffusion rate of cholesterol and possibly of the enzyme in the
membrane, the kinetics of the enzymatic reaction, buildup of cholestenone (oxidized cholesterol) in the electrode-supported lipid bilayer membrane, and the fraction of the
generated hydrogen peroxide that is oxidized at the electrode surface). However, in the
range of current densities where response correlates with cholesterol concentration (ca.
0.1-2 µA/cm2), it has been demonstrated that electrode response is strongly dependent on
the amount of enzyme immobilized on the electrode.1 Ferrocyanide characterization studies and experiments for reaction of cholesterol at the electrodes indicate that complete immobilization of enzyme requires exposure to enzyme solution for more than
10 hours (data not shown). Based on these data, it is hypothesized that the rate of
76 12.5 μM 0.2 pA 500 s 6.2 μM
3.1 μM 1.6 μM
Figure 3-3. Amperometric responses obtained at an oxidase modified electrode for
exposure to cholesterol solution and sequential dilutions. The up arrow (↑)
indicates the time of cholesterol exposure and the down arrows (↓) indicate
the times of buffer dilution. enzymatic oxidation of cholesterol is rate limiting (i.e., the reaction is not limited by mass transfer of cholesterol to the electrode).
The steady state response (over 15 min.) suggests that cholestenone (oxidized cholesterol) does not significantly accumulate in the electrode-supported lipid bilayer membrane and affect the rate of enzymatic catalysis. Under steady state turnover, cholestenone is likely released from the membrane into the cyclodextrin solution. It is noted that cholestenone is known to efflux at a faster rate than cholesterol from lipid membranes due to weaker interaction with lipids.14 The dependence of electrode response on cholesterol concentration is demonstrated by initially spiking the bulk
77 solution with an aliquot of cholesterol solution and subsequently diluting by spiking with buffer (Figure 3-3). The steady state responses indicate a clear correlation with cholesterol concentration (1.5- 12.5 μM) where higher cholesterol concentration produces larger responses.
The concentration dependence of electrode response to cholesterol is also shown
Figure 3-4. Flow injection data at an oxidase modified electrode for exposure to three
different cholesterol concentrations. The up arrows (↑) indicate the times of
cholesterol injection and the down arrows (↓) indicate the times where the
flow is reverted to buffer.
78 for flow-injection type experiments (Figure 3-4) where a continuous solution flow is
changed from buffered cyclodextrin to buffered cyclodextrin containing cholesterol and
back to buffered cyclodextrin. Data for three cholesterol concentrations (100 μM, 50
μM, 25 μM) are shown along with a replicate exposure to 25 μM. These data also indicate a clear correlation between response and cholesterol concentration. The flow injection data demonstrate baseline resolution between the individual cholesterol exposures. The response rise time and decay time are likely influenced by hydrodynamic
factors.
Due to its low aqueous solubility, cholesterol in biological systems largely
resides in lipid membranes, (e.g., the cell plasma membrane and lipid monolayer shell of
lipoproteins). Here, vesicle lipid bilayer membranes were employed as a simple model of
the cell plasma membrane. The optical micrographs shown in Figure 3-5 demonstrate the
ability to contact a vesicle with a microelectrode (contact is defined as the electrode
position that slightly deforms the spherical shape of the vesicle). It is noted that on
Figure 3-5. Photographs showing an electrode (A) positioned about 15 μm away from a
giant lipid vesicle and (B) contacting the giant lipid vesicle.
79 contacting the vesicle of interest (Figure 3-5, image B) other vesicles in the vicinity move
which brings a different set of vesicles in focus (compare Figures 3-5A and B). Various electrode responses for contacting vesicles with a 0.5 cholesterol to phopholipid ratio are shown in Figure 3-6, traces A-C. Figure 3-7 shows two control experiments (trace A: response of a oxidase modified electrode at vesicle containing no cholesterol, traces B:
response of a bare platinum electrode at vesicle containing 0.5 cholesterol to
Figure 3-6. Amperometric responses obtained at three different oxidase modified
electrodes for contacting three different giant lipid vesicles prepared with
0.5 cholesterol to phospholipid ratio. The up arrows (↑) indicate the times of
contact.
80 phospholipid ratio). The anodic current responses observed at the oxidase modified electrodes (Figure 3-6, traces A-C) are assigned to detection of cholesterol present in the vesicle lipid bilayer membrane.
It is not yet known if cholesterol present in the inner leaflet of the vesicle lipid bilayer membrane significantly contributes to electrode response. The rate of cholesterol flip-flop (transbilayer movement) across the lipid bilayer membrane is not known and discrepancies are found in the literature.15 However, sum frequency generation
Figure 3-7. Control experiments for (A) contacting a giant vesicle formed with no
cholesterol with an oxidase modified electrode and (B) for contacting a giant
vesicle with 0.5 cholesterol to phospholipid ratio with a bare platinum
electrode. The up arrows (↑) indicate the times of contact and the down
arrows (↓) indicate the times of withdrawal.
81 experiments for tracking transbilayer movement of phospholipid16 and cholesterol (John
C. Conboy; private communication) suggest that cholesterol flip-flop occurs with a t1/2 of less than a minute. This notion is consistent with work by others for gauging cholesterol flip-flop in cell plasma membranes.15 Therefore, it is proposed that cholesterol is
replenished to the electrode contact site by both transbilayer movement and lateral diffusion.
The amount of cholesterol predicted to be contained in the region of the outer
lipid leaflet that is in contact with the platinum disk electrode (100 amol of cholesterol in
the electrode footprint region of the outer lipid leaflet) corresponds to 20 pC (assuming 2
electrons per molecule of cholesterol and complete oxidation of the generated hydrogen
peroxide). The data shown in Figure 3-6, trace A, indicate oxidation of about ca. 35 % of
this amount of cholesterol (over 20 s). The lack of a continuous response decay (i.e., the
Figure 3-8. Amperometric responses of an oxidase modified electrode for two
consecutive contacts at the same giant vesicle formed with 0.66 cholesterol
to lipid ratio. The up arrows (↑) indicate the times of contact and the down
arrows (↓) indicate the times of withdrawal.
82 steady state responses) suggest that the cholesterol content of the vesicle membrane at the electrode contact site is not significantly depleted of cholesterol. It is speculated that the apparent steady state responses reflect lateral diffusion of cholesterol in the vesicle membrane to the electrode contact site as well as transbilayer movement of cholesterol from the inner lipid leaflet to the outer lipid leaflet. Assuming replenishment of cholesterol to the electrode contact site through only lateral diffusion in the outer lipid leaflet, a flux that would sustain a ca. 9 pA response is predicted (assuming a cholesterol content of 0.5 cholesterol/phospholipid ratio and a diffusion coefficient of 10-7 cm2/s).17
Because cholesterol flip-flop and lateral diffusion likely also contribute to mass transport for replenishment of cholesterol to the electrode contact site, it is reasonable to state that the electrode responses are not mass transport limited. Again, it has been demonstrated that response magnitude depends on the amount of cholesterol immobilized on the electrode surface1 and this result suggests that the rate of cholesterol exchange between
the vesicle membrane and the electrode-supported lipid bilayer membrane is not rate
limiting. It is proposed that the responses observed at vesicles are limited by enzyme
turnover rate.
Sequential experiments for contacting a vesicle with the same oxidase modified
electrode show responses of the same magnitude (Figure 3-8). This response stability is
significant in that it allows evaluation of electrode response at vesicles created with
different cholesterol content. Data collected at vesicles containing three different ratios
of cholesterol to phospholipid are shown in Figure 3-9. These data qualitatively show
larger responses for increased cholesterol content of the vesicle membrane (0.33 to 0.66
ratio of cholesterol to phospholipids). This result is consistent with data collected at
83 oocytes showing a dependence of response on plasma membrane cholesterol content.1
However, the 0.33 cholesterol/phospholipids ratio does not show a discernable response and the data are basically indistinguishable from control experiments where the vesicle membrane contains no cholesterol.
McConnell and co-workers have shown that the rate of cholesterol efflux from lipid membranes into cyclodextrin solution is nonlinear with respect to membrane cholesterol content.18 Their data indicate a sharp increase in the rate of cholesterol efflux
when the cholesterol content is increased above the 0.5 cholesterol/lipid ratio.
McConnell proposes a stoichiometric complex where one cholesterol molecule associates
with two phospholipids. The activity of cholesterol in the lipid membrane increases
Figure 3-9. Amperometric responses obtained at an oxidase modified electrode for
contacting three different vesicles containing cholesterol to lipid ratios of
(A) 0.66, (B) 0.5, and (C) 0.33. The arrow (↑) indicates the times of contact.
84 sharply when the cholesterol content is increased above this stoichiometric ratio (i.e., 0.5
cholesterol/phospholipid). This behavior may explain the inability to detect vesicle
membrane cholesterol at low cholesterol to phospholipids ratios (e.g., 0.33).
As stated above, it is proposed that cholesterol oxidation at the electrode is limited
by the rate of enzymatic oxidation of cholesterol. This hypothesis implies that the rate of
cholesterol exchange between the vesicle membrane and the electrode-supported lipid
bilayer membrane is fast relative to the rate of cholesterol oxidation. In this case, an
increase in cholesterol content of the vesicle membrane results in a higher concentration
of cholesterol in the electrode-supported lipid bilayer membrane and, thus, a faster rate of
enzymatic cholesterol oxidation. This model assumes that slow enzymatic consumption
of cholesterol in the electrode membrane does not significantly alter the cholesterol
content of the electrode membrane.
It is further noted that cholesterol contained in each leaflet of the vesicle lipid
bilayer membrane may be present in two distinct states. Small cholesterol rich domains
(ca. 10-40 nm rafts) are believed to exist in the plasma membrane of cells.19 While no
direct concrete evidence for the existence of rafts in the plasma membrane of living cells
has been provided, cholesterol has been shown to phase separate to form large domains in vesicle lipid bilayer membranes of appropriate composition.20 The vesicles used in this
work are believed to contain cholesterol in only one phase (i.e., no rafts are expected).
However, it is not yet known how cholesterol raft formation may affect the correlation
between electrode response and the net cholesterol content of the vesicle membrane.
85 3.4 CONCLUSIONS.
Platinum microelectrodes modified with a lipid bilayer membrane incorporating
cholesterol oxidase allow detection of cholesterol in vesicle lipid bilayer membranes.
The data indicate that electrode response correlates with the cholesterol content of the
vesicle membrane where higher cholesterol content yields larger responses. This behavior suggests that the electrodes will be useful in characterizing predicted changes in the cholesterol content of the macrophage cell plasma membrane during biosynthesis of high density lipoprotein (i.e., HDL). Additional experiments using lipid vesicles as a model of the cell plasma membrane are planned to study lateral diffusion of cholesterol in lipid bilayer membranes, and interbilayer and transbilayer movement of cholesterol.
