INVESTIGATING BLOOD CELL COMMUNICATION WITH NOVEL SAMPLE
HANDLING AND MICROFLUIDIC TECHNOLOGY
By
Kari B. Anderson

A DISSERTATION

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

Chemistry - Doctor of Philosophy
2013

ABSTRACT
INVESTIGATING BLOOD CELL COMMUNICATION WITH NOVEL SAMPLE
HANDLING AND MICROFLUIDIC TECHNOLOGY
By
Kari B. Anderson
Puringergic receptor signaling events in platelets is a major determinant in platelet
function.

ADP binding to the P2Y-type receptors on platelet membranes, and the

subsequent pathways evoked from this type of binding, is well-established. However, an
understanding activity of the ATP-sensitive P2X1 platelet receptor is incomplete due to
rapid receptor desensititzation.

To circumvent this problem, most studies aimed at

investigating P2X1 platelet activity require that the platelet samples contain apyrase in
order to reduce already-existing levels of ATP. Of course, one drawback to this method
is that the apyrase will also rapidly degrade any added ATP. Here, I describe a methoed
that employs the reported P2X1 inhibitor NF449 to sensitize platelets in the absence of
any added apyrase. Sensitization is verified by spectrofluorimetric determination of Ca

2+

entry into the platelets after stimulation with concentrations of ATP ranging from 67 nM
to 10 µM. Results suggest that sensitization of the P2X1 receptor by NF449 is not
necessarily dependent upon the inhibitor concentration, but rather the ratio of the
inhibitor and exogenously-added ATP concentrations. The sensitization by the NF449
was also found to be highly time-dependent.
A simple and inexpensive approach to fabricate polystyrene devices that is based
upon molding polystyrene (PS) from Petri dishes against PDMS molds was developed.
The ability to incorporate microchannels in polystyrene and integrate the resulting device

with standard laboratory equipment such as an optical plate reader for analyte readout
and pipets for fluid propulsion is described. Integration of the microfluidic device with
these off-chip functions (sample delivery and readout) enables high-throughput screens
and analyses. Furthermore, this polystyrene device was used to monitor two endothelial
cell processes. One experiment involved the fluorescence measurement of nitric oxide
(NO) produced within an endothelial cell line following stimulation with ATP. The
result was a four-fold increase in NO production (as compared to a control), with this
receptor-based mechanism of NO production verifying the maintenance of cell receptors
following immobilization onto the PS substrate. The ability to monitor cellular uptake
was also demonstrated by optical determination of Ca
stimulation with the Ca

2+

2+

into endothelial cells following

ionophore A20317. The result was a significant increase

(42%) in the calcium uptake in the presence of the ionophore, as compared to a control
(17%) (p < 0.05).
The PS device was further integrated with a polyester membrane in order to study
drug transport across the membrane. However, significant absorption of drugs still
occurred so a device made only of PS was fabricated using an epoxy mold and hydraulic
press to embed channels and seal devices. The result was a device with millimeter sized
channels, but the ability to rapidly fabricate these devices was lost. Thus, 3 Dimensional
(3D) printing, commonly used for producing design prototypes in industry to fabricate a
reusable, high throughput, 3D printed fluidic device that enables flow and incorporates a
membrane above a channel in order to study drug transport and affect cells.

Catullus 85:
Odi et amo. Quare id faciam, fortasse requiris?
Nescio, sed fieri sentio et excrucior.

iv

ACKNOWLEDGEMENTS
I would first like to thank the organizations without whom this research could not
be funded including Michigan State University (MSU) for research and travel funding,
the National Institutes of Health, the Cystic Fibrosis Foundation and Merck.
Secondly, I would like to thank my guidance committee Dr. A. Daniel Jones, Dr.
Greg Swain and Dr. Weliky who were always willing to discuss research ideas. I would
also like to thank Dr. Kathy Severin whom I taught for on numerous occasions. She was
a great teaching mentor and always willing to assist graduate students through training or
instrument usage.
The ability to collaborate with other departments at MSU has been crucial to
moving certain projects forward. I would like to extend my thanks to the MSU RTSF
Mass Spectrometry Core for instrument usage, Dr. Per Askeland in Engineering for help
obtaining scanning electron microscopy images, Dr. Brian Rook in the Composite
Materials and Structures Center for hydraulic press training, and Dr. Brian Wright in the
Department of Electrical and Computer Engineering for assistance in 3 dimensional (3D)
printing. Lastly, I would like to thank Tekna Inc. design firm in Kalamazoo, Mi for their
assistance in creating the necessary 3D files for printing.
Collaboration has also been integral to my success within the Spence group so, I
would like to specifically thank Dr. Stephen Halpin and Ms. Sarah Lockwood for help
with the polystyrene and 3D printing projects, respectively. Additionally, I would also
like to acknowledge members of the Spence research group from 2008 – 2013 for their
knowledge and cooperation. Much of the work described herein entails blood draws,
v

buffer preparation, cell culture and device fabrication and to have a group willing to
cooperate with one another on these aspects makes the work go a lot faster.
I would also like to thank Dr. Dana Spence for his role as advisor in this work.
The research environment he creates fosters not only intellectual experimental design, but
also creativity, which allows his graduate students to grow with every success and failure.
I admire his devotion to his family and the graduate students that work for him, both past
and present. I am happy at the end of my graduate career to call him mentor and friend.
Graduate school would be a hellish process to go through alone and I am grateful
for the friends I made here as they made the process more enjoyable and their support
made the hoops worth jumping through. Specifically, Suzanne Letourneau, my partner in
crime from the beginning! Suzanne and I could always count on each other for early
morning rabbit surgeries, late night study sessions and baking adventures. I would also
like to thank Yimeng Wang (Meng Meng) for being such a good friend to me and
becoming more in tune with American sarcasm every day that she can see through even
the most offhand remarks. I could not ask for a better research group, five and forever!
Frankly, none of this would have been possible without the support of my family,
specifically my parents, who have always been committed to my education and their
support was invaluable over the course of my graduate career. Finally, I would like to
thank Andy for his dedication to this work, his encouragement and willingness to let me
figure out my path in life. His patience and love are unmatched and I appreciate his
commitment to me now more than ever.

vi

TABLE OF CONTENTS

LIST OF FIGURES……………………………………………………………………...x
CHAPTER 1: INTRODUCTION…………………………………………………….....1
1.1
The Platelet and the Vascular System……………………………………………..1
1.1.1 Blood Vessel Injury, Platelet Activation and Aggregation………………….1
1.2
Platelet Inhibitors………………………………………………………………….4
1.2.1 Nitric Oxide (NO) …………………………………………………………..5
1.2.2 Prostacyclin (PGI2)………………………………………………………….5
1.2.3 Ectonucleotidase CD39……………………………………………………...5
1.3 Purinergic Receptors on the Platelet Surface …………………………………………6
1.4 Diseases with Hyperactive Platelets…………………………………………………10
1.4.1 Diabetes…………………………………………………………………….10
1.4.2 Hypertension……………………………………………………………….11
1.4.3 Cystic Fibrosis……………………………………………………………..11
1.4.4 Sickle Cell Anemia………………………………………………………...12
1.4.5 Multiple Sclerosis……………………………………….…………………12
1.4.6 Defects in RBC ATP release……………………………………………….12
1.5 Measuring Platelet Activity in Vitro…………………………………………………13
1.5.1 Aggregometry……………………………………………………………...13
1.5.2 Flow Cytometry and Fluorescence………………………………………...14
1.5.3 Microfluidic Devices………………………………………………………16
1.5.4 Apyrase and α,β-me-ATP in Platelet Preparation and Analysis…………...17
REFERENCES…………………………………………………………………………19
CHAPTER 2: PLATELET P2X1 RECEPTOR FUNCTION IN THE ABSENCE OF
EXOGENOUS APYRASE……………………………………………30
2.1 Platelet P2X1 receptor……………………………………………………………….30
2.1.1 Difficulties in measuring P2X1 receptor activity………………………….31
2.1.2 NF449 as a P2X1 receptor inhibitor……………………………………….32
2.2 Methods………………………………………………………………………………35
2.2.1 Isolation of platelets from human whole blood……………………………35
2.2.2 Fluo-4 AM probe and NF449 platelet sample preparation………………...36
2+
2.2.3 Fluorescence determination of intracellular platelet Ca ………………..37
2.2.4 Data analysis……………………………………………………………….37
2.4 Results………………………………………………………………………………..39
2+
2.4.1. NF449 sensitizes ATP-induced Ca Influx………………………………39
2+

2+

2.4.2 Effect of Ca free buffer on Ca entry into the platelet via ATP and ADP
agonists…………………………………………………………………….41
2+
2.4.3 Effect of α,β-me-ATP concentrations on Ca entry into the platelet……..43
2+

2.4.4 Effect of NF449 concentrations on Ca entry into the platelet…………...43
2.4.5 Time Dependence of NF449 Sensitization of P2X1……………………….46
vii

2+

2.4.6 P2Y1 inhibition with MRS and the effect on Ca influx…………………48
2.5 Conclusions…………………………………………………………………………..51
REFERENCES…………………………………………………………………………53
CHAPTER 3: INTEGRATION OD MULTIPLE COMPONENTS IN
POLYSTYRENE-BASED MICROFLUIDIC DEVICE……………………………..58
3.1 Introduction to Microfluidic Device Fabrication…………………………………….58
3.1.1 PDMS Microfluidic Devices……………………………………………….58
3.1.2 Polystyrene Microfluidic Devices………………………………………….59
3.2 Methods………………………………………………………………………………61
3.2.1 Fabrication of a polystyrene device with integrated channels……………..61
3.2.2 Integration of the Polystyrene Device with a Plate Reader ……………….63
3.3 Results……………………………………………………………………………….64
3.3.1 Fabrication of Polystyrene-based Devices…………………………………65
3.3.2 Alignment in Plate Reader…………………………………………………69
3.4 Conclusions…………………………………………………………………………..72
REFERENCES………………………………………………………………………….74
CHAPTER 4: ANALYSIS OF ENDOTHELIAL CELLS USING A
POLYSTYRENE-BASED MICROFLUIDIC DEVICE AND 96-WELL PLATE
READER DETECTION………………………………………………………………..79
4.1 Introduction………………………………………………………………………….79
4.1.1 Cell Analysis on Polystyrene Microfluidic Devices……………………….79
4.2 Methods………………………………………………………………………………81
4.2.1 Endothelial Cell Culture in Polystyrene-based Microchannels……………81
4.2.2 Fluorescence Detection of Intracellular Endothelial NO…………………..82
4.2.2 Fluorescence Detection of Endothelial Calcium Uptake…………………..84
4.2.3 Cell Viability Study………………………………………………………..85
4.3 Results………………………………………………………………………………..86
4.3.1 Monitoring Nitric Oxide Production, Calcium Uptake and Viability in
Endothelial Cells……………………………………………………………..86
4.4 Conclusions…………………………………………………………………………..91
REFERENCES…………………………………………………………………………95
CHAPTER 5: CONCLUSIONS AND FUTURE DIRECTIONS…………………..98
5.1 Platelet P2X1 Receptor Function………………………………………………….....98
5.2 Polystyrene Microfluidic Devices…………………………………………………..100
5.3 Introduction to 3 Dimensional Printing…………………………………………….106
5.3.1 3D Printed Device Fabrication……………………………………………107
5.3.2 Sample Preparation for Drug Transport…………………………………..108
5.3.3 Mass Spectrometric Analysis of Drug Transport…………………………108
5.3.4 Endothelial Cell Culture onto Inserts……………………………………..109
5.3.5 Sample Preparation and Experimental Procedure for Cell Viability Study111
5.4 Preliminary Results with 3D Printed Device……………………………………….111
5.4.1 Drug Transport across a Membrane…………………………………………..112
5.4.2 Cell Viability Assessment…………………………………………………….115
5.5 Conclusions on 3D Printing………………………………………………………...117
viii

REFERENCES………………………………………………………………………...122

ix

LIST OF FIGURES
Figure 1.1: Blood vessel injury. For clarity only RBCs and platelets are pictured. A
typical blood vessel is comprised of an inner layer of endothelial cells with smooth
muscle cells lining the exterior. Upon blood vessel injury, subendothelial collagen is
exposed and platelets that come into contact with the collagen become activated, undergo
a shape change that results in adhesion, and release ATP and ADP into the bloodstream.
As more platelets are recruited to the injury site, aggregation occurs and a plug over the
injury site is formed. NO, a platelet inhibitor, is also released by platelets to inhibit
further platelet recruitment to the injury site. For interpretation of the reference to color in
this and all other figures, the reader is referred to the electronic version of this
dissertation………………………………………………………………………………...2
Figure 1.2: P2 receptors on platelet membrane. The platelet membrane has three
purinergic receptors on its membrane surface. The P2X1 is a gated cation channel that is
2+
activated by ATP. Upon ATP binding there is a rapid Ca influx into the platelet.
P2Y1 and P2Y12 are both G-protein coupled receptors that bind ADP. When ADP binds
2+
P2Y1 it activates PLC, which causes intracellular Ca release and subsequent platelet
activation. When ADP binds P2Y12 it inhibits AC activity, thus, counteracting the
inhibitory effects of PGI2. P2Y12 also enhances platelet aggregation and both P2Y1 and
P2Y12 are required for full platelet aggregation………………………………………….7
Figure 1.3: Structures of P2X1 agonists. The native structure of ATP is at the top of the
image, whereas the modified structures of α,β-me-ATP and β,γ-me-ATP are shown at the
bottom of the image. The differences in structure from ATP are highlighted in red on
both structures……………………………………………………………………………..9
Figure 2.1: Structure of NF449. The P2X1 inhibitor, NF449, is pictured with
+
coordinating sodium ions (Na ), but in solution the polysulphonates are negatively
charged…………………………………………………………………………………...34
Figure 2.2: Schematic representation of experimental setup. Platelets were incubated with
2+
the Ca probe Fluo-4 AM for 30 min, then the excess probe was washed off and the
2+

platelets incubated with either buffer or NF449. Finally, measurements of Ca influx
were performed using a spectrofluorometer with a magnet setup so that samples could be
2+
stirred during measurement. Samples were kept in a Ca
free environment until
measurement……………………………………………………………………………..38
Figure 2.3: (Left) Fluorescence measurements taken over time for platelets in the
2+
presence and absence of NF449. No significant increase in Ca is measured when
CaMTB (trace C) or 2.5 uM ATP (trace B) is added to platelets alone. However, an
2+
increase in Ca was measured when 2.5 uM ATP was added to platelets containing 0.5
uM NF449 (trace A). (Right) The effect of ATP on platelet Ca

x

2+

influx with a constant

concentration of 2.5 uM NF449. As the concentration of ATP was increased from 73 nM
2+
up to 5 µM, the intracellular platelet Ca also increased…………………………...…..40
Figure 2.4: Effect of Ca

2+

free buffer on Ca

2+

influx in platelet containing 0.5 µM

NF449. (Left) 2.5 µM ATP was added as an agonist in the presence of Ca
2+

2+

as
2+

represented in trace A. The typical increase in Ca influx is seen. However, in a Ca
free environment (trace B), the addition of 2.5 µM ATP resulted in no significant
2+
2+
increase in Ca . Additionally, in the Ca free environment, ADP is still able to elicit a
2+

Ca

increase (trace C). (Right) The summarized data indicates that when Ca

2+

is not

2+

present, there is no Ca influx due to 2.5 µM ATP in platelets containing 0.5 µM NF449
or platelets alone. However, signal due to ADP stimulation is still present. Error is
standard error of the mean. N=9 (2.5 µM ATP, platelets), N=7 (2.5 µM ATP, NF449),
2+
2+
N=4 (2.5 µM ATP, Ca
free), and N=5 (2.5 µM ADP, Ca
free)……………………………………………………………………………………....42
2+

Figure 2.5: Effect of α,β-me-ATP on platelet Ca influx. Both A and B traces indicate
the presence of platelets containing 0.5 µM NF449, but the agonist for trace A is 2.5 µM
2+
ATP and the agonist for trace B is 2.5 µM α,β-me-ATP. Increases in Ca influx were
measured in the case of both agonists, but the increase due to ATP was larger than the
increase due to α,β-me-ATP. The two controls in this figure, represented as C, are
2+
platelets alone that were treated with both agonists, neither showed an increase in Ca
influx……………………………………………………………………………………..45
Figure 2.6: Effect of NF449 concentration on platelet Ca

2+

influx due to 2.5 µM ATP.

