“(Eda
.5.“ t
«fl-
5.,
. y!

 

. ..
2.} Y
x3 hi: 51v

.1. _§\4 \a.‘
. i..."fil.Mm1|1.. t
sin... .2. u

{3.13; 4
41:31 .I: 9h“. .1123314‘3
fig. H
fit ‘A:
g‘nflmfizmu
v .. .
«J? 91‘ .V
. l
‘ c Dun...
V i:
4%. 3nd ,
. :h.
{l- . .

3..
at
a

 

 

 

 

31...... a ..
5.3:. alas,
..§x!i£.x.

In)?! a. ‘3.
. I.:.!.Nc(!i¢l
a fill!

.
G s! :

..........u....... {aflitghuvt

‘slx.:d.1l§3_ai§1£

.i‘l'é

flung

I. 1!

$52.9; .91).)".
‘49.. .61.;

 

 

I .. ,
i3... .3, a). 3.
a .11.. I 1.11. o. A. (it . 7. 33...!

 

 

LIBRARY
Michigan State
L Ui’liVei‘Siiy I

 

This is to certify that the
dissertation entitled

INVESTIGATING THE MECHANISM OF HYPOXlA-INDUCED
RELEASE OF ATP FROM ERYTHROCYTES AND ITS ROLE
IN NITRIC OXIDE PRODUCTION

presented by

ANDREA FARIS

has been accepted towards fulfillment
of the requirements for the

PhD degree in Chemistry

 

 

-m\\
,,..... TY; " II '

if” - 5;“ ----,f ‘ _

( Wssor’s Signature

‘-."" -. . .—v--."f

q {30 / CC}

Date

 

MSU is an Affirmative Action/Equal Opportunity Employer

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:/Proleoc&Pres/CIRCIDatoDue.indd

INVESTIGATING THE MECHANISM OF HYPOXIA-INDUCED RELEASE
OF ATP FROM ERYTHROCYTES AND ITS ROLE IN NITRIC OXIDE
PRODUCTION
By

Andrea Nicole Faris

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Chemistry

2009

ABSTRACT
INVESTIGATING THE MECHANISM OF HYPOXIA-INDUCED RELEASE OF
ATP FROM ERYTHROCYTES AND ITS ROLE IN NITRIC OXIDE
PRODUCTION
By

Andrea F aris

Since the discovery of nitric oxide (NO) as the endothelium derived relaxing
factor (EDRF), its role in the circulation has been extensively studied; most
importantly, NO can elicit relaxation in smooth muscle cells underlying the
endothelial cells in blood vessels such as arteries and arterioles. One stimulant for NO
production is red blood cell (RBC)-derived adenosine triphosphate (ATP). Through a
series of mechanisms, ATP stimulates NO production in endothelial cells, where it
diffuses to smooth muscle cells resulting in dilation. There have been many reported
stimuli for the release of ATP from RBCS, such as, deformation, c-peptide and
hypoxia. In hypoxic conditions, RBCS release an increased level of ATP; however,
the cause of hypoxia-induced ATP release is still unknown.

Hemoglobin changes conformations from the R (relaxed) state to the T (tense)
state when exposed to a hypoxic environment. Moreover, hemoglobin is attached to
the RBC membrane through a band 3 protein. Through the differing conformations
hemoglobin undergoes, it is hypothesized a deformation is induced on the RBC
membrane, therefore, resulting in ATP release. A relationship between deformation
and hypoxia was studied using a flow-through setup. The data from these results
suggested a non-additive relationship. This led to the idea that hypoxia and

deformation-induced ATP release may be following the same signaling pathway.

A widely accepted signaling pathway for deformation-induced ATP release has
been previously reported. It starts with the activation of a G-protein and ends with the
ATP traversing the membrane through a CFTR-mediated mechanism. Through the
inhibition of each protein in this pathway and subsequent exposure to hypoxia, a
decrease in ATP release was measured. Furthermore, upon stiffening of the RBCS
with a known cell stiffener, ATP release decreased in a dose dependent manner.
These data suggest, along with those previously mentioned, hypoxia-induced ATP
release is a form of deformation.

Two theories for hypoxic vasodilation are the conversion of nitrite to NO
through deoxyhemoglobin or the transnitrosylation of hemoglobin for delivery to
respiring tissue. While these theories are extensive in their research, they lack many
details. For example, the NO released from RBCS must traverse the blood stream,
endothelium and sub-endothelium to elicit relaxation in the smooth muscle cells. This
mechanism is physiologically unlikely. From studies performed, RBCS donated NO
when placed in a hypoxic buffer; however, the NO donated is not NOS-mediated and
not ATP-dependant. Through the use of a microfluidic device and fluorescence
microscopy, measurable NO was detected from endothelial cells, and the
concentration of measurable NO was dependant on RBC-derived ATP.

Through the use of flow-through analysis techniques a non-additive
relationship was discovered for hypoxia and deformation. Furthermore, through the
inhibition of the deformation-induced signaling pathway and exposure to hypoxia, a
pathway for hypoxia-induced ATP release was elucidated. Finally, with the use of

microfluidic devices, a mechanism for hypoxic vasodilation was demonstrated.

Copyright by
ANDREA N. FARIS
2009

~ I dedicate this thesis to my dad and mom,
Nabeel and Ilene. . .my two truest blessings in life. . .I
love you more than you will ever know.

ACKNOWLEDGEMENTS

First and foremost, I would like to thank my advisor, Dr. Dana Spence. At a
time when I did not have a group, you took me into your lab. From day one you
pushed me to do my best and without your encouragement I could not have made it.
In all the years I have been in grad school and all the years to come in my career, I am
positive I will never come across a better advisor. Thank you for everything. I would
also like to thank my committee members for their time, suggestions and questions.

Throughout these last four years in grad school I have had a shining light with
me the whole time, Jen. All the laughs we have shared together are priceless to me
and I wouldn’t trade them for the world. You have always been there for me helping
me out and never wavering in your selflessness. You have been my angel, and I love
you very much and will miss you©. In thanking Jen, I would also like to thank all the
Spence group members especially Chia and Nicole. All of our “profound”
conversations have gotten me through some rough days and I was always happy to
wake up in the morning and come to lab to see you. Also, I would like to thank Ajith
for all of our talks in the scope room. . .we learned a lot about each other and had some
good discussions that will always stay with me. Wathsala, Suzanne, Kari, Adam, Paul
and Steve. . .thank you for your support, you will be missed.

Most of the encouragement throughout these years has come outside of the lab,
from my family. I dedicated this thesis to my dad and mom because they are the sole
reason I am what I am today. To my brothers, Omar and Tony. . .thank you for

supporting me my whole life and always watching out for me. In my life I have never

vi

experienced a time when I doubted whether you would be there for me, and for that I
thank you. And to the two best sister-in-laws a girl could ask for...Julie and May.
Nobody could fill the shoes of a Paris better than you two; I love you and am so
honored you are part of our family. Finally, the last addition to the Paris family, baby
Faris. . .we cannot wait to see your beautiful face! Finally, my best friend Seema Patel
has given me much of the outside support I have needed to get through these years.
All the visits and phone calls have been invaluable to me and have helped more than
you will ever know. Thank you for always being there.

Finally, I would like to thank my other family, the Matti family: Umo Zuhair,
Auntie Rawaa, Reem, Martin and Renee, Raid, Renee, Martina and Melanie. I am
truly blessed to be part of such an amazing family. Your love and support for each
other never seizes to amaze me, thank you for sharing that love and support with me.
More Specifically, I would like to thank the women from the Matti family. . .six of the
most beautiful women I know...you all have been an inspiration to me from the
moment we met and continue to do so. I love you very very much.

Last, but definitely not least, I want to thank my fiance Raid Matti. Words
cannot describe how much you mean to me and how much I truly love you.
Throughout grad school you have never missed wishing me luck on exams, teaching,
or presentations. You have been there for me through everything, never missing a
beat, and I am so grateful for your selflessness. You are nothing short of amazing to
me Raid. I don’t think I could ever express to you how much I love you or how
honored I am to be with you, but I plan on spending the rest of my life trying, I love

you.

vii

TABLE OF CONTENTS

LIST OF FIGURES ............................................................................ x
CHAPTER 1
1.1 INTRODUCTION ..................................................................... l
1.2 Nitric Oxide as a Major Vasodilator in the Circulatory System .................. 6
1.3 Proposed Mechanisms of Vasodilation in Response to Hypoxia ................. 8
1.4 ATP as a Stimulant for the Vasodilator NO ....................................... 10
1.5 RBC-Derived ATP Release .......................................................... 11
1.6 Hypoxic Vasodilation ................................................................. 17
1.7 The Effects of Hypoxia on the Vasculature ........................................ 26
1.8 Thesis Objective ........................................................................ 33
List of References ........................................................................... 35
CHAPTER 2
2.1 MEASURING TWO STIMULI OF ATP RELEASE
SIMULTANEOUSLY ............................................................... 44
2.2 EXPERIMENTAL ................................................................... 48

2.2.1 Collection of RBCS
2.2.2 Preparation of Reagants
2.3 METHODS ............................................................................ 51
2.3.1 Measurement of ATP Release
2.3.2 Flow-Based Determination of Hypoxic RBCS
2.3.3 Non-Flow Measurement of Hypoxic RBCS
2.3.4 Measurement of Oxygen Saturation

2.4 RESULTS ....................................................................................................... 54
2.5 DISCUSSION ......................................................................... 61
List of References ........................................................................... 65
CHAPTER 3
3.1 FINDING THE LINK BETWEEN HYPOXIA AND
DEFORMATION .................................................................... 67
3.2 EXPERIMENTAL .................................................................. 70

3.2.1 Collection of RBCS
3.2.2 Preparation of Reagants
3.3 METHODS ............................................................................ 74
3.3.l Measurement of ATP release
3.3.2 Non-Flow Measurement of Hypoxic RBCS
3.3.3 Absorption Measurements

3.4 RESULTS .............................................................................. 76
3.5 DISCUSSION ......................................................................... 82
List of References ........................................................................... 86

viii

CHAPTER 4
4.1 CRITICISM OF CURRENT MECHANISMS OF HYPOXIC
VASODILATION ......................................................................... 89
4.2 EXPERIMENTAL ................................................................... 93
4.2.1 Collection of RBCS
4.2.2 Preparation of Reagants
4.3 METHODS ............................................................................ 95
4.3.1 Fluorometric Determination of NO
4.3.2 Microfluidic Device Preparation and Measurement of NO

4.4 RESULTS ............................................................................ 100
4.5 DISCUSSION ........................................................................ 107
List of References ......................................................................... 110
CHAPTER 5
5.1 CONCLUSIONS .................................................................... 113
5.1.1 A Non-Additive Relationship Between Hypoxia and Deformation-
Induced ATP Release ............................................................ 109
5.1.2 Defining the Signaling Pathway for Hypoxia-Induced ATP
Release .............................................................................. 110
5.1.3 Hypoxic Vasodilation ...................................................... 110
5.2 FUTURE WORK ................................................................... 118
List of References ......................................................................... 126

ix

LIST OF FIGURES

Figure 1. Schematic of a human blood vessel. The adventitia is comprised of mostly
connective tissue. The external elastic lamina is comprised of elastic connective
tissue. The medial layer is comprised of smooth muscle cells, the site of NO-induced
vasodilation. The internal elastic lamina is comprised of elastic connective tissue.
The intima is comprised of endothelial cells, the site of NO production. RBCS flowing
through the vessel come in direct contact with this layer .................................... 2

Figure 2. Structure of hemoglobin and molecular drawing of heme structure.142 This
schematic displays the placement of the four heme groups on hemoglobin. Each chain
(alpha and beta) have a heme moiety. To the right, the molecular structure of heme is
shown. This includes the Fe atom and the porphyrin ring surrounding it. 02 binds to
iron atom and takes on a “bent” conformation with one oxygen atom protruding
out......... ........................................................................................................................ 5

Figure 3. The five electron oxidation of a guanidino nitrogen of L-arginine. Through
the conversion of L-arginine to L-citrulline, NO is produced per 2 mol of 02 and 1.5
mole of NADPH consumed. This reaction takes place in the endothelium upon
activation of eNOS.135 ........................................................................... 12

Figure 4. Diagram of the fate of ATP released from an RBC. RBC-derived ATP
release stimulates NO production in the endothelium lining vessel walls. ATP is
released from RBCS due to a number of different stimuli such as deformation and
hypoxia. Once released, ATP activates a purinergic receptor on the endothelial cell,
activating eNOS through calcium flux into the cell. eNOS catalyzes the production of
the vasodilator NO which then diffuses to the smooth muscle cells causing them to
relax, and allowing dilation of the vessel to occur .......................................... 13

Figure 5. Mechanism for ATP release proposed by Bergfeld and Forrester. ATP
leaves the RBC through the band 45 protein. The efflux of the ATP causes an
imbalance in the electrical potential of the cell. ATP has a charge of negative 4, and
upon leaving the cell, takes a Mg2+ atom with it. This leaves a negative 2 charge on
the RBC that needs to be compensated for to keep electrical balance. Band 3 imports
a chloride ion and a bicarbonate ion to keep electro-homeostasis ........................ 18

Figure 6. Schematic of hypoxia-induced ATP release. Once ATP is released from an
RBC is activates the purinergic receptor, sz, on endothelial cells. This, in turn,
activates eNOS which produces the potent vasodilator NO. NO diffuses to the smooth
muscle cells where it elicits vasodilation ..................................................... 20

Figure 7. Mechanism of hypoxic vasodilation that proposes at 50% oxygen saturation
hemoglobin acts as a nitrite reductase reducing bioavailable nitrite to NO. This NO
then diffuses through the endothelium, and the sub-endothelium to the smooth muscle
cells where it can bind to soluble guanylate cyclase and elicit vasodilation ............ 22

Figure 8. Mechanism of hypoxic vasodilation that proposes hemoglobin uptakes NO
in the lungs and conserves it on the B93 cysteine residue. While traversing the
circulation, the nitrosylated hemoglobin release NO when it enters an area of hypoxia.
This NO then diffuses through the endothelium, and the sub-endothelium to the
smooth muscle cells where it can bind to soluble guanylate cyclase and elicit
vasodilation ....................................................................................... 24

Figure 9. Cascade of events stemming from endothelial activation, the activation of
the endothelium in response to hypoxia. Upon endothelial activation, membrane
fluctuations occur, (i.e. the junction between endothelial cells open wider), CAMP and
cGMP, components for the signaling pathway for the production of NO, levels are
decreased and vasodilator and vasoconstrictor levels are increased ...................... 30

Figure 10. The two-step bioluminescence reaction between luciferin/luciferase and
ATP ................................................................................................ 50

Figure 11. Schematic of deoxygenation technique. The top right schematic in the
non-flow setup utilizing a PMT as the transducer. In this construct, only ATP release
due to hypoxia is measured. There is no deformation of the RBCS. The bottom setup
is the flow-through system using microbore tubing to induce deformation-derived
ATP release. In this setup, both hypoxia and deformation-induced ATP release is

measured simultaneously. RBCS were deoxygenated by gas purging with N2 for 4
min ................................................................................................ 53

Figure 12. Time study of percent oxygen saturation upon deoxygenation with N2 gas
at 30 kPa for 4 min with measurements taken every 15 S. A Clark-type oxygen sensor

was placed into the RBC sample during gas displacement. A thick layer of Parafilm

was placed on top of the tube holding the sensor in place along with inhibiting any
interference from the atmosphere. Results displayed are averages from the RBCS of
n=3 rabbits. All error bars are : SEM ........................................................ 56

Figure 13. ATP release from RBCS over time in a non-flow system. The initial ATP
release is from RBCS in the absence of hypoxia. Upon induction of hypoxia, the
RBC-derived ATP is determined at different time intervals up to and including 4 min.
A fresh vial of RBCS was prepared for each measurement. All error bars are : SEM.
Students t-test was performed on data to ensure the means are statistically different
(p<0.05, n=3) ..................................................................................... 57

Figure 14. A. Chemiluminescent signals from different RBC samples. The bottom
trace represents a 7% hematocrit RBC sample pumped through microbore tubing
having an inside diameter of 50 um. The middle trace represents a 7% hematocrit
RBC sample incubated for 30 min with 100 uM glibenclamide (a CFTR inhibitor)

then deoxygenated for 4 min with 30 kPa of N2 gas. This sample was then pumped
through microbore tubing having an inside diameter of 50 pm. The top trace is a 7%

hematocrit RBC sample deoxygenated for 4 min with 30 kPa of N2 gas prior to

xi

pumping through the tubing. B. Summary of the data shown in A. From the left as
shown, the black bar represents ATP release from a 7% RBC sample flowed through
microbore tubing having an inside diameter of 50 pm. The light gray bar represents

ATP release from a 7% RBC sample deoxygenated with N2 gas at 30 kPa for 4 min
prior to introduction to the tubing. The dark gray bar represents a 7% RBC sample

incubated with 100 uM glibenclamide for 30 min then deoxygenated with N2 gas at 30

kPa for 4 min. All error bars are 1 SEM. Students t-test was performed on data to
ensure the means are statistically different (p<0.001, n=3) ................................ 58

Figure 15. The ability of RBCS to release ATP in the presence of two stimuli: hypoxia
and deformation. The black bars represent RBC-derived ATP release from a 5 mL
sample of RBCS with a 7% hematocrit deoxygenated for varying times at 30 kPa. The
sample was then pumped through microbore tubing having an inside diameter of 50
pm. The gray bar represents RBC-derived ATP release of a SmL sample with a 7%
hematocrit deoxygenated for varying times at 30 kPa. The sample was placed into a
cuvette that was placed on top of a PMT and measured in the absence of flow. All
error bars are 1: SEM. Students t-test was performed on data to ensure the means are
statistically different (p<0.005, n=3) .......................................................... 60

Figure 16. The ability of hypoxic RBCS to release ATP after incubation with 40 DM
diamide, a cell stiffener. The black bar represents the ATP release of 5 mL of 3.5%
hypoxic RBCS. The gray bar represents 5 mL of 3.5% hypoxic RBCS incubated for 18
min with 40 DM diamide. Deoxygenation of all RBC samples occurred for 15 s with

N2 gas at 30 kPa. The dosage of diamide was kept constant at 200 uL. All error bars

are i SEM. Students t-test was performed on data to ensure the means are statistically
different (p<0.05, n=3) .......................................................................... 62

Figure 17. Schematic of the signaling pathway for deformation-induced ATP release.

Deformation activates the G-protein coupled receptor which activates CS or G;
through a coupling mechanism. Adenylyl cyclase becomes activated by the G protein
and catalyzes the conversion of ATP to CAMP. cAMP then activates protein kinase A

which phosphorylates CFTR by a double phosphorylation. Through the release of Cl-

ions from CFTR, ATP is released by another transmembrane protein that is
unknown .......................................................................................... 71

Figure 18. This graph represents the ability of hypoxic RBCS to release ATP after
incubation with 40 DM diamide, a cell stiffener. All hypoxic RBCS were

deoxygenated with a hypoxic buffer consisting of 1.5 mL of Oxyrase enzyme system
and 3.5 mL of PSS. The first bar represents ATP release of 3.5% normoxic RBCS.
The second bar represents the ATP release of hypoxic 3.5% RBCS. The third bar
represents hypoxic 3.5% RBCS incubated for 18 min with 40 uM diamide. The fourth
bar represents hypoxic 3.5% RBCS incubated with 80 uM diamide. All error bars are
1; SEM. Students t-test was performed on data to ensure the means are statistically
different (p<0.05, n=3) .......................................................................... 78

xii

Figure 19. (A.) Visible spectra obtained from a 0.07% solution of normoxic RBCS.
(B.) Visible spectra obtained from a 0.07% solution of hypoxic RBCS. RBCS were
made hypoxic by introduction into a hypoxic buffer solution. (C.) Visible spectra
obtained from a 0.07% solution of normoxic RBCS incubated with 100 DL of 2 mM
diamide. (D.) Visible spectra obtained from a 0.07% solution of hypoxic RBCS
incubated with 100 DL of 2 mM diamide. RBCS were made hypoxic by introduction
into a hypoxic buffer solution after incubation with diamide. This data demonstrates
that under hypoxic conditions, hemoglobin changes its conformation to the T state
structure, visible by the difference in absorbance spectrum. Moreover, RBCS

incubated with diamide and made hypoxic still release the 02 to become
deoxygenated; however, because the cell is rendered less deformable, the
conformational change does not induce as much of a deformation; therefore, there is a
decrease in ATP release as seen in Figure 18 ................................................ 79

Figure 20. This figure displays the ATP release of the RBCS under the influence of
different reagants. The first bar is the ATP release of 3.5% RBC under normoxic
conditions. The second bar represents 3.5% RBCS deoxygenated with 1.5 mL
Oxyrase® at 37 °C for 30 min. All hypoxic RBCS were deoxygenated in this manner.
The third bar represents hypoxic RBCS icubated with diamide, a known cell stiffener.
The fourth bar represents hypoxic RBCS incubated with pertussis toxin, a G protein
inhibitor. The fifth bar represents hypoxic RBCS incubated with niflumic acid, a
CFTR inhibitor. The sixth bar represents hypoxic RBCS incubated with H-89, a PKA
inhibitor. The last bar represents hypoxic RBCS incubated with MDL, an adenylyl
cyclase inhibitor. All error bars are : SEM. Students t-test was performed on all data
to ensure the means are stastically different (p<0.05) ...................................... 81

Figure 21. Production of a microfluidic array. First, PDMS is poured over a silicon
wafer (master) with 6 channels raised. The master and PDMS are heated for 20
minutes for the PDMS to cure and polymerize. The PDMS is then removed from the
master. PDMS is then poured over a second master, with no channels raised, and
heated for the same amount of time, for the same reason. This PDMS layer is also
removed from the master and 18 1/8” holes are punched through. A polycarbonate
membrane is then placedover the PDMS layer with the channels and the PDMS layer
with the holes punched through is placed on top. All three layers (PDMS with
channels, polycarbonate membrane and PDMS with wells) are combined and heated
for 30 minutes in order to irreversibly seal the two PDMS layers together. A
photograph of the finished product is displayed after this last step ....................... 98

Figure 22. Schematic of a microfluidic device well with endothelial cells. Two layers
of PDMS were irreversibly sealed with a polycarbonate membrane between. One
layer of PDMS contained channels while the other had 18 wells punched through.
Endothelial cells were cultured into the wells on the polycarbonate membrane. RBC
solutions were flowed through the channels underlying the wells. Hypoxia-induced
ATP release stimulated NO production in the endothelial cells cultured in the wells.

xiii

The endothelial cells were pre-incubated with DAF-FM DA; therefore, any NO
produced would be measured intracellularly with the fluorescence probe ............... 99

Figure 23. A. Standard curve for NO standards made from Spermine NONOate, 10
uM DAF-F M was incubated with the standards for 15 min prior to measurement.

Excitation wavelength: 485 mn. Emission wavelength: 510 nm. B. Fluorescence
spectrumof DAF-FM and NO ................................................................ 101

Figure 24. This figure displays the NO release from 7% RBCS under normoxic and
hypoxic conditions. 10 DM DAF-FM was incubated with the supernatant of the RBC
samples for 15 min prior to measurement. The increase in fluorescence intensity is
indicative of an increase in NO release from RBCS under hypoxic conditions.
Excitation wavelength: 485 nm Emission wavelength: 510 nm. All error bars are 1
SEM. Students t-test was performed on data to ensure the means are statistically
different (p<0.05, n=5) ........................................................................ 102

Figure 25. Measured NO release of 7% RBCS in normoxic and hypoxic conditions.
Upon incubation with diamide, a cell stiffener, the NO release from RBCS was
measured and found significantly similar to hypoxic RBCS alone. It has been
established RBCS possess eNOS. Upon incubation with L-NAME, an eNOS inhibitor,
the NO release did not decrease. The same trend was observed with niflumic acid, a
CFTR inhibitor. These data suggest RBC eNOS does not play a role in the production
of NO from RBCS. 10 DM DAF-FM was incubated with the supernatant of the RBCS
samples for 15 min prior to measurement. Excitation: 485 nm Emission: 510 nm. All
error bars are 1 SEM. Students t-test was performed on data to ensure the means are
statistically different (p<0.05, n=3) .......................................................... 104

Figure 26. Fluorescence intensity of NO produced from bPAECs in response to
hypoxia-induced ATP release. A. represents 3.5% normoxic RBCS. B. represents
hypoxic 3.5% RBCS. C. represents hypoxic 3.5% RBCS incubated with 100 uM
diamide, a cell stiffener. D. represents hypoxic 3.5% RBCS incubated with 200 uM
diamide. E. represents hypoxic 3.5% RBCS incubated with 20 uM niflumic acid, a
CFTR inhibitor. F. represents hypoxic 3.5% RBCS incubated with 40 uM niflumic
acid. These data suggest the dependence of endothelial-derived NO production on
hypoxia-induced ATP release. The NO released from the endothelial cells can then
diffuse to the underlying smooth muscle cells and elicit vasodilation. 100 uM DAF-
FM DA was incubated with the endothelial cells for 1 h. Data represents an n=3. All
error bars are 3: SEM. Students t-test was performed on data to ensure the means are
statistically different... ...... ... .................................................................................... 105

Figure 27. Graphical representation of the data in Figure 6. Endothelium-derived NO
increased after upon stimulation of hypoxia-induced ATP release. The addition of
diamide, a cell stiffener, decreased the measurable NO is a dose dependant manner.
Furthermore, the addition of niflumic acid, a CFTR inhibitor, decreased the
measurable NO, also in a dose dependant manner. All error bars are : SEM.

xiv

Students t-test was performed on data to ensure the means are statistically
different .......................................................................................... 106

Figure 28. This graph displays the ATP release of control and BB/ZDR/Wor rats
(Type 2 diabetic rats) exposed to a hypoxic environment. The ATP release from the
RBCS of the diabetic rats decreased by 60.7%, this demonstrates a lower ability of the
RBCS to elicit vasodilation. Furthermore, this implies the reason for worse
reperfusion injury in patients with diabetes ................................................ 121

Figure 29. Oxygen saturation curve of hypoxic RBCS and hypoxic RBCS
preincubated with 100 uM hydroxyurea. Oxygen was depleted from the RBCS
preincubated with hydroxyurea more rapidly than compared to hypoxic RBCS with no
reagent added. These data suggest oxygen is removed from hemoglobin at a more
rapid pace, indicating the removal of oxygen and the binding of NO to
hemoglobin ...................................................................................... 124

XV

CHAPTER 1
1.1 AN INTRODUCTION TO BLOOD AND BLOOD FLOW

The circulatory system is an organ system that passes nutrients, gases,
hormones and blood cells, to and from the body in order to maintain homeostasis. The
main components of the circulatory system are the heart, blood, and blood vessels. A
blood vessel consists of the adventitia, the media, the elastica interna, and the
endothelium. The last two layers comprise a layer called the intima. Most layers
consist of connective tissue; however, the media layer contains smooth muscle cells.
A schematic of a blood vessel is displayed in Figure 1. The circulation is also
sometimes classified as the pulmonary circulation, where blood is oxygenated, and the
systemic circulation, where oxygenated blood is provided to organs. An average adult
body contains about five liters of blood, which is comprised of plasma, white blood
cells, platelets and red blood cells (RBCS).