3.5 ACKNOWLEDGEMENTS.
This work was supported by the National Institute of Health (5 R21 EB003925)
and the Department of Chemistry, Case Western Reserve University. Helpful discussions
with Professors Dan Scherson and Barry Miller are also acknowledged.
86 3.6 REFERENCES.
(1) Devadoss, A.; Burgess, J. D. "Steady-State Detection of Cholesterol Contained in the Plasma Membrane of a Single Cell Using Lipid Bilayer-Modified Microelectrodes Incorporating Cholesterol Oxidase." J. Am. Chem. Soc. 2004, 126, 10214-10215
(2) Tabas, I. "Cholesterol in health and disease." J Clin Investig 2002, 110, 583-590
(3) Simons, K.; Ikonen, E. "How cells handle cholesterol." Science 2000, 290, 1721- 6.
(4) Bokoch, M. P.; Devadoss, A.; Palencsar, M. S.; Burgess, J. D. "Steady-state oxidation of cholesterol catalyzed by cholesterol oxidase in lipid bilayer membranes on platinum electrodes." Anal. Chim. Acta 2004, 519, 47-55
(5) Devadoss, A.; Burgess, J. D. "Detection of cholesterol through electron transfer to cholesterol oxidase in electrode-supported lipid bilayer membranes." Langmuir 2002, 18, 9617-9621
(6) Sampson, N. S.; Vrielink, A. "Cholesterol Oxidases: A Study of Nature's Approach to Protein Design." Acc. Chem. Res. 2003, 36, 713-722
(7) Haberland, M. E.; Reynolds, J. A. "Self-association of cholesterol in aqueous solution." Proc Natl Acad Sci U S A 1973, 70, 2313-16
(8) Phillips, M. C.; Johnson, W. J.; Rothblat, G. H. "Mechanisms and consequences of cellular cholesterol exchange and transfer." Biochim Biophys Acta 1987, 906, 223-76
(9) Meulemans, A.; Poulain, B.; Baux, G.; Tauc, L.; Henzel, D. "Micro carbon electrode for intracellular voltammetry." Anal. Chem. 1986, 58, 2088-91
(10) Menger, F. M.; Angelova, M. I. "Giant Vesicles: Imitating the Cytological Processes of Cell Membranes." Acc. Chem. Res. 1998, 31, 789-797
(11) Brown, D. A.; London, E. "Structure and function of sphingolipid- and cholesterol-rich membrane rafts." J Biol Chem 2000, 275, 17221-17224
(12) Amphlett, J. L.; Denuault, G. "Scanning Electrochemical Microscopy (SECM): An Investigation of the Effects of Tip Geometry on Amperometric Tip Response." J. Phys. Chem. B 1998, 102, 9946-9951
(13) Shao, Y.; Mirkin, M. V. "Probing Ion Transfer at the Liquid/Liquid Interface by Scanning Electrochemical Microscopy (SECM)." J. Phys. Chem. B 1998, 102, 9915-9921
87 (14) Ohvo, H.; Slotte, J. P. "Cyclodextrin-Mediated Removal of Sterols from Monolayers: Effects of Sterol Structure and Phospholipids on Desorption Rate." Biochemistry 1996, 35, 8018-8024
(15) Haynes, M. P.; Phillips, M. C.; Rothblat, G. H. "Efflux of Cholesterol from Different Cellular Pools." Biochemistry 2000, 39, 4508-4517
(16) Liu, J.; Conboy, J. C. "Direct Measurement of the Transbilayer Movement of Phospholipids by Sum-Frequency Vibrational Spectroscopy." J Am Chem Soc 2004, 126, 8376-8377
(17) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 1980.
(18) Radhakrishnan, A.; McConnell, H. M. "Chemical Activity of Cholesterol in Membranes." Biochemistry 2000, 39, 8119-8124
(19) Simons, K.; Ikonen, E. "Functional rafts in cell membranes." Nature (Lond) 1997, 387, 569-572
(20) Kahya, N.; Scherfeld, D.; Bacia, K.; Schwille, P. "Lipid domain formation and dynamics in giant unilamellar vesicles explored by fluorescence correlation spectroscopy." J Struct Biol 2004, 147, 77-89
88 CHAPTER 4.
STEADY STATE DETECTION OF CHOLESTEROL CONTAINED IN THE
PLASMA MEMBRANE OF SINGLE CELL USING LIPID BILAYER MODIFIED
MICROELECTRODES INCORPORATING CHOLESTEROL OXIDASE
4.1 INTRODUCTION.
The ability to evaluate the dynamics of intracellular cholesterol trafficking to and
from the plasma membrane would allow characterization of pathways governing
cholesterol homeostasis and, in particular, of the initial steps in atherogenesis.1 An attractive approach to track, in real time, changes in the cholesterol content of the cell plasma membrane involves the use of a microelectrode to study the behavior of a single cell. This chapter reports electrochemical detection of cholesterol contained in the plasma membrane of a single cell using platinum microelectrodes modified with a lipid bilayer membrane containing cholesterol oxidase. Of particularly importance, the steady state electrode response correlates with the cholesterol content of the cell plasma membrane.
The idealized structural model of the thiolipid/lipid bilayer membrane with cholesterol oxidase partially inserted in the outer lipid leaflet is taken from the literature describing binding of cholesterol oxidase to cell plasma membranes.2 These studies
indicate that association of the enzyme with the cell plasma membrane allows cholesterol
contained in the membrane to move directly into the enzyme active pocket without
interaction with the aqueous phase. Therefore, in the work described here, cholesterol is
89 believed to partition into the electrode-supported lipid bilayer membrane prior to
enzymatic oxidation.3
Lipid bilayer vesicles have been used as solution-phase acceptors of cholesterol in
cellular efflux studies.4 This literature provides the basis for using microelectrodes
modified with a lipid bilayer membrane containing cholesterol oxidase to extract and
detect plasma membrane cholesterol. Cholesterol molecules contained in the plasma
membrane move across an aqueous (membrane solvation) layer and partition into the
electrode-supported lipid bilayer membrane. Detection is achieved through electrochemical oxidation of hydrogen peroxide generated upon enzyme-catalyzed oxidation of cholesterol by molecular oxygen.
4.2 EXPERIMENTAL.
Fabrication and characterization of platinum microelectrodes modified with lipid
bilayer membrane and cholesterol oxidase are described in chapter 3. The nature of
oocyte contact experiments is discussed in the results and discussion section. Xenopus
oocytes were a kind gift of Prof. Jianmin Cui. To remove the vitelline layer of oocytes,
oocytes were left in a stripping medium (200 mM N-Methyl-D-glutamine aspartate, 2mM
KCl, 10 mM EDTA, 1 mM MgCl2, 10 mM HEPES, pH 7.4) and monitored under light
microscope until the vitelline layer is visible (approx. 5 mins.). The visible vitelline layer
was removed using forceps under the microscope. After the removal the oocytes were
carefully moved using a pipette to another coverslip with ND96 buffer (96 mM NaCl, 2
90 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, 5 mM HEPES, pH 7.6). All the electrochemical experiments were conducted in ND96 buffer.
4.3 RESULTS AND DISCUSSION.
Experiments for detection of cholesterol contained in a single oocyte (Xenopus frog) are conducted by initially positioning the electrode about 5 μm from the cell surface
(Figure 4-1A) for acquisition of baseline data (no cholesterol detection). Repositioning the electrode and contacting the cell plasma membrane (Figure 4-1B) results in a steady- state current response. Contact is defined as the electrode position that corresponds to maximum electrode response. As shown in Figure 4-2, positioning the electrode directly adjacent to the cell (within ca. 1 μm or partially touching) produces an intermediate response, while contacting the cell with some force yields maximum current. The response observed adjacent to the cell may reflect detection of cholesterol efflux (i.e., solution-phase cholesterol). Control experiments for contacting cells with lipid bilayer-
15 μm A B
oocyte contact no contact
Figure 4-1. Photographs showing the electrode: (A) positioned about 5 μm from the
plasma membrane and (B) contacting the plasma membrane.
91 modified electrodes containing no enzyme show no response. Additional control experiments using bare platinum electrodes also show no response.
These experiments, in conjunction with data suggesting that electrode response correlates with the cholesterol content of the cell plasma membrane (vide infra), imply that the anodic current observed at the enzyme-modified electrodes reflects detection of cellular cholesterol. Cholesterol oxidase does have activity for oxidation of other sterols.
However, cholesterol is the only sterol reported to be a constituent of the cell plasma membrane.5 Additionally, the microelectrodes have been used to detect cholesterol within
cell contact 1 pA adjacent 10 s
no contact
Figure 4-2. Amperometric data for detection of cellular cholesterol at a microelectrode
(11.5 μm diameter) modified with a lipid bilayer membrane containing
cholesterol oxidase. No contact (Figure 4-1): baseline data; no cholesterol
detection. Adjacent: data for positioning the electrode within about 1 μm of
(or partially touching) the plasma membrane. Cell contact: data for
contacting the oocyte plasma membrane. Arrows approximate the times of
changing electrode position. The buffer is 0.1 M sodium phosphate, pH 6.5.
The electrode potential is 800 mV vs NHE.
92 the lipid bilayer membrane of giant vesicles6 (Appendix 4-1).
The apparent steady-state responses suggest that cholesterol oxidation is not limited by lateral diffusion of cholesterol in the plasma membrane to the electrode contact site (i.e., no current decay for depletion is observed). The steady-state responses could reflect a slow rate of cholesterol mass transfer from the plasma membrane to the electrode-supported lipid bilayer membrane compared to the rate of replenishment to the contact site though lateral diffusion. However, this possibility is not consistent with data indicating that response magnitude is dependent on electrode pretreatment conditions.
Electrode response depends on the time allowed for immobilization of enzyme on the electrode with longer incubation (in enzyme solution) leading presumably to more
withdraw withdraw
1 pA
20 s
contact contact Figure 4-3. Replicate experiments showing amperometric responses for contacting an
oocyte with a lipid bilayer modified microelectrode containing cholesterol
oxidase. Other conditions as in Figure 4-2.
93 immobilized enzyme and larger responses (Appendix 4-1). These data suggest that response magnitude is dictated by the rate of enzyme catalysis for the cholesterol concentration reached in the electrode supported lipid bilayer membrane under steady- state turnover. It is noted that the cholesterol content of the plasma membrane (0.5 cholesterol/phospholipid ratio7) is more than sufficient to produce the observed responses. For example, the cholesterol contained in the region of the plasma membrane defined by the electrode contact area (ca.100 amol cholesterol) could produce 20 pC. The total amount of cholesterol contained in the oocyte plasma membrane is estimated at 4-6 pmol (oocyte diameter is ca. 800-1000 μm).