As NF449 concentration is increased from 0 to 0.5 µM, the Ca

2+

influx increases, but as
2+

the concentration of NF449 is further increased from 0.5 µM to 40 µM the Ca influx
decreases from the measured maximum. Error is standard error of the mean…………..45
Figure 2.7: Overcoming NF449 inhibition using 10 µM ATP. As the concentration of
2+
NF449 exceeds 0.5 µM, inhibition of Ca influx occurs when 2.5 µM ATP is added to
20 and 40 µM NF449 platelet samples (traces A). When 10 µM ATP is added, the
2+
NF449 inhibition is overcome and Ca influx occurs in 20 and 40 µM NF449 platelet
samples, traces B and C, respectively……………………………………………………47
Figure 2.8: NF449 potency over time with 2.5 µM ATP agonist. Using 62.5 nM NF449,
2+
platelet Ca influx was measured over the course of 6 hours with the data being
2+

normalized to the 5 min. percent change in signal. The Ca influx peaked at 5 min and
decreased steadily over time. Error is standard error of the mean………………………49
2+

Figure 2.9: The effect of P2Y1 inhibition on platelet Ca . As increasing concentrations
of MRS are added to the platelet solution it inhibits the Ca
xi

2+

release due to ADP binding

P2Y1. However, when NF449 is also added to the samples containing 60 µM MRS, the
2+
Ca signal from ATP and ADP recovered. P2Y1 also has some inhibitory effect on
2+

P2X1 so; the Ca signal from ATP is lower compared to the sample that was sensitized
with NF449 alone. Error is standard error of the mean. N is 4 or higher for all bars…...50
Figure 3.1: Fabrication of polystyrene devices with integrated channels. A silicon master
is coated with SU-8 50 photoresist, and exposed to UV light through a positive
photomask containing the serpentine features. The master is then developed and a PDMS
replicate of the master is made. This PDMS mold is then sealed to a piece of glass, and a
retaining ring of PDMS sealed along the outside. The PDMS is modified with 0.1 M (3mercaptopropyl)-trimethylsiloxane, and polystyrene melted into the mold at 185 °C for 9
hours. The completed device is then removed and a PDMS cover slab with punched
access holes is sealed to the polystyrene device through exposure to air
plasma.…………………………………………………………………………………...62
Figure 3.2: SEM Images of master, PDMS mold and PS device. Top left: image of
sample master with recessed channels. Top right: image of PDMS mold with raised
channels. Bottom: image of corresponding PS device made from PDMS mold. PS device
has recessed channels………………………………………………………………….....67
Figure 3.3: High-throughput micropipet pumping in polystyrene based microfluidic
device. Top-view of the device, where an 8-channel pipet can individually and
simultaneously address each channel. A pipet block is molded from immobilized 200 µL
pipet tips in PDMS. This is sealed to the cover slab of PDMS over an access hole. The
polystyrene device containing the serpentine channels is then sealed to the PDMS
components. The side view of the device shows how the pipet tips address each
serpentine channel..............................................................................................................68
Figure 3.4: Pumping, alignment and fluorescence analysis from a polystyrene device. A)
Fluorescein standards are pumped with a pipet into the device. B) The device is then
placed in a plate reader and fitted with a microplate as a guide. Fluorescence analysis is
performed with excitation of 494 nm and emission of 521 nm………………………….70
Figure 3.5: Image of fluorescein in serpentine channels. After fluorescein (ex. 494 nm,
em. 521 nm) was pumped through the serpentine channel, fluorescence microscopy was
used take an image of the channel. The “bright” portion is the fluorescein in the channel,
whereas the “dark” portion is the space between bends in the serpentine, thus, the channel
was successfully sealed and is not leaking………………………………………………71
Figure 4.1: Endothelial cell culture on a polystyrene device. To culture endothelial cells,
the device was first coated with fibronectin. Next, a suspension of endothelial cells was
pumped into the device. The device was then incubated at 37 °C, 5% CO 2 for 2 hours.
Media was changed every two hours until cell confluence was observed. A bright field
image is shown on the top right, while the image on the bottom right was obtained by
incubating the cells with CellTracker CMFDA fluorescenct probe……………………..87

xii

Figure 4.2: Intracellular measurement of NO. These cells were incubated with 10 µM
DAF-FM-DA, an intracellular NO probe, at 37 C for 30 min; excess probe was rinsed
off the cells by pumping buffer without probe through the channels. After the incubation
period, a basal fluorescence measurement was taken using a standard plate reader. Next,
0.5 µM ATP was pumped over the cells and allowed to incubate for an additional 30 min
in order to stimulate NO production in the endothelial cells. A final fluorescence
measurement was taken using the plate reader and the difference in fluorescence
calculated. This value was normalized to the difference in fluorescence for a 0 µM ATP
standard. The average changes in fluorescence for n=3 separate devices is shown in,
along with error bars representing the standard error of the mean. The changes are
significant for p<0.001…………………………………………………………………...85
Figure 4.3 Intracellular Measurement of Ca

2+

Uptake: These cells were incubated with

2+

5.7 µM Fluo 4AM, an intracellular Ca probe, at 37 C for 30 min; excess probe was
rinsed off the cells by pumping buffer without probe through the channels. After the
incubation period, a basal fluorescence measurement was taken using a standard plate
2+
reader. Next, 10 µM Ca ionophore was pumped over the cells and allowed to incubate
2+

for an additional 10 min in order to stimulate Ca uptake in the endothelial cells. A
final fluorescence measurement was taken using the plate reader and the difference in
fluorescence calculated. This value was normalized to the difference in fluorescence for
a 0 µM ionophore. The average changes in fluorescence for n=3 separate devices is
shown, along with error bars representing the standard error of the mean. The changes
are significant for p<0.05………………………………………………………………...90
Figure 4.4: Cell Viability. Saponin, a detergent that permeabilizes cell membranes, was
used to induce endothelial cell death as visualized by the nucleic acid stain Sytox Green.
Sytox Green does not label live cells thus, the image on the right indicates the presence of
dead cells after saponin administration. The bar graph shows an increase in fluorescence,
indicative of cell death, for cells treated with saponin as compared to cells treated with
PSS. The average changes in fluorescence for n=5 separate devices is shown, along with
error bars representing the standard error of the mean. The changes are significant for
p<0.001…………………………………………………………………………………..92
Figure 5.1: Membrane based polystyrene device. The bottom layer of this device is PS
containing embedded channels, while the top layer is a PDMS slab containing wells for
sample introduction (adapters), sampling (above membrane) and waste. Between these
two layers is a polyester membrane that allows for diffusion from the channel to above
the well. The luer adapters are attached to the device with a quick dry epoxy and this
allows for syringes to be used for sample pumping…………………………………….102
Figure 5.2: Polystyrene sealed to polystyrene device. The image on the left shows a two
layer device where the bottom layer has channels embedded in PS and the top blank layer
containing wells, also PS. These to pieces of PS was thermally sealed together using a
heated hydraulic press. To ensure that no channels were obstructed, fluorescein was

xiii

flowed though the channels, and the representative fluorescent image on the right shoes
that the channel is sealed, not leaking…………………………………………………..105
Figure 5.3: 3D printed device. The final 3D printed device contains adapters for syringe
based pumps, channels, membrane insertion port and outlets. The top image shows a side
view of the device where sample is pumped though the channel via tubing where it flows
under the membrane where molecules can diffuse. The bottom image shows the device
from the angle of sample entry where all 8 channels of the device can be used
simultaneously to address 8 membranes………………………………………………..113
Figure 5.4: 3D printed device side view. The side view of the device shows how the inlet
addresses the channel and allows fluid to flow under the membrane. The membrane is
part of a commercially available cell insert that is manually inserted into the port on top
of the device. Finally, there is an outlet to allow fluid to leave the device…………….114
Figure 5.5: Drug transport across a membrane. Standards of the drugs linezolid (N = 4)
and levofloxacin (N = 5) were flowed through the channels of the device; samples were
collected above the membrane and analyzed via MS. As concentration of the drug
increased so, did transport across the membrane with each concentration being
statistically different from the previous. (p value < 0.001)……………………………116
Figure 5.6: Incorporation of endothelial cells on 3D printed device. Endothelial cells
were incorporated onto the 3D printed device by culturing them onto the membrane
insert. These are commercially available cell inserts well suited for cell culture. Cells
were cultured on the inserts and stored in an incubator (37 ˚C, 5% CO2) until experiments
were performed. Cells were labeled with Hoechst dye, a nucleic acid stain, to confirm
their presence on the membrane as evidenced in the fluorescent image. Cell inserts were
then clicked into place on the device and samples flowed through the channel under the
membrane containing cells……………………………………………………………..118
Figure 5.7: Cell viability experiment. Saponin, a known cell detergent, was flowed under
endothelial cells to induce cell death. Sytox Green, which does not label live cells, was
used in order to determine the extent of cell death. The bar graph in part A shows an
increase in Sytox Green fluorescence when cells were treated with saponin compared to
buffer alone. The fluorescence images in part B represent cells treated with HBSS (left)
and cells treated with saponin (right) with the Sytox Green label……………………...119
Figure 5.8: Single channel 3D printed device prototype. The prototype shown in the top
image has the port situated in “24-well plate” format to enable future handling by
automated systems. Additionally, the bottom image shows the ability to print threads
directly into the device so adapters can be screwed in without the need of an additional
adapter or to seal with epoxy…………………………………………………………...120

xiv

CHAPTER 1: INTRODUCTION
1.1 The Platelet and the Vascular System
The human body contains between 5-6 L of whole blood, 55% of which is
comprised of plasma (water, glucose, proteins, electrolytes, hormones, clotting factors)
and 45% consisting of red blood cells (RBCs), platelets and leukocytes. The RBC’s
primary role in the bloodstream is oxygen delivery throughout the body, while leukocytes
protect the body from infection. In a blood vessel, RBCs flow in the middle of the vessel,
which forces platelets towards the edge of the vessel wall due to their size. This allows
for platelets to be closer to the site of potential vessel injuries. In general, platelets
circulate in the bloodstream in an inactive state, but upon vessel wall injury, platelets are
activated, initiating the clotting process, and forming a plug at the injury site. Platelets
are discoid in shape, anucleate and are approximately 2-4 µm in size.
9

A healthy

1

individual will have 150-400 x 10 platelets per mL of whole blood with the lifespan of
a platelet being 7-10 days.
1.1.1 Blood Vessel Injury, Platelet Activation and Aggregation
A typical blood vessel contains an inner lining of endothelial cells that serve as a
barrier between the whole blood and the outside tissue. On the outside of the vessel,
smooth muscle cells control the contraction and relaxation of the vessel. As shown in
figure 1.1, when a blood vessel is injured, sub endothelial cell collagen is exposed and
subsequently, platelets adhere to the injury, become activated and undergo a shape

1

Figure 1.1: Blood vessel injury. For clarity only RBCs and platelets are pictured. A typical blood vessel is comprised of an inner
layer of endothelial cells with smooth muscle cells lining the exterior. Upon blood vessel injury, subendothelial collagen is exposed
and platelets that come into contact with the collagen become activated, undergo a shape change that results in adhesion, and release
ATP and ADP into the bloodstream. As more platelets are recruited to the injury site, aggregation occurs and a plug over the injury
site is formed. NO, a platelet inhibitor, is also released by platelets to inhibit further platelet recruitment to the injury site. For
interpretation of the reference to color in this and all other figures, the reader is referred to the electronic version of this dissertation

2

change. Platelets initially bind to the newly exposed collagen (via the α2β1 integrin) directly, as
well as von Willebrand factor (vWF, via the GPIb-IX-V receptor), a plasma protein secreted by
endothelial cells that binds collagen.

2

Once more platelets are recruited to the injury site; the

platelets undergo a shape change where their normally smooth surface becomes adhesive and
elongated with projections. During this process, activated platelets undergo granular secretion
where additional molecules, released into the bloodstream, promote further aggregation of
activated platelets.
Platelets have three types of granules: alpha, lysosomes and dense. The alpha granules
3

secrete fibrinogen and vWF. The alpha granules also contain many proteins involved in platelet
4

adhesion, such as integrins and GPIb-IX-V.

The lysosomes contain enzymes involved in

protein, lipid and carbohydrate breakdown.

The contents of the dense granules are pro

aggregatory and include thromboxane A2 (TXA2), adenosine diphosphate (ADP), adenosine
2+

+

triphosphate (ATP), calcium (Ca ), potassium (K ) and serotonin.

5,6

Granular secretion is

2+ 7

caused by an increase in cytosolic Ca .

Once secretion occurs and more platelets are recruited to the injury site, the combination
of activating molecules, and increases in platelet Ca

2+

results in conformational changes in

integrins on the platelet surface that expose binding sites for fibrinogen.

8

Fibrinogen is what

allows the platelets to adhere to one another, but fibrin, vWF, and fibronectin can also promote
platelet aggregation. This is the start of the platelet plug, or clot, that will cover the entire injury
site and prevent further blood loss.
3

1.2 Platelet Inhibitors
In the vasculature, there are two endothelial cell derived platelet inhibitors, prostacyclin
(PGI2), and nitric oxide (NO), which result in elevated intraplatelet cyclic AMP and cyclic
GMP, respectively, while another inhibitor is CD39, an ectonucleotidase.

9

Kinases dependent

on cAMP and cGMP will affect several platelet signaling mechanisms, including Ca
10

2+

mobilization.

Ca

is an important signaling molecule in platelets and is needed for

cytoskeleton rearrangement necessary for shape change,
aggregation,

12

2+

and for NO production within the platelet.

11

integrin activation necessary for

13

1.2.1 Nitric Oxide (NO)
NO is produced when endothelial nitric oxide synthase (eNOS) converts L-arginine to Lcitrulline producing NO as a byproduct. Endothelial derived NO is not only a platelet inhibitor,
but also promotes vasodilation, or widening of the blood vessel, via smooth muscle cell
relaxation.

14,15

NO inhibits platelet function by binding to soluble guanylyl cyclase

in elevated levels of cGMP,

17, 18

which activates protein kinase G (PKG).

mechanism called “store operated Ca
medium.

19

2+

entry” where Ca

2+

9

16

resulting

PKG affects a

enters from the extracellular

Elevated cGMP levels inhibit phosphodiesterase III (PDE III) and leads to an

increase in cyclic adenosine monophosphate (cAMP) levels
22

kinase A (PKA).

20,21

causing activation of protein

Elevated levels of PKA inhibit protein kinase C (PKC) and phospholipase C

(PLC) activations and thus, intracellular Ca

2+

4

23,24

mobilization.

This inhibition triggers a

2+ 25

decrease in cytosolic Ca ,

which results in the inhibition of GPIIb-IIIa’s ability to change

conformations into its active form, which normally would allow for platelet interaction with
26,27

fibrinogen.

Furthermore, the eNOS pathway for NO synthesis also occurs in platelets.

Platelet derived NO is also released into the bloodstream before platelets are activated
during aggregation.

13,28

29

28

and

The platelet derived NO (micromolar) is released during aggregation

and prevents further platelet aggregation and adhesion to the injury site.

13,30

1.2.2 Prostacyclin (PGI2)
Prostacyclin binds to platelets via the prostacyclin receptor (IP receptor), which is only
expressed on platelets

31

and smooth muscle cells.

32

It inhibits platelet activation induced by
33

ADP, thrombin, collagen and TXA2, however, it primarily works through TXA2.

It has been suggested that P2X1-, a purinergic receptor on the platelet surface, evoked
2+

Ca

response may be less sensitive to PGI2 and NO inhibition by endothelial cells after agonist
34

exposure

when compared to the store operated Ca

2+

release.

35

This would explain why

platelets are able to respond immediately to endothelial damage given their close proximity to
endothelial cells and the inhibitors they release.
1.2.3 Ectonucleotidase CD39
The endothelial enzyme CD39 (NTPDase-1) is also able to inhibit platelet function by
converting ATP and ADP to inert AMP

36

and eventually, adenosine by CD73.

5

37

CD39 does not

directly interact with the platelet itself, but it does interact with the ADP and ATP that platelets
release from their granules.

36

1.3 Purinergic Receptors on the Platelet Surface
Purinergic receptors on the platelet surface regulate activation by ATP or ADP binding.
A simplified version of P2 receptor activation is shown in figure 1.2. Two receptors, P2Y1 and
P2Y12 are G-protein coupled and bind ADP, while the third, P2X1, binds ATP, exclusively.
ADP binds the P2Y1 receptor leading to Gq mediated activation of PLC, which leads to Ca
release intracellular stores, shape change, and reversible aggregation.

38,39

2+

When ADP binds to

Gi associated P2Y12, it leads to inhibition of adenylyl cyclase (AC) and enhanced aggregation.
PGI2 inhibits platelet activation by increasing platelet cAMP levels via AC activation.

40

38

Since

P2Y12 inhibits AC, this is in opposition to the endothelial cell release inhibitor PGI2. Activation
of both P2Y1 and P2Y12 are needed for platelets to exhibit a normal aggregation response.
Inhibition of either receptor is enough to block ADP-induced platelet aggregation.