Plasma is the yellow liquid component of whole blood in which the RBCS are
suspended and makes up about 55% of whole blood. Its contents are mostly water,
dissolved proteins, gases, glucose and hormones.

White blood cells in the circulatory system aid against infection. There are
two main types of white blood cells distinguishable by their granules: granulocytes

and agranulocytes and are often a marker for infection. A liter of adult blood contains

between 4 x 109 and 1.1 x 1010 white blood cells. Platelets are small, anucleated cells

measuring about 2-3 pm in diameter that have an average lifespan of about 8 to 12

days. The main function of platelets in the blood is

 

 

 

 

 

 

 

 

 

 

 

 

    
      

 

 

 

 

 

v. __,_ L
/ ’ / . /

 

/

adventitia ,
external elastic
lamina , . .
medial layer Internal elastic
(smooth lamina intima
muscle cells) (endothelial

cells)

Figure 1. Schematic of a human blood vessel. The adventitia is comprised of mostly
connective tissue. The external elastic lamina is comprised of elastic connective
tissue. The medial layer is comprised of smooth muscle cells, the site of NO-induced
vasodilation. The internal elastic lamina is comprised of elastic connective tissue.
The intima is comprised of endothelial cells, the site of NO production. RBCS flowing

through the vessel come in direct contact with this layer.

hemostasis (clot formation). A decrease in platelet count could lead to excessive
bleeding, while activation is thought to play a role in vessel blockage.
RBCS, the most abundant cells in blood, are anucleated and are responsible for

oxygen delivery. They measure 8 — 10 pm in diameter and are about 90 tL in volume.

Adult humans have roughly 2 x 1013 RBCS with men producing slightly more than

. . 6
women. RBCS are syntheSIzed In the bone marrow at a rate of about 2 x 10 every

second. While RBCS are developing they are known as reticulocytes and stay in this
form for about 7 days until they transform into erythrocytes. The average lifespan of
an RBC in the bloodstream is about 120 days. RBCS lack a nucleus, therefore, they
cannot synthesize their own proteins and also rely on anaerobic glycolysis for energy
production.

The RBC membrane is an ordered array of lipids and proteins. The two main
proteins found on an RBC membrane are band 3 and the glycoproteins, the latter being
responsible for a mammals’ blood type. The ordered array of lipids and proteins
affects RBCS deformability; however, while these proteins aid the RBC is
deforrnability and regulating immune recognition, hemoglobin is the most important

component of an RBC.

Hemoglobin comprises 97% of an RBCS dry content and 35% including water.

It was first discovered by Hunefeld in 1840 and the structure was determined in 1959

by Perutz,1 who used three-dimensional x-ray crystallography to resolve a structure of

horse oxyhemoglobin. It is with this knowledge, and those previously reported on

myoglobinz, prior to Perutz, that the hemoglobin structure was derived.

On a cellular level, hemoglobin is a metalloprotein responsible for ensuring
aerobic respiration in cells of vertebrates. A figure of hemoglobin’s structure is shown
in Figure 2. Each of the four globular subunits of hemoglobin consist of a heme

group. The Fe atom in these heme groups is responsible for the uptake of oxygen

throughout the body and its subsequent delivery. The Fe atom must be in the Fe2+

state to bind oxygen, which will temporarily oxidize it to the Fe3+ state. While Fe can

exist in the ferric state, it cannot bind oxygen and is referred to as methemoglobin.
Once oxygen is collected by hemoglobin, it traverses the body in search for respiring
tissue where it is then exchanged through a system of arterioles, capillaries, and
venules infused into tissue. Once metabolic demand is met, hemoglobin is carried
back to the lungs where the partial pressure of oxygen is around 100 mm Hg, a
pressure sufficient for hemoglobin to uptake oxygen; in other words, the hemoglobin
is reoxygenated in the lung bed. While the heme groups are most responsible for
oxygen transport, other structural characteristics of hemoglobin allow for improved
transport and more efficient oxygenation of vascular reperfusion.

Currently, there are two models to explain the allosteric changes

accompanying oxygen uploading that leads to cooperative binding. The KNF model,
named for the founders, Koshland, Nemethy and Filmer,3 explains the sequential
model, meaning the subunits can change their conformation to increase the binding of
oxygen to other subtmits. The other model, the MWC model,4 named after its

founders, Monod, Wyman and Changeux, describes hemoglobin as existing in two

states, the R or “relaxed” state and the T or “tense” state. The R state is the

.30 mam—550.5 88m nomaxo one 515 ”—238.8ch :28?
a so EMS was 8on :9: 9 mafia NO .3 mfiuusobsm wet Saba—om 05 v5 EOE on 05 8365 $5. .852? mm 25: mo
652:5. 33888 05 .Ewt 05 oh .3208 080: a 023 A83 28 man—av Ego scum .Enflwofion =0 858w 2:0: .58 05

me “5:5an 2: mam—ammo ocmfioaow £5. ofldhzosbm 25: mo magnum 818208 van fine—moan: mo obsoabm .N unsur—

£2060on

oEmI

520 o

 

£20
a Emzo oozaoazoa

oxygenated state, while the T state is the deoxygenated state. Equilibrium exists in
these two states such that partial binding of oxygen Shifts the equilibrium to the R state
and partial unloading of oxygen shifts back to the T state. An understanding of
cooperative binding is still not clear, however, the fact that it does occur is imperative
in oxygen transport. If hemoglobin acted like myoglobin, a globular protein which

stores oxygen, tissues would perish due to a lack of oxygenation.

1.2 Nitric Oxide as a Major Vasodilator in the Circulatory System

Oxygenated heme can only get to respiring tissue due to the RBC. The RBC
must flow through small resistance vessels, therefore, a mode of opening these vessels
is required. Dilation from a substance such as nitric oxide (N O), was first discovered

in 1987 by Ignarro, Palmer and Furchgott as the endothelium-derived relaxing factor

(EDRF) as one means of opening these vessels.34’ 35 In the discovery of NO as the

EDRF, segments of vessels were prepared from bovine intrapulmonary artery and vein
and were perfused with an oxygenated Krebs-bicarbonate solution at 37 °C. NO
released from these segments after pharmacological stimuli was measured by the
diazotization of sulfanilic acid and subsequent coupling with N-(l naphthyl)-
ethylenediamine to produce a highly colored product that can be measured
spectrophotometrically. The EDRF and NO from artery and vein segments stimulated
cyclic GMP levels. Furthermore, the cyclic GMP accumulation elicited by the EDRF
and NO was inhibited by methylene blue, oxyhemoglobin, pyrogallol (a powerful

reducing agent) and potassium. Both EDRF and NO elicited the same responses in

both artery and vein segments. This led the authors to conclude that the EDRF was
indeed NO.

Shear stress is one proposed stimulant of NO production by the endothelium.
In studies performed on resting muscle, shear stress resulted in dilation of the arteriole.
Moreover, as long as shear stress was present, dilation was produced. Upon
administration of a nitric oxide synthase (NOS, an enzyme responsible for NO

production) inhibitor, L-NMMA, arteriole dilation decreased by 85%, but remained

. . . 5
the same dIameter at basal conditions.

Studies performed by Sprague et. al. revealed shear stress alone did not induce

NO production, but rather the presence of RBCS was needed.6 Specifically, through

the administration of L-NAME, another NOS inhibitor, and measurement of
transpulmonary pressure (TPP), it was discovered that NO was a determinant of
vascular resistance; however, the inhibitory effect of L-NAME was only seen when
the lungs were perfused with RBCS. These data suggest a dependence of NO
production in the endothelium on RBCS, a discovery that potentially could modify the
aforementioned findings involving shear stress.

Due to the importance of NO in the vasculature, it may also be a determinant
in cardiovascular disease. Atherosclerosis, a condition in which the artery wall

thickens due to build up of fatty materials such as cholesterol, is associated with

compromised NO activity.7 This loss in NO activity is thought to occur in the early

stages of the disease and inflicts considerable effects on its development, such as

leukocyte diapedesis (movement of leukocytes out of the circulatory system) and

further thickening of the arterial wall through smooth muscle cell proliferation. The
origin of the decreased production of NO in atherosclerosis has not been determined.

However, one mechanism proposes that atherosclerotic lesions have a decreased NOS

expression.

1.3 Proposed Mechanisms of Vasodilation in Response to Hypoxia

Due to the demand of oxygen for aerobic respiration, it is reasonable to
conclude that blood flow must change in some way to ensure oxygen delivery to
tissues in need. During an ischemic attack, blood flow is inhibited to organs for a
certain amount of time. In this time frame, the respiring tissue is undergoing
metabolic processes, many of which are to ensure reperfusion (flow of blood), while

others further insult the already injured tissue, such as the production of reactive

oxygen species (ROS) in hypoxic injury.9 The mechanisms leading to reperfusion

have been intensely studied and can be subdivided into three categories. First, the
vesicles themselves somehow sense oxygen concentration, and dilate when this
oxygen tension is low. The second mechanism proposes that the production of ROS
can help relax vessel tone and lead to improved flow. The third proposed mechanism
suggests that a metabolite is released in respiring tissue that calls for an increase in
vascular tone. The following three paragraphs provide further information on these
mechanisms.

In vitro and in vivo studies have been performed to determine whether the

arteriole itself acts as an oxygen sensor. The work of Pittman and Duling in 1974

demonstrated that in porcine isolated carotid artery strips, the mechanical tension

declined in accordance with strip thickness and oxygen tensions, implying oxygen
tensions and arteriole diameter plays a role in vessel dilation.10 Accordingly, through
in viva measurements performed by Jackson in 1984, it was found the arterioles
themselves do not sense changes in oxygen tensions unless the arterioles are
downstream from capillaries, or venules. Furthermore, only global changes in oxygen
tension (global changes referring to a change in oxygen tension in the superfusate

flowing over the entire preparation) elicited a response in vascular diameter. In 1994,

Messina et. al. demonstrated the dependence of vascular constriction in increased

. . ll . . .
oxygen tenSIons on endothelium. Once the endothelrum was elImInated or

incubated with a prostaglandin inhibitor, indomethacin, rat cremaster arterioles
stopped constricting in oxygen tensions from 20 to 150 mm Hg and 150 to 660 mm
Hg.

As mentioned above, ROS production is the second mechanism of reperfusion.

ROS are released upon hypoxic insult.9 One source of this release may come from the

NADP(H) oxidases (NOX) NOX 1,2 and 4. These enzymes are found in the

vasculature wall and are believed to increase ROS, including superoxide and

. 8, 9 . . . . .
peroxrde. In this construct, ROS generatIon varies wrth oxygen tensron, although

this idea has been disputed.12

The third mechanism mentioned that is a plausible cause for reperfusion is the

release of a metabolite to surrounding tissue. Such a metabolite can be identified as

adenosine triphosphate (ATP). The first discovery of extracellular ATP in a perfusate

was reported by Holton in 1959.13

1.4 ATP as a Stimulant for the Vasodilator NO

In 1972, it was reported by Burnstock that ATP acts as an extracellular
signaling molecule in non-adrenergic (non adrenaline mediated) and non-cholinergic

(non acetylcholine mediated) neuromuscular transmission in the gut and urinary
bladder.33 In 1978, specific receptors mediated by adenyl nucleotides called P2 were
discovered.14 P2 receptors were then categorized on the basis of molecular structure
and function: P2X, ligand — gated ion channels, and sz, G protein — coupled
receptors.35 sz receptors respond to both purine (ATPIADP) and pyrimidine
(UTPIUDP) nucleotides; more Specifically, P2Y2, PZYII and P2Y13 are all specific to
ATP.28

When present in and around the vasculature, ATP can either activate P2X

. . . . . 14
receptors on the outsrde of vessels resulting 1n vasoconstrrctron, attenuate

sympathetic (nervous system responsible for “flight or fight”, i.e. raising blood

. . 29 . . .
pressure) vasoconstrrctron, or activate sz receptors on the vessel lumen resulting 1n

dilation.14 Once ATP activates a sz receptor on endothelial cells, intracellular [Ca-

21L] is increased through the phospholipase-C/inositol-l,4,5-triphosphate (1P3)

10

,32

31 . .
pathway. NOS enzymes respon51ble for NO production are located in the

. . . . . . 2+ .
endothelIal cells and then actrvrty 13 Increased by a Ca /calmodulIn-dependent or
. . . . 2+ . .
kinase-dependent mechanism. Upon PZY act1vatlon, Ca concentration Increases and
. . 2+ . . . . . .
thIs Influx of Ca Induces a conformational activation of eNOS, resultIng In

phosphorylation at Ser-1177.15 Once eNOS is activated, it catalyzes the five electron

oxidation of a guanidino nitrogen of L-arginine. The mechanism of this reaction is

displayed in Figure 3. Through the conversion of L-arginine to L-citrulline, NO is

produced per 2 mol of 02 and 1.5 mol of NADPH consumed. After NO is produced,

it binds to the soluble form of guanylyl cyclase, which then converts cyclic guanosine

triphosphate to cyclic guanosine monophosphate (cGMP).16 This cGMP can then
activate K+ channels leading to hyperpolarization and relaxation, increase intracellular

2+ . . . . .
Ca entry Into cells, or stimulate a cGMP-dependent protein kInase that act1vates

myosin light chain phosphatase, an enzyme that dephosphorylates myosin light chains
leading to smooth muscle relaxation. A schematic of the mechanism is displayed in
Figure 4. lankowski et. al. demonstrated that accumulation of cGMP increases in a

parallel manner to ATP-induced glomeruli relaxation suggesting yet further evidence

that ATP may be a determinant in NO production. I 7

1.5 RBC-Derived ATP Release

11

mm_.mOZo mo :ouguoa com: 82—65265 2: E 82m
moo—S .8388 watt .vofismcoo Inn—<2 CO 208 m; was NO mo 38 N Log wooswoa fl OZ 6:53:94 8 2:3me

-4 mo 868280 05 swab: 6553?..— .«o :owobE oEESBw a mo 55318 8.5026 02m 2; .m «Earn

 

 

 

 

 

 

 

 

(4:0... Hui-aim.—
O
IINI O. I
.fz
I % .0. oz.
5.5"
A...
. NIZ _
30“.!an 50“.: an.
O O my .,
2...! O u I O .. - O ..
- .7 m ”a... m .1
.fz . :z o .. :2
h".0. .2991... N3. W 75L Hind..- _
z z - - - iv" N
.le In; .2 I
£2 0: £2 0: N12

I.
NC _.. sown—LEA“. emu—Lan— Sdn. n_n_

12

.580 9 _ommo> 2: mo scum—6 wage—E 98 £22
2 EB: mfimsmo m=oo 28:8 5025 05 3 woman—bu :05 £033 OZ .5368? 05 mo accosted 05 mowbfimo mOZo .=8 65
8E 55 8261.6 nwzofiu mOZo wcugaom £3 36535 05 no SE82 omwuofiba m magnum PS. .3322 3:0 Sioux:
Ea :oumgowow mm :03 2:85 “:26th mo Spas: m 8 26 mumm 80¢ momma—8 mm AER .233 ~88.» meme: 8326585
65 E :28:on OZ 8338:... 0828 ACE EZBFUmm .Omm SW 82m momma—2 AER. mo 8mm 65 mo .52me .9 «Eur..—

 

 

-- EEO»
.. .89
aotanflom .x =oU 28:2 5255

 

 

OZ “moi—3.50.4 é
oEEthA 530. «U
_\a._/ V a :5 3.2.3.5

 

 

O
- Z4--._--_-

 

\
§

\
\

Vs

 

. : lambda A: I ...... =00 coo—m— 60M

13

ATP can be released from RBCS due to a number of different stimuli, one of which is

mechanical deformationlg'21 and pharmacological agents”.24 (iloprost). It was first

reported by Sprague et. al. in 1996 that mechanical deformation of RBCS resulted in
ATP release. A 10% hematocrit of RBCS were subjected to deformation by
introduction into a blood filtrometer, where a polycarbonate membrane with pore sizes
of 5, 8 or 12 um represented the filter. Movement of RBCS through the membrane
was induced by opening an outflow tap. If both average pore size and the hematocrit
are controlled, the transit time of the RBCS is a measure of their deforrnability or
ability to migrate through the pores. After flow, 250 uL of the RBC sample were then
collected from the filter and placed into a cuvette with 250 DL of luciferin/luciferase.
The mixture of both ATP and luciferin/luciferase results in a chemiluminescence
reaction where the amount of light is proportional to the concentration of ATP in the
sample. In rabbit RBCS, resting intracellular ATP concentration is ~ 1.6 i 0.4 mM.
Upon passage through the filter, both ATP release and RBC transit time increased
significantly. In human RBCS, resting ATP concentrations were similar; however,
human RBCS displayed a higher transit time reflecting the slightly larger diameter of
human RBCS in comparison to those of rabbits. Importantly, the ATP release from
human RBCS increased due to mechanical deformation.

In addition to deformation-induced release of ATP, there are also
pharmacological reagants that will stimulate this release. For example, prostacyclin

analogs have been used in drug therapy for patients with primary pulmonary

. . . . 25-
hypertensron; however, the mode of action of prostacyclm has not been elucrdated.

28 More specifically, the active prostacyclin analog, iloprost, has been studied

14

extensively. Iloprost binds to the IP receptor (prostacyclin receptor) that is coupled to

the GS protein and adenylyl cyclase. The mechanism by which iloprost reduces

vascular resistance and aids in treating primary pulmonary hypertension has been

proposed to be via the GS protein pathway. A signal transduction pathway involving
the GS protein for the release of ATP from RBCS under deformation and

pharmacological stimuli has been described.23 and involves the activation of adenylyl

cyclase. The binding of iloprost to rabbit and human RBCS has been previously

demonstrated;29-3l however, the only measured affect of iloprost on RBCS has been

. . . . 32, 33
an Increase In the deformabIlIty.

As previously mentioned, deformability is one stimuli for the Gs signaling

pathway for RBC-derived ATP release. This led the authors to suggest iloprost may
decrease vascular resistance through the release of ATP, a stimulant for the potent
vasodilator NO. In a study performed by Sprague et. al. , rabbit RBCS, incubated with

iloprost displayed a 332 i 72% increase in ATP release that was prevented in the

presence of an IP receptor antagonist, CAY10441.34 It was suggested the increase in

ATP release is a result of the increased deformability of the iloprost influenced RBC.
Moreover, iloprost results in ATP release from RBCS in the absence of flow.
Another stimulus for ATP release from RBCS is hypoxia or a lowering of

oxygen levels in the environment. The first study of hypoxia-induced ATP was

performed by Paddle and Burnstock.35 They investigated whether ATP was released

15

from hypoxic myocardium in guinea pig hearts by switching the oxygen concentration
of the perfusate.35 ATP concentrations were found to be increased threefold by the
end of the hypoxic exposure. Furthermore, perfusion pressure diminished, suggesting
an increase in vessel dilation. Subsequently, other studies were performed that
confirmed the results reported by Paddle and Burnstock.36

Since these discoveries, more evidence has accumulated suggesting that

hypoxia results in ATP release. In 1992, Bergfeld et. al. measured the ATP release

from human RBCS exposed to a brief period of hypoxia/hypercapnia (hypercapnia is a

condition in which there are high levels of carbon dioxide in the blood).37 The ATP

release after exposure to 50 s of hypoxia/hypercapnia at 37 °C increased from 0.45 x

106 i 0.04 to 2.67 x 106 i 0.27 molecules/RBC released.

ATP is a highly charged molecule in comparison to other biomolecules that
come into contact with the RBC membrane, therefore, Bergfeld and Forrester further
investigated the mechanism by which ATP was leaving the RBC. ATP carries a
charge of negative four; therefore, departure from the RBC results in a large potential
difference across the membrane. Therefore, it is thought that an efflux of other
charged molecules must be employed to balance the electrical equilibrium.

The band 3 protein on the erythrocyte membrane is an anionic channel

responsible for the efflux of Cl- and HCO3-; however, it can also be responsible for

the transfer of a large range of molecules from the RBC. Through inhibition of the

band 3 protein, and subsequent exposure to hypoxia, the authors concluded the band 3

16

protein plays a role in the release of ATP from the RBC as indicated by a decrease in
ATP release.

Coinciding with the results performed on the band 3 protein, were those
obtained from the inhibition of a separate nucleoside transporter, band 45. It was also
demonstrated that inhibition of band 45 protein, and successive exposure to hypoxia,
also resulted in a decrease in ATP release. These data suggest both proteins play a
role in the release of hypoxia-induced RBC-derived ATP release.

The authors proposed a model for such a situation. The ratio of band 3 to band

. . 6 4 . . .
45 protein exrsts as 1.2 x 10 to 10 , and assuming one channel per protein exrsts,

only a small portion of the band 3 protein would be utilized in the actual release of

ATP. This leaves the band 3 protein open for CH HCO3- exchange. Furthermore, the

nucleoside portion of the band 45 protein is directly involved in the release of ATP.
A schematic of the pathway proposed by Bergeld for hypoxia-induced ATP release is
shown in Figure 5.

From this study, it was concluded that the release of hypoxia-induced ATP
from human erythrocytes occurs through the band 45 protein. This study
demonstrated one mechanism of ATP release; however, it fails to explore how

hypoxia stimulates the ATP release.

1.6 Hypoxic Vasodilation

Many theories propose the RBC elicits vasodilation not only through the

release of ATP, a stimulant for NO production, but rather through the release of NO

17

ATP 4- cr HCO -,

->

 
 

3-0

 

ii
if?

Figure 5. Mechanism for ATP release proposed by Bergfeld and Forrester.37 ATP
leaves the RBC through the band 45 protein. The efflux of the ATP causes an
imbalance in the electrical potential of the cell. ATP has a charge of negative 4, and

 

 

 

 

 

   

I
I
I
I
I
I

t

*-

ATP 4_ Cl HCO -,

upon leaving the cell, takes a Mg2+ atom with it. This leaves a negative 2 charge on the
RBC that needs to be compensated for to keep electrical balance. Band 3 imports a
chloride ion and a bicarbonate ion to keep electro-homeostasis.

18

itself. Vasodilation in response to tissue hypoxia is a crucial and necessary step in

regulating vascular homeostasis. Upon hypoxic insult, vasodilation ensures the

delivery of oxygen and metabolites to respiring tissue.38 Hemoglobin, the main

carrier of oxygen throughout the body, is responsible for oxygen delivery and

therefore the RBC itself mediates this response. The RBC as a component in this

39 . .
process has been made clear; however, the mechanrsm has not been elucrdated.

4o, 41 42-44

Many proposed mechanisms exist, such as ATP release, nitrite reductase

and SNO-Hb.45’ 46

In order for hypoxia-induced ATP release to be a determinant in hypoxic
vasodilation it must be shown that ATP increases vascular caliber in vivo. Sprague et.
al. demonstrated that concentrations of ATP as low as 300 nM resulted in a decrease
in total pulmonary vascular resistance in the perfused rabbit lung. McCullough et. al.

displayed a dependence of a vasodilator response on ATP using the hamster cheek

pouch retractor muscle.47 Addition of ATP ranging from a concentration of 10'8 M to

10'4 M, induced an increase in vessel diameter from 16 to 65% with the maximum
change seen at ~150 uM. As previously demonstrated by Bergfeld, hypoxia induces
ATP release in RBCS.37 This shows a definitive relationship between hypoxia-

induced ATP release and subsequent vasodilation. A schematic of hypoxia-induced
RBC-derived ATP is displayed in Figure 6.
In addition to the theory that hypoxic vasodilation is mediated by the RBC,

others believe that another RBC-mediated mechanism exists. Specifically, studies

19

8882688. 8:23 2 80:3
mzoo 28:8 £888 2: 8 momzflmv OZ .02 892684....» 883 8: 802608 £285 mOZo 88938.. .58 8 .25. 8:3 Ezofiowco
no sham .5888 29088; 2: 88853 2 0mm 8d 88a @8828 2 Pix 880 .8528 Pix 326888093 mo otmEonom .e FEMS

8:8

283E EooEm
.2

.9
O
2

 

 

‘1' ‘

I w=®0
I E55095

\

>.
N
O.

 

 

r----------->

 

$895

20

have shown that hypoxic vasodilation include the role of nitrite reduction and

deoxyhemoglobin for the production of NO. Nitrite concentrations in the plasma and ‘

RBC are reported to be 120 nM and 290 nM, respectively.48 Moreover, nitrite has

. . . . . . . 49, 50
been recognized as a precursor for NO under aC1d1c and lschemlc conditions. At

physiological pH, addition of nitrite dilates aortic rings through the activation of

soluble guanylate cyclase (sGC).51'53 These processes suggest a biological function

for nitrite in the control of vascular homeostasis.