It is not yet known if cholesterol contained in the inner leaflet of the plasma membrane significantly contributes to electrode response. The reported t1/2 values for
0.5 pA
5 s
Figure 4-4. Amperometric responses for contacting an oocyte with a lipid bilayer
modified microelectrode containing cholesterol oxidase. Response (A) prior
to cholesterol depletion, (B) after partial cholesterol depletion, and (C) after
near complete cholesterol depletion. All data collected at the same oocyte
using the same electrode. Other conditions as in Figure 4-2.
94 cholesterol transbilayer movement (flip-flop between the leaflets of the cell plasma membrane) range from <1 s to several hours.8 This uncertainty extends to the possibility that cholesterol diffuses to the electrode contact site from membranes inside the cell.
Intracellular cholesterol trafficking between the endoplasmic reticulum, Golgi, and plasma membrane is believed to involve vesicular transport and energy consumption.8
Lateral diffusion of cholesterol in the plasma membrane to the electrode contact site is likely the primary source that feeds cholesterol oxidation. The steady-state nature of the response is consistent with the notion that cholestenone (oxidized cholesterol) does not significantly accumulate in the electrode-supported lipid bilayer membrane and affect the rate of enzyme catalysis.
As shown in Figure 4-3, replicate experiments for contacting a cell using the same electrode show reproducible responses. This result is significant because it allows an electrode to be used for evaluating changes in the cholesterol content of the cell plasma membrane. Such behavior is demonstrated by measuring electrode response at cells that have been partially depleted of plasma membrane cholesterol through incubation in cyclodextrin solution8 (1 mM). Even electrodes yielding a relatively small response (e.g.,
0.8 pA) show a discernible decrease when cholesterol has been removed from the plasma
membrane (Figure 4-4). Exposing the cell to cyclodextrin solution for 5 min results in a
decrease in response of about 50% (compare traces A and B). Exposing the cell to
cyclodextrin solution for an additional 20 min results in a further decrease in response
(trace C). Although cyclodextrin selectively removes cholesterol from the cell plasma
membrane,8 the possibility that other changes in the plasma membrane (as a result of
cyclodextrin exposure) cause the decrease in electrode response cannot conclusively be
95 ruled out. It is noted, however, that cholesterol can be delivered to the plasma membrane from a cyclodextrin solution containing cholesterol and electrode response is reestablished (Appendix 4-1).
4.4 CONCLUSIONS.
Platinum microelectrodes modified with a lipid bilayer membrane incorporating cholesterol oxidase are used for detection of cholesterol contained in the plasma membrane of a single cell. Electrode response appears to correlate with the cholesterol content of the cell plasma membrane.
4.5 ACKNOWLEDGMENT.
This work was supported by the Department of Chemistry, Case Western Reserve
University. Discussions with Professors Dan Scherson and Barry Miller are also acknowledged. Oocytes were a gift from Professor Jianmin Cui.
96 APPENDIX 4-1
Giant phospholipid vesicles containing cholesterol as a constituent of the
membrane are used as a model system for the cell plasma membrane (Figure A4-1). The
vesicle formation procedure used6 describes the preparation of large unilamellar vesicles.
However, the work reported here give no proof that the vesicles are unilamellar.
Contacting a vesicle that contains cholesterol as a constituent of the bilayer membrane
with a cholesterol oxidase modified electrode produces current responses similar to those
observed for contacting oocytes (Figure A4-2, A). It is noted that the vesicles were
formed from a cholesterol/phospholipid mixure of 0.5/1 in an effort to match the
cholesterol content of the oocyte plasma membrane. Control experiments for contacting
vesicles that are devoid of cholesterol do not show anodic current responses. However,
some control experiments show a cathodic current spike upon contact between the
Figure A4-1. Giant vesicle captured at the tip of a glass capillary. The picture also shows
small multilamellar vesicles attached to the giant vesicle.
97 withdrawal
contact 1 pA 20 s
withdrawal contact
Figure A4-2. Response of an enzyme modified electrode for contacting a vesicle that
contains cholesterol as a membrane constituent (A). Response of an
enzyme modified electrode for contacting a vesicle that contains no
cholesterol (B). Other conditions as in Figure 4-2.
electrode and the vesicles membrane which may be due to a change in electrode capacitance (Figure A4-2, B). It is noted that vesicles that contain no cholesterol (control experiments) adhere to the electrode and withdrawal of the electrode from the contact position is not always possible. Vesicles that contain cholesterol as a constituent of the bilayer membrane do not adhere to the electrode. These initial vesicle experiments, in conjunction with the control experiments conducted at the oocytes (using bare platinum electrodes and lipid bilayer modified electrodes containing no enzyme) strongly support the conclusion that plasma membrane cholesterol is detected at the cholesterol oxidase modified electrodes.
98 Single cell experiments have been conducted where the time allowed for
incorporation of enzyme into the electrode-supported lipid bilayer membrane has been
varied. These experiments suggest that electrode response is dependent on the amount of
enzyme immobilized on the electrode surface. Figure A4-3 (A) shows data for contacting
an oocyte with a lipid bilayer modified electrode that has been exposed to enzyme
B 0.3 pA 10 s A
contact
Figure A4-3. Electrode response for contacting an oocyte with a lipid bilayer modified
electrode incubated in enzyme solution for (A) 1 hour and (B) the
response of the same electrode at the same cell after incubated in enzyme
solution for an additional 2 hours. Other conditions as in Figure 4-2.
solution (for immobilization of the enzyme) for 1 hour. Figure A4-3 (B) shows a larger
response for contacting the same cell with the same electrode after exposure to enzyme
solution for an additional hour. It is speculated that longer exposure of the lipid bilayer
modified electrode to enzyme solution results in an increase in the amount of enzyme
immobilized on the electrode and that higher enzyme loading leads to larger responses.
The data clearly indicate that electrode response is influenced by electrode pretreatment
99 conditions. If the current responses were strictly limited by diffusion of cholesterol to the electrode contact site, electrode pretreatment would not be expected to affect the response.
Additionally, these data suggest that cholesterol mass transfer from the cell plasma membrane to the electrode supported lipid bilayer membrane is not strictly rate limiting. Therefore, it is proposed that current is limited by the rate of enzyme catalysis.
Response magnitude is likely a function of the amount of enzyme immobilized on the electrode surface and the cholesterol concentration reached in the electrode-supported lipid bilayer membrane under steady state turnover. This mechanistic model could explain the steady state responses and the dependence of response magnitude on the 2 pA
10 s
Figure A4-4. Response of an enzyme modified electrode for contacting an oocyte that
has been depleted of cholesterol (A). Response of the same electrode after
delivery of cholesterol to the cell plasma membrane (B). Other conditions
as in Figure 4-2.
100 cholesterol content of the plasma membrane. Cyclodextrin solutions containing
cholesterol can be used to deliver cholesterol to the cell plasma membrane. Figure A4-4 ,
A shows the lack of electrode response for contacting a cell that has been depleted of
cholesterol through exposure to 100 mM cyclodextrin. Figure A4-4 (B) show the
response of the same electrode after exposure of the cell to 100 mM cyclodextrin containing 1.25 mM cholesterol (for longer than 1 hour). The electrode response is assigned to cholesterol that has been delivered to the cell plasma membrane from solution. These data suggest that the decrease in response observed for exposure of cells to cyclodextrin solution (Figure 4-4) is due to depletion of cholesterol from the cell plasma membrane.
101 4.6 REFERENCES.
(1) Simons, K.; Ikonen, E. "Review: Cell biology: How cells handle cholesterol." Science 2000, 290, 1721-1726
(2) Chen, X.; Wolfgang, D. E.; Sampson, N. S. "Use of the Parallax-Quench Method to Determine the Position of the Active-Site Loop of Cholesterol Oxidase in Lipid Bilayers." Biochemistry 2000, 39, 13383-13389
(3) Devadoss, A.; Burgess, J. D. "Detection of cholesterol through electron transfer to cholesterol oxidase in electrode-supported lipid bilayer membranes." Langmuir 2002, 18, 9617-9621
(4) Phillips, M. C.; Johnson, W. J.; Rothblat, G. H. "Mechanisms and consequences of cellular cholesterol exchange and transfer." Biochim. Biophys. Acta 1987, 906, 223-76
(5) Luria, A.; Vegelyte-Avery, V.; Stith, B.; Tsvetkova, N. M.; Wolkers, W. F.; Crowe, J. H.; Tablin, F.; Nuccitelli, R. "Detergent-Free Domain Isolated from Xenopus Egg Plasma Membrane with Properties Similar to Those of Detergent- Resistant Membranes." Biochemistry 2002, 41, 13189-13197
(6) Karlsson, M.; Nolkrantz, K.; Davidson, M. J.; Stroemberg, A.; Ryttsen, F.; Kerman, B.; Orwar, O. "Electroinjection of Colloid Particles and Biopolymers into Single Unilamellar Liposomes and Cells for Bioanalytical Applications." Anal. Chem. 2000, 72, 5857-5862
(7) Santiago, J.; Guzman, G. R.; Rojas, L. V.; Marti, R.; Asmar-Rovira, G. A.; Santana, L. F.; McNamee, M.; Lasalde-Dominicci, J. A. "Probing the effects of membrane cholesterol in the Torpedo californica acetylcholine receptor and the novel lipid-exposed mutation aC418W in Xenopus oocytes." J. Biol. Chem. 2001, 276, 46523-46532
(8) Haynes, M. P.; Phillips, M. C.; Rothblat, G. H. "Efflux of Cholesterol from Different Cellular Pools." Biochemistry 2000, 39, 4508-4517
102 CHAPTER 5.
COVALENT MODIFICATION OF CHOLESTEROL OXIDASE TO PLATINUM
MICROELECTRODES FOR DETECTION OF CHOLESTEROL IN THE
PLASMA MEMBRANE OF SINGLE CELLS
5.1 INTRODUCTION.
Earlier works from this group demonstrated detection of plasma membrane
cholesterol in a single oocyte cell using platinum microelectrodes modified with a lipid
bilayer membrane containing immobilized cholesterol oxidase.1 This report describes
experiments for detection of oocyte plasma membrane cholesterol using platinum
electrodes modified with a sub-monolayer of covalently attached cholesterol oxidase.
The EDC scheme employed here to covalently attach cholesterol oxidase to platinum electrodes has been used by other groups to immobilize cytochrome c on solid
electrodes.2-4 Covalent linking of proteins using EDC (N-(3-Dimethylaminopropyl)-N'- ethylcarbodiimide hydrochloride) is non-specific with respect to protein orientation on the electrode surface as lysine residues are the primary target of the linker. A distribution of orientations and conformations of cholesterol oxidase is expected to exist on the platinum electrodes. No proof is provided here to verify that cholesterol oxidase is actually covalently linked to the electrode surface. However, exposure of oxidase solution to bare platinum does not produce active electrodes.