41

42

The P2X1 receptor, one of seven sub types of P2X receptors , on the platelet surface is
an ATP-gated cation channel that contains two transmembrane domains linked by a large
extracellular loop.

43,44

The extracellular loop contains the binding site for ATP and competitive

antagonists, while the transmembrane portion of the receptor forms the cation channel.

45

While

there is not a solved structure for P2X1, X-ray crystallography has been used to determine the
structure of P2X4 in the closed state

43

and with bound ATP.
6

46

P2X1 is the only P2X receptor

Figure 1.2: P2 receptors on platelet membrane. The platelet membrane has three purinergic
receptors on its membrane surface. The P2X1 is a gated cation channel that is activated by ATP.
2+
Upon ATP binding there is a rapid Ca influx into the platelet. P2Y1 and P2Y12 are both Gprotein coupled receptors that bind ADP. When ADP binds P2Y1 it activates PLC, which causes
2+
intracellular Ca release and subsequent platelet activation. When ADP binds P2Y12 it inhibits
AC activity, thus, counteracting the inhibitory effects of PGI2. P2Y12 also enhances platelet
aggregation and both P2Y1 and P2Y12 are required for full platelet aggregation.

7

found in significant quantities on the platelet surface

47

and since ATP binding is the only factor

needed for channel opening, P2X1 is also a significant source of rapid Ca
platelet following activation.

48

Additionally, the Ca

2+

influx into the

influx also leads to membrane

depolarization, which is thought to enhance signaling through P2Y1.
thought that P2X1 was not needed for proper platelet function,
this is not the case.

2+

50

49

Previously, it was

although recent reports suggest

51-53

Studying the P2X1 receptor’s contribution to platelet function is difficult because it undergoes
rapid desensitization during purification from whole blood due to spontaneously secreted ATP.

54

By combating the desensitization of P2X1 with a high dose of apyrase and using α,β-methyleneATP (α,β-me-ATP) or β,γ-methylene ATP (β,γ-me-ATP) as the agonist, activation of P2X1 leads
to rapid influx of Ca

2+

and transient shape change.

54

The structures of ATP, α,β-me-ATP, and

β,γ-me-ATP are shown in figure 1.3 where the difference from ATP is the replacement of
oxygen with –CH2 so that it cannot be broken down by apyrase. It has also been shown that
ATP can stimulate NO production in the platelet and induce platelet aggregation.
Additionally, α,β-me-ATP has also been shown to desensitize the P2X1 receptor.

55

51

There is also

literature that suggests a role for P2X1 in early activation of platelets with low concentrations of
collagen.

52, 53

coincidentally,

A number of disease states are associated with hyperactive platelets and
also

have

bioavailability

8

issues

with

regard

to

ATP.

Figure 1.3: Structures of P2X1 agonists. The native structure of ATP is at the top of the image,
whereas the modified structures of α,β-me-ATP and β,γ-me-ATP are shown at the bottom of the
image. The differences in structure from ATP are highlighted in red on both structures.

9

1.4 Diseases with Hyperactive Platelets
It has been shown that platelets exhibit hyperactivity and are prone to excessive
aggregation in certain disease states such as diabetes, pulmonary hypertension, cystic fibrosis
(CF), sickle cell anemia and multiple sclerosis (MS). This hyperactivity can lead to unnecessary
increased thrombus formation and subsequent blocking of blood vessels resulting in potential
heart attack or stroke.
1.4.1 Diabetes
Patients with diabetes have been described as having hyperactive platelets. Diabetes is
characterized by either lack of insulin secretion (type 1 diabetes) or a resistance to insulin (type 2
diabetes). Insulin is produced and stored in beta cells present in the islets of Langerhans region
of the pancreas. In response to such stimuli as elevated glucose levels, exocytosis of insulin
occurs depositing insulin in the bloodstream.

56

Patients with diabetes experience complications

associated with blood flow, heart disease and stroke.

57

The problems with blood flow may be

associated with hyperactive platelets. It has been shown that diabetic platelet activation is
enhanced with response to epinephrine, ADP and thrombin.
been shown to be resistant to inhibition by NO

61

58,59,60

and PGI2.

62

Diabetic platelets have also

Patients with diabetes also have

high platelet turnover, which results in more circulating reticulated, or young, platelets. These
reticulated platelets are larger and more reactive to activation stimuli than regular platelets.

63

Additionally, the reticulated platelets are resistant to the effects of aspirin and clopidogrel, which
are two common antiplatelet drugs.

64,65

10

1.4.2 Hypertension
Hypertension, or high blood pressure, is another condition that is characterized by
hyperactive platelets.

Platelets from patients with hypertension exhibit increased in vivo

aggregation in response to ADP, epinephrine and collagen
2+ 67

Ca

and losses in NO bioavailability.

67

66

as well as high intracellular free

The susceptibility of platelets in hypertensive

patients to aggregate can lead to increased risk of heart attack, as well as stroke.
1.4.3 Cystic Fibrosis
Cystic fibrosis (CF) is one of the most common hereditary disorders, especially in the
Caucasian population.

68

CF is a genetic disorder, which results in dysfunctional cystic fibrosis

transmembrane conductance regulator (CFTR) protein.
chloride channel present in the cell membrane.

69

68

The CFTR protein is identified as a

Certain mutations of the CFTR protein make it

inactive, which affects its functions, including chloride and ATP conductance.

70

RBCs express the CFTR protein while platelets do not and CF platelets have been
71,72

described as having increased aggregability ex vivo.

Additionally, it has been reported that
73

platelets from CF patients have increased TXA2 release
prostaglandin E1 (PGE1), a platelet inhibitor.

74

and do not respond well to

Furthermore, it has been shown that CF plasma

causes activation of CF platelets as well as healthy platelets, though activation of the CF platelets
was higher than that of the normal.

71

11

1.4.4 Sickle Cell Anemia
Sickle cell anemia is a blood disorder where RBCs sickle under deoxygenated conditions
obstructing capillaries and restricting blood flow.
“hypercoaguable state”

76

75

Sickle cell disease has been described as a

where platelet activation is abnormal. Plasma samples taken from

sickle cell patients during steady state (non-sickled RBCs) and crisis (sickled RBCs) show
increased levels of thrombin and depletion of anticoagulant proteins suggesting that the platelets
are hyperactive.

77

1.4.5 Multiple Sclerosis
A final disease state where patients exhibit platelet hyperactivity is in multiple sclerosis
(MS). MS is a neurological disease that results in demyelination and scarring of nerves in the
brain and spinal cord, which can lead to loss of nerve function, muscle weakness, and problems
78

speaking.

Abnormalities in platelets from patients with MS were first described in 1965 as

increased platelet stickiness.

79

It has been studied furthered and has been determined that MS

platelets exhibit increased platelet microparticles (PMP) and P-selectin expression when
compared to healthy controls. Both PMP and P-selectin are recognized indicators of platelet
activation.

80

1.4.6 Defects in RBC ATP release
In addition to hyperactive platelets, the diseases mentioned above all have defects in RBC
81

derived ATP release. RBCs from patients with CF , diabetes

12

82

and hypertension

83

all exhibit

decreased ATP release while those RBCs from MS

84

and sickle cell patients

85

have increased

ATP levels in the extracellular space due to RBC release and cell lysis, respectively. Platelets
circulate throughout the body in the presence of RBCs and it has been shown that RBC derived
ATP release does have an effect on platelet adhesion to an adhered endothelium at both low and
high concentrations of ATP.

51

Since P2X1 is the only ATP receptor on the platelet, it is possible

that RBC derived ATP may influence platelet activity via P2X1. The Spence group has shown
through platelet aggregation studies on washed platelets and PRP that ATP can induce platelet
aggregation; however, it is difficult to distinguish the effects of ATP from ADP, the immediate
hydrolyzed product of ATP. Ectonucleotidases in the plasma and on the platelet surface convert
ADP to ATP so; isolating the platelet response to ATP, ADP, or both is a challenge. Non
hydrolysable forms of ATP are often used to study the P2X1 receptor’s response, but these forms
of ATP do not interact with the receptor in the same manner as native ATP. Additionally, in
order to prove that hyperactive platelets may be a result of abnormal RBC derived ATP release
affecting the platelet P2X1 receptor it is necessary to consider each component of the vasculature
(RBCs, platelets, endothelial cells) and how each component contributes to platelet activation.
Consequently, it is important to have methods to study platelets in an in vitro setting where each
component of the vasculature can be isolated and probed.
1.5 Measuring Platelet Activity in Vitro
1.5.1 Aggregometry
One of the most common ways to study platelet function is by using light transmission
aggregometry (LTA). In its simplest form, LTA measures changes in optical density of platelet
solutions when exposed to individual agonists. Platelets are usually suspended in plasma, but
13

measurements can also be performed on suspensions of washed platelets. Measurements are
carried out in a clear cuvette, with stirring at 37 ˚C.

86

For setting the lower and upper limits for

LTA using samples in PRP, the 0% and 100% aggregation are calibrated to the PRP and platelet
poor plasma (PPP), respectively. Aggregation is then monitored by changes in light transmission
87

before and after addition of an agonist. .

Another form of aggregometry involves whole blood samples where changes in electrical
impedance (Ohms) are measured. In principle, a single layer of platelets attach to the electrodes
prior to aggregation and after an agonist is added, aggregating platelets attach at the electrodes
thus causing changes in impedance at the electrode.

88, 89

The inherent advantage to using whole

blood aggregometry is that all blood cells are present during the experiment, which more reflects
an in vivo situation.
For both the whole blood aggregometry and the LTA, dense granule release from
platelets can also be simultaneously monitored, specifically, ATP via a chemiluminescent
reaction with D-luciferin/firefly luciferase. The emitted light can be compared to ATP standards
for quantification purposes.

90

This technique is useful because it monitors both the

morphological changes that platelets undergo and granular release, which can further activate
platelets.
1.5.2 Flow Cytometry and Fluorescence
Flow cytometry is a powerful tool for studying platelets as it can measure multiple assays
and various cell types at once. Before analysis by flow cytometry, cells are labeled with the
desired fluorescent tag, which is often conjugated to a cell or protein specific antibody. Then the
14

cells in suspension flow one by one through a laser beam where the fluorophores are excited, and
the emitted light collected. The intensity of the emission is directly proportional to the antigen or
cell type being studied, as well as size and granularity of the cell.
One of the biggest advantages of flow cytometry is that platelets can be analyzed in the
presence of other blood components, such as RBCs and leukocytes, which can affect platelet
activation.

91,92

Furthermore, there are a number of antibodies available that bind activated

platelets and not resting platelets. A few examples include 2G5
94

fibrinogen), IAC-1

(collagen binding site exposure), and S12

95

93

(receptor site on bound

(P-selectin exposure on the

membrane).
2+

Ca , an important signaling molecule in the platelet, is also studied using flow
cytometry and fluorescence techniques. One of the most widely used probes to study Ca
Fura-2, a ratiometric dye used as an intracellular Ca

2+

probe.

96

2+

When Ca

2+

is

binds Fura-2, its

absorption shifts to a shorter wavelength, thus, the ratio of unbound to bound Fura-2 can be used
2+ 97

to calculate the concentration of Ca .
AM

98

and Fluo-4 AM

99

Other commonly used Ca

2+

probes include Fluo-3

where the AM ester group allows the dye to be loaded into the cell and

subsequently, cleaved by intracellular esterases and the resulting dye is able to bind intracellular
2+

Ca .

15

1.5.3 Microfluidic Devices
One of the inherent disadvantages of in vitro assays performed on platelet activity is the
lack of a flow component and presence of other cells common to the vasculature. Through the
use of microfluidic devices it is possible to remedy this disadvantage. Flow perfusion chambers
were the first in vitro flow devices used to study platelets, where, typically, a glass coverslip is
coated with a platelet activator or endothelial cells and thrombus formation at that interface can
be monitored at various shear rates.
thrombus formation

100

Simple devices such as this have been used to study

101

, as well as integrin activation

102

in the platelet. These flow chambers

have some disadvantages, including, lack of standardization in chamber fabrication
need for large volumes of blood (> 5 mL) to perform multiple assays.

104

103

and the

Microfluidic

technology is one way to overcome these disadvantages.
Typically, microfluidic devices are fabricated using photolithography and soft
lithography with the end device being made of polydimethylsiloxane and having channel
dimensions anywhere from 2-100 µm. PDMS is easy to use, inexpensive and biologically
compatible so, it can be used to research thrombus formation. In brief, a master is created by
photolithography, where a polymerizable SU-8 photoresist is spin coated onto a silicon wafer, a
transparent mask with a pattern of channels is placed on the coated wafer, and exposed to UV
light. The SU-8 that is exposed to the light polymerizes and the wafer developed to remove
excess photoresist resulting in a master with positive features. A mixture of PDMS and curing
agent is then poured over the master, allowed to cure, and peeled off resulting in negative
features on the PDMS. These features tend to be the channels and any inlets, access ports, or

16

waste wells are typically punched right into the PDMS. The final PDMS channel layer can be
sealed with glass or a blank layer of PDMS and the end device is able to incorporate flow across
a single or multiple channel(s).

105

Single channel microfluidic devices have been used to study collagen induced-thrombus
formation under flow conditions,
mimic,

107,108

106

anti-platelet drugs and shear gradients with a stenosis

and measurement of platelet size.

109

Microfluidics can also be used in a high

throughput manner where multiple channels are used to monitor platelet activity. Devices with
multiple channels have been used to study platelet adhesion to an endothelium,
adhesion/signaling under flow conditions,

111

113

effect of anti-platelet drugs during flow.

110,51

collagen induced thrombus formation,

112

platelet
and the

These microfluidic devices are designed to study

various aspects of thrombus formation on an in vitro platform.
1.5.4 Apyrase and α,β-me-ATP in Platelet Preparation and Analysis
As mentioned previously, the platelet P2X1 receptor is prone to rapid desensitization in
vitro due to spontaneously secreted ATP.

54

In order to reduce the effect of desensitization,

researchers generally employ apyrase, which is derived from potato and acts as an ADPase and
ATPase. In order for apyrase to protect the P2X1 receptor from desensitization, concentrations
of 0.3-1 U/ml are used during sample preparation.

53

A higher concentrations of apyrase, 5

U/mL, was used in order to study platelet shape change with α,β-me-ATP as the agonist.

114

Since apyrase is most often used when studying the P2X1 receptor, agonists that are resistant to

17

hydrolysis must be employed to probe receptor function.

Typically, α,β-me-ATP and β,γ

methylene ATP (β,γ-me-ATP) are used as P2X1 agonists, however, neither α,β-me-ATP or β,γme-ATP are as potent as native ATP when it comes to P2X1 response.

115,116

One of the inherent disadvantages in using apyrase in platelet solutions is that it breaks
down ATP/ADP, which is what forces researchers to use α,β-me-ATP or β,γ-me-ATP as these
analogues cannot be broken down by apyrase. However, these molecules do not represent the
true in vivo activation of the P2X1 receptor. It is possible that the contribution of the P2X1
receptor to platelet activation and aggregation has previously been underestimated.
In order to study the P2X1 receptor a method that uses native ATP, without the presence
of added apyrase is necessary. Furthermore, the role of RBC derived ATP (a significant source
of ATP in the bloodstream) has not been fully investigated as a platelet regulator via the P2X1
receptor.

Thus, a novel way to investigate P2X1 will be investigated, as well as novel

microfluidic technology capable of probing these cellular responses and further analyzing the
role of RBC derived ATP release in platelet activation.