It has recently been reported that oxygen sensing by the RBC is controlled

2, 43, 54

through the conversion of nitrite to NO through deoxyhemoglobin.4 NO

production from nitrite in cell-free hemoglobin or from RBCS has also been

43, 55-59

established. Crawford et. al. displayed a maximum in nitrite reductase

activity when hemoglobin is at P50; the largest quantity of NO produced at 50% —

60%.60 The conflicting physiochemical requirements of open binding sites in the T

state for nitrite to bind and the greater redox reaction rate of hemoglobin in the R state
results in this optima] conformation. In addition to increases in NO production,
deoxygenating RBCS and nitrite both stimulate various NO-mediated activities such as
vasodilation, cGMP formation, and mitochondrial respiration inhibition. These effects
were inhibited by the NO scavenger, C-PTIO. A mechanism for this process is
displayed in Figure 7.

While this mechanism has sufficiently described a possible role for nitrite in

hypoxic vasodilation, there exist weaknesses in its viability. For example, after

21

.noummnoag «6:0 Ea own—o5 32mg 0328 8 23 58
a 20:3 £60 28:8 5095. 06 8 83.058553 05 was .Essofiouco 05 swab: 33th :05 0 Z ma: .0 Z 9 85? 0325303
mamas?“ 88262 BEE a mm 38 Ego—moan: coaflsfim :omzxo o\°om “a momomoa 35 scum—60mg 2x093 mo EmgnouE .b «unwr—

28
28:8 5095

 

 

 

 

 

 

520.985;

22

deoxygenated RBCS and nitrite are combined, vasodilation takes from 2 to 7 minutes
until complete relaxation is achieved. When oxygen tension was decreased as a
function of time the vasodilation occurred more quickly. These data suggest the
discrepancy in time of relaxation is dependant upon the steady-state and non-steady
state oxygen tensions.

In 1996, Stamler and co-workers suggested a SNO-Hb (S-nitrosylated

hemoglobin) paradigm. In this construct, NO binds to oxyhemoglobin on the highly
B93

conserved Cys and is released upon transition from the R (oxy, low spin) to the T

(deoxy, high spin) state.61 The position of NO on the hemes and thiols is redistributed

in accordance with the position of the RBC in the circulation. More importantly,
interactions of the cysteine residue in the hemoglobin-binding cytoplasmic domain of

the anion exchanger AEl promotes the deoxygenated structure which serves to

relocate NO group transfer to the membrane.62 Furthermore, the NO released from

SNO-Hb has been shown to induce vasodilation in blood vessels +/- the

46, 61, 63

endothelium. A mechanism for this process is displayed in Figure 8.

While the idea of SNO—Hb inducing vessel dilation appears to be acceptable,
there exists a number of problems with this theory. First, the reaction of a reduced

thiol with NO requires a one electron oxidation; however, there is no obvious electron

. . . . 64
acceptor. The ferriheme IS an unlikely source because of the slow rate of reaction.

A second problem arises in the measurements. Specifically, many groups have

46, 64-6

reported undetectable levels of SNO-Hb in the circulation. 8 Moreover, the

23

0202:0003 20:0 000 0020.3 002.35%
03200 8 0:3 :00 2 00023 0:00 20008 50080 05 8 802050000000 05 0:0 .802050000 2: @0005 000306 :05 OZ
0:; 20893 .20 008 :0 000000 2 00:3 OZ 000200 0303080: 000030008 05 down—0020 05 $200023 22? 02208 05000.8
man 05 00 2 00300000 05 0mg: 05 5 OZ 00280: 0503060: 00000000 005 0002:0005, 2x093 m0 E25522 .w 0.5»;

2.00
0.002: 50080
\

Uflll ilv

 

 

 

 

 

 

 

 

 

 

 

TI 0:00
Q _0__0£00:0
c0_0c0-0
00m. 9.
2x095 0503080:

020.00.?

24

measurements used to quantify SNO-Hb involve the liberation of No.6971 One

method involves photolysis and chemiluminescence to detect liberated No.69

Another method involved Hg2+, which breaks down nitrosothiols, and then

subsequent chemiluminescence of the nitrogen oxide species; however results using

67, 70, 71

the latter method were only consistent in rodent RBCS and not in human. A

third problem arises from the unclear mechanism of NOx transfer from SNO-Hb to the
smooth muscle cell. As mentioned earlier, the AEl membrane protein may be
responsible for the transfer, however, because of the methodology of the experiments
involved it is not clear whether NO is responsible or other nitroso species. More
importantly, this proposed mechanism explains the transfer of NO to the intracellular
surface of the RBC but fails to explain how the NO traverses the lumen, past the
endothelial wall and elastica interna, to the smooth muscle cells. Furthermore, this
assumption excludes the properties of blood flow presented by Fahreus-Lindquist,
which states blood possesses turbulent flow in a vessel the size of an artery with a cell-
free layer forming at the vessel wall. A fourth problem, coinciding with the second,

results from a discrepancy in the values reported for SNO-Hb ranging from <lnM to
2 7

5p.M.66’ 67, 7 , 3
Nagababu et. al. has attempted to establish a relationship between these three

conflicting ideas. Through the intermediates formed from the nitrite reductase activity

of hemoglobin, SNO-Hb is formed?4 Hb(III)NO is seven times more effective at

producing SNO-Hb than Hb(II)NO. While other investigators have seen a higher

25

production of SNO-Hb with Hb(III)NO; the idea of this forming SNO-Hb was ignored

. . . . 4
because only a negllglble amount of the intermedlate was assumed to be formed. 3’ 66’

75 The authors had previously discovered the intermediate is stable through EPR and

. . 42 . . . .
chemllummescence. The authors have also established the intermediate responSIble

for SNO-Hb formation is the nitrosonium cation NO+.42

One proposed mechanism for the transfer of NO to [3-93 cysteine is through a

reaction of Hb(II)NO with H20 to form H2N02+ or HNOZ which can then diffuse

through the hydrophobic globin to the cysteine;76 however due to hydrophilic nature
of these molecules this is not a likely mechanism. Another mechanism is through the
use of nitrite as the nucleophile facilitating the release of N203.77 N203 is a

hydrophobic molecule and therefore can diffuse through the hydrophobic pocket
leaving this as a plausible mechanism. A final mechanism presented by the authors

leaves out the presence of transnitrosation but rather employs an electron transfer

between the NO+ and the B-93 cysteine.

1.7 The Effects of Hypoxia on the Vasculature

Along with the occurrence of vasodilation through a number of proposed
mechanisms, hypoxia also exerts many more effects on the vasculature. Type 1

glomus cells (a cell located in the carotid body that helps regulatebreathing) are

. . . . 78 .
responSIble for recogmzmg low oxygen tensmn. The nerve receptor responSIble for

26

innervating these cells is still debatable; however, 79 dopamine has been shown to

modulate the sensitivity of the carotid body to hypoxia.80 Acetylcholine has also

been recognized as _a possible neurotransmitter in the role of the carotid body in

recognizing hypoxia.81 After sensing hypoxia, many other changes occur in the body.

The most immediate responses include inhibition of the 02 sensitive K+ channel,

membrane depolarization, and an influx of calcium and release of neurotransmitters
from the synaptic vesicles. However, there are other physiological events in response
to hypoxia.

One major response to hypoxia is the increase in blood volume, or

polycythemia, regulated by erythropoietin, a glycoprotein synthesized and released
from the kidney during hypoxia.82 It regulates erythropoiesis, the production of red

blood cells. Under normoxic conditions, erythropoietin levels are normal; however,
under hypoxia, the levels increase. Unlike the carotid body which activates at oxygen

levels around 15%, erythropoietin levels don’t increase until oxygen saturation is

below 10%.83

Another major response to hypoxia is angiogenesis, or the grth of new
blood vessels. In tumor cells, angiogenesis occurs quickly supplying the cancerous

cells with nutrients and metabolites. Evidence shows this increase in angiogenesis is

due to an increase in the vascular endothelial growth factor (V EGF).84 VEGF is

regulated by hypoxia at the level of gene expression and is increased after

approximately 6 h of hypoxic conditions.85

27

The Pasteur effect is the change in metabolism due to a low oxygen saturation.
Aerobic respiration switches to anaerobic glycolysis, producing less ATP. Within the

first hours of hypoxia, glycolysis increases. Because anaerobic glycolysis produces

. . . 86 .
substantially less energy, its rate increases to compensate. This occurs by two

mechanisms: the increase in enzymatic activity and gene expression of the glycolytic

enzymes, and an increase in glucose uptake through increased activity of GLUT 1

87
glucose transporter.

As the cells that line vessel walls, endothelial cells are unmistakably one of the
most important aspects of the vasculature. Eukaryotic cells produce energy in aerobic
conditions. This environment yields the greatest output of energy; however, in many
diseases such as cancer oxygen levels remain low. This requires an adaptive response

of endothelials cells to a drop in oxygen tension. Not surprisingly, endothelials cells
are relatively resilient to hypoxic conditions.88 Under conditions of hypoxia for 48 h,
endothelials cells still maintain about 70% of their ATP production and metabolism.89

Endothelial activation is a coined response to the different signaling pathways

that become activated under hypoxic conditions. For example, in conditions of

hypoxia, cAMP levels are decreased,90 which may be due to a decrease in adenylyl

cyclase activity.90 Moreover, a decrease in phosphodiesterases may also contribute to
91-93 . . . .

lower cAMP levels. Followmg closely, another cyclic nucleotide With decreased

94,95

levels in endothelial dysfunction is cGMP. cGMP is a product of activation of

. . . . . 6
guanylate cyclases by NO, therefore, it IS a major component of vessel dilation.9

28

cAMP and cGMP decrease in endothelial cells are just one of many processes that
occur in response to hypoxia. Other processes such as gene transcription regulation
become altered. One major change is the membrane of endothelial cells after
exposure to 24 h hypoxic conditions. A diagram of the processes of endothelial
activation is displayed in Figure 9.

The membranes of endothelial cells permit molecules to pass through by a

process called restricted diffusion.97 This type of diffusion allows for small molecules

to pass while larger molecules need transport molecules to aid in their crossing. In

conditions of 24 h hypoxia, the membrane of endothelial cells loses its selectivity
process and allows for the passage of molecules, large or small to diffuse through.98
This change in membrane permeability highly contributes to the procoagulant state as
exposure of the endothelial substrata, rich in collagen and tissue factor, is exposed.99
These processes are reversible, with a return to normal restrictive diffusion 24 h after

. 99
exposure to normal oxygen tenswn.

Adhesion receptors are surface proteins present on endothelial cells and
leukocytes. They work to maintain the proper position of these two cell types.

Hypoxic conditions lead to an increased expression of adhesion receptors, which leads

. . . . 100 . .
to further expressmn of addltlonal adhesmn receptors. More spec1fically, selectm

E is expressed on both endothelial cells and leukocytes and is involved in leukocyte

adherence to the endothelial wallwl Once adhered, leukocytes are able to migrate

29

Hyploxia

Endothelial Cells

—>

 

I
V

Endothelial Activation

  
 

Membrane

I Vasodilators
Fluctuations

Vasoconstrictors

i cAMP
cGMP

Figure 9. Cascade of events stemming from endothelial activation, the activation
of the endothelium in response to hypoxia. Upon endothelial activation,
membrane fluctuations occur, (i.e. the junction between endothelial cells open
wider), cAMP and cGMP, components for the signaling pathway for the
production of NO. levels are decreased and vasodilator and vasoconstrictor levels

30

through the endothelial membrane by the adhesion molecule vascular endothelial

(VE)-cadherin.]02’ ‘03

It has been established the endothelium plays a role in the regulation of

vascular tone. Through the release of such molecules as NO, endothelial cells can

2,104

. . . . 5
induce vasodilation on underlying smooth muscle cells. Conversely, under

conditions of hypoxia, the endothelium can elicit a vasoconstrictive response. The

. . . . . 105 . .
most potent vasoconstrictor 1n the endothellum IS endothelm-l. Hypox1a induces

. . . 106 . . .
the expressmn and secretion of endothelm-l. Upon secretion, endothelm-l binds

to receptors on vascular smooth muscle cells, increases calcium influx and activates

. . . . . . . 107
protein kinase C which phosphorylates the myosm leading to vasoconstnction. It

has been shown by the same group that NO suppresses this effect. 108

One major protein in hypoxia pathway is hypoxia-inducible factor (HIF), a

transcriptional factor composed of a HIFlB subunit and one of the HIFa subunits. Of

these, HIF-la is the most extensively studiedlogu111 HIF-l a is regulated by an 02-

. . . . - 4
dependent pathway: the hydroxylatlon of prolme or asparagine reSIdues.112 11 In

well-oxygenated cells, HIFa is short lived (< 5 min).115 HIF-la is synthesized

continuously but is degraded quickly by the ubiquitin-proteasome system through a

. . 116-118 . .
process that requires two 02-dependent domains. Conversely, in hypox1c

119,120

conditions, HIFa accumulates. HIF-la is unique to HIF-l and is the primary

determinant in DNA binding and transcriptional activity whereas HIP-10 is unaffected

31

116,121,122

by hypoxia. One process controlled by HIF-la is the negative regulation

of hepcidin, the key hormonal regulator of iron homoeostasis.123 Hepcidin is

synthesized in the liver and inhibits active transmembrane export of iron by blocking

ferroportin, an iron transporter.124 Through the stabilization of HIF-la, both hypoxia

. . . . . . . . . 123
and non defic1ency suppress hep1c1d1n and therefore increase bioavailable iron.

Reactive oxygen species (ROS) are formed as a result of oxidative reactions.

Enzymes exist to protect cells from the deleterious effects of ROS, such as catalase,

superoxide dismutase and glutathione peroxidase.125 ROS becomes harmful when

released in large amounts and in the presence of catalytic iron ions, which are released

. 126 . . .
under hypoxm. In the presence of hypox1a, the antiox1dant system may not

counteract ROS this would lead to lipid peroxidation in RBCS and cell membrane

damage.127 In a present study done by Devi et. al. the effect of hypobaric hypoxia

. . . . . . . . 128 .
and antiox1dants 1n improvmg ox1dat1ve stress in the RBCS of rats was tested. This

was accomplished by evaluating the antioxidant defense enzymes, assaying the lipid
peroxidation, and determining the status of protein oxidation. The antioxidant
enzymes evaluated were superoxide dismutase, catalase and glutathione peroxidase.
The markers of oxidative stress assayed were hemolysis, osmotic fragility and
membrane malondialdehyde, a marker of lipid peroxidation. Finally, to assess protein
oxidation, membrane protein carbonyl content was measured, advanced oxidized
protein products and membrane sulphydryl groups. Results indicate an increase in the

activities of superoxide dismutase and catalase under intermittent hypoxia. Hemolysis

32

caused by free radicals can be characterized by lipid peroxidation and redistribution of

. . . . 129 . . .
ox1dized band 3 Within the membrane. The authors noted an increase in hemolySis,

induced by hydrogen peroxide, that may be due, in part, to the higher occurrence of
these events in hypobaric-hypoxia rats. In conclusion, oxidative stress was higher in
rats with intermittent hypobaric hypoxia which was offset with supplementation of

antioxidants such as vitamin C, vitamin E and L—carnitine.

1.8 Thesis Objective

From the binding of oxygen to hemoglobin in the lungs to the distribution in
respiring tissue, it is abundantly clear the processes involved in hypoxia are very
complex. Findings from 150 years ago, such as NO binding to heme, are still being
elucidated. Many processes occur in the vasculature to ensure correct dilation and
vasoconstriction of blood vessels. NO is a major determinant of these processes in the
vasculature. Through the binding of NO to soluble guanylate cyclase, smooth muscle
cells that underlie the endothelium dilate. NO can be induced by shear stress or by a
metabolite released into the blood stream, such as ATP, which can activate purinergic
receptors on endothelial cells that, in turn, activate eNOS, the enzyme responsible for
NO production in the endothelium. Through various stimuli, such as mechanical
deformation, or hypoxia, RBCS release ATP; however, the mechanism due to hypoxia
remains incomplete. The release of ATP from RBCS under hypoxic conditions has

been demonstrated; however, the mechanism is still under speculation.

33

This dissertation will provide evidence that the different conformational
changes hemoglobin undergoes in differing oxygen tensions induces a deformation on
the RBC membrane, which is the origin of ATP release. In chapter 2, the ATP release
from both hypoxia and deformation were tested simultaneously to establish whether an
additive or non-additive relationship exists. The data obtained provided the
hypothesis for the research in chapter 3. Through the inhibition of the deformation-
induced signaling pathway and subsequent hypoxia and the addition of a cell
stiffening agent, it was concluded hypoxia-induced ATP release is really a form of
deformation on the RBC membrane. Furthermore, the mechanism of hypoxic
vasodilation is described in chapter 4 starting from the RBC and ending with

endothelium-derived NO.

34

References

l.

10.

ll.

12.
13.
14.

15.

16.

17.

18.

M. F. Perutz, L. K. Steinrauf, A. Stockell and A. D. Bangham, J. Mol. Biol.,
1959, 1, 402-404.

H. Scouloudi, Nature (London, United Kingdom), 1959, 183, 374-376.

D. E. Koshland, Jr., G. Nemethy and D. Filmer, Biochemistry, 1966, 5, 365-
385.

J. Monod, J. Wyman and J. P. Changeux, J. Mol. Biol., 1965, 12, 88-118.

I. A. Arenas, Y. Xu and S. T. Davidge, American Journal of Physiology, 2006,
290, H1259-H1263.

R. S. Sprague, A. H. Stephenson, R. A. Dimmitt, N. A. Weintraub, C. A.
Branch, L. McMurdo and A. J. Lonigro, American Journal of Physiology,
1995, 269, H1941-H1948.

U. Forstennann, A. Mugge, U. Alheid, A. Haverich and J. C. F rolich,
Circulation Research, 1988, 62, 185-190.

P. W. Shaul, J. Physio]. (Cambridge, U. K. ), 2003, 547, 21-33.

M. Younes and O. Strubelt, Res. Commun. Chem. Pathol. Pharmacol, 1988,
59, 369-381.

R. N. Pittman and B. R. Duling, Microvasc Res, 1973, 6, 202-211.

E. J. Messina, D. Sun, A. Koller, M. S. Wolin and G. Kaley, Microvasc Res,
1994, 48,151-160.

M. S. Wolin and S. A. Gupte, Circulation Research, 2005, 97, 612-614.
P. Holton, J. Physiol. (Cambridge, U. K), 1959, 145, 494-504.
G. Burnstock and C. Kennedy, Gen. Pharmacol, 1985, 16, 433-440.

S. Dimmeler, I. Fleming, B. Fissithaler, C. Hermann, R. Busse and A. M.
Zeiher, Nature (London), 1999, 399, 601-605.

P. A. Marsden, T. A. Brock and B. J. Ballermann, American Journal of
Physiology, 1990, 258, F1295-F1303.

M. Jankowski, M. Szczepanska-Konkel, L. Kalinowski and S. Angielski, J.
Physiol. (Cambridge, U. K), 2001, 530, 123-130.

R. S. Sprague, M. L. Ellsworth, A. H. Stephenson and A. J. Lonigro, American
Journal of Physiology, 1996, 271, H2717-H2722.

35

19.
20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

A. G. Gilman, Annu Rev Biochem, 1987, 56, 615-649.

E. J. Diamantopoulos, C. Kittas, D. Charitos, M. Grigoriadou, G. Ifanti and S.
A. Raptis, Hormone and Metabolic Research, 2004, 36, 142-147.

S. Carroll Jamie, C.-J. Ku, W. Karunarathne and M. Spence Dana, Anal Chem
2007, 79, 5133-5138.

T. Nakajima, K. Wakarnatsu and H. Mukai, Animal Toxins, 2000, 116-126.

J. J. Olearczyk, A. H. Stephenson, A. J. Lonigro and R. S. Sprague, Med Sci
Monit 2001, 7, 669-674.

I. Schwaner, R. Seifert and G. Schultz, Eicosanoids, 1992, 5 Suppl, $10-12.

F. T. Kaneko, A. C. Arroliga, R. A. Dweik, S. A. Comhair, D. Laskowski, R.
Oppedisano, M. J. Thomassen and S. C. Erzurum, Am J Respir Crit Care Med,
1998, 158, 917-923.

K. Laliberte, C. Ameson, R. Jeffs, T. Hunt and M. Wade, J. Cardiovasc.
Pharmacol, 2004, 44, 209-214.

V. V. McLaughlin, S. P. Gaine, R. J. Barst, R. J. Oudiz, R. C. Bourge, A.
Frost, I. M. Robbins, V. F. Tapson, M. D. McGoon, D. B. Badesch, J. Sigman,
R. Roscigno, S. D. Blackburn, C. Ameson, L. J. Rubin and S. Rich, J.
Cardiovasc. Pharmacol, 2003, 41, 293-299.

M. Ozkan, R. A. Dweik, D. Laskowski, A. C. Arroliga and S. C. Erzurum,
Lang, 2002, 179, 233-243.

A. K. Dutta-Roy and A. K. Sinha, Biochim. Biophys. Acta, Biomembr., 1985,
812, 671-678.

A. K. Dutta-Roy, L. Hoque and B. J. Paterson, Eur. J. Biochem, 1993, 213,
1167-1173.

C. Willems, H. V. Stel, W. G. Van Aken and J. A. Van Mourik, Br. J.
Haematol, 1983, 54, 43-52.

R. A. Korbut, T. Adamek-Guzik, J. Madej and R. Korbut, J. Physio].
Pharmacol, 2002, 53, 65 5-665.

I. B. Kovacs and J. O'Grady, Agents Actions, 1984, 14, 306-310.

R. S. Sprague, E. A. Bowles, M. S. Hanson, E. A. Dufaux, M. Sridharan, S.
Adderley, M. L. Ellsworth and A. H. Stephenson, Microcirculation (N. Y., NY,
U. S.), 2008, 15, 461-471.

B. M. Paddle and G. Burnstock, Blood Vessels, 1974, 11, 110-119.

36

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

W. G. Nayler, P. A. Poole-Wilson and A. Williams, J. Mol. Cell. Cardiol,
1979, 11, 683-706.

G. R. Bergfeld and T. Forrester, Cardiovascular Research, 1992, 26, 40-47.

J. Gonzalez-Alonso, D. B. Olsen and B. Saltin, Circulation Research, 2002,
91, 1046-1055.

M. L. Ellsworth, T. Forrester, C. G. Ellis and H. H. Dietrich, Am J Physiol
1995, 269, H2155-2161.

R. S. Sprague, J. J. Olearczyk, D. M. Spence, A. H. Stephenson, R. W. Sprung
and A. J. Lonigro, American Journal of Physiology, 2003, 285, H693-H700.

M. L. Ellsworth, Medicine & Science in Sports & Exercise, 2004, 36, 35-41.

E. Nagababu, S. Ramasamy, D. R. Abemethy and J. M. Rifldnd, Journal of
Biological Chemistry, 2003, 278, 46349-46356.

K. Cosby, K. S. Partovi, J. H. Crawford, R. P. Patel, C. D. Reiter, S. Martyr, B.
K. Yang, M. A. Waclawiw, G. Zalos, X. Xu, K. T. Huang, H. Shields, D. B.
Kim-Shapiro, A. N. Schechter, R. O. Cannon and M. T. Gladwin, Nature
Medicine (New York, NY, United States), 2003, 9, 1498-1505.

M. T. Gladwin, J. H. Crawford and R. P. Patel, Free Radical Biol. Med, 2004,
36, 707-717.

D. J. Singel and J. S. Stamler, Annual Review of Physiology, 2005, 67, 99-145.

J. S. Stamler, L. Jia, J. P. Eu, T. J. McMahon, I. T. Demchenko, J.
Bonaventura, K. Gemert and C. A. Piantadosi, Science (Washington, D. C),
1997, 276, 2034-2037.

W. T. McCullough, D. M. Collins and M. L. Ellsworth, American Journal of
Physiology, 1997, 272, H1886-H1891.

A. Dejam, C. J. Hunter, M. M. Pelletier, L. L. Hsu, R. F. Machado, S. Shiva,
G. G. Power, M. Kelm, M. T. Gladwin and A. N. Schechter, Blood, 2005, 106,
734-739.

J. O. Lundberg and E. Weitzberg, Arterioscler., Thromb, Vasc. Biol., 2005,
25, 915-922.

J. L. Zweier, P. Wang, A. Samouilov and P. Kuppusamy, Nat. Med. (N. Y. ),
1995, 1, 804-809.

L. J. Ignarro and C. A. Gruetter, Biochimica et Biophysica Acta, General
Subjects, 1980, 631, 221-231.

37

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.
63.

64.

65.

R. F. F urchgott and S. Bhadrakom, Journal of Pharmacology and
Experimental Therapeutics, 1953, 108, 129-143.

W. P. Arnold, C. K. Mittal, S. Katsuki and F. Murad, Proceedings of the
National Academy of Sciences of the United States of America, 1977, 74, 3203-
3207.

V. P. Reutov and E. G. Sorokina, Biochemistry (Moscow), 1998, 63, 874-884.

Z. Huang, S. Shiva, D. B. Kim-Shapiro, R. P. Patel, L. A. Ringwood, C. E.
Irby, K. T. Huang, C. Ho, N. Hogg, A. N. Schechter and M. T. Gladwin,
Journal of Clinical Investigation, 2005, 115, 2099-2107.

K. T. Huang, A. Keszler, N. Patel, R. P. Patel, M. T. Gladwin, D. B. Kim-
Shapiro and N. Hogg, Journal of Biological Chemistry, 2005, 280, 31126-
31131.

M. P. Doyle, R. A. Pickering, T. M. DeWeert, J. W. Hoekstra and D. Pater,
Journal of Biological Chemistry, 1981, 256, 12393-12398.