Detection of cholesterol contained in the cell plasma membrane involves
positioning the electrode in contact with the plasma membrane. A thin aqueous layer
103 between the hydrophilic electrode surface and the plasma membrane is predicted due to
lipid head group solvation. Cholesterol has a finite solubility in the aqueous phase (CMC
≈ 30 nM)5 and efflux of cell plasma membrane cholesterol into water (containing no
cholesterol acceptor such as cyclodextrin) is slow with a t1/2 of days. The electrode
consumes aqueous phase cholesterol that is directly adjacent to, and in equilibrium with,
the cell plasma membrane. This consumption of aqueous phase cholesterol at the cell
surface causes further cholesterol efflux from the plasma membrane so that cholesterol
mass transport occurs from the high concentration in the plasma membrane to the lower
concentration at the electrode surface. Electrode response is highly dependent on the
force applied on the oocyte cell plasma membrane by the electrode and some speculation
regarding this result is given below.
5.2 EXPERIMENTAL.
Fabrication of platinum microelectrodes and the electrochemical instrumentations
are described in chapter 3. EDC and Triton X-100 were a kind gift of Prof. Salomon,
Department of Chemistry, Case Western Reserve University. Cholesterol oxidase (Wako
Pure Chemicals Inc.) and cholesterol (Sigma Inc.) were dissolved in 100 mM sodium
phosphate buffer, pH 7.4. Platinum microelectrodes were immersed in 1 mM ethanolic
solution of MUA-11 for 2 hrs. MUA (11- mercapto undecanoic acid, Sigma, Inc.)
modified electrodes were immersed in 1mM EDC dissolved in water for 1 hr and then
transferred to 2 mg/ml solution of cholesterol oxidase for 3 hrs. Cholesterol was dissolved in chloroform and dried under nitrogen before preparing the solutions of
104 required concentrations. The dried cholesterol film was added with the required amount
of sodium phosphate buffer and 1 % (v/v) of Triton X-100 and sonicated to get a final
required concentration of cholesterol.
Solution flow analyses were conducted in a custom made flow cell. A
microscope glass slide with a circular silicone barrier in the middle which could hold a
droplet with a volume of ca. 200 μl was used. The flow rate used was 500 μl/min. The
flow was controlled using a six-way valve (Rheodyne Model 5020, Supelco Corp.) and
two syringe pumps (Harvard Apparatus) to direct buffer Triton X-100 or buffered Triton
X-100 containing cholesterol over the electrode. Injection times were controlled
manually.
5.3 RESULTS AND DISCUSSION.
Cyclic voltammetry of potassium ferrocyanide at a bare platinum electrode is
compared to the voltammogram after deposition of MUA-11 (Figure 5-1 A and B). The
decreases in the peak currents suggest deposition of a sub-monolayer of MUA-11 to the
platinum electrode. The structure of the MUA sub-monolayer is not known and it may form islands or be unifrmly distributed on the electrode surface. There is no discernable
change in the voltammetry for reaction of EDC with the MUA sub-monolayer. Figure 5-
1C shows a ferrocyanide voltammogram after modifying the electrode with cholesterol
oxidase. The additional decreases in peak currents are consistent with cholesterol
oxidase being immobilized on the electrode surface. The cholesterol oxidase
modification step is complete within 3 hrs. Extending the modification time to 12 hrs
105 does not show a further decrease in peak current or any increase in activity towards cholesterol oxidation (data not shown).
Figure 5-2 shows the current response obtained for flow injection exposure of
cholesterol to an electrode modified with choelsterol oxidase where the flow is changed
from buffered Triton X-100 to buffered Triton X-100 containing cholesterol and then
reverted to Triton X-100. Three cholesterol concentrations are shown and the replicate
injections (concentrations of 1.0 mM, 0.5 mM and 0.2 mM) demonstrate response
stability over hours. Control experiments for flow injection exposure of cholesterol at an
electrode modified with only MUA-11 and EDC (containing no immobilized cholesterol
140 A 120 100 B C 80 60 40 20 current (nA) 0 -20 -40 -60 -200 0 200 400 600 800 potential (mV)
Figure 5-1. Cyclic voltammogram of 5 mM potassium ferrocyanide at A) a bare platinum
electrode with 100 μm diameter B) same electrode after modified with
MUA-11 and C) same electrode after modified with cholesterol oxidase.
Scan rate 50 mVs-1.
106 oxodase) show no response. The current response measured at the oxidase modified electrode is attributed to enzymatic catalysis of cholesterol oxidation by molecular oxygen and electrochemical oxidation of the generated hydrogen peroxide at the platinum electrode surface. The current response is believed to be limited by enzyme turnover rate. The electrode responses vary by ± 20% between different electrodes.
Figure 5-3A shows the cholesterol oxidase modified electrode positioned 200 μm away from the plasma membrane of the oocyte for collection of baseline data. Figure 5-
3B shows the electrode contacting the plasma membrane to yield an electrode response.
Figure 5-4A shows that the contact force applied between the electrode and the plasma membrane significantly affects electrode response with larger force yielding a larger response. The cell plasma membrane has significant surface roughness (e.g., coated pits).
1mM 1mM
0.5mM 100pA 200s 0.5mM
0.2mM 0.2mM
Figure 5-2. Flow injection amperometry data for injection of various concentration of
cholesterol dissolved with Triton X-100. The (↑) shows the time when the
flow is changed from buffer to cholesterol and (↓) shows the time when the
flow is changed from cholesterol to buffer.
107 Higher contact force likely creates in a thinner aqueous layer between the electrode surface and the plasma membrane and perhaps a more planer region of plasma membrane at the electrode contact site. Figure 5-4B1 is a control experiment for contacting a cell with a bare platinum electrode. No response is observed even under increased contact force Figure 5-4B2. This suggests that cholesterol is selectively detected at the enzyme modified electrode and that other species are not oxidized at the platinum electrode at a rate that produces measurable current.
The data shown in Figure 5-5 demonstrates the reproducibility of the electrode
response for replicate contact of the plasma membrane (with the same contact force).
The stability of the current response for consecutive contacts of plasma membrane also
suggests that cholestenone (oxidized cholesterol) does not significantly affect the
electrode activity.
A B
200 μm
Figure 5-3. Photographs showing cholesterol oxidase modified A) positioned ca.150 μm
away from oocyte plasma membrane B) contacting the plasma membrane of
oocyte.
108 A calculation of the number of cholesterol molecules present in the region of the
plasma membrane that is directly adjacent to the 100 μm diameter electrode (e.g.,
electrode footprint) gives an equivalent of approximately 1.6 nC (0.33 mole fraction cholesterol). Assuming 100 % oxidation of the generated hydrogen peroxide, even the
largest response obtained (66 pA) suggests that the electrode oxidizes only about 4% of cholesterol available in the electrode footprint (over 20 s). Lateral diffusion of cholesterol in the plasma membrane to the electrode contact site could replenish this loss
3
A 4 20 pA 20 25 s 2
1
B 4 pA 50 s 1 2
Figure 5-4. Amperometry data obtained when cholesterol oxidase modified electrode (A)
1) contacting the plasma membrane of the oocyte 2) contacting the plasma
membrane with additional force 3) withdrawn to an applied force closer to
postion 1 4) withdrawn away from the plasma membrane (B) Bare platinum
electrode 1) contacting the plasma membrane 2) contacting the plasma
membrane with additional force.
109 in cholesterol.
McConnell and co-workers have calculated cholesterol release rates in the
presence of cyclodextrin (a well documented aqueous phase cholesterol acceptor of
cellular efflux) as a function of mole fraction of cholesterol in phospholipid monolayers.
Taking the surface roughness of the plasma membrane (4-20 surface roughness factor)
into account, this efflux would produce currents of 5-100 pA. This result is not
surprising given that the estimated concentration of cholesterol in the cell plasma
membrane is ca. 130 pmol/cm2 for a monolayer1 and that lateral diffusion and flip-flop
provide replenishment of cholesterol to the electrode contact site. It is not yet known if
cholesterol present in the inner leaflet of the vesicle lipid bilayer membrane significantly
contributes to electrode response. The rate of cholesterol flip-flop (transbilayer
movement) across the lipid bilayer membrane is not known and discrepancies are found
in the literature.6 However, second harmonic generation experiments for tracking
transbilayer movement of phospholipid7 and cholesterol (John C. Conboy; private
communication) suggest that cholesterol flip-flop occurs with a t1/2 of less than a minute.
This notion is consistent with work by others for gauging cholesterol flip- flop in cell
plasma membranes.6 Therefore, it is proposed that cholesterol is replenished to the electrode contact site by both transbilayer movement and lateral diffusion along the lipid bilayer membrane.
5.4 CONCLUSION.
Cholesterol oxidase is covalently immobilized on platinum disk electrodes.
Ferrocyanide cyclic voltammogram studies suggest that a sub-monolayer of cholesterol
110 oxidase is attached to the platinum surface. Hydrogen peroxide generated by enzyme catalyzed oxidation of cholesterol by molecular oxygen is electrochemically oxidized at the platinum electrode electrochemically. Flow injection experiments for exposure of cholesterol solution to the electrode show that response correlates with cholesterol concentration. The electrodes allow detection of cholesterol contained in the plasma membrane of a single oocyte cell where response is found to depend on contact force between the electrode and the plasma membrane. This behavior is consistent with experiments for detection of oocyte plasma membrane using platinum electrodes modified with a thiolipid/lipid bilayer incorporating cholesterol oxidase. The lipid bilayer electrode architecture produces larger current densities for detection of plasma membrane cholesterol compared to the covalent attachment scheme.
5.5 ACKNOWLEDGEMENTS.
This work was supported by the National Institute of Health (5 R21
EB003925)and the Department of Chemistry, Case Western Reserve University. Helpful discussions with Professors Dan Scherson and Barry Miller are also acknowledged.
111 5.6 REFERENCES.