18

REFERENCES

19

REFERENCES
1.
A. D. Michelson, How Platelets Work: Platelet Function and Dysfunction. J.
Thromb. Thrombolysis 2004, 16. 7-12.
2.
V. T. Turitto, H. J. Weiss, T. S. Zimmerman, I. I. Sussman, Factor VIII/von
Willebrand factor in subendothelium mediates platelet adhesion. Blood 1985, 65. 823-31.
3.
H. R. Gralnick, S. B. Williams, L. P. Mckeown, D. M. Krizek, B. C. Shafer, M. E.
Rick, Platelet Vonwillebrand-Factor - Comparison with Plasma Vonwillebrand-Factor.
Thromb. Res. 1985, 38. 623-633, DOI: Doi 10.1016/0049-3848(85)90205-1.
4.
D. M. Maynard, H. F. G. Heijnen, M. K. Horne, J. G. White, W. A. Gahl,
Proteomic analysis of platelet alpha-granules using mass spectrometry. J. Thromb.
Haemostasis 2007, 5. 1945-1955, DOI: DOI 10.1111/j.1538-7836.2007.02690.x.
5.
H. Holmsen, H. J. Weiss, Secretable storage pools in platelets. Annu. Rev. Med.
1979, 30. 119-34.
6.
C. S. Mckinney, E. D. Ozkan, R. Zafar, P. M. Kelley, D. Njus, D. A. Walz,
Characterization of Serotonin Transport in Human Platelet Dense Granules. Thromb
Haemost 1993, 69. 549-549.
7.
D. E. Knight, M. C. Scrutton, Direct Evidence for a Role for Ca2+ in Amine
Storage Granule Secretion by Human-Platelets. Thromb. Res. 1980, 20. 437-446, DOI:
Doi 10.1016/0049-3848(80)90282-0.
8.
S. J. Shattil, C. Kim, M. H. Ginsberg, The final steps of integrin activation: the
end game. Nat. Rev. Mol. Cell Biol. 11. 288-300.
9.
U. R. Schwarz, U. Walter, M. Eigenthaler, Taming platelets with cyclic
nucleotides. Biochemical Pharmacology 2001, 62. 1153-1161.
10.
D. Varga-Szabo, A. Braun, B. Nieswandt, Calcium signaling in platelets. J.
Thromb. Haemostasis 2009, 7. 1057-1066.
11.
D. R. Hathaway, R. S. Adelstein, Human platelet myosin light chain kinase
requires the calcium-binding protein calmodulin for activity. Proceedings of the National
Academy of Sciences of the United States of America 1979, 76. 1653-7.
12.
S. J. Shattil, L. F. Brass, Induction of the fibrinogen receptor on human platelets
by intracellular mediators. Journal of Biological Chemistry 1987, 262. 992-1000.
13.
J. E. Freedman, J. Loscalzo, M. R. Barnard, C. Alpert, J. F. Keaney, Jr., A. D.
Michelson, Nitric oxide released from activated platelets inhibits platelet recruitment.
Journal of Clinical Investigation 1997, 100. 350-356.
20

14.
L. J. Ignarro, G. M. Buga, K. S. Wood, R. E. Byrns, G. Chaudhuri, Endotheliumderived relaxing factor produced and released from artery and vein is nitric oxide.
Proceedings of the National Academy of Sciences of the United States of America 1987,
84. 9265-9.
15.
L. J. Ignarro, R. E. Byrns, G. M. Buga, K. S. Wood, Endothelium-derived
relaxing factor from pulmonary artery and vein possesses pharmacologic and chemical
properties identical to those of nitric oxide radical. Circ. Res. 1987, 61. 866-79.
16.
T. C. Bellamy, J. Garthwaite, The receptor-like properties of nitric oxideactivated soluble guanylyl cyclase in intact cells. Molecular and Cellular Biochemistry
2002, 230. 165-176, DOI: Doi 10.1023/A:1014204426331.
17.
F. Murad, C. K. Mittal, W. P. Arnold, S. Katsuki, H. Kimura, Guanylate cyclase:
activation by azide, nitro compounds, nitric oxide, and hydroxyl radical and inhibition by
hemoglobin and myoglobin. Adv Cyclic Nucleotide Res 1978, 9. 145-58.
18.
T. D. Warner, Influence of endothelial mediators on the vascular smooth muscle
and circulating platelets and blood cells. International angiology : a journal of the
International Union of Angiology 1996, 15. 93-9.
19.
J. A. Rosado, S. Jenner, S. O. Sage, A role for the actin cytoskeleton in the
initiation and maintenance of store-mediated calcium entry in human platelets. Evidence
for conformational coupling. Journal of Biological Chemistry 2000, 275. 7527-7533.
20.
J. A. Beavo, Cyclic nucleotide phosphodiesterases: functional implications of
multiple isoforms. Physiological Reviews 1995, 75. 725-48.
21.
H. H. Schmidt, S. M. Lohmann, U. Walter, The nitric oxide and cGMP signal
transduction system: regulation and mechanism of action. Biochimica et Biophysica Acta
1993, 1178. 153-75.
22.
D. A. Wink, J. B. Mitchell, Chemical biology of nitric oxide: Insights into
regulatory, cytotoxic, and cytoprotective mechanisms of nitric oxide. Free Radical
Biology & Medicine 1998, 25. 434-56.
23.
D. R. Riddell, J. S. Owen, Nitric oxide and platelet aggregation. Vitam Horm
1999, 57. 25-48.
24.
R. Feil, S. M. Lohmann, H. de Jonge, U. Walter, F. Hofmann, - Cyclic GMPdependent protein kinases and the cardiovascular system. Circ Res 2003, 93. 907-16.
25.
G. H. Rao, S. Krishnamurthi, L. Raij, J. G. White, Influence of nitric oxide on
agonist-mediated calcium mobilization in platelets. Biochemical Medicine and Metabolic
Biology 1990, 43. 271-5.

21

26.
M. E. Mendelsohn, S. Oneill, D. George, J. Loscalzo, Inhibition of Fibrinogen
Binding to Human Platelets by S-Nitroso-N-Acetylcysteine. Journal of Biological
Chemistry 1990, 265. 19028-19034.
27.
A. D. Michelson, S. E. Benoit, M. I. Furman, W. L. Breckwoldt, M. J. Rohrer, M.
R. Barnard, J. Loscalzo, Effects of nitric oxide/EDRF on platelet surface glycoproteins.
Am J Physiol 1996, 270. H1640-8.
28.
M. W. Radomski, R. M. Palmer, S. Moncada, An L-arginine/nitric oxide pathway
present in human platelets regulates aggregation. Proceedings of the National Academy of
Sciences of the United States of America 1990, 87. 5193-7.
29.
Q. S. Zhou, G. R. Hellermann, L. P. Solomonson, Nitric-Oxide Release from
Resting Human Platelets. Thromb. Res. 1995, 77. 87-96, DOI: Doi 10.1016/00493848(95)90868-G.
30.
J. E. Freedman, R. Sauter, E. M. Battinelli, K. Ault, C. Knowles, P. L. Huang, J.
Loscalzo, Deficient platelet-derived nitric oxide and enhanced hemostasis in mice lacking
the NOSIII gene. Circ. Res. 1999, 84. 1416-1421.
31.
A. K. Dutta-Roy, A. K. Sinha, Purification and properties of prostaglandin
E1/prostacyclin receptor of human blood platelets. J Biol Chem 1987, 262. 12685-91.
32.
R. L. Jones, Y. M. Qian, H. N. C. Wong, H. W. Chan, A. P. C. Yim, Prostanoid
action on the human pulmonary vascular system. Clinical and Experimental
Pharmacology and Physiology 1997, 24. 969-972, DOI: DOI 10.1111/j.14401681.1997.tb02730.x.
33.
J. M. Manganello, Y. Djellas, C. Borg, K. Antonakis, G. C. Le Breton, Cyclic
AMP-dependent phosphorylation of thromboxane A2 receptor-associated Gα13. Journal
of Biological Chemistry 1999, 274. 28003-28010.
34.
C. Y. E. Fung, S. Jones, A. Ntrakwah, K. M. Naseem, R. W. Farndale, M. P.
Mahaut-Smith, Platelet Ca2+ responses coupled to glycoprotein VI and Toll-like
receptors persist in the presence of endothelial-derived inhibitors: roles for secondary
activation of P2X1 receptors and release from intracellular Ca2+ stores. Blood 119. 36133621.
35.
J. Geiger, C. Nolte, E. Butt, S. O. Sage, U. Walter, Role of cGMP and cGMPdependent protein kinase in nitrovasodilator inhibition of agonist-evoked calcium
elevation in human platelets. Proceedings of the National Academy of Sciences of the
United States of America 1992, 89. 1031-5.
36.
A. J. Marcus, M. J. Broekman, J. H. F. Drosopoulos, N. Islam, T. N.
Alyonycheva, L. B. Safier, K. A. Hajjar, D. N. Posnett, M. A. Schoenborn, K. A.
Schooley, R. B. Gayle, C. R. Maliszewski, The endothelial cell ecto-ADPase responsible
22

for inhibition of platelet function is CD39. Journal of Clinical Investigation 1997, 99.
1351-1360.
37.
H. Zimmermann, 5'-Nucleotidase: molecular structure and functional aspects.
Biochem J 1992, 285 ( Pt 2). 345-65.
38.
M. Cattaneo, The P2 receptors and congenital platelet function defects. Semin.
Thromb. Hemostasis 2005, 31. 168-173.
39.
S. O. Sage, R. Reast, T. J. Rink, ADP evokes biphasic calcium influx in fura-2loaded human platelets. Evidence for calcium entry regulated by the intracellular calcium
store. Biochemical Journal 1990, 265. 675-80.
40.
M. Cattaneo, A. Lecchi, Inhibition of the platelet P2Y12 receptor for adenosine
diphosphate potentiates the antiplatelet effect of prostacyclin. J. Thromb. Haemostasis
2007, 5. 577-582.
41.
J. Jin, S. P. Kunapuli, Coactivation of two different G protein-coupled receptors is
essential for ADP-induced platelet aggregation. Proceedings of the National Academy of
Sciences of the United States of America 1998, 95. 8070-8074.
42.
R. A. North, Molecular physiology of P2X receptors. Physiological Reviews
2002, 82. 1013-1067.
43.
T. Kawate, J. C. Michel, W. T. Birdsong, E. Gouaux, Crystal structure of the
ATP-gated P2X4 ion channel in the closed state. Nature (London, United Kingdom)
2009, 460. 592-598.
44.
A. J. Brake, M. J. Wagenbach, D. Julius, New structural motif for ligand-gated
ion channels defined by an ionotropic ATP receptor. Nature (London, United Kingdom)
1994, 371. 519-23.
45.
L. E. Browne, L.-H. Jiang, R. A. North, New structure enlivens interest in P2X
receptors. Trends in Pharmacological Sciences 31. 229-237.
46.
M. Hattori, E. Gouaux, Molecular mechanism of ATP binding and ion channel
activation in P2X receptors. Nature (London, United Kingdom) 485. 207-212.
47.
L. Wang, O. Oestberg, A. K. Wihlborg, H. Brogren, S. Jern, D. Erlinge,
Quantification of ADP and ATP receptor expression in human platelets. J. Thromb.
Haemostasis 2003, 1. 330-336.
48.
S. O. Sage, T. J. Rink, The kinetics of changes in intracellular calcium
concentration in fura-2-loaded human platelets. Journal of Biological Chemistry 1987,
262. 16364-9.

23

49.
J. Martinez-Pinna, I. S. Gurung, C. Vial, C. Leon, C. Gachet, R. J. Evans, M. P.
Mahaut-Smith, Direct Voltage Control of Signaling via P2Y1 and Other Gαq-coupled
Receptors. Journal of Biological Chemistry 2005, 280. 1490-1498.
50.
S. Takano, J. Kimura, I. Matsuoka, T. Ono, No requirement of P2X1
purinoceptors for platelet aggregation. European Journal of Pharmacology 1999, 372.
305-309.
51.
W. Karunarathne, C. J. Ku, D. M. Spence, The dual nature of extracellular ATP as
a concentration-dependent platelet P2X1 agonist and antagonist. Integr Biol 2009, 1. 655663, DOI: Doi 10.1039/B909873a.
52.
C. Y. Fung, C. A. Brearley, R. W. Farndale, M. P. Mahaut-Smith, A major role
for P2X1 receptors in the early collagen-evoked intracellular Ca2+ responses of human
platelets. Thromb Haemost 2005, 94. 37-40.
53.
C. Y. E. Fung, C. Cendana, R. W. Farndale, M. P. Mahaut-Smith, Primary and
secondary agonists can use P2X1 receptors as a major pathway to increase intracellular
Ca2+ in the human platelet. J. Thromb. Haemostasis 2007, 5. 910-917.
54.
M. G. Rolf, C. A. Brearley, M. P. Mahaut-Smith, Platelet shape change evoked by
selective activation of P2X1 purinoceptors with alpha ,beta -methylene ATP. Thromb
Haemost 2001, 85. 303-308.
55.
H. H. Dalziel, G. A. Gray, R. M. Drummond, B. L. Furman, P. Sneddon,
Investigation of the selectivity of alpha , beta -methylene ATP in inhibiting vascular
responses of the rat in vivo and in vitro. Br J Pharmacol 1990, 99. 820-4.
56.
F. Chimienti, A. Favier, M. Seve, ZnT-8, a pancreatic beta-cell-specific zinc
transporter. Biometals 2005, 18. 313-317, DOI: DOI 10.1007/s10534-005-3687-9.
57.
A. I. Vinik, T. Erbas, T. S. Park, R. Nolan, G. L. Pittenger, Platelet dysfunction in
type 2 diabetes. Diabetes Care 2001, 24. 1476-1485.
58.
D. J. Betteridge, K. E. El Tahir, J. P. Reckless, K. I. Williams, Platelets from
diabetic subjects show diminished sensitivity to prostacyclin. Eur J Clin Invest 1982, 12.
395-8.
59.
H. E. Harrison, A. H. Reece, M. Johnson, Effect of insulin treatment on
prostacyclin in experimental diabetes. Diabetologia 1980, 18. 65-8.
60.
J. Kutti, H. Wadenvik, B. Henestam, G. Stenstrom, Evaluation of platelet
reactivity in diabetes mellitus. Acta Med Scand 1986, 219. 195-9.
61.
N. N. Chan, P. Vallance, H. M. Colhoun, Nitric oxide and vascular responses in
Type I diabetes. Diabetologia 2000, 43. 137-47.
24

62.
G. Davi, G. B. Rini, M. Averna, S. Novo, G. Difede, A. Pinto, A. Notarbartolo, A.
Strano, Thromboxane-B2 Formation and Platelet Sensitivity to Prostacyclin in InsulinDependent and Insulin-Independent Diabetics. Thromb. Res. 1982, 26. 359-370, DOI:
Doi 10.1016/0049-3848(82)90254-7.
63.
D. Tschoepe, P. Roesen, J. Esser, B. Schwippert, H. K. Nieuwenhuis, B. Kehrel,
F. A. Gries, Large Platelets Circulate in an Activated State in Diabetes-Mellitus. Semin.
Thromb. Hemostasis 1991, 17. 433-438, DOI: DOI 10.1055/s-2007-1002650.
64.
S. Guthikonda, E. I. Lev, R. Patel, T. Delao, A. L. Bergeron, J. F. Dong, N. S.
Kleiman, Reticulated platelets and uninhibited COX-1 and COX-2 decrease the
antiplatelet effects of aspirin. J. Thromb. Haemostasis 2007, 5. 490-496, DOI: DOI
10.1111/j.1538-7836.2007.02387.x.
65.
C. L. Alviar, M. Vaduganathan, A. Tellez, K. Rammohan, B. Nure, T. Delao, M.
Arikan, J. F. Granada, E. I. Lev, N. S. Kleiman, S. Guthikonda, Role of reticulated
platelets and platelet size in modifying anti-platelet effects of aspirin and clopidogrel in
patients with stable coronary artery disease. Circ. Res. 2007, 101. 1210-1210.
66.
L. Somova, J. Mufunda, Platelet activity and salt sensitivity in the pathogenesis of
systemic (essential) hypertension in black Africans. Clin Exp Hypertens 1993, 15. 78196.
67.
M. El Haouari, J. A. Rosado, Platelet function in hypertension. Blood Cells, Mol.,
Dis. 2009, 42. 38-43.
68.
G. R. Cutting, Cystic fibrosis: using genetic association to identify modifiers of
disease variability in cystic fibrosis. European Journal of Human Genetics 2006, 14. 890891.
69.
M. R. Knowles, M. J. Stutts, A. Spock, N. Fischer, J. T. Gatzy, R. C. Boucher,
Abnormal ion permeation through cystic fibrosis respiratory epithelium. Science
(Washington, DC, United States) 1983, 221. 1067-70.
70.
P. M. Quinton, Chloride impermeability in cystic fibrosis. Nature 1983, 301. 4212.
71.
B. P. O'Sullivan, M. D. Linden, A. L. Frelinger, III, M. R. Barnard, M. SpencerManzon, J. E. Morris, R. O. Salem, M. Laposata, A. D. Michelson, Platelet activation in
cystic fibrosis. Blood 2005, 105. 4635-4641.
72.
R. J. Stead, M. A. Barradas, D. P. Mikhailidis, J. Y. Jeremy, M. E. Hodson, J. C.
Batten, P. Dandona, Platelet hyperaggregability in cystic fibrosis. Prostaglandins Leukot
Med 1987, 26. 91-103.