B. P. Luchsinger, E. N. Rich, A. J. Gow, E. M. Williams, J. S. Stamler and D.
J. Singel, Proceedings of the National Academy of Sciences of the United
States of America, 2003, 100, 461-466.

B. P. Luchsinger, E. N. Rich, Y. Yan, E. M. Williams, J. S. Stamler and D. J.
Singel, J. Inorg. Biochem, 2005, 99, 912-921.

J. H. Crawford, T. S. Isbell, Z. Huang, S. Shiva, B. K. Chacko, A. N.
Schechter, V. M. Darley-Usmar, J. D. Kerby, J. D. Lang, Jr., D. Kraus, C. Ho,
M. T. Gladwin and R. P. Patel, Blood, 2006, 107, 566-574.

L. J ia, C. Bonaventura, J. Bonaventura and J. S. Stamler, Nature (London),
1996, 380, 221-226.

J. R. Pawloski, D. T. Hess and J. S. Stamler, Nature, 2001, 409, 622-626.

T. J. McMahon, A. E. Stone, J. Bonaventura, D. J. Singel and J. S. Stamler,
Journal of Biological Chemistry, 2000, 275, 16738-16745.

M. T. Gladwin, J. R. Lancaster, B. A. Freeman and A. N. Schechter, Nature
Medicine (New York, NY, United States), 2003, 9, 496-500.

T. J. McMahon, R. E. Moon, B. P. Luschinger, M. S. Carraway, A. E. Stone,
B. W. Stolp, A. J. Gow, J. R. Pawloski, P. Watke, D. J. Singel, C. A.
Piantadosi and J. S. Stamler, Nature Medicine (New York, NY, United States),
2002, 8, 711-717.

38

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.
81.

82.

A. J. Hobbs, M. T. Gladwin, R. P. Patel, D. L. H. Williams and A. R. Butler,
Trends Pharmacol. Sci, 2002, 23, 406-411.

T. Rassaf, N. S. Bryan, R. E. Maloney, V. Specian, M. Kelm, B.
Kalyanaraman, J. Rodriguez and M. Feelisch, Nature Medicine (New York NY,
United States), 2003, 9, 481-482.

N. S. Bryan, T. Rassaf, J. Rodriguez and M. Feelisch, Nitric Oxide, 2004, 10,
221-228.

J. S. Stamler, O. Jaraki, J. Osborne, D. 1. Simon, J. Keaney, J. Vita, D. Singel,
C. R. Valeri and J. Loscalzo, Proceedings of the National Academy of Sciences
of the United States of America, 1992, 89, 7674-7677.

B. Saville, 1958, 83, 670-672.

M. Feelisch, T. Rassaf, S. Mnaimneh, N. Singh, N. S. Bryan, D. Jourd'heuil
and M. Kehn, FASEB Journal, 2002, 16, 1775-1785.

P. Ferranti, A. Malorni, G. Mamone, N. Sannolo and G. Marino, FEBS Lett. ,
1997, 400, 19-24.

E. F. Funai, A. Davidson, S. P. Seligman and T. H. Finlay, Biochemical and
Biophysical Research Communications, 1997, 239, 875-877.

E. Nagababu, S. Ramasamy and J. M. Rifldnd, Nitric Oxide, 2006, 15, 20-29.

J. M. Robinson and J. R. Lancaster, Jr., American Journal of Respiratory Cell
and Molecular Biology, 2005, 32, 257-261.

S. Herold and G. Roeck, Journal of Biological Chemistry, 2003, 278, 6623-
6634.

M. S. Joshi, T. B. Ferguson, Jr., T. H. Han, D. R. Hyduke, J. C. Liao, T.
Rassaf, N. Bryan, M. F eelisch and J. R. Lancaster, Jr., Proceedings of the
National Academy of Sciences of the United States of America, 2002, 99,
10341-10346.

J. Lopez-Lopez, C. Gonzalez, J. Urena and J. Lopez-Barneo, J. Gen. Physiol,
1989, 93,1001-1015.

C. Gonzalez, L. Almaraz, A. Obeso and R. Rigual, Physiol. Rev., 1994, 74,
829-898.

S. R. Chiocchio, M. E. B. Rendo and J. H. Tramezzani, 1981, 237-244.
A. Jackson and c. A. Nurse, J. Neurochem., 1998, 70, 653-662.

W. Jelkmann, Physiol. Rev. , 1992, 72, 449-489.

39

83.

84.

85.

86.

87.

88.
89.

90.

91.

92.

93.

94.

95.

96.
97.

98.

K. U. Eckardt, S. T. Koury, C. C. Tan, S. J. Schuster, B. Kaissling, P. J.
Ratcliffe and A. Kurtz, Kidney Int, 1993, 43, 815-823.

D. Shweiki, A. Itin, D. Soffer and E. Keshet, Nature (London, United
Kingdom), 1992, 359, 843-845.

A. Minchenko, T. Bauer, S. Salceda and J. Caro, Lab. Invest, 1994, 71, 374-
379.

D. Wolfle and K. Jungermann, Eur J Biochem, 1985, 151, 299-303.

C. L. Doria-Medina, D. D. Lund, A. Pasley, A. Sandra and W. I. Sivitz,
American Journal of Physiology, 1993, 265, E454-E464.

K. K. Graven and H. W. Farber, Kidney Int, 1997, 51, 426-437.

D. J. Pinsky, S.-F. Yan, C. Lawson, Y. Naka, J .-X. Chen, E. S. Connolly and
D. M. Stern, Semin. Cell Biol., 1995, 6, 283-294.

S. Ogawa, S. Koga, K. Kuwabara, J. Brett, B. Morrow, S. A. Morris, J. P.
Bilezikian, S. C. Silverstein and D. Stern, American Journal of Physiology,
1992, 262, C546-C554.

N. Suttorp, P. Ehreiser, S. Hippenstiel, M. Fuhrmann, M. Kruell, H. Tenor and
C. Schudt, Lung, 1996, 174, 181-194.

N. Suttorp, S. Hippenstiel, M. Fuhrmann, M. Kruell and T. Podzuweit,
American Journal of Physiology, 1996, 270, C778-C785.

N. Suttorp, U. Weber, T. Welsch and C. Schudt, Journal of Clinical
Investigation, 1993, 91, 1421-1428.

D. J. Pinsky, Y. Naka, N. C. Chowdhury, H. Liao, M. C. Oz, R. E. Michler, E.
Kubaszewski, T. Malinski and D. M. Stern, Proceedings of the National
Academy of Sciences of the United States of America, 1994, 91, 12086-12090.

D. J. Pinsky, M. C. Oz, S. Koga, Z. Taha, M. J. Broekman, A. J. Marcus, H.
Liao, Y. Naka, J. Brett and et.al., Journal of Clinical Investigation, 1994, 93,
2291-2297.

N. A. Berdina and I. M. Rodionov, Pfluegers Arch, 1980, 383, 245-248.

R. C. Schaeffer, Jr., F. Gong and M. S. Bitrick, Jr., American Journal of
Physiology, 1992, 263, L27-L36.

X. Hong, Z.-Y. Yin, X.-B. Sun and Y.-D. Lu, Zhongguo Yingyong Shenglixue
Zazhi, 2000, 16, 41-44.

40

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

S. Ogawa, H. Gerlach, C. Esposito, A. Pasagian-Macaulay, J. Brett and D.
Stern, Journal of Clinical Investigation, 1990, 85, 1090-1098.

I. Ginis, S. J. Mentzer, X. Li and D. V. Faller, J. Immunol, 1995, 155, 802-
810.

J. Wang, M.-x. Zhang, P.-w. Zhuang, Y. Feng and Y.-j. Zhang, Tianjin
Zhongyiyao Daxue Xuebao, 2007, 26, 146-149.

E. Dejana, G. Bazzoni and M. G. Lampugnani, Exp. Cell Res, 1999, 252, 13-
19.

P. L. Hordijk, B. Anthony, F. P. J. Mul, R. Rientsma, L. C. J. M. Oomen and
D. Roos, J. Cell Sci, 1999, 112, 1915-1923.

L. J. Ignarro, G. M. Buga, K. S. Wood, R. E. Byms and G. Chaudhuri,
Proceedings of the National Academy of Sciences of the United States of
America, 1987, 84, 9265-9269.

M. Yanagisawa, H. Kurihara, S. Kimura, Y. Tomobe, M. Kobayashi, Y.
Mitsui, Y. Yazaki, K. Goto and T. Masaki, Nature (London, United Kingdom),
1988, 332, 411-415.

S. Kourembanas, P. A. Marsden, L. P. McQuillan and D. V. Faller, Journal of
Clinical Investigation, 1991, 88, 1054-1057.

A. Encabo, M. Ferret, J. Marin, J. Villamor and G. Balfagon, Gen.
Pharmacol, 1992, 23, 263-267.

S. Kourembanas, L. P. McQuillan, G. K. Leung and D. V. Faller, Journal of
Clinical Investigation, 1993, 92, 99-104.

M. Ema, S. Taya, N. Yokotani, K. Sogawa, Y. Matsuda and Y. Fujii-
Kuriyarna, Proceedings of the National Academy of Sciences of the United
States of America, 1997, 94, 4273-4278.

Y.-Z. Gu, S. M. Moran, J. B. Hogenesch, L. Wartman and C. A. Bradfield,
Gene Expression, 1998, 7, 205-213.

G. L. Wang, B. H. Jiang, E. A. Rue and G. L. Semenza, Proceedings of the
National Academy of Sciences of the United States of America, 1995, 92, 5510-
5514.

M. Ivan, K. Kondo, H. Yang, W. Kim, J. Valiando, M. Ohh, A. Salic, J. M.
Asara, W. S. Lane and W. G. Kaelin, Jr., Science (Washington, DC, U. S.),
2001, 292, 464-468.

41

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.
126.
127.

P. Jaakkola, D. R. Mole, Y.-M. Tian, M. 1. Wilson, J. Gielbert, S. J. Gaskell,
A. von Kriegsheim, H. F. Hebestreit, M. Mukheiji, C. J. Schofield, P. H.
Maxwell, C. W. Pugh and P. J. Ratcliffe, Science (Washington, DC, U. S. ),
2001, 292, 468-472.

D. Lando, D. J. Peet, J. J. Gorman, D. A. Whelan, M. L. Whitelaw and R. K.
Bruick, Genes Dev, 2002, 16, 1466-1471.

L. E. Huang, J. Gu, M. Schau and H. F. Bunn, Proceedings of the National
Academy of Sciences of the United States of America, 1998, 95, 7987-7992.

L. E. Huang, Z. Arany, D. M. Livingston and H. F. Bunn, Journal of
Biological Chemistry, 1996, 271, 32253-32259.

C. W. Pugh, J. F. O'Rourke, M. Nagao, J. M. Gleadle and P. J. Ratcliffe,
Journal of Biological Chemistry, 1997, 272, 11205-11214.

S. Salceda and J. Caro, Journal of Biological Chemistry, 1997, 272, 22642-
22647.

W. G. Kaelin, Nat. Rev. Cancer, 2002, 2, 673-682.

P. H. Maxwell, M. S. Wiesener, G.-W. Chang, S. C. Clifford, E. C. Vaux, M.
E. Cockman, C. C. Wykoff, C. W. Pugh, E. R. Maher and P. J. Ratcliffe,
Nature (London), 1999, 399, 271-276.

B.-H. Jiang, E. Rue, G. L. Wang, R. Roe and G. L. Semenza, Journal of
Biological Chemistry, 1996, 271, 17771-17778.

G. L. Semenza, B.-H. Jiang, S. W. Leung, R. Passantino, J .-P. Concordet, P.
Maire and A. Giallongo, Journal of Biological Chemistry, 1996, 271, 32529-
32537.

C. Peyssonnaux, A. S. Zinkemagel, R. A. Schuepbach, E. Rankin, S. Vaulont,
V. H. Haase, V. Nizet and R. S. Johnson, Journal of Clinical Investigation,
2007, 117, 1926-1932.

B. Nemeth, M. S. Tuttle, J. Powelson, M. B. Vaughn, A. Donovan, D. M.
Ward, T. Ganz and J. Kaplan, Science (Washington, DC, U. S. ), 2004, 306,
2090-2093.

M. R. McCall and B. Frei, Free Radical Biol. Med, 1999, 26, 1034-1053.
B. W. Askew, Toxicology, 2002, 180, 107-119.

T. Akaike, K. Sato, S. Ijiri, Y. Miyamoto, M. Kohno, M. Ando and H. Maeda,
Arch. Biochem. Biophys, 1992, 294, 55-63.

42

128. S. A. Devi, R. Vani, M. V. V. Subramanyarn, S. S. Reddy and K. Jeevaratnam,
Cell Biochem. F unct., 2007, 25, 221-231.

129. Y. Sato, S. Kamo, T. Takahashi and Y. Suzuki, Biochemistry, 1995, 34, 8940-
8949.

43

CHAPTER 2

2.1 MEASURING TWO STIMULI OF ATP RELEASE
SIMULTANEOUSLY

Hypoxia is a condition in which a region of the body, or the body as a whole, is
deprived of a sufficient oxygen supply. Oxygen transport in mammalian species is

carried out by hemoglobin, a metalloprotein connected to the acidic amino-terminal

segment of the band 3 membrane protein of RBCS by weak binding.1 The hemoglobin

molecule itself is comprised of four subunits: two (ii-chains of 141 amino acid residues

. . . . 2 . .
each and two B-chains, each containing 146 reSidues. Each chain carries a heme

group that is held in pockets formed by several helical and non-helical segments. In
its deoxygenated state, hemoglobin is in a (T) tense conformation. Upon oxygenation,

hemoglobin changes to a relaxed (R) conformation. The change in conformations is

responsible for its cooperative effect that ensures uptake and release of 02 over the

. . 3
range of partial 02 pressures between the lungs and tissues.

Related to its role in oxygen transport, Ellsworth et. al. have reported that the

RBC may actually serve as an oxygen sensor.4 One mechanism to explain the RBC’s

internal oxygen sensory is through the release of sub-micromolar amounts of ATP

upon exposure to brief periods of hypoxia.5 It has been shown that ATP has the

ability to stimulate NO production in both cultured endothelial cells and the

pulmonary endothelium?8 Importantly, NO is known to be a potent vasodilator and,

as such, relaxes the smooth muscle cells surrounding the blood vessel enabling an

44

increase in blood flow to organs and tissue. It is in this construct that the RBC 13 not

only a deliverer of oxygen, but is also a determinant of blood flow due to the ability of
hypoxia-induced ATP release to stimulate NO production.

In addition to hypoxia, there exist other methods of stimulating ATP release
from RBCS. For example, it is well-established that mechanical deformation of RBCS
results in the release of ATP. Sprague et. al. have demonstrated that deformation of

the RBC occurs when RBCS traverse porous membranes having diameters similar to
. _ , 9 . .

those of reSistance vessels in vivo. Moreover, it has also been demonstrated in an

indirect manner (through the monitoring of vessel relaxation) that the introduction of

RBCS through the isolated rabbit lung also results in ATP release.10 Finally, by using

microbore tubing or the channels in a microfluidic device to approximate the

dimensions of resistance vessels in vivo, we have demonstrated the ability to measure

deformation-induced ATP release from the RBC.1 1’ 12

While hypoxia and deformation, as well as other methods (e.g.
phannacological),l3’ 14 are known to induce ATP release from the RBC, none of these

stimuli have been examined in combination. Here, quantitative data are reported from

a determination in which RBCS were pumped through a microflow system similar to

. . 15 . .
those studies preViously reported. However, in the studies reported here, the RBCS

were also subjected to a brief period of hypoxia prior to their introduction into the
microbore tubing. Such a protocol enabled the determination of ATP release from

RBCs due to deformation and/or varying levels of hypoxia. Therefore, the work

45

described here is the first attempt to differentiate ATP release from RBCS due to
hypoxia from that due to deformation. This conclusion would prove useful in the
attempt to elicit the pathway resulting in hypoxia-induced ATP release. These studies
should impact such diseases as diabetes where the RBCS do not respond as well to
hypoxic conditions as those RBCS obtained from healthy controls.

However, measuring these effects simultaneously required the development of
a system to determine both ATP release due to hypoxia and deformation. As
previously mentioned, microbore tubing and microfluidic devices have been employed
to mimic resistance vessels in vivo. Fluid dynamics are crucial in measurements as
they replicate the circulation and introduce many variables not seen in a static system,
such as shear stress, the formation of a cell-free layer and pressure changes. Previous
work in the Spence group has resulted in instrumentation that includes flow and
mimics the diameter of vessels in vivo.

Since the early ‘90’s, microfluidic technology has flourished. In microfluidics,
small volumes of samples traverse micron size channels embedded in a chip.
Microfluidic devices, initially, were designed with simple channel layouts; however,

they have developed to allow for much more complicated processes such as sample

. . . . 16, 17 . . . . .
and reagent mixmg, and purification. Microfluidic deVices can be fabricated

from many different substrates such as glass and silicon or polymers, more
specifically, polydimethylsiloxane (PDMS). These devices will be more prevalent in
studies described in chapter 4.

Here, an instrumental setup that mimics resistance vessels in vivo is described

that includes the use of fused silica microbore tubing. A set up described by Edwards

46

et. al. used a temperature-controlled high pressure syringe pump to introduce

18 . . . . . . .
reagants. Luicferin/luCiferase was pumped through fused Silica microbore tubing

with an outside diameter of 365 pm. The capillary tubing ended at a mixing tee with
an internal volume of 29 nL. PSS was pumped through fused silica microbore tubing
also having an inside diameter of 365 pm. The capillary tubing containing the PSS
ended at a 4-port injection valve where RBCS were injected. The RBCS were then
flowed through capillary tubing of varying lengths and diameters (25-75 uM). These
inside diameters were utilized in order to induce a deformation on the RBCS. RBCS
then mixed with luciferin/luciferase in the mixing tee and a chemiluminescence
reaction occurred. The sample was then flowed through another piece of microbore
tubing with an inside diameter of 50 pm. The polyimide coating was burned off the
tubing with a flame leaving a window open for the light to be detected by a
photomultiplier tube (PMT). The light was converted to voltage and measured on a
computer. RBC-derived ATP release decreased with an increase in the capillary’s
inside diameter when the flow rate was held constant at 15 uL/min. This setup not
only displays deformation-induced ATP release, but is an excellent mimic of
physiological blood flow.

Through the use of flow-through instrumentation, we were able to induce
deformation and hypoxia simultaneously on RBCS. The novelty of this system is the
simultaneous measurement of both stimuli. Previous work measured deformation and
hypoxia-induced ATP release in a static system after the stimuli was applied. This

allowed us to determine an additive or non-additive relationship between the two

47

stimuli. This would further suggest if hypoxia and deformation release ATP through

the same signaling pathway and if hypoxia is really a form of deformation.
2.2 EXPERIMENTAL
2.2.1 Collection of RBCS

Male New Zealand White rabbits (2.0-2.5 kg) were anesthetized with ketamine

(8 mg kg], intramuscular injection) and xylazine (1 mg kg], i.m.) followed by
pentobarbital sodium (15 mg kg.1 , intravenous injection). A cannula was placed in
the trachea and the animals were ventilated with room air at 20 breaths min-1 and a

tidal volume of 20 mL kg-l. A catheter was placed into a carotid artery for

administration of heparin and for phlebotomy. After heparin (500 units, i.v.) animals
were exsanguinated. Blood was centrifuged at 500 g at 4 ° C for 10 min. The plasma
and buffy coat were removed for other experiments. RBCS were then resuspended and
washed three times in a physiological salt solution (PSS). The PSS was made by
combining 25 mL of TRIS buffer [prepared by mixing 50.9 g of TRIS in l L of

distilled and deionized water (DDW)] and 25 mL of Ringer’s solution (164.2 g NaCl,

7.0 g KCl, 5.9 g CaC12-2H20 and 2.83 g MgSO4 in 1 L of DDW). After the addition

of 0.50 g of dextrose and 2.50 g of albumin bovine fraction V (fatty-acid free) to the

TRIS-Ringer’s mixture, the entire solution was diluted to 500 mL with DDW and the

48

pH was adjusted to 7.35-7.45. The PSS was then filtered three times using a 0.45 pm

filter.

2.2.2 Preparation of Reagants

Diamide, a known cell stiffening agent, was prepared for incubation with RBCS to
render them less deformable. A 2 mM solution was prepared by dissolving 0.009 g of
diamide in 25 mL of PSS.

Glibenclamide inhibits the cystic fibrosis transmembrane regulator, the last
protein in the pathway for deformation-induced ATP release. It was prepared by
adding 0.049 g of glibenclamide to 2 mL of 0.1 M NaOH. 8 mL of a solution
containing 1 g dextrose and 20 mL DDW was added to the glibenclamide and NaOH
solution, heated to 50 °C and stirred until dissolved, resulting in a 0.1 M solution of
glibenclamide.

A 100 uM stock solution of ATP was prepared by adding 0.0551 g of ATP to
1000 mL of DDW. ATP standards with concentrations ranging between 0 and 1.5 uM
were then prepared in PSS from the stock.

To prepare the luciferin-luciferase mixture involved in the measurement of
ATP by chemiluminescence, 5 mL of DDW were added to a vial containing luciferase
and luciferin. In order to enhance the sensitivity of the assay, 2 mg of luciferin were
added to the vial. All reagants were purchased from Sigma Chemical (St. Louis, MO)
The chemiluminescence reaction of lucierfin/lucierfase with ATP is displayed in

Figure 10.

49

 

Luciferin Adenosine Triphosphate
Firefly
Lucif erase

+ Mg®®

 

Luciferin a; N
02 Adenosme Monophosphate <N ‘ \‘
Luminescence \N / N
G
\ N N O NH2
/' /E>'CK + H20 + co.

('3 s s
Oxyluciferin

Figure 10. The two-step bioluminescence reaction between luciferin/luciferase and
ATP.

50

2.3 METHODS
2.3.1 Measurement of ATP Release

For ATP measurements involving RBCS, all samples were diluted to a 7%
hematocrit in PSS. In order to determine the ATP release due to deformation, the
luciferin-luciferase mixture was placed in a 500 uL syringe. Either the ATP standards
or the RBCS were placed in another 500 uL syringe next to the luciferin-luciferase
mixture. The syringes were loaded on to a syringe pump and were connected to 50 cm
of microbore tubing with an internal diameter of 50 um and an outer diameter of 362

pm. The RBCS or ATP standards were then pumped through the microbore tubing at

. -1 . . . .
a rate of 6.7 uL min . The two separate microbore tubes contaimng either the

luciferin-luciferase mixture or the RBCS/ATP standards were combined at a T-
junction. The now combined contents flowed through a 10 cm section of microbore
tubing having an inside diameter of 75 um, where the resultant chemiluminescence
was detected by a photomultiplier tube (PMT). The PMT was housed in a light-
excluding box to minimize ambient light. The tubing over the PMT had its polyimide
coating removed in order to create a quartz window, through which the

chemiluminescence was detected.

2.3.2 Flow-based Determination of Hypoxic RBCS

51

RBCS were deoxygenated by purging N2 gas through 5 mL of a 7% RBC

sample at 30 kPa for varying times resulting in oxygen saturations from 6.2% to
66.3%. The hypoxic RBCS were then placed into a 500 uL syringe and loaded on to
the syringe pump next to a 500 uL syringe filled with the luciferin-luciferase mixture.
The sample and reagent met at the T-junction and produced a chemiluminescence
reaction, from which the emission which was measured by the PMT. A schematic of

the instrumental setup is displayed in Figure 11.

2.3.3 Non-flow Measurement of Hypoxic RBCS

A setup similar to the flow technique described above was used for studies
involving non-flow techniques. A PMT served as the transducer; however, the sample
container was a plastic cuvette placed over the PMT. Samples were prepared using
addition of RBCS and ATP standards of concentrations 0.25-1.5 uM. To perform the
assay, 100 uL of ATP standard and 100 uL of luciferin/luciferase were pipetted into a
plastic cuvette. Upon mixing of the ATP and luciferin/luciferase solutions, 15 s
elapsed prior to measurement with the PMT setup. This allowed for improved
reproducibility and helped to ensure that all measurements were performed at the same

time interval. A schematic of the instrumental setup is displayed in Figure 11.

Next, 5 mL of 7% RBCS were then deoxygenated with N2 gas at 30 kPa for

varying times. 100 uL of the hypoxic RBCS were pipetted into the cuvette with 100
uL of the luciferin/luciferase solution prior to measurement of the chemiluminescence

by the PMT.

52

as v 50 NZ 5?

@2qu 00» an 0000:0w800v 0003 00mm 503008025820 0080008 2 000200 AER. 0003088288830 080 08093 509
.9300 25 E 000200 mh< 00>tov-8ou0§omow 00308 8 mafia 0.890028 m8? 880% $385-32.“ 05 2 @800 888$
0E. 00% 05 mo 820080200 0: 2 0005. 0080008 02 00893 8 030 000200 nines baa s8.2.0.0800 25 5 800200805
05 00 ESE 0 mafia: @800 Boméo: 06 8 05080200 Bat 98 2:. 01082000. Gorgowbagu mo 0:080:0m .: 0.59m

 

      

 

  

  

 

 

 

 

 

 

 

 

 

  

 

0930.500
0
0mm._0t03_\c_._0t034
momm o\ok
.520. 00mm. {ext i
1% mmmct>w
000 91.722
XOD QEUEOXO “cm:
«2
.52.”.
- 030240

H9

 

couJQEOO xon OEUEOXO 59...

53

2.3.4 Measurement of Oxygen Saturation

In order to construct the oxygen saturation curve, a 7% solution of RBCS was
prepared. A Clark-type oxygen electrode was immersed into the solution with
constant stirring. A Clark-type oxygen electrode has four main components: a teflon
membrane, a platinum cathode, a silver anode and an electrolyte solution (KCl
solution). Oxygen
diffuses through the teflon membrane to the platinum cathode where it is reduced.