(1) Devadoss, A.; Burgess, J. D. "Steady-State Detection of Cholesterol Contained in the Plasma Membrane of a Single Cell Using Lipid Bilayer-Modified Microelectrodes Incorporating Cholesterol Oxidase." J. Am. Chem. Soc. 2004, 126, 10214-10215
(2) Scheller, F. W.; Wollenberger, U.; Lei, C.; Jin, W.; Ge, B.; Lehmann, C.; Lisdat, F.; Fridman, V. "Bioelectrocatalysis by redox enzymes at modified electrodes." Rev. Mol. Biotech. 2002, 82, 411-424
(3) Hedges, D. H. P.; Richardson, D. J.; Russell, D. A. "Electrochemical control of protein monolayers at indium tin oxide surfaces for the reagentless optical biosensing of nitric oxide." Langmuir 2004, 20, 1901-1908
(4) Collinson, M.; Bowden, E. F.; Tarlov, M. J. "Voltammetry of covalently immobilized cytochrome c on self-assembled monolayer electrodes." Langmuir 1992, 8, 1247-50
(5) Haberland, M. E.; Reynolds, J. A. "Self-association of cholesterol in aqueous solution." Proc Natl Acad Sci U S A 1973, 70, 2313-16
(6) Haynes, M. P.; Phillips, M. C.; Rothblat, G. H. "Efflux of Cholesterol from Different Cellular Pools." Biochemistry 2000, 39, 4508-4517
(7) Liu, J.; Conboy, J. C. "Direct Measurement of the Transbilayer Movement of Phospholipids by Sum-Frequency Vibrational Spectroscopy." J Am Chem Soc 2004, 126, 8376-8377
112 CHAPTER 6.
SCANNING FORCE MICROSCOPY IMAGES OF CHOLESTEROL OXIDASE
IMMOBILIZED IN SUPPORTED LIPID BILAYER MEMBRANE
6.1 INTRODUCTION.
A goal of this research group is the further development and implementation of
microelectrodes modified with a lipid bilayer membranes containing immobilized
cholesterol oxidase for detection of cholesterol contained in the plasma membrane of
single cells.1-3 This electrode architecture is believed to retain an active enzyme
conformation by mimicking the natural interaction of the bacterial enzyme with the cell
plasma membrane of mammalian cells. This note reports a tapping mode-scanning force microscopic (TM-SFM) study aimed at characterizing the state (e.g., monomers vs. aggregates) of the enzyme associated with supported lipid bilayer membranes.
The enzyme immobilization scheme is supported by structural studies of
cholesterol oxidase, as well as reports indicating that the enzyme associates with lipid bilayer vesicles and cell plasma membranes during oxidation of membrane-resident
cholesterol. From X-ray crystallographic data4, Sampson et al. have identified two Ω
loops (flexible structures) on the enzyme surface that cap the substrate-binding cavity and
sequester it from bulk solvent. It has been proposed that the Ω loops open due to
hydrophobic interactions upon association of the enzyme with the membrane. These
conformational changes are believed to provide a hydrophobic path through which
cholesterol molecules in the membrane can move into the active site of the enzyme. It is
113 likely that cholesterol oxidase associates with the membrane in part by the amino acid side chains of these Ω loops, along with a more extensive contact surface.5
This hypothesis is supported by a mutation study in which one Ω-loop residue
was labeled with an environment-sensitive probe (acrylodan). The fluorescence change
observed upon exposure of the labeled enzyme to lipid bilayer vesicles indicated
movement of the acrylodan probe to a more hydrophobic environment (i.e., association
with the membrane interior).6 It is also noted that cholesterol oxidase has been used
widely to probe the cholesterol content of cell membranes.7 Sampson has reported a
disassociation constant (KD) of 57 ± 21 μM for binding of cholesterol oxidase (from
Streptomyces Sp.) to mixed phospholipid/cholesterol vesicles.5 At room temperature, the
enzyme used in this work (Pseudomonas Sp.) disassociates at a slow rate (t1/2 ca. 2 days)
from supported lipid bilayer membranes formed on mica and on various electrode
materials.3 It is noted that the enzyme more quickly disassociates from supported lipid
bilayer membranes at 37°C (t1/2 ca. 5 min.).
It is well documented that vesicle fusion can be used to deposit lipid bilayer
membranes on cleaved mica and there is much scanning probe microscopy data available
in the literature on this system.8-11 This is the method used to prepare supported lipid
bilayer structures for imaging immobilized cholesterol oxidase. The images of deposited
lipid bilayer membranes on mica reported here are, in general, very similar to those
published by other groups.8,9 The surfaces contain large continuous regions of lipid
bilayer coverage with no bilayer step edges (i.e., lipid bilayer plateaus). Upon exposure
to cholesterol oxidase solution, features that are assigned monomers and small aggregates
of cholesterol oxidase are observed on lipid bilayer plateaus. The features also exhibit a
114 difference in phase contrast compared to the surrounding lipid bilayer plateau and the
possibility that the features are small islands of lipid bilayer is ruled out based on height
measurements and phase data. The TM-SFM images pictorially support the model
proposed by Sampson and co-workers where cholesterol oxidase binds to the surface of
the lipid bilayer membrane partially inserting into the hydrophobic membrane interior.
6.2 EXPERIMENTAL SECTION.
Sample Preparation. Cholesterol oxidase (Wako Pure Chemical Industries, Ltd.,
biochemistry grade) (2 mg/ml) was dissolved in 50 mM sodium phosphate buffer with a
pH 7.4. DPPC (Avanti Polar Lipids, Inc.) dissolved in chloroform is dried under nitrogen
and sonicated in buffer until clear to obtain a 1mM vesicle solution. A freshly cleaved
mica surface is exposed to the vesicle solution for 30 mins. The substrate with DPPC membrane is exposed to cholesterol oxidase solution for 15 mins and dried to image cholesterol oxidase in lipid membranes. The samples were rinsed with water and dried before imaging.
Instrumentation. All images were acquired in air using MAC mode- Scanning force
microscope (Molecular Imaging, Inc.). The tapping frequency of the probes ( Type II
MAClevers) were in the range of 60-85 kHz. The images were scanned at a range of 4 –
15 μm/s.
115 6.3 RESULTS AND DISCUSSION.
Figure 6-1 A shows a topographical image of DPPC deposited on cleaved mica by vesicle fusion. The topography of the multilamellar membrane consists of plateaus that differ in the height associated with the thickness of the lipid bilayer (ca. 5 nm). The images are very similar to those reported in the literature for lipid membrane deposition on mica by vesicle fusion. Others have reported heights of 4-5 nm9 for lipid bilayer step edges and discrepancies with the predicted height based on the structures of the lipid have been explained by sample deformation by the tip. Topographical images collected in water are similar and the step edges appeared to be 4.5 nm.
Figure 6-1 B is the phase contrast image collected simultaneously with the topography image shown in Figure 6-1A. Note that all of the lipid membrane plateau regions exhibit nearly the same degree of phase contrast irrespective of the height. This is expected because phase contrast is primarily a function of tip-sample interactions and changes in height are effectively rejected in these data. The flat phase contrast image is
A B
Å Deg 100 12 nm nm nm nm 2000 2000 2000 2000 1000 1000 1000 1000 0 0 Figure 6-1. TM-SFM images of lipid bilayer membrane modified mica surface A)
topographical image B) phase contrast image.
116 consistent with the notion that the entire surface is covered by a multilamellar lipid
membrane.
Upon exposure of the lipid membrane to cholesterol oxidase solution,
topographical images show step edges that are predominately the thickness of a lipid
bilayer in height. After exposure to enzyme solution, much of the surface showed lipid
bilayer plateaus decorated with hemispherical features that appear to be 2-5 nm in height
and 10-50 nm in diameter (e.g., Figure 6-2A). It is noted that lateral diminutions of small
particles can appear to be more than double in size due to the aspect ratio of the tip.12
Thus, a single cholesterol oxidase molecule partially inserting into the lipid membrane and having dimensions of 7.3 x 6.3 x 5.1 nm (taken from the crystal structure of the
Streptomyces Sp. enzyme)4 may appear as a feature that is ca. 10-15 nm in diameter.
Small islands of enzyme molecules that are laterally spaced in a monolayer structure
likely explain the features with larger lateral deminsions. Thus, the sizes of the features
imaged on lipid plateau regions of the surface are consistent with the dimensions of the
enzyme and its predicted insertion in the lipid membrane. Images collected in water are
A B
Deg Å 12 100 nm nm nm nm 800 800 800 800 400 400 400 400 0 0 Figure 6-2. TM-SFM images of cholesterol oxidase immobilized in lipid bilayer
membrane on mica surface A) topographical image B) phase contrast image.
117 similar and the features assigned to immobilized enzyme are less clearly observed.
Figure 6-2B shows the phase contrast image collected simultaneously with the
topographical image shown in Figure 6-2A. The lipid plateaus exhibit a phase shift that
is in agreement with that measured prior to exposure to enzyme (e.g., Figure 6-1B).
Importantly, the hemispherical features surrounded by lipid membrane exhibit a negative
phase shift compared to the lipid membrane. This large difference in phase relative to the
surrounding lipid membrane suggests that the tip is interacting with a material with different properties (e.g., elasticity). These data further support the proposed assignment of the features to cholesterol oxidase monomers and aggregates. The aggregates are likely 2-5 cholesterol oxidase molecules arranged in a monolayer structure with each enzyme partially inserted into the lipid membrane. Phase contrast images collected in water also show a negative phase shift for immobilized enzyme.
The images of cholesterol oxidase in multilamellar lipid membranes on
cleaved mica may not be strictly relevant to the electrode supported thiolipid/lipid bilayer
membranes formed on platinum. Nevertheless, the data suggest that aggregate formation
may be an issue for immobilization of this enzyme in supported lipid bilayer membranes.
It is noted that the thiolipid/lipid bilayer modified platinum electrodes are exposed to
enzyme for hours to produce surfaces that exhibit measurable enzymatic activity.2,3 TM-
SFM images for longer exposure of lipid membrane modified mica to enzyme solution suggest that the surface becomes covered by aggregates of cholesterol oxidase and it is not possible to clearly discern the existence of lipid membrane plateaus.
118 6.4 CONCLUSIONS.
TM-SFM images of a multilamellar lipid membrane on cleaved mica show
membrane plateaus and step edges that are one lipid bilayer thicknesses in height.
Exposure of cholesterol oxidase solution to the lipid membrane results in immobilization
of enzyme in lipid membrane plateaus. The immobilized enzyme yields a negative phase
shift compared to the surrounding lipid membrane plateau. Based on the size and shape
of the features assigned to enzyme, it is proposed that cholesterol oxidase exist as
monomers and aggregates partially inserted in the lipid membrane. The images
pictorially support the proposed model for interaction of the bacterial enzyme with the
plasma membrane of mammalian cells.
6.5 ACKNOWLEDGEMENTS.
This work was supported by the National Institute of Health (5 R21 EB003925)
and the Department of Chemistry, Case Western Reserve University. Helpful discussions with Professors Dan Scherson and Barry Miller are also acknowledged
119 6.6 REFERENCES.