25

73.
D. P. Mikhailidis, R. J. Stead, M. A. Barradas, M. E. Hodson, J. C. Batten, P.
Dandona, Platelet abnormalities in patients with cystic fibrosis and obligate
heterozygotes. Haematologica 1990, 75. 137-40.
74.
G. Agam, M. Aviram, M. Zilberman-Kaufman, A. Rothstein, A. A. Livne, Cyclic
AMP-related and cation-affected human platelet chloride transport regulation. Eur J Clin
Chem Clin Biochem 1995, 33. 329-35.
75.
P. A. Lane Sickle cell disease; Department of Pediatrics, University of Colorado
School of Medicine, Denver, USA: United States, 1996.
76.
R. B. Francis, Jr., Platelets, coagulation, and fibrinolysis in sickle cell disease:
their possible role in vascular occlusion. Blood Coagul Fibrinolysis 1991, 2. 341-53.
77.
K. I. Ataga, E. P. Orringer, Hypercoagulability in sickle cell disease: a curious
paradox. Am. J. Med. 2003, 115. 721-728.
78.
A. Compston, A. Coles, Multiple sclerosis. Lancet 2008, 372. 1502-1517, DOI:
Doi 10.1016/S0140-6736(08)61620-7.
79.
H. P. Wright, R. H. Thompson, K. J. Zilkha, Platelet adhesiveness in multiple
sclerosis. Lancet 1965, 2. 1109-10.
80.
W. A. Sheremata, W. Jy, L. L. Horstman, Y. S. Ahn, J. S. Alexander, A. Minagar,
Evidence of platelet activation in multiple sclerosis. J. Neuroinflammation 2008, 5. No pp
given.
81.
R. S. Sprague, M. L. Ellsworth, A. H. Stephenson, M. E. Kleinhenz, A. J.
Lonigro, Deformation-induced ATP release from red blood cells requires CFTR activity.
American Journal of Physiology 1998, 275. H1726-H1732.
82.
J. Carroll, M. Raththagala, W. Subasinghe, S. Baguzis, T. D. a. Oblak, P. Root, D.
Spence, An altered oxidant defense system in red blood cells affects their ability to
release nitric oxide-stimulating ATP. Mol. BioSyst. 2006, 2. 305-311.
83.
R. S. Sprague, A. H. Stephenson, M. L. Ellsworth, C. Keller, A. J. Lonigro,
Impaired release of ATP from red blood cells of humans with primary pulmonary
hypertension. Exp. Biol. Med. (Maywood, NJ, U. S.) 2001, 226. 434-439.
84.
S. Letourneau, L. Hernandez, A. N. Faris, D. M. Spence, Evaluating the effects of
estradiol on endothelial nitric oxide stimulated by erythrocyte-derived ATP using a
microfluidic approach. Anal. Bioanal. Chem. 397. 3369-3375.
85.
A. Dorn-Beineke, T. Frietsch, Sickle cell disease - pathophysiology, clinical and
diagnostic implications. Clin. Chem. Lab. Med. 2002, 40. 1075-1084.

26

86.
G. V. Born, Aggregation of blood platelets by adenosine diphosphate and its
reversal. Nature 1962, 194. 927-9.
87.
G. V. Born, M. J. Cross, The Aggregation of Blood Platelets. J Physiol 1963, 168.
178-95.
88.
D. C. Cardinal, R. J. Flower, Electronic Aggregometer - Novel Device for
Assessing Platelet Behavior in Blood. J Pharmacol Methods 1980, 3. 135-158, DOI: Doi
10.1016/0160-5402(80)90024-8.
89.
D. L. McGlasson, G. A. Fritsma, Whole Blood Platelet Aggregometry and
Platelet Function Testing. Semin. Thromb. Hemostasis 2009, 35. 168-180, DOI: DOI
10.1055/s-0029-1220325.
90.
R. D. Feinman, J. Lubowsky, I. Charo, M. P. Zabinski, Iumi-Aggregometer - New
Instrument for Simultaneous Measurement of Secretion and Aggregation by Platelets.
Journal of Laboratory and Clinical Medicine 1977, 90. 125-129.
91.
M. T. Santos, J. Valles, A. J. Marcus, L. B. Safier, M. J. Broekman, N. Islam, H.
L. Ullman, A. M. Eiroa, J. Aznar, Enhancement of platelet reactivity and modulation of
eicosanoid production by intact erythrocytes. A new approach to platelet activation and
recruitment. J Clin Invest 1991, 87. 571-80, DOI: 10.1172/JCI115032.
92.
M. R. Barnard, L. A. Krueger, A. L. Frelinger, 3rd, M. I. Furman, A. D.
Michelson, Whole blood analysis of leukocyte-platelet aggregates. Curr Protoc Cytom
2003, Chapter 6. Unit 6 15, DOI: 10.1002/0471142956.cy0615s24.
93.
C. Zamarron, M. H. Ginsberg, E. F. Plow, Monoclonal antibodies specific for a
conformationally altered state of fibrinogen. Thromb Haemost 1990, 64. 41-6.
94.
A. Schoolmeester, K. Vanhoorelbeke, S. Katsutani, H. Depraetere, H. B. Feys, J.
M. Heemskerk, M. F. Hoylaerts, H. Deckmyn, Monoclonal antibody IAC-1 is specific for
activated alpha2beta1 and binds to amino acids 199 to 201 of the integrin alpha2 Idomain. Blood 2004, 104. 390-6, DOI: 10.1182/blood-2003-12-4224.
95.
E. Larsen, A. Celi, G. E. Gilbert, B. C. Furie, J. K. Erban, R. Bonfanti, D. D.
Wagner, B. Furie, PADGEM protein: a receptor that mediates the interaction of activated
platelets with neutrophils and monocytes. Cell 1989, 59. 305-12.
96.
R. Y. Tsien, M. Poenie, Fluorescence Ratio Imaging - a New Window into
Intracellular Ionic Signaling. Trends in Biochemical Sciences 1986, 11. 450-455, DOI:
Doi 10.1016/0968-0004(86)90245-8.
97.
R. Y. Tsien, T. Pozzan, T. J. Rink, Measuring and Manipulating Cytosolic Ca-2+
with Trapped Indicators. Trends in Biochemical Sciences 1984, 9. 263-266, DOI: Doi
10.1016/0968-0004(84)90160-9.
27

98.
A. Minta, J. P. Kao, R. Y. Tsien, Fluorescent indicators for cytosolic calcium
based on rhodamine and fluorescein chromophores. J Biol Chem 1989, 264. 8171-8.
99.
K. R. Gee, K. A. Brown, W. N. U. Chen, J. Bishop-Stewart, D. Gray, I. Johnson,
Chemical and physiological characterization of fluo-4 Ca2+-indicator dyes. Cell Calcium
2000, 27. 97-106, DOI: DOI 10.1054/ceca.1999.0095.
100. K. S. Sakariassen, R. Muggli, H. R. Baumgartner, Measurements of platelet
interaction with components of the vessel wall in flowing blood. Methods in Enzymology
1989, 169. 37-70.
101. S. P. Jackson, W. S. Nesbitt, E. Westein, Dynamics of platelet thrombus
formation. Journal of thrombosis and haemostasis : JTH 2009, 7 Suppl 1. 17-20, DOI:
10.1111/j.1538-7836.2009.03401.x.
102. J. M. Cosemans, B. F. Iserbyt, H. Deckmyn, J. W. Heemskerk, Multiple ways to
switch platelet integrins on and off. Journal of thrombosis and haemostasis : JTH 2008,
6. 1253-61, DOI: 10.1111/j.1538-7836.2008.03041.x.
103. M. Roest, A. Reininger, J. J. Zwaginga, M. R. King, J. W. Heemskerk, Flow
chamber-based assays to measure thrombus formation in vitro: requirements for
standardization. Journal of thrombosis and haemostasis : JTH 2011, 9. 2322-4, DOI:
10.1111/j.1538-7836.2011.04492.x.
104. R. Van Kruchten, J. M. Cosemans, J. W. Heemskerk, Measurement of whole
blood thrombus formation using parallel-plate flow chambers - a practical guide.
Platelets 2012, 23. 229-42, DOI: 10.3109/09537104.2011.630848.
105. D. C. Duffy, J. C. McDonald, O. J. A. Schueller, G. M. Whitesides, Rapid
Prototyping of Microfluidic Systems in Poly(dimethylsiloxane). Analytical Chemistry
1998, 70. 4974-4984.
106. D. P. Sarvepalli, D. W. Schmidtke, M. U. Nollert, Design Considerations for a
Microfluidic Device to Quantify the Platelet Adhesion to Collagen at Physiological Shear
Rates. Ann Biomed Eng 2009, 37. 1331-1341, DOI: DOI 10.1007/s10439-009-9708-z.
107. W. S. Nesbitt, E. Westein, F. J. Tovar-Lopez, E. Tolouei, A. Mitchell, J. Fu, J.
Carberry, A. Fouras, S. P. Jackson, A shear gradient-dependent platelet aggregation
mechanism drives thrombus formation. Nat. Med. (N. Y., NY, U. S.) 2009, 15. 665-673.
108. F. J. Tovar-Lopez, G. Rosengarten, E. Westein, K. Khoshmanesh, S. P. Jackson,
A. Mitchell, W. S. Nesbitt, A microfluidics device to monitor platelet aggregation
dynamics in response to strain rate micro-gradients in flowing blood. Lab on a Chip
2010, 10. 291-302, DOI: 10.1039/b916757a.

28

109. D. W. Inglis, K. J. Morton, J. A. Davis, T. J. Zieziulewicz, D. A. Lawrence, R. H.
Austin, J. C. Sturm, Microfluidic device for label-free measurement of platelet activation.
Lab Chip 2008, 8. 925-931.
110. C.-J. Ku, T. D'Amico Oblak, D. M. Spence, Interactions between Multiple Cell
Types in Parallel Microfluidic Channels: Monitoring Platelet Adhesion to an
Endothelium in the Presence of an Anti-Adhesion Drug. Anal. Chem. (Washington, DC,
U. S.) 2008, 80. 7543-7548.
111. E. Gutierrez, B. G. Petrich, S. J. Shattil, M. H. Ginsberg, A. Groisman, A.
Kasirer-Friede, Microfluidic devices for studies of shear-dependent platelet adhesion.
Lab Chip 2008, 8. 1486-1495.
112. K. B. Neeves, S. F. Maloney, K. P. Fong, A. A. Schmaier, M. L. Kahn, L. F.
Brass, S. L. Diamond, Microfluidic focal thrombosis model for measuring murine platelet
deposition and stability: PAR4 signaling enhances shear-resistance of platelet aggregates.
J Thromb Haemost 2008, 6. 2193-201.
113. S. F. Maloney, L. F. Brass, S. L. Diamond, P2Y12 or P2Y1 inhibitors reduce
platelet deposition in a microfluidic model of thrombosis while apyrase lacks efficacy
under flow conditions. Integr Biol 2. 183-192.
114. C. Oury, E. Toth-Zsamboki, C. Thys, J. Tytgat, J. Vermylen, M. F. Hoylaerts, The
ATP-gated P2X1 ion channel acts as a positive regulator of platelet responses to collagen.
Thromb Haemost 2001, 86. 1264-1271.
115. R. A. North, A. Surprenant, Pharmacology of cloned P2X receptors. Annual
Review of Pharmacology and Toxicology 2000, 40. 563-580.
116. M. P. Mahaut-Smith, S. Jones, R. J. Evans, The P2X1 receptor and platelet
function. Purinergic Signal. DOI: 10.1007/s11302-011-9224-0.

29

CHAPTER 2: PLATELET P2X1 RECEPTOR FUNCTION IN THE ABSENCE OF
EXOGENOUS APYRASE
2.1 Platelet P2X1 receptor
Purinergic receptors on the surface of the platelet are important in mediating the
clotting process that occurs after vascular injury, most notably the ADP-sensitive
1-4

receptors, P2Y1/P2Y12, and the ATP-sensitive, P2X1 receptor.

Though the roles for

P2Y-type receptors in platelet activation and thrombus formation are well established, the
role of the P2X1 receptor in platelet function is less clear. For example, P2X1 activation
5

has been shown to induce transient platelet shape change, and is also involved with
collagen-induced aggregation events.

6-9

It has also been shown that ATP acting through

the P2X1 receptor is important as a secondary platelet agonist in the activation process.
Conversely, other studies report a minimal role for P2X1 in platelet activation.
Activation of the P2X1 receptor by ATP is responsible for extracellular Ca
the platelet. Not only is Ca

2+

11

entry into

important for P2X1 response, but also ADP dependent

P2Y response where extracellular Ca
platelet aggregation.

2+

10

2+

dependent ectonucleotidases affect ADP induced

12

A review on the P2X1 receptor stresses its importance in thrombotic events and
the need to study the receptor as a potential therapeutic target.

13

More specifically, an in

vitro study performed on platelets from P2X1 receptor-deficient mice showed that
collagen-induced platelet aggregation and adhesion to a collagen coated surface were
30

reduced. The collagen coated surface was mimicking a vascular injury and thrombus
growth on that coated region was also reduced in the P2X1 deficient platelets.

7

Additionally, in an in vivo study on these P2X1 deficient mice, bleeding times were
normal, but there was reduced thrombus formation following laser induced vascular
injury and reduced mortality in models of thromboembolism.

7

With the information

garnered from both in vitro and in vivo studies on platelet P2X1 receptor activity, it is
possible that P2X1 is a potential therapeutic target for anti-thrombotic drugs.

2.1.1 Difficulties in measuring P2X1 receptor activity
The P2X1 receptor is difficult to study because it is rapidly desensitized during in
vitro experimentation, especially when the platelets have been separated from plasma and
14-15

washed.

In order to overcome this rapid receptor desensitization,

16-18

apyrase,

which catalyzes the dephosphorylation of adenine nucleotides, is commonly used during
platelet purification from whole blood to reduce the number of phosphate-containing
nucleotides already present in the bulk solution. Unfortunately, the addition of apyrase to
the bulk solution results in the breakdown of exogenously added ATP to ADP, which is
unable to activate the P2X1 receptor.
Another factor contributing to the complexity of platelet P2X1 activation is
distinguishing effects due to ATP from those arising from ADP, especially after
dephosphorylation from ATP to ADP. For example, when P2X1 is stimulated with ATP,
2+

Ca

enters the platelet from the sample matrix and can be measured with a fluorescent

probe such as Fura-2 or Fluo-4 AM. However, ADP can also contribute to this Ca
31

2+

signal, as ADP binding to the P2Y1 receptor results in the release of Ca
contained within the platelet.

19-21

2+

from granules

Therefore, the degradation of ATP to ADP, or ADP

contamination in the ATP agonist, leads to inconclusive results regarding the source of
2+

the increased platelet Ca . However, there are techniques that can be used to distinguish
2+

the two receptors. For example, in the case of measuring Ca , one can simply perform
2+

the studies in a Ca -free buffer. Such a system would prevent Ca

2+

due to P2X1 activation by ATP, but not P2Y activation, since the Ca
intracellular. In other words, any measured Ca

2+

free buffer would likely be due to release of Ca

entering the platelet
2+

release would be
2+

signal during experiments with a Ca -

2+

from intracellular stores. This release

is indicative of ADP binding to the platelet.
2+

In addition to using Ca -free buffers to differentiate receptor activity,

many

investigators employ the use of ,-methylene-ATP (,-me-ATP), a stable ATP
analogue that cannot be enzymatically dephosphorylated to ADP. The use of such a
stable ATP analogue ensures that any measured Ca

2+

flux into the platelet is due to
5

P2X1. However, ,-me-ATP leads only to transient platelet shape change and low
levels of aggregation

22

and could potentially be under representing P2X1 activity that

would be measured with authentic ATP as discussed in Chapter 1.

2.1.2 NF449 as a P2X1 receptor inhibitor

32

To facilitate the understanding of receptor action and response, inhibitors of the
platelet purinergic receptors can be employed.
antagonists described
synthesized.

24-27

23

Suramin was one of the first P2X

and subsequent derivatives of suramin have also been

One of those derivatives, 4,4’,4’’,4’’’-(carbonyl(imino-5,1,3-

benzenetriyl-bis(carbonylimino)))tetrakis-benzene-1,3-disulphonic

acid

or

more

commonly, NF449, has been shown to be a potent and specific antagonist of P2X1.

27

The structure of NF449 is shown in figure 2.1. Previously, the structure of the P2X4
28

receptor from zebrafish had been solved in the absence of bound ATP , but more
29

recently, the structure of P2X4 with bound ATP has been solved , giving more
information about the ATP binding pocket on P2X type receptors. The mechanism of
activation involves ATP binding in the extracellular loop, which induces conformational
changes that prompts flexing of the transmembrane domains and allowing for “lateral
fenestration” of cations.

29

It is proposed that NFF49 binds at or near the ATP binding

30

site on P2X receptors , and due to steric hindrance, prevents the channel pore form
29

opening.

However, structural studies involving P2X-bound NF449 have yet to be

performed.
NF449 is one of the most widely used platelet P2X1 inhibitors and it has been
shown to reduce Ca
me-ATP.

25, 31-32

2+

influx and reduce activation of the receptor by both ATP and ,-

Interestingly, in contrast to its inhibitory actions on the P2X1 receptor,

NF449 has been reported to enhance P2X1 sensitization in frog oocytes.
33

25

A detailed

Figure 2.1: Structure of NF449.