The equation for this reduction is displayed below:

02 + 2H20 + 2e‘ 9 H202 + on'
Electrons for the platinum cathode are provided by oxidation of the silver anode after
a -700 mV potential is applied. The current from the electrons reducing the oxygen is

measured and this reading is the indicative of oxygen in solution. The buffer solution

is used as a conductive species in between the two electrodes.

2.4 RESULTS

It has been established that RBCS, due to brief periods of hypoxia, release

4, 5,19

ATP It is also known that RBCS subjected to mechanical deformation release

9 . . . .
ATP; however, to date there has been no attempt to investigate these two stimuli

simultaneously. By mimicking resistance vessels in vivo and inducing both hypoxia

and deformation it could be determined if these stimuli of ATP release were additive.

54

Times for deoxygenation were determined from the oxygen saturation plot as
displayed in Figure 12. The largest decrease in oxygen saturation occurs within the
first 15 s with a decrease in oxygen saturation from 66.3 to 22.3%. Figure 13 shows

the ATP release over the same time span as the oxygenation saturation curve in a non-

flow setup (no flow-induced deformation). Hypoxia was induced with N2 gas at 30

kPa for varying times up to 4 min. The ATP release displayed in Figure 13 is
inversely proportional to the oxygen saturation in Figure 12. As the percent oxygen
saturation decreased, the ATP release from the RBCS due to hypoxia increased. Not
surprisingly, the largest difference in oxygen saturation (between 66.3 and 22.3%)
results in the largest incremental change in ATP release (from 0.107 jg 0.005 to 0.387
: 0.016 uM). Upon cell lysis, intracellular ATP is released from the RBCS, greatly
increasing the chemiluminescent signal. In order to demonstrate that ATP release is
indeed due to hypoxia and deformation, and not cell lysis, glibenclamide was added to
the RBCS prior to exposure to hypoxia. Glibenclamide inhibits the cystic fibrosis
transmembrane regulator (CFTR) protein located in the RBC membrane. Sprague and

co-workers have shown that this protein is involved in the signalling pathway of RBC-

derived ATP release.20 By inhibiting this protein, the ATP release in response to

stimuli should decrease if cell lysis has not occurred. Figure 14 A. shows raw data
obtained from RBCS, hypoxic RBCs and RBCS incubated with 100 uM glibenclamide
for 30 min. Figure 14 B. shows the average release of ATP from the RBCS of n=3
rabbits. In accordance with expected results, the ATP does decrease after incubation
with glibenclamide providing evidence that the measured ATP release is not a result of

cell lysis.

55

.Emm H 000 000m. 0080 =< .0238 mM: .20 00mm 05 80¢ 00w80>0 000 008—006 022.00% 000200080 08 80¢
000000.20008 .30 @3528 2:3 w020 0020 8 000000 05 $220: 008 05 m0 000 :0 0002a 00? @850H0m .«0 00x2 225 <

808002020 00m @280 20800 0mm 05 38 00020 003 000000 00308 0Qb2020 < .0 2 820 00200 0808080008
53> 88 v 00.2 0%— Om .0 00m N2 :23 0030003800 000: 000.8300 00308 80800 .20 >030 08C. .3 0.52...—

20000002 05;

Dew mmm 9m .02 ow? wow 02 mm? 03 m9. om mh ow we on me o

. . p _ h _ _ . P _ . b _ L b _ o

 

or

ON.

on

9.

cm

cm

on

 

 

on

56

 

N .N
O 01
I n A l L

7"
01

T‘
0

1L]

ATP Release (uM)

.0
01

  

 

RBCS 1 5 30 60 120 1 80 240

 

.0
0

seconds

Figure 13. ATP release from RBCS over time in a non-flow System. The initial
ATP release is from RBCs in the absence of hypoxia Upon induction of hypoxia,
the RBC-derived ATP is determined at different time intervals up to and including 4
min. A fresh vial of RBCS was prepared for each measurement. All error bars are i
SEM. Students t-test was performed on data to ensure the means are statistically

different (p<0.05, n=3)

57

Amie .0883 050.000 0080003...
000 08008 08 080000 00 0000 000 00808000 003
0000-0 08208m .EMm H 000 0000 00000 =< .88 0 00¢
002 cm 00 00m~ NZ :23 0000003800 :08 88 cm 00¢
20800—00088 21 2: 823 00008008 20800 0mm
£0 0 080000000 000 30% 2000 05. @808 08 8
00000800008 00 02.8 88 0 00¢ 0%— om 0 00m Nz :83
000000030800 20800 0mm e\e\i 0 80¢ 000200 0.5..
080000000 000 >00w 8w: 02H .81 cm .00 00008000
2208 00 w8>02 w808 00090028 080005 00320
20800 0mm 0x00 0 80¢ 000200 0.30 080000000
000 2003 08 83080 00 ¢2 0:... 80000 ea. 8
003000 0000 08 m0 300880m .m 0808 08 030008
808882 00 0200 00m Nz m0 0.3 on :83 88 0 00¢
000000300020 20800 0mm 000000080: $0. 0 02 00000
000 00:. .81 cm m0 00008020 2208 :0 w8>08 858
00000028 205005 000800 :08 003 20800 00.2.
.00m NZ m0 0A2 cm :83 88 0 00¢ 000000800020
55 023080 8.00 3 80805000
21 2: 0E3 88 Om 00¢ 00008008 20800 0mm
000000080: 0x00 0 080000000 00000 0:228 00:. .81
cm .00 00008000 2208 00 w8>02 @058 00000028
080005 000808 20800 0mm 80000080: $0 0
080000000 00000 800000 05. 020800 0mm 8008.80
80¢ 20080 800008888020 .0. .3 0.880—

mm on mN ON mr or m

 

 

 

 

 

 

 

Wood

. .
00
:5
m

V F

'1'!!!
(D
?

 

 

Amocoommv GEE.

LO

- .- P F - F

 

 

:0 + mom“ X000 l
womm X000 III!
momm I

.000
000.0

iNOd

I 1 r1

.mOd

H.000

 

.000

 

58

As mentioned previously, both hypoxia and deformation were applied to the RBCS to
determine if an additive relationship exists between the two stimuli. In order to

provide both stimuli, RBCS were pumped through microbore tubing having an inside

diameter of 50 um after hypoxia was induced by gas exchange with N2 at 30 kPa for

varying times. As stated above, the window of oxygen saturation studies is in
agreement with the largest incremental change (66.3-22.3%) in ATP release.

The non-flow set up described above was used to measure the effect of RBC
derived ATP release due to hypoxia alone. This ensures that there is no deformation
due to flow acting o the RBCS. Figure 15 shows the results of both hypoxia and

deformation, and hypoxia alone on ATP release. Most notably, at around 25.0%

oxygen saturation (15 s of N2) the total ATP release does not appear to be additive. If

an additive relationship did exist, the ATP release due to deformation and hypoxia
(black bars) would always be larger than the corresponding gray bar (ATP release due
to hypoxia alone). The data in Figure 15 suggests that hypoxia dominates as the main
source of RBC-derived ATP release when oxygen saturation is less than 25%.
Furthermore, the apparent non-additive feature also suggests that hypoxia and
deformation may release ATP through the same mechanism.

Hemoglobin, the main oxygen carrier in vivo, is a component of the RBC

membrane. To examine the possibility that hypoxia induces deformation of the RBCS,

21
the cells were stiffened with diamide [(CH3)2NC(O)N=NC(O)N(CH3)2] and then

exposed to hypoxic conditions for 15 s. The conformational changes that hemoglobin

undergoes from its oxygenated to deoxygenated state may be responsible for hypoxia-

59

 

 

 

 

 

 

« - Flow
0.6 L - Non-flow
E 0.55
3 l
g 0.4? , '
2 1 FL
(D .
D: 0.3 2
E r *
‘ l
0.1 - : !
: l '
0.0 ‘ '-

 

 

 

RBCs 5 1O 15 3O 60
Time (seconds)

Figure 15. The ability of RBCS to release ATP in the presence of two stimuli:
hypoxia and deformation. The black bars represent RBC-derived ATP release
from a 5 mL sample of RBCS with a 7% hematocrit deoxygenated for varying
times at 30 kPa. The sample was then pumped through microbore tubing having
an inside diameter of 50 pm. The gray bar represents RBC-derived ATP release
of a SmL sample with a 7% hematocrit deoxygenated for varying times at 30
kPa. The sample was placed into a cuvette that was placed on top of a PMT and
measured in the absence of flow. All error bars are 1 SEM. Students t-test was
performed on data to ensure the means are statistically different (p<0.05, n=3).

6O

induced ATP release through deformation of the membrane. When incubated with
RBCS, diamide oxidizes glutathione (the most abundant non-enzymatic antioxidant) to
its dimer form. This results in a decrease in antioxidant defenses rendering the
underlying spectrin more susceptible to oxidant attack. Moreover, an oxidant attack
on the spectrin causes disulfide linkages, leaving the cell stiff. Diamide is favored

over other cell stiffeners because it can oxidize the spectrin without affecting the

cytosol of the RBCS.2]

The non-flow setup used for these measurements involving diamide ensures
that hypoxia is the only stimulant of ATP release. The data in Figure 16 shows that
when hypoxic RBCS are incubated for 18 min with 40 uM diamide, ATP release
decreased from 0.225 1 0.040 uM to 0.102 t 0.010 M, thereby suggesting that
hypoxia evokes less ATP release from stiffened RBCS. These results suggest that
hypoxia-induced ATP release from RBCS may actually be a type of deformation on

the RBC membrane.

2.5 DISCUSSION

Mechanical deformation is known to stimulate ATP release from RBCS

through a G-protein mediated pathway.9 When RBCS traverse the circulatory system,

this ATP is released due, in part, to the shear stress imparted to the RBC. It is also

known that hypoxia induces ATP release.5 ATP is a stimulus of the endothelial

synthesis of NO, which diffuses to the smooth muscle cells lining the outer vascular

61

 

 

 

 

 

 

 

 

 

 

 

- Deox RBCS
m Deox RBCS + Diamide
0.20 .
2 .
3
a) 0.15 -
(I)
(U
9
q) .
0: 0.10 4
E 1
< I
0.05 :
0.00 ‘

Figure 16. The ability of hypoxic RBCS to release ATP after incubation with 40
M diamide, a cell stiffener. The black bar represents the ATP release of 5 mL of
3.5% hypoxic RBCS. The gray bar represents 5 mL of 3.5% hypoxic RBCS
incubated for 18 min with 40 pM diamide. Deoxygenation of all RBC samples

occurred for 15 s with N2 gas at 30 kPa. The dosage of diamide was kept constant
at 200 uL. All error bars are : SEM. Students t-test was performed on data to
ensure the means are statistically different (p<0.05, n=3).

62

wall. This initiates a signalling pathway involving cyclic guanine monophosphate

9

(cGMP) that ultimately results in vasodilation. '10 This effect of hypoxia-induced

vasodilation ensures uptake of oxygen and other metabolites to deprived portions of
tissue, for example, post-ischemic tissue.

Here, a non-additive relationship between hypoxia-induced ATP release and
deformation-induced ATP release is established. Due this relationship, ATP released
from both stimuli may follow the same signalling pathway for release from the RBC.
By using a flow-through system that mimics vessels in vivo, both deformation- and
hypoxia-induced ATP release from the RBC sample could be determined
simultaneously. It was determined that at percent oxygen saturations greater than
around 25 .0%, deformation due to flow-induced shear is a primary stimulant of ATP
release from the RBC. However, below 25 .0%, the RBC-derived ATP seems to be
primarily due to hypoxia. Therefore, our results suggest that, not only are the two
factors non-additive, but that they also may have similar mechanisms of release.

As a possible explanation of this result, it is important to note that
haemoglobin is attached to the RBC membrane and undergoes conformational
changes due to differing oxygenation states. It is possible that these changes in
conformational states induce a deformation on the RBC membrane, therefore,
resulting in ATP release. The data in Figure 16 suggest that this is a possibility,
although further studies are requires to confirm such a speculation.

Hypoxia-induced ATP release may be a determinant in such diseases as

hypertension and diabetes. It has been shown that the RBCS of diabetic and

15,
pulmonary hypertension patients release less ATP than those of healthy controls.

63

- 4 _ . . - '
22 2 As prev10usly mentioned, RBC-derived ATP release stimulates NO productlon

ultimately resulting in vasodilation. The decrease in ATP release from RBCS of
patients with these conditions may be a factor in why diabetic patients are twice as

likely to have a stroke and why their post-stroke complications are often increased by

a factor of four.25 As reported by Ellsworth and co-workers, if the RBC is an oxygen

sensor (in addition to delivering oxygen to tissues), then the inability of certain RBCS

to respond to hypoxia may impede RBC flow to tissue deprived of needed oxygen.

64

References

10.

ll.

12.

13.

14.

15.

16.

X. Q. Mi, J. Y. Chen and L. W. Zhou, Journal of Photochemistry and
Photobiology, B: Biology, 2006, 83, 146-150.

M. F. Perutz, A. J. Wilkinson, M. Paoli and G. G. Dodson, Annu Rev Biophys
Biomol Struct 1998, 27, 1-34.

F. B. Jensen, Acta Physiologica Scandinavica, 2004, 182, 215-227.

M. L. Ellsworth, T. Forrester, C. G. Ellis and H. H. Dietrich, Am J Physiol
1995, 269, H2155-2161.

G. R. Bergfeld and T. Forrester, Cardiovascular Research, 1992, 26, 40-47.

J. A. Angus and T. M. Cocks, Proceedings of the Australian Physiological and
Pharmacological Society, 1984, 15, 59-65.

R. G. Bogle, S. B. Coade, S. Moncada, J. D. Pearson and G. E. Mann,
Biochemical and Biophysical Research Communications, 1991, 180, 926-932.

L. J. Ignarro, G. M. Buga, K. S. Wood, R. E. Byms and G. Chaudhuri,
Proceedings of the National Academy of Sciences of the United States of
America, 1987, 84, 9265-9269.

R. S. Sprague, M. L. Ellsworth, A. H. Stephenson and A. J. Lonigro, American
Journal of Physiology, 1996, 271, H2717-H2722.

R. S. Sprague, J. J. Olearczyk, D. M. Spence, A. H. Stephenson, R. W. Sprung
and A. J. Lonigro, American Journal of Physiology, 2003, 285, H693-H700.

S. Carroll Jamie, C.-J. Ku, W. Karunarathne and M. Spence Dana, Anal Chem
2007, 79, 5133-5138.

L. 1. Genes, N. V. Tolan, M. K. Hulvey, R. S. Martin and D. M. Spence, Lab
on a Chip, 2007, 7, 1256-1259.

J. J. Olearczyk, M. L. Ellsworth, A. H. Stephenson, A. J. Lonigro and R. S.
Sprague, Journal of Pharmacology and Experimental Therapeutics, 2004, 309,
1079-1084.

I. Schwaner, R. Seifert and G. Schultz, Eicosanoids, 1992, 5 Suppl, $10-12.

J. Carroll, M. Raththagala, W. Subasinghe, S. Baguzis, T. D. a. Oblak, P. Root
and D. Spence, Molecular BioSystems, 2006, 2, 305-311.

D. Erickson and D. Li, Anal. Chim. Acta, 2004, 507, 11-26.

65

17.

18.

19.

20.

21.

22.

23.

24.

25.

K. Sato, A. Hibara, M. Tokeshi, H. Hisamoto and T. Kitamori, Adv. Drug
Delivery Rev., 2003, 55, 379-391.

J. Edwards, R. Sprung, D. Spence and R. Sprague, Analyst (Cambridge, United
Kingdom), 2001, 126, 1257-1260.

J. Olearczyk Jeffrey, L. Ellsworth Mary, H. Stephenson Alan, J. Lonigro
Andrew and S. Sprague Randy, J Pharmacol Exp T her 2004, 309, 1079-1084.

J. J. Olearczyk, A. H. Stephenson, A. J. Lonigro and R. S. Sprague, Med Sci
Monit 2001, 7, 669-674.

N. S. Kosower, E. M. Kosower, B. Wertheim and W. S. Correa, Biochemical
and Biophysical Research Communications, 1969, 37, 593-596.

J. A. Meyer, J. M. Froelich, G. E. Reid, W. K. A. Karunarathne and D. M.
Spence, Diabetologia 2008, 51, 175-182.

R. S. Sprague, A. H. Stephenson, E. A. Bowles, M. S. Stumpf and A. J.
Lonigro, Diabetes, 2006, 55, 3588-3593.

R. S. Sprague, A. H. Stephenson, M. L. Ellsworth, C. Keller and A. J. Lonigro,
Experimental Biology and Medicine (Maywood, NJ, United States), 2001, 226,
434-439.

J. Weinberger, V. Biscarra, M. K. Weisberg and J. H. Jacobson, Stroke 1983,
14, 709-712.

66

CHAPTER 3

3.1 FINDING THE LINK BETWEEN HYPOXIA AND DEFORMATION

Ischemic tissue in need of oxygen will be replenished by the RBC, the oxygen
carrier in vivo. However, this role for the RBC (oxygen carrier) is not singular. As

mentioned in chapter 1, the RBC is a determinant in blood flow in both the systemic

and pulmonary circulation via its ability to release adenosine triphosphate (ATP).1

Upon binding of ATP to the P2Y purinergic receptor located on the endothelial cell

surface, endothelial nitric oxide synthase (eNOS) is stimulated resulting in the

increase of the vasodilator nitric oxide (NO) and subsequent increased vessel

7

dilation.

The RBC has been shown to release ATP in response to various chemical and
physical stimuli. For example, pharmacological agents such as iloprost have been

shown to increase the release of ATP from the RBC due to activation of the
prostacyclin (IP) receptor found on the RBC.4-6 Other G-protein activators such as
mastoporan have also been shown to stimulate the release of ATP from the RBC.7

Recently, our group has demonstrated that metal-activated C-peptide, a substance co-

released with insulin, also has the ability to stimulate ATP release from the RBC

through an increase in cellular glycolysis.8

67

In addition to the stimuli listed above, it is well-established that hypoxia and

deformation of the RBC will result in the release of ATP from these cells.]’ 5’ 9‘” A

‘ mechanism describing the secretion of ATP from RBCs has been proposed for

deformation-induced release.5 In chapter 2, we reported a method for quantitatively

measuring the effects of hypoxia- and deformation-induced RBC-derived ATP.12

From these studies, it was concluded that ATP release from the RBC due to hypoxia
and deformation were not additive, suggesting a possible synergy between the two
stimuli. Subsequent studies demonstrated that a stiffened RBC released a decreased
amount of ATP upon being subjected to a hypoxic environment.

Ellsworth reported over a decade ago that the RBC may serve as both an

. l . . .
oxygen deliverer and oxygen sensor. Such conclusrons, coupled With our prevrous

work involving hypoxia-induced ATP release from stiffened RBCS shown in the
previous chapter, suggest that hemoglobin (the carrier of oxygen in the RBC) may be a

determinant in the ATP release mechanism. Hemoglobin changes conformational

. 13
states upon deoxygenatron from the R (relaxed) state to the T (tense) state. Due to

its binding to the RBC membrane through the acidic amino-terminal segment of the

band 3 protein, it may be possible that the conformational change of hemoglobin

induces a temporary deformation of the RBC membrane.14 These findings suggest

that RBCS subjected to hypoxia undergo a deformation that results in G-protein

activation and subsequent release of ATP via a pathway previously proposed by

Sprague.5

68

In 1996, Sprague et. al. determined mechanical deformation results in ATP
release from rabbit and human RBCS. Since that discovery, the signaling pathway or
mechanism for the release of ATP through an RBC membrane has been extensively

studied. It has been suggested that ATP leaves the RBC through an “ATP binding

cassette”,15 with the first ATP binding cassette suggested to be CFTR.16

It was first discovered in 1995 that CFTR may play a role in the release of

9

ATP from RBCs;l7 however, its role as either an ion channel regulating ATP effluxl

. . . 20-22 .
or havrng no 1nvolvement in ATP release was incomplete. Regardless,

through the use of varying filter pore sizes to induce deformation on the RBC and the

inhibition of CFTR with pharmacological agents, Sprague et. al. determined that

CFTR was required in the release of deformation-induced ATP release.16

CFTR is reported to be stimulated by an increase in cAMP.23-25 cAMP is

generated by adenylyl cyclase, which is regulated by the heterotrimeric G-proteins.
29 Therefore, a widely accepted signaling pathway was elucidated starting with the

heterotrimeric G protein GS, and ending with the transmembrane protein CFTR.

Through the use of western blot analysis and various pharmacological agents, a

complete signaling pathway for deformation-induced ATP release was reported.5

G-proteins, Gs and G, were identified as the first proteins in the pathway.

This G-protein is activated through a G-protein coupled receptor (GPCR) mechanism.

Upon activation, the GPCR and G-protein couple together activating the G-protein.30

69

Once activated, the G protein regulates adenylyl cyclase, which is responsible for the

conversion of ATP to cAMP. The formed cAMP then binds to two regulatory sites on

. . . . . . . . 31 . .
protein kinase A facrlrtatrng its release from the catalytic subumts. This act1vates

protein kinase A and the subsequent phosphorylation of CFTR. CFTR aids in the
release of ATP through chloride exchange; however, the protein ATP uses to traverse
the membrane is not known. A schematic of this pathway is displayed in Figure 17.

In chapter 2, we defined a non-additive relationship between the ability of
hypoxia and deformation to stimulate ATP release. the two stimuli. Moreover, upon
stiffening of the RBCs with diamide and subsequent exposure to hypoxia, the ATP
release decreased. These data, along with the knowledge of hemoglobin attached to
the
RBC membrane and changing conformations upon exposure to varying oxygen
tensions, led to the hypothesis that hypoxia is a form of deformation on the RBC.
Moreover, this leads to the assumption that hypoxia-induced ATP is released through
the deformation-induced signaling pathway for ATP release.

Here we provide data suggesting hypoxia-induced ATP release from RBCS
follows the same signaling pathway as deformation-induced ATP release. Moreover,
the origin of this claim stems from the idea that hypoxia causes a deformation on the
RBC membrane through the conformational changes hemoglobin undergoes during

varying oxygen tensions.

3.2 EXPERIMENTAL

70

.5352: E 85 £8on oSEpEoEmwa .8605 3 comma—2 fl Air‘s JELHO Bot 32 -5 («o ammo—o.— ofi :msoEH douflbofimonm

03:8 w E #75 moan—bonmmofi 5033 < 3.22 58on $3282“ :2: m2<o .méo 9 ER mo 865280 05 mouth—Sac use
5805 O 05 3 US$38“ 3883 3.298 332% .EmEanooE mam—ago m swab: 5 .8 $0 8338‘“ £023 8382 @2980

E8890 05 8:958 coumgomom .0828 ACE woosufiéonmgomow com 3358 95¢:me on..— mo 03anan .5 «Laura

mh<
@526

 

 

 

 

 

 

«0&0

 

 

o<

 

 

 

 

 

 

 

mhmo

 

 

 

 

 

‘-------------------

71

3.2.1 Collection of RBCS

Male New Zealand White rabbits (2.0-2.5 kg) were anesthetized with ketamine

-l . . . . . - .
(8 mg kg , 1ntramuscular injection) and xylazrne (1 mg kg 1, i.m.) followed by
pentobarbital sodium (15 mg kg.1 , intravenous injection). A cannula was placed in
the trachea and the animals were ventilated with room air at 20 breaths min-1 and a

tidal volume of 20 mL kg'l. A catheter was placed into a carotid artery for

administration of heparin and for phlebotomy. After heparin (500 units, i.v.) animals
were exsanguinated. Blood was centrifuged at 500 g at 4 ° C for 10 min. The plasma
and buffy coat were removed for other experiments. RBCS were then resuspended and
washed three times in a physiological salt solution (PSS). The PSS was made by
combining 25 mL of TRIS buffer [prepared by mixing 50.9 g of TRIS in l L of

distilled and deionized water (DDW)] and 25 mL of Ringer’s solution (164.2 g NaCl,

7.0 g KCl, 5.9 g CaC12-2H20 and 2.83 g MgSO4 in 1 L of DDW). After the addition

of 0.50 g of dextrose and 2.50 g of albumin bovine fraction V (fatty-acid free) to the
TRIS-Ringer’s mixture, the entire solution was diluted to 500 mL with DDW and the
pH was adjusted to 7.35-7.45. The PSS was then filtered three times using a 0.45 pm

filter.

3.2.2 Preparation of Reagants

72

Diamide, a known cell stiffening agent, was prepared for incubation with RBCS to
render them less deformable. A 2 mM solution was prepared by dissolving 0.009 g of
diamide in 25 mL of PSS. Next, 100 uL of the 2 mM diamide was incubated with a
3.5% solution of RBCS. After 15 min of incubation, the RBCS were centrifuged at
500 g at 4 °C for 3 min and added to 5 mL of PSS with 1.5 mL of Oxyrase® that was
incubated for 30 min at 37 °C. 100 uL of the RBC solution and 100 uL of
luciferin/luciferase was added to a cuvette and 15 s was allowed before the

measurement was taken using a PMT housed in a light excluding box.
G protein (Gs) inhibition was performed by incubation of the RBCS with

pertussis toxin (PTX).32 A concentration of 1 ng/mL of PTX was incubated with a

3.5% hematocrit of RBCS for 2 h. After 2 h, the PTX and RBC mixture was
centrifuged at 500 g at 4 °C for 3 min. The supernatant was disposed of and the
remaining solution of RBCS was added to 5 mL of PSS containing 1.5 mL of
Oxyrase® that had incubated for 30 min at 37 °C . 100 uL of the hypoxic RBCS with

PTX were pipetted into the cuvette to determine ATP release as described above.