(1) Devadoss, A.; Burgess, J. D. "Detection of cholesterol through electron transfer to cholesterol oxidase in electrode-supported lipid bilayer membranes." Langmuir 2002, 18, 9617-9621
(2) Devadoss, A.; Burgess, J. D. "Steady-State Detection of Cholesterol Contained in the Plasma Membrane of a Single Cell Using Lipid Bilayer-Modified Microelectrodes Incorporating Cholesterol Oxidase." J. Am. Chem. Soc. 2004, 126, 10214-10215
(3) Bokoch, M. P.; Devadoss, A.; Palencsar, M. S.; Burgess, J. D. "Steady-state oxidation of cholesterol catalyzed by cholesterol oxidase in lipid bilayer membranes on platinum electrodes." Anal. Chim. Acta 2004, 519, 47-55
(4) Yue, Q. K.; Kass, I. J.; Sampson, N. S.; Vrielink, A. "Crystal structure determination of cholesterol oxidase from Streptomyces and structural characterization of key active site mutants." Biochemistry 1999, 38, 4277-4286
(5) Sampson, N. S.; Kass, I. J.; Ghoshroy, K. B. "Assessment of the Role of an W Loop of Cholesterol Oxidase: A Truncated Loop Mutant Has Altered Substrate Specificity." Biochemistry 1998, 37, 5770-5778
(6) Chen, X.; Wolfgang, D. E.; Sampson, N. S. "Use of the Parallax-Quench Method to Determine the Position of the Active-Site Loop of Cholesterol Oxidase in Lipid Bilayers." Biochemistry 2000, 39, 13383-13389
(7) MacLachlan, J.; Wotherspoon, A. T. L.; Ansell, R. O.; Brooks, C. J. W. "Cholesterol oxidase: sources, physical properties and analytical applications." Journal of Steroid Biochemistry and Molecular Biology 2000, 72, 169-195
(8) Dufrene, Y. F.; Lee, G. U. "Advances in the characterization of supported lipid films with the atomic force microscope." Biochim Biophys Acta 2000, 1509, 14- 41
(9) Janshoff, A.; Steinem, C. "Scanning force microscopy of artificial membranes." ChemBioChem 2001, 2, 798-808
(10) Richter, R. P.; Brisson, A. R. "Following the formation of supported lipid bilayers on mica: A study combining AFM, QCM-D, and ellipsometry." Biophys J 2005, 88, 3422-3433
(11) Schoenherr, H.; Johnson, J. M.; Lenz, P.; Frank, C. W.; Boxer, S. G. "Vesicle Adsorption and Lipid Bilayer Formation on Glass Studied by Atomic Force Microscopy." Langmuir 2004, 20, 11600-11606
120 (12) Takano, H.; Kenseth, J. R.; Wong, S.-S.; O'Brien, J. C.; Porter, M. D. "Chemical and biochemical analysis using scanning force microscopy." Chem. Rev. (Washington, D. C.) 1999, 99, 2845-2890
121 BIBLIOGRAPHY
Abe, T.; Lau, Y. Y.; Ewing, A. G. "Characterization of glucose microsensors for intracellular measurements." Anal Chem 1992, 64, 2160-3
Aleksandrovsky, Y. A.; Titov, V. N. "Influence of the structural organization of cholesterol aggregates in water-organic media on cholesterol oxidation catalyzed by cholesterol oxidase." Journal of Chemical and Biochemical Kinetics 1991, 1, 337-45
Amphlett, J. L.; Denuault, G. "Scanning Electrochemical Microscopy (SECM): An Investigation of the Effects of Tip Geometry on Amperometric Tip Response." J. Phys. Chem. B 1998, 102, 9946-9951
Atrat, P. G.; Wagner, B.; Wagner, M.; Schumann, G. "Localization of the cholesterol oxidase in Rhodococcus erythropolis IMET 7185 studied by immunoelectron microscopy." J Steroid Biochem Mol Biol 1992, 42, 193-200
Axup, A. W.; Albin, M.; Mayo, S. L.; Crutchley, R. J.; Gray, H. B. "Distance dependence of photoinduced long-range electron transfer in zinc/ruthenium-modified myoglobins." J Am Chem Soc 1988, 110, 435-9
Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 1980.
Bertrand, C.; Coulet, P. R.; Gautheron, D. C. "Enzyme electrode with collagen- immobilized cholesterol oxidase for the microdetermination of free cholesterol." Anal Lett 1979, 12, 1477-88
Bianco, P.; Haladjian, J. "Electrochemistry of cytochrome c3 at a lipid-modified graphite electrode." J. Electroanal. Chem. 1994, 367, 79-84
Bianco, P.; Haladjian, J. "Control of the electron transfer reactions between c-type cytochromes and lipid-modified electrodes." Electrochim. Acta 1994, 39, 975-82
Bokoch, M. P.; Devadoss, A.; Palencsar, M. S.; Burgess, J. D. "Steady-state oxidation of cholesterol catalyzed by cholesterol oxidase in lipid bilayer membranes on platinum electrodes." Anal Chim Acta 2004, 519, 47-55
Bongiovanni, C.; Ferri, T.; Poscia, A.; Varalli, M.; Santucci, R.; Desideri, A. "An electrochemical multienzymatic biosensor for determination of cholesterol." Bioelectrochemistry 2001, 54, 17-22
Brown, D. A.; London, E. "Structure and function of sphingolipid- and cholesterol- rich membrane rafts." J Biol Chem 2000, 275, 17221-17224
122
Buoninsegni, F. T.; Herrero, R.; Moncelli, M. R. "Alkanethiol monolayers and alkanethiol|phospholipid bilayers supported by mercury: an electrochemical characterization." J. Electroanal. Chem. 1998, 452, 33-42
Burgess, J. D.; Jones, V. W.; Porter, M. D.; Rhoten, M. C.; Hawkridge, F. M. "Scanning force microscopy images of cytochrome c oxidase immobilized in an electrode-supported lipid bilayer membrane." Langmuir 1998, 14, 6628-6631
Burgess, J. D.; Rhoten, M. C.; Hawkridge, F. M. "Cytochrome c Oxidase Immobilized in Stable Supported Lipid Bilayer Membranes." Langmuir 1998, 14, 2467-2475
Burgess, J. D.; Rhoten, M. C.; Hawkridge, F. M. "Observation of the Resting and Pulsed States of Cytochrome c Oxidase in Electrode-Supported Lipid Bilayer Membranes." J Am Chem Soc 1998, 120, 4488-4491
Burgess, J. D.; Hawkridge, F. M. "Direct electrochemistry of proteins and enzymes at electrodes." Electroanalytical Methods for Biological Materials 2002, 109-142
Cheillan, F.; Lafont, H.; Termine, E.; Fernandez, F.; Sauve, P.; Lesgards, G. "Molecular characteristics of the cholesterol oxidase and factors influencing its activity." Biochim Biophys Acta 1989, 999, 233-8
Chen, X.; Wolfgang, D. E.; Sampson, N. S. "Use of the Parallax-Quench Method to Determine the Position of the Active-Site Loop of Cholesterol Oxidase in Lipid Bilayers." Biochemistry 2000, 39, 13383-13389
Collinson, M.; Bowden, E. F.; Tarlov, M. J. "Voltammetry of covalently immobilized cytochrome c on self-assembled monolayer electrodes." Langmuir 1992, 8, 1247-50
Crockett, E. L.; Hazel, J. R. "Sensitive assay for cholesterol in biological membranes reveals membrane-specific differences in kinetics of cholesterol oxidase." J Exp Zool 1995, 271, 190-5
Cullison, J. K.; Hawkridge, F. M.; Nakashima, N.; Yoshikawa, S. "A Study of Cytochrome c Oxidase in Lipid Bilayer Membranes on Electrode Surfaces." Langmuir 1994, 10, 877-82
Devadoss, A.; Burgess, J. D. "Detection of cholesterol through electron transfer to cholesterol oxidase in electrode-supported lipid bilayer membranes." Langmuir 2002, 18, 9617-9621
Devadoss, A.; Burgess, J. D. "Steady-State Detection of Cholesterol Contained in the Plasma Membrane of a Single Cell Using Lipid Bilayer-Modified Microelectrodes Incorporating Cholesterol Oxidase." J Am Chem Soc 2004, 126, 10214-10215
123 Ding, L.; Li, J.; Dong, S.; Wang, E. "Supported phospholipid membranes: comparison among different deposition methods for a phospholipid monolayer." J. Electroanal. Chem. 1996, 416, 105-112
Dolfi, A.; Tadini-Buoninsegni, F.; Moncelli, M. R.; Guidelli, R. "Photocurrents Generated by Bacteriorhodopsin Adsorbed on Thiol/Lipid Bilayers Supported by Mercury." Langmuir 2002, 18, 6345-6355
Donley, C.; Dunphy, D.; Paine, D.; Carter, C.; Nebesny, K.; Lee, P.; Alloway, D.; Armstrong, N. R. "Characterization of Indium-Tin Oxide Interfaces Using x-ray Photoelectron Spectroscopy and Redox Processes of a Chemisorbed Probe Molecule: Effect of Surface Pretreatment Conditions." Langmuir 2002, 18, 450-457
Doukyu, N.; Aono, R. "Two moles of O2 consumption and one mole of H2O2 formation during cholesterol peroxidation with cholesterol oxidase from Pseudomonas sp. strain ST-200." Biochem J 1999, 341, 621-7
Dufrene, Y. F.; Lee, G. U. "Advances in the characterization of supported lipid films with the atomic force microscope." Biochim Biophys Acta 2000, 1509, 14-41
Edmiston, P. L.; Saavedra, S. S. "Molecular orientation distributions in protein films: III. Yeast cytochrome c immobilized on pyridyl disulfide-capped phospholipid bilayers." Biophys J 1998, 74, 999-1006
Edmiston, P. L.; Saavedra, S. S. "Molecular Orientation Distributions in Protein Films. 4. A Multilayer Composed of Yeast Cytochrome c Bound through an Intermediate Streptavidin Layer to a Planar Supported Phospholipid Bilayer." J Am Chem Soc 1998, 120, 1665-1671
Elbicki, J. M.; Morgan, D. M.; Weber, S. G. "Theoretical and practical limitations on the optimization of amperometric detectors." Anal Chem 1984, 56, 978-85
Fahey, P. F.; Koppel, D. E.; Barak, L. S.; Wolf, D. E.; Elson, E. L.; Webb, W. W. "Lateral diffusion in planar lipid bilayers." Science (Washington, DC, U. S.) 1977, 195, 305-6
Fischer, B.; Heyn, S. P.; Egger, M.; Gaub, H. E. "Antigen binding to a pattern of lipid-anchored antibody binding sites measured by surface plasmon microscopy." Langmuir 1993, 9, 136-40
Gao, H.; Luo, G. A.; Feng, J.; Ottova, A. L.; Ti Tien, H. "Fabrication and photoelectric properties of self-assembled bilayer lipid membranes on conducting glass." J. Photochem. Photobiol., B 2000, 59, 87-91