The P2X1 inhibitor, NF449, is pictured with

+

coordinating sodium ions (Na ), but in solution the polysulphonates are negatively
charged.

34

examination of these reports involving NF449 inhibition and sensitization reveals that the
conditions (concentrations of NF449, number of cells used in the studies and incubation
time of the NF449 with the cells prior to measurement) differed greatly. These reports
led me to the hypothesis that NF449 may exhibit sensitizing and inhibitory actions on the
platelet P2X1 receptor that are dependent upon the ratio of agonist and inhibitor
employed during platelet preparation. This is important because if NF449 can be used to
sensitize platelets, instead of apyrase, there would not be a need to use a nonhydrolysable form of ATP during experiments, enabling the effects of ATP versus ADP
to be distinguished with regard to platelet activity. This is significant, as discussed in
Chapter 1; because certain diseases (cystic fibrosis, diabetes, hypertension, sickle cell
anemia) have defects in RBC derived ATP release, as well as increased platelet activity.
The correlation between these two cell types would be the ATP activation of P2X1
receptor on the platelets so, a method to study platelet response to ATP that does not
involve apyrase or ,-me-ATP would be extremely useful.

2.2 Methods
2.2.1 Isolation of platelets from human whole blood
Whole blood was obtained via venipuncture under consent from donors and collected
into heparinized tubes. In general, 10 tubes of whole blood were collected from each
donor (~30 mL). After the whole blood was centrifuged at 500 g for 10 min at room
temperature, the platelet-rich plasma (PRP) was decanted for subsequent platelet
purification.

35

Platelets were isolated from the PRP by adding 1 mL of acid citrate dextrose (ACD;
in mM, 41.6 citric acid anhydrous, 76.7 monosodium citric acid anhydrous, 122.1
dextrose) to every 9 mL of PRP and centrifuging at 1500 g for 10 min at room
temperature. The harvested platelets were then washed twice with 10% ACD calcium
free Modified Tyrodes Buffer (MTB; in mM, 12 NaHCO3, 0.32 NaH2PO4, 10 HEPES,
137 NaCl, 2.7 KCl, 0.5 MgCl2, and 5.5 dextrose, Sigma-Aldrich, St. Louis, MO) solution
with centrifugation in between each wash. The washed platelets were then re-suspended
in 10% ACD calcium free MTB. No apyrase was used in this purification. The washed
platelets were counted using a hemacytometer and adjusted to a concentration of 2.0 x
9

-1

10 platelets mL

in 1 mL 10% ACD calcium free MTB.

Platelet samples were

prepared and investigated the same day as isolation.

2.2.2 Fluo-4 AM probe and NF449 platelet sample preparation
A 2.3 mM stock of the Ca

2+

specific probe Fluo-4 AM (Molecular

Probes/Invitrogen Eugene, OR) was prepared by dissolving 50 g of the probe in 20 L
of anhydrous DMSO. A 5 L aliquot of the stock was mixed with 5 L of a 200 mg/mL
pluronic F-127 surfactant (Molecular Probes/Invitrogen Eugene, OR) solution to enhance
the probe’s penetration capacity. Finally, this probe solution was diluted to 1 mL in 10%
ACD calcium free MTB to create an 11.4 M working solution. This 1 mL solution of
the fluorescent probe was then added to the 1 mL platelet solution for a final
concentration of 5.7 M Fluo 4 AM and incubated on ice for 30 min.

The platelet

solution containing the probe was then centrifuged at 1500 g for 5 min at room
36

temperature and washed twice with 10% ACD calcium free MTB to remove any excess
probe. Finally, the platelets were resuspended in 2 mL of 10% ACD calcium free MTB.
9

-1

Aliquots of the Fluo-4 AM loaded platelet solution (1.0 x 10 platelets mL )
were incubated for 30 min on ice with NF449 (Tocris Bioscience, Ellisville, MO) at final
concentrations ranging from 0-40 M, or MRS 2179 (Tocris Bioscience) at final
concentrations ranging from 0-60 M.

2.2.3 Fluorescence determination of intracellular platelet Ca

2+

Calcium measurements were performed with a spectrofluorometer (HORIBA
Jobin Yvon, Edison, NJ) having 2 nm slit widths using excitation at 496 nm and emission
at 516 nm. The sample was contained in a 4 mL 21N-Q-10 far UV quartz cuvette. A 40
2+

L aliquot of platelet sample was added to 1940 L of MTB with calcium (2 mM Ca )
at 37 C and a fluorescence baseline was collected for 15 s prior to the addition of 20 L
of agonist prepared in Ca

2+

free MTB. Agonists included ATP, ADP and ,-methylene

ATP (Sigma-Aldrich, St. Louis, MO) at stock concentrations ranging from 0-1000 M
resulting in final concentrations of 0-10 µM in the cuvette. Fluorescence measurements
were collected for an additional 90 s; samples were stirred throughout the measurement.
A schematic of sample preparation and subsequent fluorescence measurement is shown
in figure 2.2.

2.2.4 Data analysis

37

Figure 2.2: Schematic representation of experimental setup. Platelets were incubated with
2+

the Ca

probe Fluo-4 AM for 30 min, then the excess probe was washed off and the
2+

platelets incubated with either buffer or NF449. Finally, measurements of Ca influx
were performed using a spectrofluorometer with a magnet setup so that samples could be
stirred during measurement. Samples were kept in a Ca
measurement.

38

2+

free environment until

Included in the results and discussions section are spectra that represent raw fluorescence
data collected over time, whereas the summarized data are represented as a percent
change in fluorescence. This percent change was calculated by subtracting the quotient
of the average baseline fluorescence intensity and average peak fluorescence intensity
from 1 and multiplying by 100 to obtain a percent. N = human blood draws.

2.4 Results
2.4.1. NF449 sensitizes ATP-induced Ca

2+

Influx

ATP was added as an agonist to platelets, in the presence and absence of NF449, to
evaluate its effect on P2X1 mediated Ca

2+

influx. Shown in figure 2.3 (left), adding

CaMTB to platelets alone has no effect on the fluorescence signal. Furthermore, the
addition of ATP to the platelet sample resulted in no significant increase in fluorescence
signal (compared to baseline) due to Ca
significant increase in Ca
NF449.

2+

2+

entry into the platelets. However, there was a

entry when ATP was added to platelets containing 0.5 µM

Furthermore, as shown in figure 2.3 (right), Ca

2+

entry into the platelets

containing 2.5 µM NF449 increased as the concentration of ATP was increased from 67
nM to 5 µM, although the first significant change in signal (in comparison to baseline)
did not occur until the ATP concentration reached 310 nM. Thus, activation of the P2X1
receptor and subsequent Ca

2+

influx is ATP concentration dependent under NF449

sensitizing conditions.

39

C

Fluorescence Intensity (Ca2+)

Fluorescence Intensity (Ca2+)

4000

3500

B

3000

2500

2000

A

5000
4500
4000
3500
3000
2500
2000

1500
10

20

30

40

50

10

60

20

30

40

50

Time (s)

Time (s)

Figure 2.3: (Left) Fluorescence measurements taken over time for platelets in the presence and absence of NF449. No significant
2+

increase in Ca
2+

Ca

is measured when CaMTB (trace C) or 2.5 uM ATP (trace B) is added to platelets alone. However, an increase in

was measured when 2.5 uM ATP was added to platelets containing 0.5 uM NF449 (trace A). (Right) The effect of ATP on
2+

platelet Ca

influx with a constant concentration of 2.5 uM NF449. As the concentration of ATP was increased from 73 nM up to 5

µM, the intracellular platelet Ca

2+

also increased.
40

2.4.2 Effect of Ca

2+

free buffer on Ca

2+

entry into the platelet via ATP and ADP

agonists
A possible explanation for the increase in fluorescence emission shown in figure
2.3 is that the ATP contained ADP impurities that were stimulating P2Y-type receptors,
resulting in the release of Ca

2+

stores in the platelet granules. Therefore, studies were

2+

also performed in a Ca -free buffer. In this construct, activation of the P2X1 receptor
with ATP should not result in an increase of fluorescence emission due to Ca
the platelet because there is no Ca

2+

2+

entry into

in the buffer to cross the platelet membrane. The

data in figure 2.4 (left) show that ATP additions to platelets containing NF449 in a Ca

2+

free environment result in no increase in fluorescence signal. The top trace in figure 2.4
(left) is another aliquot of platelets from the same sample. However, this trace was
2+

obtained using a Ca -containing buffer and, as shown, Ca

2+

entry into the platelet was

detected. Finally, when ADP was added to NF449-containing platelets, there was an
2+

increase in fluorescence signal even in the absence of Ca , which is due to Ca

2+

release

from granular stores in the platelets. The summarized data in figure 2.4 (right) suggests
that the ATP induced Ca

2+

entry into NF449-containing platelets is through P2X1

activation and not solely a result of an ADP impurity contained in the ATP agonist.

41

2+

Figure 2.4: Effect of Ca

free buffer on Ca

agonist in the presence of Ca

2+

2+

influx in platelet containing 0.5 µM NF449. (Left) 2.5 µM ATP was added as an

as represented in trace A. The typical increase in Ca

2+

influx is seen. However, in a Ca
2+

environment (trace B), the addition of 2.5 µM ATP resulted in no significant increase in Ca . Additionally, in the Ca
environment, ADP is still able to elicit a Ca

2+

increase (trace C). (Right) The summarized data indicates that when Ca

2+

2+

2+

2+

free
free

is not

present, there is no Ca influx due to 2.5 µM ATP in platelets containing 0.5 µM NF449 or platelets alone. However, signal due to
ADP stimulation is still present. Error is standard error of the mean. N=9 (2.5 µM ATP, platelets), N=7 (2.5 µM ATP, NF449), N=4
2+
2+
(2.5 µM ATP, Ca free), and N=5 (2.5 µM ADP, Ca free).
42

2.4.3 Effect of α,β-me-ATP concentrations on Ca

2+

entry into the platelet

The P2X1 receptor is most commonly studied using a form of ATP that is stable,
such as ,-me-ATP. This form of ATP ensures that breakdown to ADP does not occur,
thus minimizing effects from P2Y-type events.

Figure 2.5 shows a Ca

2+

influx due to

the addition of ,-me-ATP to NF449 containing platelets (18.7 ± 0.9%), a significantly
weaker signal (P < 0.001) in comparison to authentic ATP stimulation (39.3 ± 0.8%). No
calcium influx was seen when ATP or ,-me-ATP were added to platelets alone.
Unlike ATP, higher concentrations of ,-me-ATP did not elicit Ca

2+

influx (data not

shown), suggesting the potential for ,-me-ATP to desensitize the P2X1 receptor more
readily than ATP itself.

2.4.4 Effect of NF449 concentrations on Ca

2+

entry into the platelet

NF449 is generally regarded as a competitive inhibitor of the P2X1 receptor and at
concentrations greater than 1 µM has been shown to have non-specific inhibitory effects
on P2Y receptors. However, data in figures 2.2-2.4 suggest receptor sensitization at a
concentration of 0.5 and 2.5 µM NF449. Therefore, it was anticipated that concentrations
of NF449 less than 1 µM would also have the same sensitizing effects. Figure 2.6
summarizes the effect of varying NF449 concentrations from 62.5 nM up to 40 µM in the
presence of 2.5 µM ATP as the agonist for P2X1 stimulation and subsequent Ca

2+

entry

into the platelet. As shown, the signal resulting from the fluorescence emission due to
2+

Ca

entry into the platelet begins to increase with increments of NF449. However, as

43

Fluorescence Intensity (Ca2+)

5500

A

5000
4500
4000

B
3500
3000
2500

C
2000
10

20

30

40

50

60

Time (s)
2+

Figure 2.5: Effect of α,β-me-ATP on platelet Ca influx. Both A and B traces indicate
the presence of platelets containing 0.5 µM NF449, but the agonist for trace A is 2.5 µM
2+

ATP and the agonist for trace B is 2.5 µM α,β-me-ATP. Increases in Ca influx were
measured in the case of both agonists, but the increase due to ATP was larger than the
increase due to α,β-me-ATP. The two controls in this figure, represented as C, are
platelets alone that were treated with both agonists, neither showed an increase in Ca
influx.

44

2+

50

N = 3 (62.5 nM NF449)
N = 4 (125 nM and 250 nM NF449)
N = 6 (Platelets, 0.5 M and 10 M NF449)
N = 5 (20 M and 40 M NF449)

Percent Change

40

30

20

10

Figure 2.6: Effect of NF449 concentration on platelet Ca

2+

9

9

F4
4

40

M

N

N

F4
4

9
M

N
20

M

N

F4
4
10

M

F4
4

9

9
N

M
0

0.
25

0.
5

F4
4
N

nM
5

F4
4

9

9
F4
4

N

nM
12

62

.5

Pl
a

te
l

et
s

0

influx due to 2.5 µM ATP.

As NF449 concentration is increased from 0 to 0.5 µM, the Ca

2+

influx increases, but as

the concentration of NF449 is further increased from 0.5 µM to 40 µM the Ca
decreases from the measured maximum. Error is standard error of the mean.

45

2+

influx

the concentration of NF449 continues to increase, an inhibitory effect emerges (for an
ATP concentration of 2.5 µM), resulting in reduced Ca

2+

entry into the platelet.

Importantly, when the NF449 concentrations reached inhibitory levels (of the 2.5 µM
ATP), the addition of higher concentrations of ATP were able to restore the Ca

2+

entry

into the platelets. For example, at an NF449 concentration of 0.5 µM, the percent change
in fluorescence emission upon the addition of 2.5 µM ATP was 39.3 ± 2.4%. This value
dropped to 13.1 ± 2.2% when the NF449 concentration was increased to 20 µM.
However, when the ATP concentration was raised to 10 µM, the percent change in
fluorescence emission recovered to a value of 34.1 ± 2.8%. Sample spectra are shown in
figure 2.7. These results suggest that the ratio between ATP and NF449 is key in
stimulating or inhibiting Ca

2+

increase through the P2X1 receptor. Therefore, if the ATP

concentration in a particular study is 2.5 µM, the optimum sensitizing concentration of
NF449 is ~ 0.5 µM, while inhibitory concentrations of NF449 would be obtained at
concentrations > 10 µM. Of course, these values of NF449 would change if the ATP
concentrations were changed.

2.4.5 Time Dependence of NF449 Sensitization of P2X1
Previously, NF449 has been shown to sensitize P2X1, although the effect was
measured at picomolar concentrations on single oocytes and only lasted for a few
seconds. In studies reported here, nanomolar to low micromolar concentrations exhibited
this same sensitization feature even though the NF449 was allowed to incubate with the

46

Fluorescence Intensity (Ca 2+)

7000

B

6500
6000

C

5500
5000
4500
4000

A

3500
3000
10

20

30

40

50

60

Time (s)

Figure 2.7: Overcoming NF449 inhibition using 10 µM ATP. As the concentration of
2+
NF449 exceeds 0.5 µM, inhibition of Ca influx occurs when 2.5 µM ATP is added to
20 and 40 µM NF449 platelet samples (traces A). When 10 µM ATP is added, the
2+
NF449 inhibition is overcome and Ca influx occurs in 20 and 40 µM NF449 platelet
samples, traces B and C, respectively.

47

platelets for 30 minutes prior to the addition of agonist.

Therefore, a study was

performed to determine if the ability of NF449 to sensitize the platelet P2X1 receptor was
time dependent. The time study was executed using platelets incubated with 62.5 nM
NF449 followed by the addition of 2.5 µM ATP as the agonist. The results of this study,
shown in figure 2.8, indicate that the fluorescence signal from Ca

2+

entry due to addition

of the ATP agonist decreased with time. The increase at 1 min is thought to be less than
that of the 5 min signal due to NF449 not having enough time to diffuse to the platelet
surface. The subsequent decrease in signal suggests that the sensitizing effect of NF449
is indeed time dependent, consistent with the previous report.

27

However, it appears that

increasing the NF449 concentration prolongs its sensitization abilities.

2.4.6 P2Y1 inhibition with MRS and the effect on Ca

2+

influx

In order to investigate the P2Y1 receptor action in the presence of ADP and ATP,
MRS2179 was employed. MRS2179 is a competitive antagonist of P2Y1, and is more
selective towards P2Y1 than P2X1.

33-34

As shown in figure 2.9, with increasing

concentration of MRS2179, the intracellular Ca

2+

signal decreases in response to ADP.

However, when NF449 is also added to the highest concentration of MRS2179, some of
the ADP signal is recovered as well as Ca

2+

influx due to ATP. Though, the Ca

2+

signal

from ATP is not as high as for samples that were only sensitized with NF449. This could
suggest a relationship between the sensitization of P2X1 and the ability of P2Y1 to

48

Figure 2.8: NF449 potency over time with 2.5 µM ATP agonist. Using 62.5 nM NF449,
2+

platelet Ca

influx was measured over the course of 6 hours with the data being
2+

normalized to the 5 min. percent change in signal. The Ca influx peaked at 5 min and
decreased steadily over time.
Error is standard error of the mean.