Inhibition of CFTR was performed by incubating the RBCS with niflumic acid.
Niflumic acid was prepared as a 2 mM solution by dissolving 0.141 g in 25 mL of
DDW. It was further diluted such that the final concentration incubated with the 3.5%
RBCS was 20 1.1M for 30 minutes after which it was centrifuged at 500 g at 4°C for 3
min. The supernatant was disposed of and the RBCS were added to a solution of 5 mL
PSS with 1.5 mL Oxyrase® that was incubating for 30 min at 37 °C. The ATP release

was then measured as described above.

73

H-89, a known protein kinase A inhibitor, was brought to a concentration of
1.8 mM by dissolving 5 mg in 5 mL DDW. 108 uL of H-89 was incubated with a
3.5% solution of RBCS for 2 h. The RBCS were centrifuged at 500 g at 4 °C for 3 min.
The supernatant was disposed of and the RBCS were added to 5 mL of PSS with 1.5
mL of Oxyrase® that was incubated for 30 min at 37 °C. The ATP release was then
measured as described above.

N-(Cis-2-phenyl-cyclopentyl)azacyclotridecan-Z-imine-hydrochloride (MDL
12330A), a known adenylyl cyclase inhibitor, was prepared at a concentration of 400
uM by dissolving 2.5 mg in 66.3 mL of DDW. 190 uL of MDL was incubated with a
3.5% solution of RBCS for 20 min. The RBCs were centrifuged at 500 g at 4 °C for 3
min. The supernatant was disposed of and the RBCS were added to 5 mL of PSS with
1.5 mL of Oxyrase® that was incubated for 30 min at 37 °C. The ATP release was
then measured as described above.

A 100 1.1M stock solution of ATP was prepared by adding 0.0551 g of ATP
to 1000 mL of DDW. ATP standards with concentrations ranging between 0 and 1.5
“M were then prepared in PSS from the stock.

To prepare the luciferin-luciferase mixture used to measure the ATP by
chemiluminescence, 5 mL of DDW were added to a vial containing luciferase and
luciferin. In order to enhance the sensitivity of the assay, 2 mg of luciferin were added

to the vial.

3.3 METHODS

74

3.3.l Measurement of ATP Release

For ATP measurements involving RBCS, all RBC samples were diluted to a
7% hematocrit. In order to determine the ATP release due to deformation, the
luciferin-luciferase mixture was placed in a 500 uL syringe. Either the ATP standards
or the RBCS were placed in another 500 uL syringe next to the luciferin-luciferase
mixture. The syringes were loaded on to a syringe pump and were connected to 50 cm
of microbore tubing with an internal diameter of 50 um and an outer diameter of 362

um. The RBCS or ATP standards were then pumped through the microbore tubing at

. -1 . . . .
a rate of 6.7 uL mm . The two separate microbore tubes contarnmg either the

luciferin-luciferase mixture or the RBCS/ATP standards were combined at a T-
junction. The now combined contents flowed through a 10 cm section of microbore
tubing having an inside diameter of 75 um, where the resultant chemiluminescence
was detected by a photomultiplier tube (PMT). The PMT was housed in a light-
excluding box to minimize ambient light. The tubing over the PMT had its polyimide
coating removed in order to create a quartz window, through which the

chemiluminescence was detected.
3.3.2 Non-flow Measurement of Hypoxic RBCS
A setup similar to the flow technique described above was used for studies

involving non-flow techniques. A PMT served as the transducer; however, the sample

container was a plastic cuvette placed over the PMT. Samples were prepared using

75

addition of RBCS and ATP standards of concentrations 0.25-1.5 uM. To perform the
assay, 100 [AL of ATP standard and 100 uL of luciferin/luciferase were pipetted into a
plastic cuvette. Upon mixing of the ATP and luciferin/luciferase solutions, 15 s
elapsed prior to measurement with the PMT setup. This allowed for improved
reproducibility and helped to ensure that all measurements were performed at the same

time interval.

Next, 5 mL of 7% RBCS were then deoxygenated with N2 gas at 30 kPa for

varying times. 100 uL of the hypoxic RBCS were pipetted into the cuvette with 100
11L of the luciferin/luciferase solution prior to measurement of the chemiluminescence

by the PMT.

3.3.3 Absorption Measurements

Due to the different conformational changes hemoglobin undergoes as a result
of differing oxygenation states, hemoglobin absorbs light at different wavelengths. By
obtaining the absorbance spectrum of hemoglobin, its conformation state can be
determined. Oxyhemoglobin’s maximum absorbtion occurs at 542 nm and 578 nm
while deoxyhemoglobin’s maximum absorption occurs at 570 nm. Spectra were
obtained for 0.07% RBCS, 0.07% deoxygenated RBCS, 0.07% RBCS incubated with

diamide, 0.07% deoxygenated RBCS incubated with diamide.

3.4 RESULTS

76

To further investigate the possibility that hypoxia-induced release of ATP is
actually a form of deformation, the ATP release from normal and stiffened RBCS
after exposure to a hypoxic buffer was measured. The data in Figure 18 shows an
increase of ATP release from RBCS from 15 nM to 492 nM when exposed to a
hypoxic environment; however, when a separate aliquot of these RBCS were
incubated in diamide, the stiffened RBCS decreased back to a value of 243 nM.
While this value is still significantly higher than RBCS not exposed to hypoxia, the
decrease in ATP release for the stiffened RBCS is also significant. The data in
Figure 18 also demonstrates that RBCS stiffened with diamide release ATP in a
dosage dependant manner, suggesting that RBC deformability may be a
determinant in the cell’s ability to release hypoxia-induced ATP.

The possibility exists that the stiffened RBC or diamide itself may prevent
the release of oxygen from the RBC; therefore, absorbance spectra were obtained.
for aliquots of control and stiffened RBCs in normoxic and hypoxic buffers to
determine if the hemoglobin in a stiffened RBC would release oxygen differently
than a healthy, control RBC. Figure l9a-d display spectra obtained for a solution
of 0.07% RBCS, 0.07% RBCs exposed to hypoxia, 0.07% RBCS stiffened with
diamide, and 0.07% RBCS stiffened with diamide and then exposed to hypoxia.
The spectra cleary demonstrate that the hemoglobin is able to release oxygen when
the RBCs are exposed to a hypoxic buffer. The conformational change to the T
state, indicative of oxygen loss, is evident by the single absorbance band at about
560 nm. These spectra also provide evidence that, whether or not an RBC

membrane is in the normal state or stiffened, the exposure to hypoxia results in the

77

 

:0
o:

 

 

.0
01
1.1.;

.0
4:.

 

ATP Release (pM)
.0 .0
N 0)

.0
_L
I

 

 

 

.0
o
|

 

Figure 18. This graph represents the ability of hypoxic RBCS to release ATP
after incubation with 40 pM diamide, a cell stiffener. All hypoxic RBCS were
deoxygenated with a hypoxic buffer consisting of 1.5 mL of Oxyrase enzyme
system and 3.5 mL of PSS. The first bar represents ATP release of 3.5%
normoxic RBCS. The second bar represents the ATP release of hypoxic 3.5%
RBCS. The third bar represents hypoxic 3.5% RBCS incubated for 18 min with
40 M diamide. The fourth bar represents hypoxic 3.5% RBCS incubated with
80 M diamide. All error bars are i SEM. Students t-test was performed on
data to ensure the means are statistically different (p<0.05, n=3).

78

 

 

Absorbance
Absorbance

 

 

 

 

 

 

o
.U

 

 

Absorbance
Absorbance

 

 

 

 

 

Figure 19. (A.) Visible spectra obtained from a 0.07% solution of
normoxic RBCs. (B.) Visible spectra obtained from a 0.07% solution of
hypoxic RBCS. RBCs were made hypoxic by introduction into a hypoxic
buffer solution. (C.) Visible spectra obtained from a 0.07% solution of
normoxic RBCs incubated with 100 uL of 2 mM diamide. (D.) Visible
spectra obtained from a 0.07% solution of hypoxic RBCS incubated with
100 uL of 2 mM diamide. RBCS were made hypoxic by introduction into a
hypoxic buffer solution after incubation with diamide. This data
demonstrates that under hypoxic conditions, hemoglobin changes its
conformation to the T state structure, visible by the difference in absorbance
spectrum. Moreover, RBCS incubated with diamide and made hypoxic still

release the 02 to become deoxygenated; however, because the cell is
rendered less deformable, the conformational change does not induce as
much of a deformation; therefore, there is a decrease in ATP release as seen
in Figure 18.

79

loss of oxygen from the RBC.

In order to determine if the pathway for hypoxia-induced ATP release is
similar to that described for deformation-induced release, studies were performed
to examine the role of G protein and CFTR, two proteins whose activities are
required for ATP release from deformed RBCS. Hypoxic RBCS were incubated
with 1 ng/mL PTX and the ATP release was measured. PTX inhibits G protein
through an ADP ribosylation of the a subunit preventing the interaction of the G-
protein coupled receptor with the G protein. The ATP release decreased from
0.850 : 0.098 to 0.027 is 0.014 uM, shown in Figure 20. This decrease suggests
that G-protein activation is involved in the pathway by which hypoxia induces

ATP release from the RBC.
Niflumic acid inhibits CFTR by plugging the channel pore.33 Hypoxic

RBCS were incubated for 30 min with 20 uM niflumic acid, The ATP release after
incubation with niflumic acid decreased from 0.850 1 0.098 to 0.511 t 0.074 uM
as displayed in Figure 20.

H-89 inhibits protein kinase A, the kinase responsible for the
phosphorylation of CFTR. H-89 inhibits PKA by competitively binding to the
ATP-binding site in a

1:1 ratio.34 Hypoxic RBCS were incubated with 1.8 mM H-89 for 2 h. The ATP

release after incubation with H-89 decreased from 0.850 1 0.098 to 0.065 1 0.033
uM as displayed in Figure 20.
Adenylyl cyclase is an enzyme regulated by G proteins. It catalyzes the

conversion of ATP to cAMP; therefore supplying the cAMP for protein kinase A.

80

_x
O

 

‘4‘;

.0
00

p
O)
J A l - L‘AAJ

.0
4:.

 

ATP Release (uM)

 

.0
N

 

 

.0
o

 

Figure 20. This figure displays the ATP release of the RBCS under the
influence of different reagants. The first bar is the ATP release of 3.5% RBC
under normoxic conditions. The second bar represents 3.5% RBCS
deoxygenated with 1.5 mL Oxyrase® at 37 °C for 30 min. All hypoxic
RBCS were deoxygenated in this manner. The third bar represents hypoxic
RBCS icubated with diamide, a known cell stiffener. The fourth bar
represents hypoxic RBCS incubated with pertussis toxin, a G protein
inhibitor. The fiflh bar represents hypoxic RBCS incubated with niflumic
acid, a CFTR inhibitor. The sixth bar represents hypoxic RBCS incubated
with H-89, a PKA inhibitor. The last bar represents hypoxic RBCS incubated
with MDL, an adenylyl cyclase inhibitor. All error bars are 1 SEM.
Students t-test was performed on all data to ensure the means are statistically
different (p<0.05).

81

MDL inhibits adenylyl cyclase by binding to the hydrophobic region of the plasma

membrane and altering the enzyme, more specifically, interacting with the catalytic

site of adenylyl cyclase.35 Hypoxic RBCS were incubated with 400 uM MDL for

20 min. The ATP release after incubation with MDL decreased from 0.850 3:

0.098 to 0.281 : 0.092 uM as displayed in Figure 20.

3.5 DISCUSSION

Upon oxygenation and deoxygenation, the membrane - bound protein
hemoglobin (the main oxygen carrier in the body) changes conformation. The
structural difference hemoglobin undergoes in varying oxygen saturations may impact

the RBC membrane. Interestingly, it has been reported both deformation and hypoxia

1,5,9]

induce ATP release. - 1 It is believed that hypoxia-induced ATP release is a form

of deformation-induced ATP release. This RBC-derived ATP serves as a source for
NO production. After NO is produced it diffuses to underlying smooth muscle cells
where a pathway occurs resulting in vasodilation and subsequent reperfusion of
oxygen and metabolites. This response is especially important in oxygenating
respiring tissue during stroke or ischemia.

In order to identify hypoxia-induced ATP release as that of deformation,
proteins in the signalling pathway for deformation-induced ATP release were inhibited
and the RBCS were then exposed to hypoxia. Upon inhibition of every RBC protein in
the signalling pathway, the ATP release of hypoxic RBCs decreased significantly.

These data suggest dependence for hypoxia-induced ATP release on the signaling

82

pathway proposed for deformation-induced ATP release. Importantly, these data
suggest hypoxia-induced ATP release is a result of deformation of the RBC membrane
due to different hemoglobin conformations in carrying oxygen tensions.

The results from this study have implications in the role of the RBC in diseases
such as primary pulmonary hypertension and diabetes mellitus. It has been shown that

the RBCS of patients with diabetes are less deformable when compared to those of a

healthy.36 Moreover, it has been shown that the RBCS of patients with diabetes and

. 37, 38 . . .
hypertensron release less ATP. Interestingly, the occurrence of stroke 1n patrents

with diabetes is 1.5-3 times worse, with a greater reperfusion (flow of blood back to

9, 40

respiring tissue) injury.3 If hypoxia is a form of deformation and the RBCS of

patients with diabetes are less deformable and release less ATP, the reason for
increased stroke occurrence (or post-stroke complications) is so how could be due to
the RBC not responding well (in terms of ATP release) in hypoxic conditions. When
RBCS were made less deformable with diamide, less ATP was released in a hypoxic
environment. A decrease in ATP release suggests a resultant decrease in NO
production, a potent vasodilator. Without the dilation of the blood vessels, more
oxygen and metabolites are not able to nourish respiring tissue, therefore, leading to
higher reperfusion injury.

These findings also have an impact on the role of drug discovery for patients

with diabetes. It was shown that drugs such as Trental and Iloprost increase the

release of ATP. Interestingly, Trental rs admmrstered to patrents wrth drabetes to

help with peripheral vascular disease; however, Trental has been shown to induce ATP

83

release only after deformation was induced.4 If hypoxia-induced ATP release is a
form of deformation, the administration of Trental to patients with a high risk of stroke
may lower their chances of ischemic attack and decrease the reperfusion injury,
because of the cell-softening ability of Trental.4

Other substances are also known to improve deformability. C-peptide is a 31

amino acid peptide co-released in equimolar amounts with insulin. In patients with

type 1 diabetes (no insulin production) C-peptide is not bio-available. It has been

previously demonstrated that C-peptide renders an RBC more deformable.41 It has

also been shown C-peptide increases ATP release;8 therefore, in patients with type 1

diabetes, C-peptide therapy may produce more hypoxia-induced ATP by making the
cell more deformable.

Another important drug used for the treatment of sickle cell disease is
hydroxyurea. Sickle cell disease is a life long blood disorder in which the RBCS take

on a sickle or rod-like shape. Interestingly, patients with sickle cell disease are more

. . 42 .
susceptible to stroke and hypertensron. Moreover, it has been suggested

hydroxyurea renders RBCS more deformable.43 By making the RBC less stiff, these

cells may respond to hypoxia more efficiently, therefore, an increase in ATP release
and the vasodilator NO.

Conversely, a stiffened RBC, like those of patients with hypertension, will not
respond to a drug such as mastoparan 7, which activates the G protein independent of

flow-induced shear. If the G protein is activated in a stiffened cell, the deformation

84

will not be as notable; this would mean the levels of ATP would not be increased.
Such therapy would not be useful for patients with stiffened RBCS.

The data presented here demonstrates hypoxia-induced ATP release as a form
of deformation-induced ATP release. Moreover, this finding has implications in
diseases where the RBC has been found to be less deformable. Interestingly, the
occurrence of stroke and post-stroke complications is greater in these diseases. With
drug therapy that targets the RBC and renders it more deformable, hypoxia can have a
more profound effect on the RBC increasing the ATP release and subsequently
dilating vessels through NO production. This would decrease the occurrence of stroke

and aid in reperfusion, limiting the injury of tissue post-ischemic attack.

85

References

10.

ll.
12.

13.

14.

15.

l6.

17.

M. L. Ellsworth, T. Forrester, C. G. Ellis and H. H. Dietrich, Am J Physiol
1995, 269, H2155-2161.

R. G. Bogle, S. B. Coade, S. Moncada, J. D. Pearson and G. E. Mann,
Biochemical and Biophysical Research Communications, 1991, 180, 926-932.

L. J. Ignarro, G. M. Buga, K. S. Wood, R. E. Byms and G. Chaudhuri,
Proceedings of the National Academy of Sciences of the United States of
America, 1987, 84, 9265-9269.

S. Carroll Jamie, C.-J. Ku, W. Karunarathne and M. Spence Dana, Anal Chem
2007, 79, 5133-5138.

J. J. Olearczyk, A. H. Stephenson, A. J. Lonigro and R. S. Sprague, Med Sci
Monit 2001, 7, 669-674.

I. Schwaner, R. Seifert and G. Schultz, Eicosanoids, 1992, 5 Suppl, 810-12.
T. Nakajima, K. Wakamatsu and H. Mukai, Animal Toxins, 2000, 116-126.

J. A. Meyer, J. M. Froelich, G. E. Reid, W. K. A. Karunarathne and D. M.
Spence, Diabetologia 2008, 51, 175-182.

J. Carroll, M. Raththagala, W. Subasinghe, S. Baguzis, T. D. a. Oblak, P. Root
and D. Spence, Molecular BioSystems, 2006, 2, 305-311.

R. S. Sprague, M. L. Ellsworth, A. H. Stephenson and A. J. Lonigro, American
Journal of Physiology, 1996, 271, H2717-H2722.

G. R. Bergfeld and T. Forrester, Cardiovascular Research, 1992, 26, 40-47.
A. Paris and D. M. Spence, 2008, 133, 678-682.

M. F. Perutz, A. J. Wilkinson, M. Paoli and G. G. Dodson, Annu Rev Biophys
Biomol Struct 1998, 27, 1-34.

X. Q. Mi, J. Y. Chen and L. W. Zhou, Journal of Photochemistry and
Photobiology, B: Biology, 2006, 83, 146-150.

G. E. Tusnady, E. Bakos, A. Varadi and B. Sarkadi, FEBS Lett, 1997, 402, 1-
3. '

R. S. Sprague, M. L. Ellsworth, A. H. Stephenson, M. E. Kleinhenz and A. J.
Lonigro, American Journal of Physiology, 1998, 275, H1726-H1732.

Q. Al-Awqati, Science (Washington, D. C. ), 1995, 269, 805-806.

86

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.
28.
29.
30.
31.

32.

33.

34.

A. G. Prat, I. L. Reisin, D. A. Ausiello and H. F. Cantiello, American Journal
of Physiology, 1996, 270, C538-C545.

I. L. Reisin, A. G. Prat, E. H. Abraham, J. F. Amara, R. J. Gregory, D. A.
Ausiello and H. F. Cantiello, Journal of Biological Chemistry, 1994, 269,
20584-20591.

R. Grygorczyk, J. A. Tabcharani and J. W. Hanrahan, J. Membr. Biol., 1996,
151, 139-148.

C. Li, M. Ramjeesingh and C. E. Bear, Journal of Biological Chemistry, 1996,
271, 11623-11626.

M. M. Reddy, P. M. Quinton, C. Haws, J. J. Wine, R. Grygorczyk, J. A.
Tabcharani, J. W. Hanrahan, K. L. Gunderson and R. R. Kopito, Science
(Washington, D. C.), 1996, 271, 1876-1879.

D. Carling, A. Woods, C. Thornton, P. C. F. Cheung, F. C. Smith, M. Ponticos
and S. C. Stein, Biochem. Soc. Trans, 1997, 25, 1224-1228.

A. M. Dulhanty and J. R. Riordan, Biochemistry, 1994, 33, 4072-4079.

J. L. Meinkoth, A. S. Alberts, W. Went, D. Fantozzi, S. S. Taylor, M.
Hagiwara, M. Montrniny and J. R. Ferarnisco, Molecular and Cellular
Biochemistry, 1993, 127/128, 179-186.

R. K. Sunahara, C. W. Dessauer and A. G. Gilman, Annu. Rev. Pharmacol.
Toxicol, 1996, 36, 461-480.

N. Mons and D. M. F. Cooper, Trends Neurosci, 1995, 18, 536-542.

A. G. Gilman, Annu Rev Biochem, 1987, 56, 615-649.

S. Offermanns and M. I. Simon, Cancer Surv., 1996, 27, 177-198.

W. M. Oldharn and H. E. Hamm, Nat. Rev. Mol. Cell Biol. , 2008, 9, 60-71.

H. M. Berman, L. F. Ten Eyck, D. S. Goodsell, N. M. Haste, A. Komev and S.
S. Taylor, Proceedings of the National Academy of Sciences of the United
States of America, 2005, 102, 45-50.

G. S. Kopf and M. J. Woolkalis, Methods in Enzymology, 1991, 195, 257-266.

T. S. Scott-Ward, H. Li, A. Schmidt, Z. Cai and D. N. Sheppard, Molecular
Membrane Biology, 2004, 21, 27-38.

A. Lochner and J. A. Moolman, Cardiovascular Drug Reviews, 2006, 24, 261-
274.

87

35.

36.

37.

38.

39.

40.

41.

42.

43.

G. Guellaen, J. L. Mahu, P. Mavier, P. Berthelot and J. Hanoune, Biochim
Biophys Acta FIELD Full Journal TitlesBiochimica et biophysica acta, 1977,
484, 465-475.

E. J. Diamantopoulos, C. Kittas, D. Charitos, M. Grigoriadou, G. Ifanti and S.
A. Raptis, Hormone and Metabolic Research, 2004, 36, 142-147.

R. S. Sprague, A. H. Stephenson, E. A. Bowles, M. S. Stumpf and A. J.
Lonigro, Diabetes, 2006, 55, 3588-3593.

R. S. Sprague, A. H. Stephenson, M. L. Ellsworth, C. Keller and A. J. Lonigro,
Experimental Biology and Medicine (Maywood, NJ, United States), 2001, 226,
434-439.

J. Weinberger, V. Biscarra, M. K. Weisberg and J. H. Jacobson, Stroke 1983,
14, 709-712.

B. Giorda Carlo, A. Avogaro, M. Maggini, F. Lombardo, E. Mannucci, S.
Turco, S. Alegiani Stefania, R. Raschetti, M. Velussi and E. Ferrannini, Stroke
FIELD Full Journal TitlesStroke; a journal of cerebral circulation, 2007, 38,
l 1 54-1 160.

T. Hach, T. Forst, T. Kunt, K. Ekberg, A. Pfuetzner and J. Wahren, Exp.
Diabetes Res. , 2008, No pp given.

G. J. Kato, M. Hsieh, R. Machado, J. V. I. Taylor, J. Little, J. A. Butman, T.
Lehky, J. Tisdale and M. T. Gladwin, Am. J. Hematol, 2006, 81, 503-510.

G. Athanassiou, A. Moutzouri, A. Kourakli and N. Zoumbos, Clin.
Hemorheol. Microcirc., 2006, 35, 291-295.

88

CHAPTER 4

4.1 CRITICISM OF CURRENT MECHANISMS OF HYPOXIC
VASODILATION

Chapter 1 described two conflicting theories that exist to explain the

mechanism of hypoxic vasodilation. The first theory orginates from Gladwin et. al.

. . . . 1 .
and describes the hemoglobin molecule as a nitrite reductase. Under this construct,

the hemoglobin molecule reduces nitrite, a bioavailable molecule (~290 nM in RBCS),

to NO. In support of this theory, Nagababu and coworkers reported an NO formation

when nitrite was added to deoxyhemoglobin and purged with argon; however,2 it

could not be deduced whether the NO formation was due to nitrite reduction. The
authors believed that the NO has a weak complex with ferriheme and purging with
argon displaced the NO, allowing it to be measured by chemiluminescence.
Furthermore, Nagababu states that Gladwin’s theory does not account for the high

scavenging of NO by hemoglobin. NO has a lifetime of 065-15 3 in the presence of

hemoglobin;3 therefore, the mechanism of NO leaving the RBC from ferriheme

predicts it would merely re-attach to another deoxyheme.
There have been other studies suggesting a role for nitrite. Upon injection of

nitrite into human forearm, blood flow increased according to a study performed by

4 . . . . . .
Cosby et. al. A problem arises in the concentrations of nitrite used and if these

concentrations are physiologically relevant. When concentrations used were 200 nM,
Cosby found nitrite concentrations in the venous blood rose from 176 nM to 2.6 nM.

The values found in venous blood are physiologically relevant; however, the

89

concentration at the site of injection is more concentrated as nitrite becomes diluted

when combined with blood. Another study performed using the same parameters by

5 .
Lauer and coworkers dld not observe the same effect; however, they measured blood

velocity 1 min after injection while Cosby and coworkers waited 5 min after injection.
One major problem with the idea of nitrite reduction to NO from hemoglobin
is the amount of NO that is actually released (not scavenged) and its ability to induce

vasodilation. It has been reported by Cosby et. al., the concentration of NO release

from RBCS is 47 pM/s or (10-3- 10.2 N0 molecules rim-2 - 3-1).4 It has also been
reported the amount of NO released by the endothelium is on the order of 103 — 104

N0 molecules um.2 - 3'1. 6 These data suggest the NO released from RBCS, due to

reduction from hemoglobin, is not sufficient enough to elucidate vasodilation.
Moreover, along with being too small of a quantity, the NO released from RBCS must
traverse the cell-free layer present in vessels, the endothelium, the sub-endothelium
and make it to the underlying smooth muscle cells where it can elicit vasodilation.
Physiologically, this seems highly improbable.

In addition to conversion of nitrite to NO, other theories exist involving the
role of RBC and NO release. In 1996, Stamler and co-workers suggested the role of

SNO-Hb (S-nitrosylated hemoglobin) in regulating vascular tone under conditions of

hypoxia.7 In short, hemoglobin uptakes NO in the lung along with 02 and dispenses it

in areas of low oxygen saturation. The NO is conserved on a BCys-93 residue and the
NO is transnitrosated to the anion exchanger 1 protein (AE1) and then to glutathione

or albumin for release from the RBC. It then diffuses to the smooth muscle cell where

90

it can bind to soluble guanylate cyclase and induce vasodilation. While SNO-Hb has

. . . . . 7-10 .
been detected in concentratrons of 2.5 uM 1n the human crrculatron, there exrst

many disputes with the physiological feasibility of this model.