Ghoshroy, K. B.; Zhu, W.; Sampson, N. S. "Investigation of membrane disruption in the reaction catalyzed by cholesterol oxidase." Biochemistry 1997, 36, 6133-6140
124
Greenplate, J. T.; Duck, N. B.; Pershing, J. C.; Purcell, J. P. "Cholesterol oxidase: An oostatic and larvicidal agent active against the cotton boll weevil, Anthonomus grandis." Entomol Exp Appl 1995, 74, 253-8
Groves, J. T.; Ulman, N.; Boxer, S. G. "Micropatterning fluid lipid bilayers on solid supports." Science (Washington, D. C.) 1997, 275, 651-653
Guidelli, R.; Aloisi, G.; Becucci, L.; Dolfi, A.; Moncelli, M. R.; Tadini Buoninsegni, F. "Bioelectrochemistry at metal|water interfaces." J. Electroanal. Chem. 2001, 504, 1-28
Gyoervary, E.; Wetzer, B.; Sleytr, U. B.; Sinner, A.; Offenhaeusser, A.; Knoll, W. "Lateral Diffusion of Lipids in Silane-, Dextran-, and S-Layer-Supported Mono- and Bilayers." Langmuir 1999, 15, 1337-1347
Haberland, M. E.; Reynolds, J. A. "Self-association of cholesterol in aqueous solution." Proc Natl Acad Sci U S A 1973, 70, 2313-16
Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part 1. An adsorption-controlled mechanism." Electrochim. Acta 1997, 43, 579-588
Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part II: effect of potential." Electrochim. Acta 1998, 43, 2015-2024
Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part IV: phosphate buffer dependence." Electrochim. Acta 1999, 44, 4573-4582
Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part III: effect of temperature." Electrochim. Acta 1999, 44, 2455-2462
Hall, S. B.; Khudaish, E. A.; Hart, A. L. "Electrochemical oxidation of hydrogen peroxide at platinum electrodes. Part V: inhibition by chloride." Electrochim. Acta 2000, 45, 3573-3579
Haynes, M. P.; Phillips, M. C.; Rothblat, G. H. "Efflux of Cholesterol from Different Cellular Pools." Biochemistry 2000, 39, 4508-4517
Hedges, D. H. P.; Richardson, D. J.; Russell, D. A. "Electrochemical control of protein monolayers at indium tin oxide surfaces for the reagentless optical biosensing of nitric oxide." Langmuir 2004, 20, 1901-1908
125 Hillebrandt, H.; Wiegand, G.; Tanaka, M.; Sackmann, E. "High Electric Resistance Polymer/Lipid Composite Films on Indium-Tin-Oxide Electrodes." Langmuir 1999, 15, 8451-8459
Hin, B. F. Y. Y.; Lowe, C. R. "Amperometric response of polypyrrole entrapped bienzyme films." Sens. Actuators, B 1992, B7, 339-42
Huang, L. "Incorporation of acylated antibody into planar lipid multilayers: characterization and cell binding." Biochemistry 1985, 24, 29-34
Janshoff, A.; Steinem, C. "Scanning force microscopy of artificial membranes." ChemBioChem 2001, 2, 798-808
Kahya, N.; Scherfeld, D.; Bacia, K.; Schwille, P. "Lipid domain formation and dynamics in giant unilamellar vesicles explored by fluorescence correlation spectroscopy." J Struct Biol 2004, 147, 77-89
Kajiya, Y.; Tsuda, R.; Yoneyama, H. "Conferment of cholesterol sensitivity on polypyrrole films by immobilization of cholesterol oxidase and ferrocenecarboxylate ions." J. Electroanal. Chem. 1991, 301, 155-64
Kanzaki, Y.; Hayashi, M.; Minami, C.; Inoue, Y.; Kogure, M.; Watanabe, Y.; Tanaka, T. "Intercalation and bilayer formation of phospholipids in layered synthetic mica. 2. Solvent effect of the intercalation reaction of natural and reduced-type phosphatidylcholines." Langmuir 1997, 13, 3674-3680
Karlsson, M.; Nolkrantz, K.; Davidson, M. J.; Stroemberg, A.; Ryttsen, F.; Kerman, B.; Orwar, O. "Electroinjection of Colloid Particles and Biopolymers into Single Unilamellar Liposomes and Cells for Bioanalytical Applications." Anal Chem 2000, 72, 5857-5862
Karube, I.; Hara, K.; Matsuoka, H.; Suzuki, S. "Amperometric determination of total cholesterol in serum with use of immobilized cholesterol esterase and cholesterol oxidase." Anal Chim Acta 1982, 139, 127-32
Kondo, T.; Takechi, M.; Sato, Y.; Uosaki, K. "Adsorption behavior of functionalized ferrocenylalkane thiols and disulfide onto Au and ITO and electrochemical properties of modified electrodes: effects of acyl and alkyl groups attached to the ferrocene ring." J. Electroanal. Chem. 1995, 381, 203-9
Kumar, A.; Malhotra, R.; Malhotra, B. D.; Grover, S. K. "Co-immobilization of cholesterol oxidase and horseradish peroxidase in a sol-gel film." Anal Chim Acta 2000, 414, 43-50 Lang, H.; Duschl, C.; Vogel, H. "A new class of thiolipids for the attachment of lipid bilayers on gold surfaces." Langmuir 1994, 10, 197-210
126 Lange, Y. "Tracking cell cholesterol with cholesterol oxidase." J Lipid Res 1992, 33, 315-21
Lee, S. Y.; Rhee, H. I.; Tae, W. C.; Shin, J. C.; Park, B. K. "Purification and characterization of cholesterol oxidase from Pseudomonas sp. and taxonomic study of the strain." Appl Microbiol Biotechnol 1989, 31, 542-6
Lewis, R. N. A. H.; Sykes, B. D.; McElhaney, R. N. "Thermotropic phase behavior of model membranes composed of phosphatidylcholines containing cis- monounsaturated acyl chain homologs of oleic acid: differential scanning calorimetric and phosphorus-31 NMR spectroscopic studies." Biochemistry 1988, 27, 880-7
Li, J.; Vrielink, A.; Brick, P.; Blow, D. M. "Crystal structure of cholesterol oxidase complexed with a steroid substrate: Implications for flavin adenine dinucleotide dependent alcohol oxidases." Biochemistry 1993, 32, 11507-15
Liu, K.; Maddaford, T. G.; Ramjiawan, B.; Kutryk, M. J. B.; Pierce, G. N. "Effects of cholesterol oxidase on cultured vascular smooth muscle cells." Mol Cell Biochem 1991, 108, 39-48
Liu, J.; Conboy, J. C. "Direct Measurement of the Transbilayer Movement of Phospholipids by Sum-Frequency Vibrational Spectroscopy." J Am Chem Soc 2004, 126, 8376-8377
Luria, A.; Vegelyte-Avery, V.; Stith, B.; Tsvetkova, N. M.; Wolkers, W. F.; Crowe, J. H.; Tablin, F.; Nuccitelli, R. "Detergent-Free Domain Isolated from Xenopus Egg Plasma Membrane with Properties Similar to Those of Detergent-Resistant Membranes." Biochemistry 2002, 41, 13189-13197
MacLachlan, J.; Wotherspoon, A. T. L.; Ansell, R. O.; Brooks, C. J. W. "Cholesterol oxidase: sources, physical properties and analytical applications." J Steroid Biochem Mol Biol 2000, 72, 169-195
Masoom, M.; Townshend, A. "Determination of cholesterol by flow injection analysis with immobilized cholesterol oxidase." Anal Chim Acta 1985, 174, 293-7
Matsumoto, N.; Chen, X.; Wilson, G. S. "Fundamental Studies of Glucose Oxidase Deposition on a Pt Electrode." Anal Chem 2002, 74, 362-367
Menger, F. M.; Angelova, M. I. "Giant Vesicles: Imitating the Cytological Processes of Cell Membranes." Acc. Chem. Res. 1998, 31, 789-797
Meulemans, A.; Poulain, B.; Baux, G.; Tauc, L.; Henzel, D. "Micro carbon electrode for intracellular voltammetry." Anal Chem 1986, 58, 2088-91
Meyerhoff, J. B.; Ewing, M. A.; Ewing, A. G. "Ultrasmall enzyme electrodes with
127 response time less than 100 milliseconds." Electroanalysis 1999, 11, 308-312
Motonaka, J.; Faulkner, L. R. "Determination of cholesterol and cholesterol ester with novel enzyme microsensors." Anal Chem 1993, 65, 3258-61
Nakaminami, T.; Ito, S.-i.; Kuwabata, S.; Yoneyama, H. "Amperometric Determination of Total Cholesterol at Gold Electrodes Covalently Modified with Cholesterol Oxidase and Cholesterol Esterase with Use of Thionin as an Electron Mediator." Anal Chem 1999, 71, 1068-1076
Nakanishi, M. "Rapid reconstitution of a transmembrane protein into supported planar lipid membranes." FEBS Lett 1984, 176, 385-8
Naumann, R.; Jonczyk, A.; Hampel, C.; Ringsdorf, H.; Knoll, W.; Bunjes, N.; Graeber, P. "Coupling of proton translocation through ATPase incorporated into supported lipid bilayers to an electrochemical process." Bioelectrochemistry and Bioenergetics 1997, 42, 241-247
Niemz, A.; Imbriglio, J.; Rotello, V. M. "Model Systems for Flavoenzyme Activity: One- and Two-Electron Reduction of Flavins in Aprotic Hydrophobic Environments." J Am Chem Soc 1997, 119, 887-892
Ohvo, H.; Slotte, J. P. "Cyclodextrin-Mediated Removal of Sterols from Monolayers: Effects of Sterol Structure and Phospholipids on Desorption Rate." Biochemistry 1996, 35, 8018-8024
Ottova, A. L.; Ti Tien, H. "Self-assembled bilayer lipid membranes: from mimicking biomembranes to practical applications." Bioelectrochemistry and Bioenergetics 1997, 42, 141-152
Parikh, A. N.; Beers, J. D.; Shreve, A. P.; Swanson, B. I. "Infrared Spectroscopic Characterization of Lipid-Alkylsiloxane Hybrid Bilayer Membranes at Oxide Substrates." Langmuir 1999, 15, 5369-5381
Phillips, M. C.; Johnson, W. J.; Rothblat, G. H. "Mechanisms and consequences of cellular cholesterol exchange and transfer." Biochim Biophys Acta 1987, 906, 223-76
Pierrat, O.; Lechat, N.; Bourdillon, C.; Laval, J.-M. "Electrochemical and surface plasmon resonance characterization of the step-by-step self-assembly of a biomimetic structure onto an electrode surface." Langmuir 1997, 13, 4112-4118
Plant, A. L. "Self-assembled phospholipid/alkanethiol biomimetic bilayers on gold." Langmuir 1993, 9, 2764-7