49

2+

Figure 2.9: The effect of P2Y1 inhibition on platelet Ca . As increasing concentrations
2+

of MRS are added to the platelet solution it inhibits the Ca release due to ADP binding
P2Y1. However, when NF449 is also added to the samples containing 60 µM MRS, the
2+

Ca

signal from ATP and ADP recovered. P2Y1 also has some inhibitory effect on
2+

P2X1 so; the Ca signal from ATP is lower compared to the sample that was sensitized
with NF449 alone. Error is standard error of the mean. N is 4 or higher for all bars.

50

2+

mobilize Ca

in response to ADP. It should be noted that ATP could not evoke Ca

2+

influx without the presence of NF449 to sensitize the platelets prior to study.
2.5 Conclusions
Collectively, the data in figures 2.5 and 2.7 suggest something more profound
about the effect of NF449 on P2X1 and the receptor itself. Specifically, the data in
figures 2.5 and 2.8 demonstrate that the effect of NF449 on P2X1 is both concentrationand time-dependent, respectively. Moreover, both of these effects can be immediately
overcome by high concentrations of ATP. In addition, the data in figure 2.6 show that
the overall signal from the addition of ATP is higher when both P2X1 and P2Y1 are
open. This suggests that the addition of ATP to the platelet sample results in an increase
in P2Y1 activity. It is possible that once ATP is bound to P2X1 and Ca

2+

enters, the

ATP is hydrolyzed to ADP, which can subsequently activate P2Y1 or P2Y12. The fate
of ATP after it is bound to P2X1 is not known. If such a relationship exists, it gives more
weight to the effect of ATP on platelet activity since initially ATP would bind P2X1, but
then its degradation product could go on to activate the P2Y receptors, which are also
involved in platelet activation.
The P2X1 receptor activity can explain why disease states with altered RBC ATP
release also have hyperactive platelets. Resting platelets are constantly coming into
contact with RBCs in the vasculature so, it follows that ATP released from these RBCs
would be binding P2X1 and causing low levels of activation without the presence of an
injured endothelium. The Spence group showed that ATP affected platelets adhering to
an uninjured endothelium and that RBCs incubated with an ATP release inhibitor
affected platelet adhesion.

35

It is a possibility that in vivo when an injury occurs, platelets
51

are activated from exposed collagen and granular release from the platelets, but also the
ATP released from RBCs. The ability to study the platelet without apyrase or the stable
,-me-ATP allows for the effects of native ATP on the platelet to be studied without the
risk of apyrase affecting the amount of available ATP or ,-me-ATP derived
desensitization. These sample handling tools can be used in conjunction with studies on
washed platelets or platelets in the presence of RBC’s in order to study P2X1 activity.

52

REFERENCES

53

REFERENCES

1.
S. P. Kunapuli, Z. Ding, R. T. Dorsam, S. Kim, S. Murugappan, T. M. Quinton,
ADP receptors - targets for developing antithrombotic agents. Curr. Pharm. Des. 2003, 9.
2303-2316.
2.
M. P. Mahaut-Smith, S. J. Ennion, M. G. Rolf, R. J. Evans, ADP is not an agonist
at P2X1 receptors: evidence for separate receptors stimulated by ATP and ADP on
human platelets. Br J Pharmacol 2000, 131. 108-114.
3.
S. Murugappan, S. P. Kunapuli, The role of ADP receptors in platelet function.
Front. Biosci. 2006, 11. 1977-1986.
4.
C. Vial, B. Hechler, C. Leon, J. P. Cazenave, C. Gachet, Presence of P2X1
purinoceptors in human platelets and megakaryoblastic cell lines. Thromb Haemost 1997,
78. 1500-1504.
5.
M. G. Rolf, C. A. Brearley, M. P. Mahaut-Smith, Platelet shape change evoked by
selective activation of P2X1 purinoceptors with alpha ,beta -methylene ATP. Thromb
Haemost 2001, 85. 303-308.
6.
C. Y. E. Fung, C. A. Brearley, R. W. Farndale, M. P. Mahaut-Smith, A major role
for P2X1 receptors in the early collagen-evoked intracellular Ca2+ responses of human
platelets. Thromb Haemost 2005, 94. 37-40.
7.
B. Hechler, N. Lenain, P. Marchese, C. Vial, V. Heim, M. Freund, J.-P. Cazenave,
M. Cattaneo, Z. M. Ruggeri, R. Evans, C. Gachet, A role of the fast ATP-gated P2X1
cation channel in thrombosis of small arteries in vivo. J. Exp. Med. 2003, 198. 661-667.
8.
M. P. Mahaut-Smith, G. Tolhurst, R. J. Evans, Emerging roles for P2X1 receptors
in platelet activation. Platelets 2004, 15. 131-144.
9.
C. Oury, E. Toth-Zsamboki, C. Thys, J. Tytgat, J. Vermylen, M. F. Hoylaerts, The
ATP-gated P2X1 ion channel acts as a positive regulator of platelet responses to collagen.
Thromb Haemost 2001, 86. 1264-1271.
10.
C. Y. E. Fung, C. Cendana, R. W. Farndale, M. P. Mahaut-Smith, Primary and
secondary agonists can use P2X1 receptors as a major pathway to increase intracellular
Ca2+ in the human platelet. J. Thromb. Haemostasis 2007, 5. 910-917.
11.
S. Takano, J. Kimura, I. Matsuoka, T. Ono, No requirement of P2X1
purinoceptors for platelet aggregation. European Journal of Pharmacology 1999, 372.
305-309.

54

12.
S. Jones, R. J. Evans, M. P. Mahaut-Smith, Extracellular Ca2+ modulates ADPevoked aggregation through altered agonist degradation: implications for conditions used
to study P2Y receptor activation. Br J Haematol 153. 83-91.
13.
M. P. Mahaut-Smith, S. Jones, R. J. Evans, The P2X1 receptor and platelet
function. Purinergic Signal. DOI: 10.1007/s11302-011-9224-0.
14.
S. Valera, N. Hussy, R. J. Evans, N. Adami, R. A. North, A. Suprenant, G. Buell,
A new class of ligand-gated ion channel defined by P2X receptor for extracellular ATP.
Nature (London, United Kingdom) 1994, 371. 516-19.
15.
C. J. Lewis, R. J. Evans, Lack of run-down of smooth muscle P2X receptor
currents recorded with the amphotericin permeabilized patch technique, physiological
and pharmacological characterization of the properties of mesenteric artery P2X receptor
ion channels. Br J Pharmacol 2000, 131. 1659-1666.
16.
A. Baurand, A. Eckly, N. Bari, C. Leon, B. Hechler, J.-P. Cazenave, C. Gachet,
Desensitization of the platelet aggregation response to ADP: differential down-regulation
of the P2Y1 and P2cyc receptors. Thromb Haemost 2000, 84. 484-491.
17.
C. Gachet, P2 receptors, platelet function and pharmacological implications.
Thromb Haemost 2008, 99. 466-472.
18.
A. R. Hardy, P. B. Conley, J. Luo, J. L. Benovic, A. W. Poole, S. J. Mundell,
P2Y1 and P2Y12 receptors for ADP desensitize by distinct kinase-dependent
mechanisms. Blood 2005, 105. 3552-3560.
19.
J. Jin, S. P. Kunapuli, Coactivation of two different G protein-coupled receptors is
essential for ADP-induced platelet aggregation. Proceedings of the National Academy of
Sciences of the United States of America 1998, 95. 8070-8074.
20.
P. Savi, P. Beauverger, C. Labouret, M. Delfaud, V. Salel, M. Kaghad, J. M.
Herbert, Role of P2Y1 purinoceptor in ADP-induced platelet activation. FEBS Letters
1998, 422. 291-295.
21.
C. Leon, B. Hechler, M. Freund, A. Eckly, C. Vial, P. Ohlmann, A. Dierich, M.
LeMeur, J.-P. Cazenave, C. Gachet, Defective platelet aggregation and increased
resistance to thrombosis in purinergic P2Y1 receptor-null mice. Journal of Clinical
Investigation 1999, 104. 1731-1737.
22.
J. A. Erhardt, K. Pillarisetti, J. R. Toomey, Potentiation of platelet activation
through the stimulation of P2X1 receptors. J Thromb Haemost 2003, 1. 2626-35.
23.
P. M. Dunn, A. G. H. Blakeley, Suramin - a Reversible P2-Purinoceptor
Antagonist in the Mouse Vasdeferens. Br J Pharmacol 1988, 93. 243-245.

55

24.
K. Braun, J. Rettinger, M. Ganso, M. Kassack, C. Hildebrandt, H. Ullmann, P.
Nickel, G. Schmalzing, G. Lambrecht, NF449: a subnanomolar potency antagonist at
recombinant rat P2X1 receptors. Naunyn Schmiedebergs Arch Pharmacol 2001, 364.
285-290.
25.
M. Hulsmann, P. Nickel, M. Kassack, G. Schmalzing, G. Lambrecht, F.
Markwardt, NF449, a novel picomolar potency antagonist at human P2X1 receptors.
European Journal of Pharmacology 2003, 470. 1-7.
26.
M. U. Kassack, K. Braun, M. Ganso, H. Ullmann, P. Nickel, B. Boing, G. Muller,
G. Lambrecht, Structure-activity relationships of analogues of NF449 confirm NF449 as
the most potent and selective known P2X1 receptor antagonist. Eur. J. Med. Chem. 2004,
39. 345-357.
27.
J. Rettinger, K. Braun, H. Hochmann, M. U. Kassack, H. Ullmann, P. Nickel, G.
Schmalzing, G. Lambrecht, Profiling at recombinant homomeric and heteromeric rat P2X
receptors identifies the suramin analogue NF449 as a highly potent P2X1 receptor
antagonist. Neuropharmacology 2005, 48. 461-468.
28.
T. Kawate, J. C. Michel, W. T. Birdsong, E. Gouaux, Crystal structure of the
ATP-gated P2X4 ion channel in the closed state. Nature (London, United Kingdom)
2009, 460. 592-598.
29.
M. Hattori, E. Gouaux, Molecular mechanism of ATP binding and ion channel
activation in P2X receptors. Nature (London, United Kingdom) 485. 207-212.
30.
S. El-Ajouz, D. Ray, R. C. Allsopp, R. J. Evans, Molecular basis of selective
antagonism of the P2X1 receptor for ATP by NF449 and suramin: contribution of basic
amino acids in the cysteine-rich loop. Br J Pharmacol 2012, 165. 390-400, DOI:
10.1111/j.1476-5381.2011.01534.x.
31.
B. Hechler, S. Magnenat, M. L. Zighetti, M. U. Kassack, H. Ullmann, J.-P.
Cazenave, R. Evans, M. Cattaneo, C. Gachet, Inhibition of platelet functions and
thrombosis through selective or nonselective inhibition of the platelet P2 receptors with
increasing
doses
of
NF449
[4,4',4'',4'''-(carbonylbis(imino-5,1,3benzenetriylbis(carbonylimino)))tetrakis-benzene-1,3-disulfonic acid octasodium salt]. J.
Pharmacol. Exp. Ther. 2005, 314. 232-243.
32.
M. Hohenegger, M. Waldhoer, W. Beindl, B. Boing, A. Kreimeyer, P. Nickel, C.
Nanoff, M. Freissmuth, Gsalpha -selective G protein antagonists. Proceedings of the
National Academy of Sciences of the United States of America 1998, 95. 346-351.
33.
J. L. Boyer, A. Mohanram, E. Camaioni, K. A. Jacobson, T. K. Harden,
Competitive and selective antagonism of P2Y1 receptors by N6-methyl 2'deoxyadenosine 3',5'-bisphosphate. Br J Pharmacol 1998, 124. 1-3, DOI:
10.1038/sj.bjp.0701837.
56

34.
J. L. Boyer, T. Romero-Avila, J. B. Schachter, T. K. Harden, Identification of
competitive antagonists of the P2Y1 receptor. Mol. Pharmacol. 1996, 50. 1323-1329.
35.
W. Karunarathne, C. J. Ku, D. M. Spence, The dual nature of extracellular ATP as
a concentration-dependent platelet P2X1 agonist and antagonist. Integr Biol 2009, 1. 655663, DOI: Doi 10.1039/B909873a.

57

CHAPTER 3: INTEGRATION OF MULTIPLE COMPONENTS IN
POLYSTYRENE-BASED MICROFLUIDIC DEVICES

3.1 Introduction to Microfluidic Device Fabrication
As discussed in Chapter 1, soft lithography is a fast and easy method for rapid
fabrication of microfluidic devices.

1-2

Furthermore, microfluidic devices are popular for

use in studying biological systems due to their ability to perform fast/high-throughput
analysis,

3-4

utilize small sample volumes,

5

and integrate such processes.

6-9

The

incorporation of cells onto microfluidic devices has become popular because dimensions
mimicking in vivo environments can be fabricated and that both adherent and nonadherent cells can be investigated separately or simultaneously.

10-13

3.1.1 PDMS Microfluidic Devices
One of the most commonly used elastomeric polymers, polydimethyl siloxane
(PDMS), is optically transparent, gas permeable, pliable and has been useful for various
cell based applications. PDMS based microfluidic devices have been used to study
interactions between multiple vascular cell types,
underlying adhesion proteins,
permeability,

20

17

14-16

cell migration,

18

transendothelial electrical resistance,

analytes released by cells.

21

interactions between cells and

automated cell culture,

19

cell

and electrochemical detection of

6, 22

While PDMS has been a successful material for creating cell based experiments and
platforms, it has some disadvantages.

PDMS, a crosslinked polymer composed of
58

hydrophobic dimethyl-siloxane oligomers, has been shown to absorb small hydrophobic
23

molecules , leach un-crosslinked oligomers into solution,
through bulk material.

25

24

and allow evaporation

Furthermore, efforts have been made to make microfluidic

devices amenable for use with standard lab equipment and easier to use in general. This
includes integration with passive pumping and automated liquid handlers,
ability to make microfluidic arrays capable of plate reader detection,

26-27

3, 28

and the

but these

devices are still academic prototypes and they do not fit with all standard lab equipment.
Though this is a good start to making microfluidic devices integrated with standard
equipment more work in this area is needed.

3.1.2 Polystyrene Microfluidic Devices
Devices made from materials more established for cell culture, such as polystyrene
(PS), have the potential to be more compatible for cell culture on microfluidic devices
than PDMS.

11

Multiple groups have focused on the fabrication of polystyrene devices

for microarrays or cell-based assays.

29-34

Takayama et al. tested several “hard top, soft

bottom” microfluidic devices using a combination of the following rigid and elastomeric
polymers: PS, glycol-modified polyethylene terephthalate (PETG), or cyclic olefin
copolymer (COC) and polyurethane (PU) or PDMS coated with parylene C. To fabricate
the rigid layer of the devices, a PDMS master mold was made using conventional soft
1

lithography techniques and high temperature epoxy was poured onto the mold and
allowed to cure. The result is an epoxy mold with positive features. Then, via hot
59

embossing, the epoxy mold was pressed into PS, PETG or COC. Finally, the soft bottom
of the device was bonded to the rigid plastic via plasma bonding. These devices had
lower evaporation and lower oxygen permeability when compared to devices made solely
out of PDMS.

35

Beebe et al. took this approach one step further and fabricated a device whose top and
bottom are made out of polystyrene. Using a similar technique based on soft lithography
to make epoxy molds and hot embossing to create channel structures on PS, arrayed
devices were fabricated. Though a device made completely of PS was fabricated, a
limitation of the method still exists. This limitation includes the potential for channels to
collapse during the thermal bonding of PS to PS when aspect ratios are too high.

30

A

feature common to both of these previous methods is the use of a heated hydraulic press
and multiple molding steps.
Allbritton et al. developed a method that only involves the production of a PDMS
mold via soft lithography, without the need for a heated press to produce channel
structure, as gamma-butyrolactone (GBL) was utilized to dissolve PS onto the mold.
Importantly, GBL dissolved PS without swelling the PDMS mold. Using this dissolving
process, devices with microwell and micropost arrays were fabricated and used to study
adherent and non-adherent cells. However, to seal the devices, a blank PS sheet was
made (using the same GBL process) and then sealed to the other PS piece via hot plate
heating and manual pressure.

36

Another study investigated the biocompatibility of cells on different substrates, where
a hot embosser was used to create a chamber that cells were cultured in and studied.