Based on the widely accepted kinetics of NO, it has been established that NO

11,12

will react with dioxygen to form nitrate, and that it favors this reaction. In

testing this theory, humans that inhaled NO exhibited a blood concentration of ~80

uM methemoglobin (a form of hemoglobin in which Fe is in the Fe3+ state not the
Fe2+, therefore, cannot bind oxygen) and nitrate, but only 1 uM iron-nitrosyl-

hemoglobin and almost undetectable levels of SNO-Hb.13 This is inconsistent with
studies performed by Stamler and coworkers, which contested NO favored iron-
. l4
nitrosyl heme and SNO-Hb.
One major aspect of the SNO-Hb paradigm involves a role for glutathione
(GSH) in the vasodilation effects of SNO-Hb potentiated by hypoxialo’ 15 This

current model is under dispute as recent research utilizing physiologically relevant
concentrations of GSH displays an independence on oxygen tensions. In the presence

of 1 mM GSH, SNO-Hb induced aortic ring relaxation under normal and hypoxic

.. 16-19 . . . . .
conditions. This increase observed in vessel relaxation under hypoxrc

conditions10 may be due to an increase in NO donors induced by hypoxia rather than
the conformational changes trans- nitrosylated hemoglobin undergoes to release the

N 0.20-23

91

Finally, it has been previously mentioned that there exists a need for

endothelial nitric oxide synthase (eNOS) in the production of NO. Upon injection

with an NO synthase inhibitor, blood flow in human forearm decreased by 25-30%.24-

These data suggest a dependence of vasodilation on an NO synthase. In order for

vasodilation to occur under Stamler’s theory, the NO from SNO-Hb must be released
and diffuse through the RBCS, the endothelium and smooth muscle cells to elicit
vasodilation (because they contest it is not eNOS-mediated).

These two theories are very complex in their explanations and extensive in
their reasoning; however, they still lack depth in many areas. For example, the trans-

nitrosylation reported by Stamler et. al. requires the removal of one electron, although

7,

this mechanism has yet to be shown. 8 In addition, the diffusion of NO through the

RBCS, the endothelium and smooth muscle cells does not appear physiologically
likely. NO is a reactive species and, as such, would be scavenged before this could
occur. The lack of clarity in these areas suggests that another mechanism, such as,
ATP-induced vasodilation, may be feasible. Through the use of microfluidic
technology and fluorescence microscopy, combined with standard fluorescence
measurements, a clear pathway has been revealed involving hypoxia-induced ATP
release and subsequent endothelium-derived NO production.

In 2007, a microfluidic array was constructed that allowed for the interaction

of multiple cell types, more specifically, between endothelial cells and red blood

cells.30 The array was fabricated with two individual PDMS molds irreversibly sealed

together. One PDMS mold had 18 inlet holes punched through, 6 lines of 3 holes

92

each. A polycarbonate membrane available with varying pore sizes was situated
underneath this layer. The second layer of PDMS was placed on the other side of the
polycarbonate membrane so there were two layers of PDMS with a polycarbonate
membrane embedded between them. The bottom PDMS layer had channels fabricated
within and waste ports punched through. RBC flow was introduced into these
channels. Bovine pulmonary arterial endothelial cells (bPAECS) were cultured on the
polycarbonate membrane in the inlet holes punched through the top layer of PDMS.
Flow was introduced with a displacement syringe pump through Tygon tubing. All
channels had a width of 150 um and a depth of 100 pm. 7% RBCs incubated with
iloprost, a pharmacological agent that stimulates ATP release, were flowed in the
bottom channels at a flow rate of 1.0 uL/min. ATP released from the RBCS diffused
through the polycarbonate membrane and stimulated NO production from the
endothelial cells cultured on the same membrane. Endothelial cells were pre-
incubated with DAF-FM DA, an intracellular fluorescence probe for the detection of
NO. Fluorescence intensity of the bPAECs increased from 16.5 i 1.0 intensity units
to 23.0 i 3.5. This increase in fluorescence intensity can be attributed to an increase
in NO production by the bPAECs in response to the increase in RBC-derived, iloprost-
induced ATP release. While the fabrication of a vascular endothelium array is a novel
finding, the use of this device in drug efficacy proves it all the more innovative. The
contributions to a blood vessel mimic are indispensable as multiple cell types can be

monitored simultaneously and with the attribution of flow.

4.2 EXPERIMENTAL

93

4.2.1 Collection of RBCS

Male New Zealand White rabbits (2.0-2.5 kg) were anesthetized with ketamine

-l . . . . . - .
(8 mg kg , intramuscular injection) and xylazrne (1 mg kg 1, i.m.) followed by
pentobarbital sodium (15 mg kg-1 , intravenous injection). A cannula was placed in
the trachea and the animals were ventilated with room air at 20 breaths min.1 and a

tidal volume of 20 mL kg-l. A catheter was placed into a carotid artery for

administration of heparin and for phlebotomy. After heparin (500 units, i.v.) animals
were exsanguinated. Blood was centrifuged at 500 g at 4 ° C for 10 min. The plasma
and buffy coat were removed for other experiments. RBCS were then resuspended and
washed three times in a physiological salt solution (PSS). The PSS was made by
combining 25 mL of TRIS buffer [prepared by mixing 50.9 g of TRIS in l L of

distilled and deionized water (DDW)] and 25 mL of Ringer’s solution (164.2 g NaCl,

7.0 g KCl, 5.9 g CaClz-ZHZO and 2.83 g MgSO4 in l L of DDW). After the addition

of 0.50 g of dextrose and 2.50 g of albumin bovine fraction V (fatty-acid free) to the
TRIS-Ringer’s mixture, the entire solution was diluted to 500 mL with DDW and the
pH was adjusted to 7.3 5-7.4S. The PSS was then filtered three times using a 0.45 pm

filter.

4.2.2 Preparation of Reagants

94

Diamide, a known cell stiffening agent, was prepared for incubation with RBCS to
render them less deformable. A 2 mM solution was prepared by dissolving 0.009 g of
diamide in 25 mL of PSS. Next, 250 uL of this 2 mM diamide was incubated with a
3.5% solution of RBCS. After 15 min of incubation, the RBCs were centrifuged at
500 g at 4 °C for 3 min and added to 5 mL of PSS containing 1.5 mL of Oxyrase® that
was incubated for 30 min at 37 °C. The RBCS were then loaded into a gas-tight

syringe and placed onto a syringe pump.

Inhibition of CFTR was performed by incubating the RBCS with niflumic acid.
Niflumic acid was brought to a concentration of 2 mM by dissolving 0.141 g in 25 mL
of DDW. It was further diluted so the final concentration incubated with 3.5% RBCs
was 20 M for 30 minutes and then centrifuged at 500 g at 4°C for 3 min. The
supernatant was disposed of and the RBCS were added to a solution of 5 mL PSS with
1.5 mL Oxyrase® that was incubating for 30 min at 37 °C. The ATP release was then

measured as described above.

10 mg of spermine NONOate was diluted in 0.01 M NaOH to a concentration
of 380 nM. 10 uL of this solution was then added to 1 mL of phosphate buffered
saline (PBS). Varying volumes of this solution were then added to Hank’s Balanced
Salt Solution (HBSS) to bring to a final volume of 1 mL for each standard. Standards

were then incubated at 37 °C for 30 min to initiate the release of NO.

4.3 METHODS

4.3.1 Fluorometric Determination of NO

95

After incubation of standards, 100 uL of 10 uM 4-Amino-5-methylamino-2',7'-
difluorofluorescein (DAF-FM) was added to each standard and incubated at 37 °C for
another 15 min. 700 uL of each standard were then placed into a quartz cuvette and a
standard curve was generated on a fluorometer with excitation at 485 nm and
emmission at 510 nm. For NO measurements involving RBCS, the RBC samples were
prepared as a hematocrit of 3.5%. Normoxic RBCS were centrifuged at 500 g at 4 °C
for 3 min. 900 uL of supernatant were removed and incubated with 100 uL of 10 uM
DAF-FM. 700 uL of the sample were then placed into a quartz cuvette and the NO
was measured with the same parameters used for standardization. Deoxygenated

RBCS were made hypoxic by incubation with a hypoxic buffer that contained 1.5 mL

Oxyrase® enzyme system and 3.5 mL PSS. They were then incubated at 37 °C for 30

min. The hypoxic RBCS were centrifuged at 500 x g at 4 °C for 3 min. 900 uL of
supernatant were removed and incubated with 100 uL of DAF-FM. 700 uL of the
sample were then placed into a quartz cuvette and the NO was measured as described

above.

4.3.2 Microfluidic Device Preparation and Measurement of NO

A six-channel microfludic array was fabriated using poly(dimethylsiloxane)
(PDMS) to incorporate both flowing RBCS and cultured endothelial cells into the
same device. Two individual layers of PDMS were thermocured around a track-

etched polycarbonate membrane with pore diameters of 1.0 pm as previously

described.30 Briefly, each individual layer was obtained by baking 10 g of a degassed

96

20:1 elastomer base to curing agent mixture at 37 °C for 20 min on two seperate
silicon wafer masters, one comprised of 6 parallel channels and one free from any
raised features. The channels are approximately 200 pm in width by 100 mm high and
3 cm long and obtained using standard photolithographic techniques. A second 10 g
degassed 5:1 mixture was poured ontop of each wafer and baked for an additional 20
min before being removed. At which time, inlet holes and waste ports were punched

though the channel layer using a 20 gague luer stub adapter (Becton Dickinson, NJ)

and a 1/g” hole punch, respectively. An array of 1/3” wells, 18 total, were punched

through the blank layer of PDMS and after aligning the two layers, they were
thermocured around the polycarbonate membrane by baking for an additional 30 min
at 37 °C. A schematic of the production of a microfluidic device is displayed in Figure

21 along with a schematic of a microfluidic device displayed in Figure 22.

After completion of the microfluidic array, 10 uL of a 100 ug/mL solution of
bovine plasma fibronectin (Invitrogen, CA) was pipetted into each well and incubated
for 30 min then placed under UV light for an additional 15 min. before removing

remaining liquid. Next, 10 uL of a bovine pulmonary artery endothelial cell (bPAEC)

solution (averaging 2.16 x 107 cells/mL) were pipetted into each well, then incubated
at 5% C02 and 37 °C for 1 hr before removing and replacing the endothelial cell

media (ECM). Endothelial cells were then allowed to incubate at 5% C02 and 37 °C

for approximately 4 hrs before removing the ECM and replacing it with 10 nL of 5

mM L-arginine (4.4 mg in 5 mL phosphate-bufered (PBS) saline solution) and

incubated for 1 hr at 5% C02 and 37 °C. The solution was replaced with 10 ML of a

97

dBm $2 25
Home @PQEwQ mm 8:35 @onmam 05 mo nafimouoan < .Hofiomop meet?— mEDm oz: 05 Row bflmugoE 8 890 E 3558 cm
How @833 @5 @oEano can @203 £15 mEQm @5 235808 Smashing—om $3585 .23 mEDnc weeks ooh: =< d8 no
@829 E 3:85 @onoqsm mac—o: 05 5g 8?; mEnE 05 @5 £030 05 £3» ~32 mEQm 05 ~25 @083 :05 mm 285808
8338033 < .nmsoht @28an one 33: #3 M: @5 H838 0% Set @2682 83 mm Hobs: wEDm ma A838 25%
05 com .08: we Esofim 05mm 05 com @88: @5 63.3: 205.30 on at? £338 @808 a S>o @ocsoa 58. E wEDm .8935
05 80% @2682 55 fl mEQm 2:. .ontofibom @5 0:5 8 mEQm 05 How $358 on 8% @833 8a mzflm @5 H885
22. @0ch $0585 0 63, ©8925 Home? nooEm a $5 @258 mm mEDm Jmhm 58.3 22338:: a .«o .8326on AN PSME

83% 22:282.: BoEEoo
. $533 mamEEoE
@2QO 29232: 2209323

      
  

All

‘lll'll

    
        

  

.3325 5:;
35:3 :82; xcg
9w m=o>> .m: m um>m_@mm_.‘_,_:%m xcgm

 

Eot uo>oEe

2.5.2. £3 + £55
21335 . ,.

 

@230
2 229.
‘I . ‘IJIIIIIII
.5.me ../ 22:35
so: 2:852 ./ «.33..
w E Q n— 2; 8285 /:t...é..H.m. w m a Lg” “wumaaofi
23 +
amuse; wzoa a _>_ ewing .3me

microfluidic
device

 

Figure 22. Schematic of a microfluidic device well with
endothelial cells. Two layers of PDMS were irreversibly sealed
with a polycarbonate membrane between. One layer of PDMS
contained channels while the other had 18 wells punched through.
Endothelial cells were cultured into the wells on the
polycarbonate membrane. RBC solutions were flowed through
the channels underlying the wells. Hypoxia-induced ATP release
stimulated NO production in the endothelial cells cultured in the
wells. The endothelial cells were pre-incubated with DAF-FM
DA; therefore, any NO produced would be measured
intracellularly with the fluorescence probe.

99

100 uM 4-amino-5-methylamino-2’,7’-difluorofluorescein diacetate (DAF-FM DA)

soltuion in PBS and allowed to incubate for 1 hr at 5% C02 and 37 °C. The DAF-FM

DA was removed and the bPAECs were washed with PBS once, then 10 uL PBS was

pipetted into each well.

Each RBC sample was placed in a separate 500 uL gas-tight syringe and a
displacement syringe pump was used to pump each solution for 30 min at a rate of 1.0
uL/min through the underlying microfluidic channels. Fluorescence intensities
relating to the intracellular NO producion were imaged using an Olympus MVX
fluorescence macro stereomicroscope (Center Valley, PA), and each well was

analyzed for pixel intensities. The MVX was fitted with a ET GFP Sputtered Filter Set

(Chroma, VT) having Xmax excitation and emission wavelengths of 450 and 525 nm,

respectively.

4.4 RESULTS

In order to clarify the role of the RBC as an NO donor, RBCS were made
hypoxic and NO was measured spectrofluorometrically with DAF-FM as the
probe. A standard curve of NO standards was generated and is displayed in Figure
23. The NO released from hypoxic 7% RBCS is displayed in Figure 24.

It was demonstrated in chapters 2 and 3 the RBC releases ATP in reponse to
hypoxia. Moreover, it has recently been discovered the eNOS is expressed on

RBCs; therefore, the release of ATP from hypoxic RBCS could activate eNOS on

100

.02 ea :25 mo 8288...
8588035 .m .8: 3m ”Swan—oi; c2835 .8: mwv ”Em—H8022,» noumzoxm 85805808 8 8ch EE 2 8% $3983

8: .33 @8335 we; EmJFd 21 3 .8eOZOZ 65::on Sod o@mE m@8@:8m OZ com 956 @S@=8m .< .MN ouswmm

 

 

 

€555.08, E52.
B So 8m 8m omv 8e 8e 8v 8m 0
p p p . _ p . . p — . p - . Pip L p l % o P _ _ _ §.m
-N w
. r... m+mrm
a m
. o
.4 m
H M34 0 w E.”
m
m a
H m m .
u w e
. u
. B. m .
H «M - 903”
flow a
. N. 058

 

 

 

 

 

 

 

101

 

2500.

1500

10005

NO Release (nM)

 

 

 

Figure 24. This figure displays the NO release from 7% RBCS under normoxic
and hypoxic conditions. 10 nM DAF-FM was incubated with the supernatant of
the RBC samples for 15 min prior to measurement. The increase in fluorescence
intensity is indicative of an increase in NO release from RBCS under hypoxic
conditions. Excitation wavelength: 485 nm Emission wavelength: 510 um. All
error bars are : SEM. Students t-test was performed on data to ensure the means
are statistically different (p<0.05, n=5).

102

other RBCS in the blood stream resulting in NO production from the RBC. Figure
25 displays hypoxic RBCS incubated with L-NAME, an eNOS inhibitor. Upon
inhibition of eNOS, NO production is not statistically different from hypoxic
RBCS. Moreover, incubation with various inhibitors of hypoxia-induced ATP
release, such as diamide and niflumic acid, did not signigicantly alter NO release
when compared to hypoxic RBCS. These data suggest an independence of NO
release from RBC eNOS and further displays NO is simply released by the RBC
rather than produced due to stimulation by ATP.

One of the main objectives of this work was to demonstrate that hypoxia-
induced release of ATP was capable of stimulating endothelium-derived NO. A

microfluidic device fabricated from soft-lithographic technology, and similar in

design to that previously reported by our group,30 was employed to investigate the

intercellular communication between the RBCS and an immobilized endothelium.
The images and summarized data in Figures 26 and 27 indicate that endothelium-
derived NO is increased in the presence of hypoxic RBCS. Specifically, a 3.5%
RBC sample incubated in hypoxic buffer stimulated greater than a 2-fold increase
in fluorescence intensity (which is a measure of intracellular NO production within
the endothelial cells) compared to RBCS alone. The fluorescence intensity
resulting from the RBCS hypoxic environment is decreased by 28.6% and 40.4%
when RBCS are first incubated with 100 uM diamide or 20 uM niflumic acid prior
to introduction to the hypoxic buffer. This trend was dose dependent as the higher
concentrations of diamide and niflumic acid resulted in a greater decrease in

fluorecence intensity. Illustrating that the intracellular endothelial NO is

103

 

llljl

Alllll

NO Release (nM)
0 see 3% E g

In.

1“

 

Allll

 

 

 

Q96 gfyxdéétg’g f ebb
M «of of

O

le\°

Figure 25. Measured NO release of 7% RBCS in normoxic and hypoxic
conditions. Upon incubation with diamide, a cell stiffener, the NO release from
RBCs was measured and found significantly similar to hypoxic RBCS alone. It
has been established RBCS possess eNOS. Upon incubation with L-NAME, an
eNOS inhibitor, the NO release did not decrease. The same trend was observed
with niflumic acid, a CFTR inhibitor. These data suggest RBC eNOS does not
play a role in the production of NO from RBCS. 10 M DAF-FM was incubated
with the supernatant of the RBCS samples for 15 min prior to measurement.
Excitation: 485 nm Emission: 510 nm. All error bars are i SEM. Students t-test
was performed on data to ensure the means are statistically different (p<0.05,
n=3).

104

3.5% RBCS

   
  
   

Deox RBCS + IOOuM
Diamid

Deox RBCS + 200uM
Diamid-

Deox RBCS + 20uM
Niflumic acid

Deox RBCS + 40uM
Niflumic acid

Figure 26. Fluorescence intensity of NO produced from bPAECs in response to
hypoxia-induced ATP release. A. represents 3.5% normoxic RBCS. B. represents
hypoxic 3.5% RBCS. C. represents hypoxic 3.5% RBCS incubated with 100 uM
diamide, a cell stiffener. D. represents hypoxic 3.5% RBCS incubated with 200
uM diamide. E. represents hypoxic 3.5% RBCS incubated with 20 uM niflumic
acid, a CFTR inhibitor. F. represents hypoxic 3.5% RBCS incubated with 40 uM
niflumic acid. These data suggest the dependence of endothelial-derived NO
production on hypoxia-induced ATP release. The NO released from the
endothelial cells can then difi‘use to the underlying smooth muscle cells and elicit
vasodilation. 100 uM DAF -F M DA was incubated with the endothelial cells for 1
h. Data represents an n=3. All error bars are : SEM. Students t-test was
performed on data to ensure the means are statistically different.

105

 

N
01

:"=3 *1 *p<0.001
1 p<0.05

_u _L N
O 01 O
n I I . n A I . l A I

P
01

 

 

Normalized Fluorescence Intensity

0.0 3

 

Figure 27. Graphical representation of the data in Figure 6. Endothelium-
derived NO increased after upon stimulation of hypoxia-induced ATP release.
The addition of diamide, a cell stiffener, decreased the measurable NO is a dose
dependant manner. Furthermore, the addition of niflumic acid, a CFT R
inhibitor, decreased the measurable NO, also in a dose dependant manner. All
error bars are i SEM. Students t-test was performed on data to ensure the means
are statistically different.

106

dependent upon the RBC-derived ATP release, which is suggested here to be a
function of the hypoxic environment. Moreover, the NO released from RBCS was
not eNOS-mediated as NO concentration did not decrease after inhibition of eNOS

with L-NAME.
4.5 DISCUSSION

The topic of hypoxic vasodilation has been disputed for many years. Two

. . . . . . . . 21
main, conflicting theories involve hemoglobin as a mtrlte reductase or the

nitrosylation of NO to become SNO-Hb.7 Both theories present valid reasoning for

vasodilation in hypoxic conditions; however, both theories have yet to describe a full

. . . . l9
mechanlsm. For example, Stamler presents data usmg an aortic ring assay. The

aortic ring assay has been used to study vessel tone for many years. In 1998, Louis
Ignarro won the Nobel Prize in medicine or physiology for his finding of NO as the
EDRF. The aortic ring assay led to this discovery. It consists of a segment of a aortic
ring mounted on a stirrup housed in tissue baths filled with a PSS. Reagents are added
to the PSS containing the vessel and changes in isometric tension are measured with a
pressure transducer and recorded with a polygraph. In many studies, the vessel can be
modified to elude a certain effect of a reagant, for example, in Stamler’s study the
vessel was denuded of an endothelimn and NO was pumped through the vessel to
measure dilation and prove the endothelium is unnecessary in eliciting a dilatory
response. However, in the absence of an endothelium the NO would have less

interference when diffusing to the smooth muscle cells; therefore, an even greater

107

dilation would occur because all NO would be available for the smooth muscle cells to
use for relaxation. In a sense, trying to disprove the theory of ATP-stimulated, NO-
mediated hypoxic vasodilation, it was actually further concluded. Furthermore, the

vessel is in a bath of PSS in which reagents are added. However, ATP released in the

lumen of a vessel activates ng, which stimulates eNOS and ends in relaxation;

however, ATP applied to the outside of a vessel activates P2x which results in

vasoconstriction. Such sensitivities to analytes and the location of their application

must be carefully monitored in an aortic ring assay.

The data presented here suggests that ATP released from hypoxic RBCS
stimulates NO production in endothelial cells. Through the use of a microfluidic
array, endothelial cells were cultured on wells on polycarbonate membrane. 3.5%

hypoxic RBCS were flowed in underlying channels. The ATP released from the RBCS

due to hypoxia stimulated the P2), receptor on the endothelial cells and subsequent NO

production. The NO production was decreased in the presence of pharmacological
agents that have been shown to decrease hypoxia-induced ATP release. This suggests
a dependence of endothelial-derived NO on hypoxia-induced ATP release. As
previously shown in Figure 24, RBCS release NO under hypoxic conditions, therefore,
the NO released from the hypoxic RBCS pumped under the endothelial cells may have
diffused through the endothelium, reacted with DAF-FM DA and was measured along
with NO produced by the endothelium. This may be the case; however, we have also

demonstrated inhibitors of hypoxia-induced ATP release (diamide, niflumic acid) do

108

not affect NO release from RBCs; however, they do affect the concentration of

measurable NO in the endothelial cells.

The results shown here in chapter 4 do not validate the method by which RBCS
release NO. It does, however, suggest another pathway for hypoxic vasodilation,
specifically hypoxia-induced ATP release and subsequent production of NO. This
also elucidates a pathway for hypoxic vasodilation as the NO produced from the
endothelial cells then diffuses luminally and ablurninally. Abluminally, it relaxes
smooth muscle cells leading to vessel dilation. The establishment of a mechanism for
hypoxia-induced vasodilation may help in the development of therapeutic agents for
hypoxia-associated diseases such as sickle cell anemia, stroke and pulmonary

hypertension, as discussed in the previous chapter.

109

References

10.

ll.
12.

13.

J. H. Crawford, T. S. Isbell, Z. Huang, S. Shiva, B. K. Chacko, A. N.
Schechter, V. M. Darley-Usmar, J. D. Kerby, J. D. Lang, Jr., D. Kraus, C. Ho,
M. T. Gladwin and R. P. Patel, Blood, 2006, 107, 566-574.

E. Nagababu, S. Ramasamy, D. R. Abemethy and J. M. Rifkind, Journal of
Biological Chemistry, 2003, 278, 46349-46356.

V. S. Shanna, T. G. Traylor, R. Gardiner and H. Mizukami, Biochemistry,
1987, 26, 3837-3843.

K. Cosby, K. S. Partovi, J. H. Crawford, R. P. Patel, C. D. Reiter, S. Martyr, B.
K. Yang, M. A. Waclawiw, G. Zalos, X. Xu, K. T. Huang, H. Shields, D. B.
Kim-Shapiro, A. N. Schechter, R. O. Cannon and M. T. Gladwin, Nature
Medicine (New York, NY, United States), 2003, 9, 1498-1505.

T. Lauer, Ml. Preik, T. Rassaf, B. E. Strauer, A. Deussen, M. Feelisch and M.
Kelm, Proceedings of the National Academy of Sciences of the United States of
America, 2001, 98, 12814-12819.

N. M. Tsoukias, M. Kavdia and A. S. Popel, American Journal of Physiology,
2004, 286, H1043-H1056.

L. J ia, C. Bonaventura, J. Bonaventura and J. S. Stamler, Nature (London),
1996, 380, 221-226.

J. S. Stamler, L. Jia, J. P. Eu, T. J. McMahon, I. T. Demchenko, I.
Bonaventura, K. Gemert and C. A. Piantadosi, Science (Washington, D. C. ),
1997, 276, 2034-2037.