Plant, A. L. "Supported Hybrid Bilayer Membranes as Rugged Cell Membrane Mimics." Langmuir 1999, 15, 5128-5135
128
Pollegioni, L.; Gadda, G.; Ambrosius, D.; Ghisla, S.; Pilone, M. S. "Cholesterol oxidase from Streptomyces hygroscopicus and Brevibacterium sterolicum: effect of surfactants and organic solvents on activity." Biotechnol Appl Biochem 1999, 30, 27- 33
Pollegioni, L.; Wels, G.; Pilone, M. S.; Ghisla, S. "Kinetic mechanisms of cholesterol oxidase from Streptomyces hygroscopicus and Brevibacterium sterolicum." Eur J Biochem 1999, 264, 140-151
Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. "Spontaneously organized molecular assemblies. 4. Structural characterization of n-alkyl thiol monolayers on gold by optical ellipsometry, infrared spectroscopy, and electrochemistry." J Am Chem Soc 1987, 109, 3559-68
Radhakrishnan, A.; McConnell, H. M. "Chemical Activity of Cholesterol in Membranes." Biochemistry 2000, 39, 8119-8124
Ram, M. K.; Bertoncello, P.; Ding, H.; Paddeu, S.; Nicolini, C. "Cholesterol biosensors prepared by layer-by-layer technique." Biosens Bioelectron 2001, 16, 849- 856
Rhoten, M. C.; Burgess, J. D.; Hawkridge, F. M. "Temperature and pH effects on cytochrome c oxidase immobilized in an electrode-supported lipid bilayer membrane." Electrochim. Acta 2000, 45, 2855-2860
Rhoten, M. C.; Burgess, J. D.; Hawkridge, F. M. "The reaction of cytochrome c from different species with cytochrome c oxidase immobilized in an electrode supported lipid bilayer membrane." J. Electroanal. Chem. 2002, 534, 143-150
Ropers, M.-H.; Bilewicz, R.; Stebe, M.-J.; Hamidi, A.; Miclo, A.; Rogalska, E. "Fluorinated and hydrogenated cubic phases as matrices for immobilisation of cholesterol oxidase on electrodes." Phys. Chem. Chem. Phys. 2001, 3, 240-245
Ross, E. E.; Bondurant, B.; Spratt, T.; Conboy, J. C.; O'Brien, D. F.; Saavedra, S. S. "Formation of Self-Assembled, Air-Stable Lipid Bilayer Membranes on Solid Supports." Langmuir 2001, 17, 2305-2307
Rothblat, G. H.; De la Llera-Moya, M.; Atger, V.; Kellner-Weibel, G.; Williams, D. L.; Phillips, M. C. "Cell cholesterol efflux: integration of old and new observations provides new insights." J Lipid Res 1999, 40, 781-796 Sackmann, E. "Supported membranes: scientific and practical applications." Science (Washington, D. C.) 1996, 271, 43-8
Salamon, Z.; Tollin, G. "Reduction of cytochrome c at a lipid bilayer modified electrode." Bioelectrochemistry and Bioenergetics 1991, 25, 447-54
129
Salamon, Z.; Gleason, F. K.; Tollin, G. "Direct electrochemistry of thioredoxins and glutathione at a lipid bilayer-modified electrode." Arch Biochem Biophys 1992, 299, 193-8
Salamon, Z.; Tollin, G. "Direct electrochemistry of spinach plastocyanin at a lipid bilayer-modified electrode: cyclic voltammetry as a probe of membrane-protein interactions." Arch Biochem Biophys 1992, 294, 382-7
Salamon, Z.; Hazzard, J. T.; Tollin, G. "Direct measurement of cyclic current-voltage responses of integral membrane proteins at a self-assembled lipid-bilayer-modified electrode: cytochrome f and cytochrome c oxidase." Proc Natl Acad Sci U S A 1993, 90, 6420-3
Salamon, Z.; Macleod, H. A.; Tollin, G. "Surface plasmon resonance spectroscopy as a tool for investigating the biochemical and biophysical properties of membrane protein systems. II: Applications to biological systems." Biochim Biophys Acta 1997, 1331, 131-152
Sampson, N. S.; Kass, I. J.; Ghoshroy, K. B. "Assessment of the Role of an W Loop of Cholesterol Oxidase: A Truncated Loop Mutant Has Altered Substrate Specificity." Biochemistry 1998, 37, 5770-5778
Sampson, N. S.; Vrielink, A. "Cholesterol Oxidases: A Study of Nature's Approach to Protein Design." Acc. Chem. Res. 2003, 36, 713-722
Santiago, J.; Guzman, G. R.; Rojas, L. V.; Marti, R.; Asmar-Rovira, G. A.; Santana, L. F.; McNamee, M.; Lasalde-Dominicci, J. A. "Probing the effects of membrane cholesterol in the Torpedo californica acetylcholine receptor and the novel lipid- exposed mutation aC418W in Xenopus oocytes." J Biol Chem 2001, 276, 46523- 46532
Santiago, J.; Guzman, G. R.; Rojas, L. V.; Marti, R.; Asmar-Rovira, G. A.; Santana, L. F.; McNamee, M.; Lasalde-Dominicci, J. A. "Probing the effects of membrane cholesterol in the Torpedo californica acetylcholine receptor and the novel lipid- exposed mutation alpha C418W in Xenopus oocytes." J Biol Chem 2001, 276, 46523- 32
Satoh, I.; Karube, I.; Suzuki, S. "Enzyme electrode for free cholesterol." Biotechnol Bioeng 1977, 19, 1095-9
Scheller, F. W.; Wollenberger, U.; Lei, C.; Jin, W.; Ge, B.; Lehmann, C.; Lisdat, F.; Fridman, V. "Bioelectrocatalysis by redox enzymes at modified electrodes." Reviews in Molecular Biotechnology 2002, 82, 411-424
Schoenherr, H.; Johnson, J. M.; Lenz, P.; Frank, C. W.; Boxer, S. G. "Vesicle
130 Adsorption and Lipid Bilayer Formation on Glass Studied by Atomic Force Microscopy." Langmuir 2004, 20, 11600-11606
Shao, Y.; Mirkin, M. V. "Probing Ion Transfer at the Liquid/Liquid Interface by Scanning Electrochemical Microscopy (SECM)." J. Phys. Chem. B 1998, 102, 9915- 9921
Simons, K.; Ikonen, E. "Functional rafts in cell membranes." Nature (Lond) 1997, 387, 569-572
Simons, K.; Ikonen, E. "Review: Cell biology: How cells handle cholesterol." Science (Washington, D. C.) 2000, 290, 1721-1726
Sinner, E.-K.; Knoll, W. "Functional tethered membranes." Curr Opin Chem Biol 2001, 5, 705-711
Steinem, C.; Janshoff, A.; Ulrich, W.-P.; Sieber, M.; Galla, H.-J. "Impedance analysis of supported lipid bilayer membranes: a scrutiny of different preparation techniques." Biochim Biophys Acta 1996, 1279, 169-80
Stroeve, P.; Miller, I. "Lateral diffusion of cholesterol in monolayers." Biochim Biophys Acta 1975, 401, 157-67
Tabas, I. "Cholesterol in health and disease." J Clin Investig 2002, 110, 583-590
Takano, H.; Kenseth, J. R.; Wong, S.-S.; O'Brien, J. C.; Porter, M. D. "Chemical and biochemical analysis using scanning force microscopy." Chemical Reviews (Washington, D. C.) 1999, 99, 2845-2890
Teng, J. I.; Smith, L. L. "Sterol peroxidation by Pseudomonas fluorescens cholesterol oxidase." Steroids 1996, 61, 627-33
Tiede, D. M. "Incorporation of membrane proteins into interfacial films: model membranes for electrical and structural characterization." Biochim Biophys Acta 1985, 811, 357-79
Tien, H. T.; Salamon, Z. "Formation of self-assembled lipid bilayers on solid substrates." Bioelectrochemistry and Bioenergetics 1989, 22, 211-18
Tien, H. T.; Ottova, A. L. "Supported planar lipid bilayers (s-BLMs) as electrochemical biosensors." Electrochim. Acta 1998, 43, 3587-3610 Vengatajalabathy Gobi, K.; Mizutani, F. "Layer-by-layer construction of an active multilayer enzyme electrode applicable for direct amperometric determination of cholesterol." Sens. Actuators, B 2001, B80, 272-277
Vidal, J.-C.; Garcia-Ruiz, E.; Castillo, J.-R. "Design of a multilayer cholesterol
131 amperometric biosensor for preparation and use in flow systems." Electroanalysis 2001, 13, 229-235
Vrielink, A.; Lloyd, L. F.; Blow, D. M. "Crystal structure of cholesterol oxidase from Brevibacterium sterolicum refined at 1.8 .ANG. resolution." J Mol Biol 1991, 219, 533-54
Wagner, M. L.; Tamm, L. K. "Tethered polymer-supported planar lipid bilayers for reconstitution of integral membrane proteins: silane-polyethyleneglycol-lipid as a cushion and covalent linker." Biophys J 2000, 79, 1400-1414
Wang, H.; Mu, S. "Bioelectrochemical characteristics of cholesterol oxidase immobilized in a polyaniline film." Sens. Actuators, B 1999, B56, 22-30
Wiegand, G.; Arribas-Layton, N.; Hillebrandt, H.; Sackmann, E.; Wagner, P. "Electrical properties of supported lipid bilayer membranes." J. Phys. Chem. B 2002, 106, 4245-4254
Wollenberger, U.; Kuehn, M.; Scheller, F.; Deppmeyer, V.; Jaenchen, M. "Amperometric enzyme sequence electrodes for cholesterol." Bioelectrochemistry and Bioenergetics 1983, 11, 307-17
Wu, Z.; Wang, B.; Cheng, Z.; Yang, X.; Dong, S.; Wang, E. "A facile approach to immobilize protein for biosensor: self-assembled supported bilayer lipid membranes on glassy carbon electrode." Biosens Bioelectron 2001, 16, 47-52
Yan, C.; Zharnikov, M.; Goelzhaeuser, A.; Grunze, M. "Preparation and Characterization of Self-Assembled Monolayers on Indium Tin Oxide." Langmuir 2000, 16, 6208-6215
Yin, Y.; Sampson, N. S.; Vrielink, A.; Lario, P. I. "The presence of a hydrogen bond between asparagine 485 and the p system of FAD modulates the redox potential in the reaction catalyzed by cholesterol oxidase." Biochemistry 2001, 40, 13779-13787
Yue, Q. K.; Kass, I. J.; Sampson, N. S.; Vrielink, A. "Crystal structure determination of cholesterol oxidase from Streptomyces and structural characterization of key active site mutants." Biochemistry 1999, 38, 4277-4286
Zhang, Z.; Nassar, A.-E. F.; Lu, Z.; Schenkman, J. B.; Rusling, J. F. "Direct electron injection from electrodes to cytochrome P450cam in biomembrane-like films." J. Chem. Soc., Faraday Trans. 1997, 93, 1769-1774
132