A. J. Gow and J. S. Stamler, Nature (London), 1998, 391, 169-173.

T. J. McMahon, R. E. Moon, B. P. Luschinger, M. S. Carraway, A. E. Stone,
B. W. Stolp, A. J. Gow, J. R. Pawloski, P. Watke, D. J. Singel, C. A.
Piantadosi and J. S. Stamler, Nature Medicine (New York, NY, United States),
2002, 8, 711-717.

M. P. Doyle and J. W. Hoekstra, J. Inorg. Biochem., 1981, 14, 351-358.
S. Herold, M. Exner and T. Nauser, Biochemistry, 2001, 40, 3385-3395.

M. T. Gladwin, F. P. Ognibene, L. K. Pannell, J. S. Nichols, M. E. Pease-Fye,
J. H. Shelhamer and A. N. Schechter, Proceedings of the National Academy of
Sciences of the United States of America, 2000, 97, 9943-9948.

110

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

A. J. Gow, B. P. Luchsinger, J. R. Pawloski, D. J. Singel and J. S. Stamler,
Proceedings of the National Academy of Sciences of the United States of
America, 1999, 96, 9027-9032.

T. J. McMahon and J. S. Stamler, Methods in Enzymology, 1999, 301, 99-114,
111 plate.

S. Deem, M. T. Gladwin, J. T. Berg, M. E. Kerr and E. R. Swenson, Am J
Respir Crit Care Med, 2001, 163, 1164-1170.

S. Deem, J.-U. Kim, B. N. Manjula, A. S. Acharya, M. E. Kerr, R. P. Patel, M.
T. Gladwin and E. R. Swenson, Circulation Research, 2002, 91, 626-632.

M. T. Gladwin, X. Wang, C. D. Reiter, B. K. Yang, E. X. Vivas, C.
Bonaventura and A. N. Schechter, Journal of Biological Chemistry, 2002, 277,
27818-27828.

T. J. McMahon, A. E. Stone, J. Bonaventura, D. J. Singel and J. S. Stamler,
Journal of Biological Chemistry, 2000, 275, 16738-16745.

P. Agvald, L. C. Adding, A. Artlich, M. G. Persson and L. E. Gustafsson, Br.
J. Pharmacol, 2002, 135, 373-382.

J. H. Crawford, C. R. White and R. P. Patel, Blood, 2003, 101, 4408-4415.

A. J. Hobbs, M. T. Gladwin, R. P. Patel, D. L. H. Williams and A. R. Butler,
Trends Pharmacol. Sci, 2002, 23, 406-411.

M. Wolzt, R. J. MacAllister, D. Davis, M. Feelisch, S. Moncada, P. Vallance
and A. J. Hobbs, Journal of Biological Chemistry, 1999, 274, 28983-28990.

J. L. Amezcua, R. M. J. Palmer, B. M. De Souza and S. Moncada, Br. J.
Pharmacol, 1989, 97, 1119-1124.

A. Chu, D. E. Chambers, C. C. Lin, W. D. Kuehl, R. M. J. Palmer, S. Moncada
and F. R. Cobb, Journal of Clinical Investigation, 1991, 87, 1964-1968.

M. T. Gladwin, J. H. Shelhamer, A. N. Schechter, M. E. Pease-F ye, M. A.
Waclawiw, J. A. Panza, F. P. Ognibene and R. O. Cannon, III, Proceedings of
the National Academy of Sciences of the United States of America, 2000, 97,
11482-11487.

J. A. Panza, P. R. Casino, C. M. Kilcoyne and A. A. Quyyumi, Circulation,
1993, 87, 1468-1474.

D. D. Rees, R. M. J. Palmer, H. F . Hodson and S. Moncada, Br. J. Pharmacol. ,
1989, 96, 418-424.

111

29.

30.

A. A. Quyyumi, N. Dakak, N. P. Andrews, S. Husain, S. Arora, D. M.
Gilligan, J. A. Panza and R. O. Cannon, 3rd, J Clin Invest, 1995, 95, 1747-
1755.

L. 1. Genes, N. V. Tolan, M. K. Hulvey, R. S. Martin and D. M. Spence, Lab
on a Chip, 2007, 7, 1256-1259.

112

CHAPTER 5

5.1 CONCLUSIONS

For many years, the RBC has been viewed as a “sack of hemoglobin”. Since
then, its role in disease pathogenesis, as a therapeutic target, and as a regulator of
blood flow has been extensively studied. It has been shown the RBC can control
vascular caliber through a number of different mechanisms. One mechanism is the

release of a metabolite into the blood stream, adenosine triphosphate (ATP). Multiple

stimuli such as pharmacological agents (iloprost, mastoparan),l.3 C-peptide4 and

deformations have been reported to increase ATP release from the RBC. Here, the

role of hypoxia as an important stimulant for ATP release from the RBC has been

. . 6
investigated.

Hypoxia, an inadequate supply of oxygen to tissue or an organ as a whole,
induces many different processes on the vasculature. For example, membrane
integrity is compromised in endothelial cells subjected to hypoxia (leading to exposure

of the collagen and fibrin rich under layer). Moreover, this compromised membrane

9

integrity results in an increase in the procoagulant state.

Erythropoiesis, an increase in RBC production, has also been linked to low

. 9 . . . .
oxygen tensrons. When oxygen saturation 15 less than 10%, erythropmetm levels

begin to increase, resulting in increased RBC production, of course this leads to more

oxygen delivery. 10

113

As mentioned previously, one major mechanism for reperfusion (blood flow)

under hypoxic conditions is the release of ATP. Once ATP is released into the blood

stream, it activates the purinergic receptor P2), on the endothelial cells lining the vessel
wall.11 Upon activation, intracellular [Ca2+] is increased through the phospholipase-

C/inositol-l,4,5-triphosphate pathway.12’ 13 This increase in [Ca2+] induces a

conformational change on endothelial nitric oxide synthase (eNOS), an enzyme
responsible for the production of NO. This NO can diffuse to the underlying smooth

muscle cells, where it activates soluble guanylate cyclase, which converts cyclic
triphosphate to cyclic guanosine monophosphate (cGMP).l4 Subsequently, the cGMP
results in smooth muscle relaxation. It is under this construct that hypoxia-induced
ATP release can elicit vessel relaxation, thereby rendering the RBC as both an oxygen

. . 15
sensor and oxygen delivery mechanism.

A signaling pathway for deformation-induced ATP release has been

proposed.2 Through the activation of a G-protein, GS or G1, adenylyl cyclase is

activated and converts ATP to cAMP. The cAMP then activates protein kinase A
which phosphorylates the cystic fibrosis transmembrane regulator (CFTR), which is a
required component for the release of ATP from the RBC. Since deformation releases
ATP, and stiffened RBCS are associated with certain diseases where perfusion is also a
problem, hypoxia seems more problematic with stiffened cells.

Under hypoxic conditions, hemoglobin changes conformations from the R

(relaxed) to the T (tense) state. Hemoglobin is connected to the RBC membrane

114

through the band 3 protein. We hypothesize the change in conformational states of
hemoglobin induces a deformation on the RBC membrane. This deformation is what
results in the ATP release; therefore, we hypothesize hypoxia-induced ATP release is

really a form of deformation.

5.1.1 A Non-Additive Relationship Between Hypoxia and Deformation-Induced

ATP Release

To test the hypothesis if hypoxia and deformation are non-additive, we set up a
microvessel mimic that induced both hypoxia and deformation simultaneously on the
RBCS was used to investigate total ATP release. Through the use of fused silica

microbore tubing having an inside diameter of 50 um, ATP release was induced by

deformation. Hypoxia was induced by gas exchange with N2 gas for 4 min at 30 kPa.

The RBCS were placed in a syringe that was placed on a syringe pump and flowed at a
rate of 6.7 uL/min. Another syringe was filled with a luciferin/luciferase mixture and
flowed along side the RBCS. Both mixtures met at a T-junction and a
chemiluminescence reaction occurred between the released ATP and the
luciferin/luciferase mixture. The mixed samples continued to flow from the T-
junction over a PMT. A window was etched out of the microbore tubing so the light
given off by the chemiluminescence reaction could be measured.

To determine the ATP release from hypoxia alone, RBCS were injected into a
cuvette mounted over a PMT. An aliquot of the luciferin/luciferase mixture was

injected into the cuvette along with the RBCS. After 15 s, the measurement was

115

obtained. The results from this study are displayed in chapter 2. Hypoxia and

deformation displayed a non-additive relationship, suggesting that both stimuli follow

the same signaling pathway. 16

5.1.2 Defining the Signaling Pathway for Hypoxia-Induced ATP Release

As mentioned previously, a signaling pathway for deformation-induced ATP

release has been proposed.2 This signaling pathway was shown in chapter 3. To

determine if this pathway is the same for hypoxia-induced ATP release, all proteins or
enzymes in the pathway were inhibited and exposed to hypoxic conditions. If hypoxia
appears to be a form of deformation as the ATP release decreased upon inhibition of
any protein in the proposed deformation-induced pathway. As shown in chapter 3,
inhibition of the signaling mechanism and subsequent exposure to hypoxia resulted in
a decrease in ATP release. Moreover, upon stiffening of the RBCS with diamide, a
known cell stiffener, and exposure to a hypoxic environment, the ATP release

decreased in a dosage dependant manner. This data further confirms our hypothesis.
5.1.3 Hypoxic Vasodilation

Hypoxic vasodilation is a phenomenon that has been under intense
investigation for the past 15 years. Two conflicting theories have been proposed.

First, the theory of hemoglobin acting as a nitrite reductase has been proposed by

Gladwin.” The authors report that deoxyhemoglobin, at p50, converts nitrite, already

116

in high concentrations in the blood, to NO. This NO then diffuses through the
endothelium, sub-endothelium, and to the underlying smooth muscle cells. The
second theory proposed by Stamler suggests hemoglobin becomes nitrosylated in the
lungs and conserves the NO on a B93 cysteine residue. Upon entering a hypoxic
environment, hemoglobin changes its conformation in order to induce the release of

the NO. The NO then diffuses through the endothelium, the sub-endothelium and to

. . . . . l8 . .
the smooth muscle cells where it elic1ts vasodilation. While both these theories are

extensive in their reasoning, they raise some questions, physiologically.

Here, it was hypothesized that hypoxic vasodilation occurs due to an ATP
release from the RBC as opposed to NO donation. To test this theory, microfluidic
devices were utilized. Microfluidic devices have gained increasing popularity since
their discovery in the early ‘905. Due to their small size and the ability to mimic
vessel diameters in a real time flow-through system, they are ideal for mimicking the
vasculature.

A microfluidic array designed and fabricated by our lab was utilized. RBCS
and hypoxic RBCS were delivered through channels underlying a porous
polycarbonate membrane. Bovine pulmonary arterial endothelial cells OaPAECs) were
cultured on the membrane. RBCS incubated with diamide and niflumic acid, a CFTR
inhibitor, were also delivered in the underlying channels (chapter 3 showed these two
reagants inhibit hypoxia-induced ATP release). Upon inhibition of ATP release, NO
production was also decreased, suggesting that production of the vasodilator NO is

dependant on ATP release from RBCS under hypoxia.

117

Summarily, this thesis demonstrates the pathway for hypoxia-induced ATP
release is deforrnation-based and that hypoxic vasodilation is stimulated by this

release, as opposed to RBC donation of NO.

5.2 FUTURE WORK

It has been previously demonstrated that the ATP release from RBCS obtained
from patients with varying diseases increased or decreased. For example, the Spence
group has measured the ATP release from patients with multiple sclerosis and found it
increased when compared to RBCS of normal patients. Moreover, it has been

established that patients with multiple sclerosis have higher NO levels in their cerebral

spinal fluid that may lead to demyelination.19

Another disease where ATP plays a role is diabetes mellitus. It has been

established the RBCS of patients with diabetes are less deformable20 and release less

ATP.2]’ 22 This may decrease the production of NO and its vasodilatory effects and

may be a reason for the higher occurrence of stroke and hypertension in patients with

diabetes.23' Based on these ATP release profiles, the RBC itself may be used as a

biomarker for diabetes and related diseases such as cardiovascular disease.

Diabetes mellitus, literally translated to “sweet urine” is a syndrome of
disordered metabolism. It is characterized by having a fasting blood glucose level of
greater than 140 mg/dl and a non-fasting blood glucose level of greater than 200

mg/dl. In 2007, the Center for Disease Control and Prevention (CDC) estimated about

118

23.6 million people in the United States (about 7.8% of the population) had

diabetes.24 In patients with diabetes, blood glucose levels are higher than normal due

to a decreased sensitivity to insulin or a complete deficiency of the hormone. In liver,
muscle, and fat cells insulin is used to uptake glucose where it is stored in the form of
glycogen or used in the production of energy through glycolysis. There are three
classifications of diabetes: type 1 diabetes, type 2 diabetes, and gestational diabetes.

Type 1 diabetes mellitus (TlDM) results from immune-mediated destruction of
the pancreatic islet beta cells. The cause of TlDM is not entirely known, although
there are over 20 genes that influence susceptibility to type 1 diabetes, namely, the
human leukocyte antigen (HLA) and the insulin gene locus. Therapy for TlDM
includes injection of exogenous insulin.

Type 2 diabetes mellitus (T2DM) is described as an “insulin resistant” form of
diabetes. Insulin resistance is defined as the inability of insulin to lower circulating
glucose. It occurs in about 90% of patients with T2DM and gradually progresses and
worsens over the years. Insulin resistance often arises from metabolic abnormalities
known as the metabolic syndrome. Other conditions compose the metabolic syndrome
including dyslipidemia, hypertension, abdominal obesity, and endothelial dysfunction
to name a few. Although the role of insulin resistance to the origin of the metabolic
syndrome is unclear, they are strongly associated. Currently, the World Health

Organization (WHO) has approved two oral pharmaceutical therapies for T2DM:

glibenclamide, a sulfonylurea, and metforrnin, a biguanide.25

Gestational diabetes mellitus (GDM) is a glucose intolerance that appears

during pregnancy. Although insulin resistance is apparent in all pregnancies, to a

119

certain degree, some women cannot neutralize the insulin resistance and therefore

develop GDM.26 In order to diagnose GDM, a 1 hour, 50 g. glucose test is performed

and if the resulting plasma glucose level is lower than 140 mg/dL the patient is not
considered to have GDM. However, if the plasma glucose level is greater than 140
mg/dL, a second test is performed consisting of a 3 hour, 100 g. glucose test. After 1,
2, and 3 hours the plasma glucose levels are checked and if they are above 180, 155,
and 140 mg/dL the patient is diagnosed with GDM. While this is the standard for
GDM testing in the United States, the WHO uses a different method for diagnosis.27

29

It has been established that microvascular and macrovascular complications are

present in prediabetic patients“).37 Early detection of diabetes allows for the finding

. . 9-13
of increased vascular disease.

As previously mentioned, the occurrence of stroke in patients with diabetes is

1.5-3 times higher with a worse reperfusion injury.23 We have demonstrated hypoxia-

induced ATP release from the RBCS of diabetic rats is 39.3% less when compared to
those of control rats. The results of this study are displayed in Figure 28. For future
studies, testing the ATP release in viva from animal models with diabetes alter the
occurrence of stroke would further elucidate if ATP release plays a role in their
reperfusion injury and occurrence of ischemic attacks. Moreover, adding a
pharmacological stimulant, such as pentoxyfilline, that increases ATP release could

determine a drug therapy.

120

 

 

 

ATP Release

 

 

 

Figure 28. This graph displays the ATP release of control and BB/ZDR/Wor
rats (Type 2 diabetic rats) exposed to a hypoxic environment. The ATP release
from the RBCS of the diabetic rats decreased by 39.%, this demonstrates a
lower ability of the RBCS to elicit vasodilation. Furthermore, this implies the
reason for worse reperfusion injury in patients with diabetes.

121

Moreover, if the RBCS of patients with diabetes are not able to deliver oxygen,
this would aid in their higher reperfusion injury. 2,3-DPG is an organic phosphate,
produced by glycolysis, that is present on hemoglobin. 2,3-DPG allows hemoglobin

to release bound oxygen to respiring tissue. Under conditions of hypoxia, glycolysis

is increased; therefore, more 2.3-DPG.38 If the RBCS of patients with diabetes do not

respond as well to hypoxia with ATP release, it may be possible their glycolysis
production doesn’t increase either. Importantly, many of the glycolytic enzymes are
attached to the band 3 protein, the same protein hemoglobin is attached to. Testing the
2,3-DPG levels in the RBCS of patients with diabetes could lead to a better
understanding of where the oxygen is being delivered and if it is delivered prematurely
in patients with diabetes.

Recent research has been focusing on the mechanism of hydroxyurea, a proven
treatment in sickle cell disease. Sickle cell disease is one of the most common genetic
diseases in the United States. It occurs from RBCS taking on an abnormal, rigid,
sickle shape. Interestingly, this sickling event occurs only when hemoglobin is in the
deoxygenated form.

The prevalence of sickle cell disease in the United States is about 1 in every
5,000 people, mostly affecting African-Americans. Due to a single amino acid

mutation in hemoglobin, hemoglobin S is formed. This form of hemoglobin is the

cause of the physiological conditions presented in this disease.39 There are four types

of sickle cell disease: sickle-cell anemia, sickle-hemoglobin C, sickle beta-plus-

thalassemia and sickle-beta-zero thalassemia. The last three forms are very rare.

122

Previous studies report the mechanism of action for hydroxyurea is through an

increase in production of fetal hemoglobin (HbF).40 Fetal hemoglobin has a higher

capacity for oxygen, therefore, more tissue is perfused and less sickling occurs.

Moreover, an improvement in the deformability of RBCS in patients receiving

hydroxyurea has been reported.41 Due to the finding that hypoxia is a form of

deformation, and hydroxyurea increases deformability, it was hypothesized that the
ATP release from RBCS under the influence of both hypoxia and hydroxyurea will
increase. Data obtained from this study suggested otherwise. The ATP release from
RBCS due to hypoxia and hydroxyurea was at the same concentration as the ATP
release from hypoxia alone. Interestingly, our lab has recently discovered an increase
in NO production from RBCS pre-incubated with hydroxyurea (submitted data). This
led to the idea that NO (which has a higher affinity for hemoglobin than oxygen) was
attaching to the heme, releasing the oxygen atom. This would prevent hemoglobin
from changes conformations and not allow the deformation to occur; thereby not
releasing increased ATP. As shown in Figure 29, the depletion of oxygen after RBCS
were preincubated with hydroxyurea occurs more readily. This suggests that NO is

replacing the oxygen, leaving free oxygen in the buffer ready to be removed by my

®
Oxyrase enzyme system. These measurements were taken on a Clark-type electrode.

Further studies can provide a more detailed mechanism for this hypothesis. For

123

52882 3 oz do means
2: we“ combs do 3552 2: 953me down 2&2 22: a «a :52on0: Bob 3.682 mm cowxxo “momma—m See 22E. .335
Home“: 0: 5:5 mum—M 2x33 8 839:8 55 5292 208 mogxxohc»: 2:3 vegansoemoa momm 2: 82m code—mow we?
nomxxO 82:58:33 .21 2: :23 “8233255 momd 2893 93 momd 2x093 mo ego coumhsmm nowxxO 6N 25E

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 we:
88 82 88 8.: 8.: 8a a
p p I n p n a
I]!!!
a
2 o
X
m
U
it an o
0
U
3 m
m.
an m
a... u
DI + momm .x0 089 F
- S.
DI oz .momm .xO exams + k?
a at
. a...

 

124

example, measuring the concentration of NO on hemoglobin through EPR or IR, or

creating a miniaturized oxygen electrode to better detect oxygen saturation.

125

References

1.

10.

11.

12.

13.

14.

T. Nakajima, K. Wakamatsu and H. Mukai, Animal Toxins, 2000, 116-126.

J. J. Olearczyk, A. H. Stephenson, A. J. Lonigro and R. S. Sprague, Med Sci
Monit 2001, 7, 669-674.

I. Schwaner, R. Seifert and G. Schultz, Eicosanoids, 1992, 5 Suppl, S10-12.

J. A. Meyer, J. M. Froelich, G. E. Reid, W. K. A. Karunarathne and D. M.
Spence, Diabetologia 2008, 51, 175-182.

R. S. Sprague, M. L. Ellsworth, A. H. Stephenson and A. J. Lonigro, American
Journal of Physiology, 1996, 271, H2717-H2722.

G. R. Bergfeld and T. Forrester, Cardiovascular Research, 1992, 26, 40-47.

X. Hong, Z.-Y. Yin, X.-B. Sun and Y.-D. Lu, Zhongguo Yingyong Shenglixue
Zazhi, 2000, 16, 41-44.

S. Ogawa, H. Gerlach, C. Esposito, A. Pasagian-Macaulay, J. Brett and D.
Stern, Journal of Clinical Investigation, 1990, 85, 1090-1098.

W. Jelkmann, Physiol. Rev., 1992, 72, 449-489.

K. U. Eckardt, S. T. Koury, C. C. Tan, S. J. Schuster, B. Kaissling, P. J.
Ratcliffe and A. Kurtz, Kidney Int. , 1993, 43, 815-823.

G. Burnstock and C. Kennedy, Gen. Pharmacol, 1985, 16, 433-440.

A. S. Awad, R. L. Webb, R. M. Carey and H. M. Siragy, J Hypertens, 2004,
22,1571-1577.

K. J. Smith, R. Kapoor and P. A. Felts, Brain Pathol, 1999, 9, 69-92.

P. A. Marsden, T. A. Brock and B. J. Ballermann, American Journal of
Physiology, 1990, 258, F l295-F 1303.

126

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

M. L. Ellsworth, T. Forrester, C. G. Ellis and H. H. Dietrich, Am J Physiol
1995, 269, H2155-2161.

A. Paris and D. M. Spence, 2008, 133, 678-682.

J. H. Crawford, C. R. White and R. P. Patel, Blood, 2003, 101, 4408-4415.

L. J ia, C. Bonaventura, J. Bonaventura and J. S. Stamler, Nature (London),
1996, 380, 221-226.

J. M. Encinas, L. Manganas and G. Enikolopov, Current Neurology and
Neuroscience Reports, 2005, 5, 232-238.

E. J. Diamantopoulos, C. Kittas, D. Charitos, M. Grigoriadou, G. Ifanti and S.
A. Raptis, Hormone and Metabolic Research, 2004, 36, 142-147.

J. Carroll, M. Raththagala, W. Subasinghe, S. Baguzis, T. D. a. Oblak, P. Root
and D. Spence, Molecular BioSystems, 2006, 2, 305-311.

R. S. Sprague, A. H. Stephenson, E. A. Bowles, M. S. Stumpf and A. J.
Lonigro, Diabetes, 2006, 55, 3588-3593.

J. Weinberger, V. Biscarra, M. K. Weisberg and J. H. Jacobson, Stroke 1983,
14, 709-712.

C. Cowie Catherine, F. Rust Keith, D. Byrd-Holt Danita, S. Eberhardt Mark,
M. F legal Katherine, M. Engelgau Michael, H. Saydah Sharon, E. Williams
Desmond, S. Geiss Linda and W. Gregg Edward, Diabetes Care, 2006, 29,
1263-1268.

N. A. Kasim, M. Whitehouse, C. Ramachandran, M. Bermejo, H. Lennemaes,
A. S. Hussain, H. E. Junginger, S. A. Stavchansky, K. K. Midha, V. P. Shah
and G. L. Amidon, M01. Pharm, 2004, 1, 85-96.

A. H. Xiang, R. K. Peters, E. Trigo, S. L. Kjos, W. P. Lee and T. A. Buchanan,
Diabetes, 1999, 48, 848-854.

127

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

R. A. Amadin, O. O. F amuyiwa and B. O. Adelusi, Cent Afl J Med, 1989, 35,
486-489.

D. J. Pettitt, P. H. Bennett, R. L. Hanson, K. M. Narayan and W. C. Knowler,
Diabetes Care, 1994, 17, 1264-1268.

C. Deerochanawong, C. Putiyanun, M. Wongsuryrat, S. Serirat and P. Jinayon
Diabetologia, 1996, 39, 1070-1073.

3

T. Nguyen Thanh, J. Wang lie and Y. Wong Tien, Diabetes Care, 2007, 30,
2708-2715.

D. M. Nathan, E. Chew, C. A. Christophi, M. D. Davis, S. Fowler, B. J.
Goldstein, R. F. Hamman, L. D. Hubbard, W. C. Knowler and M. E. Molitch,
Diabetic Med, 2007, 24, 137-144.

C. J. Sumner, S. Sheth, J. W. Griffin, D. R. Cornblath and M. Polydeflds,
Neurology, 2003, 60, 108-111.

H. Herman William, Diabetes Care, 2007, 30, 1912-1919.

B. Barrett-Connor and A. Ferrara, Diabetes Care, 1998, 21, 1236-1239.

S. M. Hafiner, S. Lehto, T. Ronnemaa, K. Pyorala and M. Laakso, N Engl J
Med, 1998, 339, 229-234.

A. I. Adler, H. A. Neil, S. E. Manley, R. R. Holman and R. C. Turner, Am
Heart J, 1999, 1138, S353-359.

M. Nathan David, A. Cleary Patricia, C. Backlund Jye-Yu, M. Genuth Saul,
M. Lachin John, J. Orchard Trevor, P. Raskin and B. Zinman, N Engl J Med,
2005, 353, 2643-2653.

L. E. Costa and I. M. De Miranda, Acta Physiol. Lat. Am., 1976, 26, 115-121.

M. Aksoy, Blood, 1963, 22, 757-769.

128

40.

41.

O. S. Platt, S. H. Orkin, G. Dover, G. P. Beardsley, B. Miller and D. G.
Nathan, Hydroxyurea enhances fiatal hemoglobin production in sickle cell

anemia, United States, 1984.

G. Athanassiou, A. Moutzouri, A. Kourakli and N. Zoumbos, Clin.
Hemorheol. Microcirc. , 2006, 35, 291-295.

129

11111111

1

3 03063

3