60

29

While these previous fabrication strategies were useful in demonstrating the advantages
of using polystyrene as a biocompatible microfluidic device material, there has been a
lack of integration with standard lab equipment such as pipets or instrumentation. This
chapter will focus on a method to fabricate a polystyrene microfluidic device that is easy,
capable of cell culture and integration with standard lab equipment.

3.2 Methods
3.2.1 Fabrication of a polystyrene device with integrated channels
The general strategy for fabricating polystyrene devices with integrated channels
is shown in figure 3.1. A PDMS mold with raised features was fabricated based on
previously established soft lithography techniques,

37-38

except a positive mask was used

to produce recessed features in the master. Masters were fabricated on clean silicon
wafers with SU-8 50 photoresist (Microchem Corporation, Newton, MA, USA) spun at
500 rpm for 15 s and then 1000 rpm for 30 s, eventually leading to features that were 100
µm tall. The coated wafer was then exposed to ultraviolet light through a transparency
positive mask containing 250 µm wide channels in a serpentine pattern.
development, this resulted in a master with recessed features.

After

A 10:1 ratio of

elastomer:curing agent (Sylgard 184, Ellsworth Adhesives, Germantown, WI, USA) was
then poured over the completed master, cured at 75 ˚C, and subsequently removed from
the master and sealed to a glass slide using a thin layer of uncured PDMS. A retaining
ring, made from PDMS, was then sealed to the mold to eventually hold melted PS. A
surface modification on the PDMS was then performed by preparing a 0.1 M solution of
(3-mercaptopropyl)-trimethylsiloxane (Sigma-Aldrich, St. Louis, MO, USA) in
61

Figure 3.1: Fabrication of polystyrene devices with integrated channels. A silicon master
is coated with SU-8 50 photoresist, and exposed to UV light through a positive
photomask containing the serpentine features. The master is then developed and a PDMS
replicate of the master is made. This PDMS mold is then sealed to a piece of glass, and a
retaining ring of PDMS sealed along the outside. The PDMS is modified with 0.1 M (3mercaptopropyl)-trimethylsiloxane, and polystyrene melted into the mold at 185 °C for 9
hours. The completed device is then removed and a PDMS cover slab with punched
access holes is sealed to the polystyrene device through exposure to air plasma.

62

The acetonitrile was allowed to evaporate, and then approximately 5 g of polystyrene
from Petri dishes were melted in the PDMS mold via contact heating for 9 hours on a
hotplate set to 185 ˚C. The hotplate was then turned off and the PS device cooled to room
temperature before it was removed from the mold. Once the PS device was removed
from the mold, the channels were air plasma sealed to a 10:1 PDMS layer with an inlet
(1/8 in) at each channel. This PDMS layer was made by pouring the elastomer and curing
agent mixture onto a blank silicon wafer and after curing, inlets were punched. In order
to achieve air plasma sealing, the PDMS and PS layers were cleaned with isopropyl
alcohol and doubly deionized water before being dried under vacuum for 5 min and then
exposed to air plasma for 1 min.

37

Immediately following exposure to plasma, the two

layers were aligned by hand and sealed. To ensure permanent sealing, the device was
placed in a 75 ˚C oven for 30 min.
A PDMS injection block was also fabricated for sample delivery. Pipet tips were
suspended in a 10:1 ratio of elastomer:curing agent and cured for 30 min at 75 ˚C. The
pipet tips were then removed and inlets were punched with a 20 gauge luer stub adapter.
The PDMS injection block was then cut to the size of the polystyrene chip and air plasma
sealed to the device in the same manner as described above.

3.2.2 Integration of the Polystyrene Device with a Plate Reader
In order to make the device amenable for measurement in a plate reader (Spectramax
M4, Molecular Devices, Sunnyvale, CA, USA) a conventional, black, 96-well plate
(Greiner Bio-One, Monroe, NC, USA) was modified by removing the bottom of wells
with a rotary tool. This plate was then aligned over the serpentine channels in the
63

fabricated polystyrene microfluidic device, and the assembly was inserted into the plate
reader for analysis.

3

Fluorescein standards were then pipeted into the device using the

pipet interface block described in the methods section above. Fluorescence measurements
were obtained using an excitation wavelength of 494 nm and an emission wavelength of
521 nm.

Fluorescein standards were prepared in phosphate buffered saline at

concentrations of 0, 18, 37, 75, and 150 µM. Standards were pumped through individual
channels on the polystyrene device using a 100 L pipet set to 40 L.
3.3 Results
Although PDMS has been widely used as a substrate material for microfluidic
devices, limitations of its use for biological assays include poor cell adhesion, absorption
of hydrophobic molecules, and swelling in organic solvents.

11

There have been several

reports of fabricating polystyrene-based microfluidic devices with hot embossing
methods.

29-30, 34

However, many groups that routinely fabricate microfluidic devices do

not necessarily have access to all the tools needed for hot embossing so, being able to
utilize the common soft lithography equipment already in place would be beneficial.
Additionally, hot embossing techniques do not allow for any integration of analytical
tools during the fabrication process, such as electrodes or pumping mechanisms. In the
approach described by Allbritton’s group, where GBL was used to dissolve PS and then
poured onto a PDM mold, resulted in devices that are thin (0.3 mm) and requires full
evaporation of the solvent used to dissolve the PS.

64

36

Thicker PS layers are needed (3

mm) to allow for integrated channels as well as be a robust foundation for integrating the
PDMS injection block.
This led to the development of a different strategy based upon melting of polystyrene
from Petri dishes on a conventional hotplate against a PDMS mold. The goal of this
method is to produce thicker, robust devices that can be fabricated in a day with no
special equipment, bubble formation or solvent evaporation. Furthermore, these devices
were further integrated with pipets for pumping and plate reader detection schemes.
Additionally, the Martin group has shown that these devices can be similarly fabricated
with electrodes and suitable for electrochemical detection schemes.

39

3.3.1 Fabrication of Polystyrene-based Devices
Figure 3.1 outlines the process by which PDMS molds were created for subsequent
channel formation in polystyrene devices. The process of fabricating a master, where a
lithographically-derived master with recessed features was used to create the PDMS mold
with raised features. A retaining wall fabricated from PDMS was then sealed to this
mold. It was determined that the mold and retaining ring should be of the same material
to ensure equal thermal expansion. Thus, when pieces of a polystyrene-based Petri dish
were melted onto the PDMS with retaining walls, serpentine channels were formed in the
polystyrene. While upper and lower channel width limits were not determined here, in
this study we routinely utilized channel widths ranging from 200 to 300 µm. In order to
seal the channels on the polystyrene device, a 10:1 PDMS layer with an inlet and waste
well for each serpentine channel was subjected to air plasma and irreversibly sealed to the
PS.
65

Images of the master, PDMS mold and corresponding PS device were taken using
scanning electron microscopy (SEM). The PDMS mold and PS device were sputter
coated with gold prior to imaging. The master contains recessed channels, which
correspond to raised channels on the PDMS mold. The PS device, which is made from
the PDMS mold, also has recessed channels. As shown in figure 3.2, the SEM images
illustrate the successful transfer of features from the PDMS mold to the PS device as the
feature on the mold is replicated onto the PS.
The last step in device construction involved the creation of an injection block
that was molded out of PDMS using suspended pipet tips. The PDMS block was cut to
size and aligned over the inlets that were punched in the PDMS layer sealed to the
polystyrene. The block was then air plasma sealed in a similar manner to the PDMS
layer and the final device is shown in figure 3.3. A unique feature of the injection block
is that it is molded to support the use of pipets, thus enabling a standard micropipet or
fluidic injection system to serve as the device pump. Pipet tips, suitable for a 100 or 200
µL pipet, were used to create the mold and thus, deliver solutions to the serpentine
channels and out the waste well (figure 3.3). This allows for the use of pipets to induce
flow rather than syringe-based pumps that involve multiple syringes and intensive rinsing
procedures. A multi-channel pipet can be used to pump solution through 8 channels
simultaneously and as shown in a side view of a device (figure 3.3), each pipet tip
directly introduces fluid to the channel. It has also been shown that this device is capable
of enabling blood flow,

39

which is a complex biological fluid often studied in clinical

laboratories. The ability to propel small volumes of biological samples through a

66

Figure 3.2: SEM Images of master, PDMS mold and PS device. Top left: image of
sample master with recessed channels. Top right: image of PDMS mold with raised
channels. Bottom: image of corresponding PS device made from PDMS mold. PS device
has recessed channels.

67

Figure 3.3: High-throughput micropipet pumping in polystyrene based microfluidic
device. Top-view of the device, where an 8-channel pipet can individually and
simultaneously address each channel. A pipet block is molded from immobilized 200 µL
pipet tips in PDMS. This is sealed to the cover slab of PDMS over an access hole. The
polystyrene device containing the serpentine channels is then sealed to the PDMS
components. The side view of the device shows how the pipet tips address each
serpentine channel.

68

microfluidic device in a high throughput manner could be useful in laboratories where
such fluids are investigated on a routine basis.
3.3.2 Alignment in Plate Reader
An optical plate reader is standard equipment in most high throughput screening
laboratories. The serpentine channels on the polystyrene device were fabricated to have
the same width and distance apart as wells on a standard 96-well plate so that they could
be analyzed on a standard plate reader. Fluorescein standards were prepared in phosphate
buffered saline at concentrations of 0, 18, 37, 75, and 150 µM and then used to optimize
alignment of the polystyrene device in the plate reader. Standards were pumped through
individual channels on the polystyrene device using a 100 L pipet set to 40 L (figure
3.4A). In order to make the device amenable for measurement in the plate reader, a 96well plate was aligned on top of the device so that drilled-out wells corresponded to each
channel’s serpentine (figure 3.4B). Fluorescence measurements were then taken using an
excitation wavelength of 494 nm and an emission wavelength of 521 nm.

It was

important to determine whether or not the device was aligned properly in the plate reader
by testing the linearity of the fluorescein standard curve. The alignment in the plate
2

reader proved successful as fluorescence intensity was linear (R =0.99) with
concentration (figure 3.4C). To ensure that the top PDMS layer had sufficiently sealed to
the channel containing PS layer, fluorescence microscopy was used to show that the
channels were not leaking. As shown in figure 3.5, when fluorescein is pumped through
the device using a pipet and imaged using a fluorescein isothiocyanate (FITC) filter on a
fluorescence microscope, the edges of the channel are clearly visible, indicating the
channel is sealed.
69

Figure 3.4: Pumping, alignment and fluorescence analysis from a polystyrene device. A)
Fluorescein standards are pumped with a pipet into the device. B) The device is then
placed in a plate reader and fitted with a microplate as a guide. Fluorescence analysis is
performed with ex. 494 nm and em. 521 nm.

70

Figure 3.5: Image of fluorescein in serpentine channels. After fluorescein (ex. 494 nm,
em. 521 nm) was pumped through the serpentine channel, fluorescence microscopy was
used take an image of the channel. The “bright” portion is the fluorescein in the channel,
whereas the “dark” portion is the space between bends in the serpentine, thus, the channel
was successfully sealed and is not leaking.

71

One of the other concerns about these devices was the potential for analyte absorption
into the material. Absorption was investigated in PDMS and native polystyrene devices
using the anti-platelet drug, clopidogrel.

It was found that clopidogrel absorbs

significantly in a PDMS device after 30 minutes, when compared to standards. However,
for the PS device, clopidogrel was recovered at statistically equivalent concentrations to
the standards. These studies were performed using mass spectrometry for analysis.

39

3.4 Conclusions
Microfluidic devices can be fabricated from polystyrene using standard softlithographic procedures and equipment without the need for embossing procedures or
solvent evaporation wait time. This new fabrication strategy allows for any lab already
set up to do standard lithography to create a polystyrene device. In addition, the rigidity
of polystyrene makes it more amenable to integration with equipment associated with
high- throughput screening, such as plate readers and pipets for fluid delivery.
The ultimate goal for these microfluidic devices would be to incorporate multiple
cells types on one device and integrate multiple detection schemes. In the Spence group,
this would involve integrating more blood components, such as red blood cells, platelets
and endothelial cells, to study these cells together as a system, rather than individually, in
order to better imitate the in vivo environment. Furthermore, when multiple cell types are
used there is often the desire to measure multiple analytes related to the cell system being
studied. In the example of blood components, such analytes as ATP, NO, Ca

2+

would be

relevant to study, but would employ detections techniques of chemiluminescence,
electrochemistry, and fluorescence. However, the ability to integrate multiple detection
72

schemes is often limited by the reliability of the microfluidic device being used. When
multiple cell types, detection schemes, and pumping mechanism are used on one device
the need for exact timing and proper function is important. If one detection scheme
should fail, or one cell type fails to culture, then a whole experiment may be deemed a
failure. Thus, reliability of the devices being used has to be examined.
PDMS is a suitable material for device prototyping, but its flexibility does not make it
suitable for robust device designs. PS has a low glass transition temperature, which
allows it to be molded at lower temperatures and it is the material commonly used for cell
culture. PS is also a rigid material, which lends itself to integration with pumping
mechanism and detection schemes as PS can be molded into a 96-well plate format
suitable for plate reader detection. For these reasons, the PS device describe herein has
the potential to be used to study biological process occurring in cells as polystyrene is a
common material used in cell culture and can be integrated with pumping mechanism and
detection schemes. Chapter 4 will investigate the use of these devices in studying cellular
behavior.

73

REFERENCES

74

REFERENCES

1.
D. C. Duffy, J. C. McDonald, O. J. A. Schueller, G. M. Whitesides, Rapid
Prototyping of Microfluidic Systems in Poly(dimethylsiloxane). Analytical Chemistry
1998, 70. 4974-4984.
2.
G. M. Whitesides, E. Ostuni, S. Takayama, X. Jiang, D. E. Ingber, Soft
lithography in biology and biochemistry. Annu. Rev. Biomed. Eng. 2001, 3. 335-373.
3.
S. T. Halpin, D. M. Spence, Direct Plate-Reader Measurement of Nitric Oxide
Released from Hypoxic Erythrocytes Flowing through a Microfluidic Device. Anal.
Chem. 2010, 82. 7492-7497, DOI: 10.1021/ac101130s.
4.
S. C. Jacobson, C. T. Culbertson, J. E. Daler, J. M. Ramsey, Microchip Structures
for Submillisecond Electrophoresis. Anal. Chem. 1998, 70. 3476-3480, DOI:
10.1021/ac980349t.
5.
M. W. Li, A. L. Bowen, N. G. Batz, R. S. Martin, in Lab on a Chip Technology,
Part II: Fluid Control and Manipulation, ed. K. E. Herold, A. Rasooly. Caister Academic
Press: Nofolk, UK, 2009, pp 385-403.
6.
A. L. Bowen, R. S. Martin, Integration of On-Chip Peristaltic Pumps and
Injection Valves with Microchip Electrophoresis and Electrochemical Detection.
Electrophoresis 2010, 31. 2534–2540.
7.
L. I. Genes, N. V. Tolan, M. K. Hulvey, R. S. Martin, D. M. Spence, Addressing a
vascular endothelium array with blood components using underlying microfluidic
channels. Lab Chip 2007, 7. 1256-1259.
8.
L. C. Mecker, R. S. Martin, Integration of Microdialysis Sampling and Microchip
Electrophoresis with Electrochemical Detection. Anal. Chem 2008, 80. 9257-9264.
9.
S. D. Noblitt, G. S. Lewis, Y. Liu, S. V. Hering, J. L. Collett, C. S. Henry,
Interfacing Microchip Electrophoresis to a Growth Tube Particle Collector for
Semicontinuous Monitoring of Aerosol Composition. Anal. Chem. 2009, 81. 1002910037, DOI: 10.1021/ac901903m.
10.

J. El-Ali, P. K. Sorger, K. F. Jensen, Cells on chips. Nature 2006, 442. 403-411.

11.
M. L. Kovarik, P. C. Gach, D. M. Ornoff, Y. Wang, J. Balowski, L. Farrag, N. L.
Allbritton, Micro Total Analysis Systems for Cell Biology and Biochemical Assays. Anal
Chem 2012, 84. 516-540.

75

12.
R. S. Martin, P. D. Root, D. M. Spence, Microfluidic technologies as platforms
for performing quantitative cellular analyses in an in vitro environment. Analyst 2006,
131. 1197-1206.
13.
D. M. Spence, N. J. Torrence, M. L. Kovarik, R. S. Martin, Amperometric
Determination of Nitric Oxide in An Endothelial Cell-coated Microchip Channel. Analyst
2004, 129. 995-1000.
14.
L. I. Genes, N. V. Tolan, M. K. Hulvey, R. S. Martin, D. M. Spence, Addressing a
vascular endothelium array with blood components using underlying microfluidic
channels. Lab Chip 2007, 7. 1256-1259.
15.
T. D. A. Oblak, P. Root, D. M. Spence, Fluorescence Monitoring of ATPStimulated, Endothelium-Derived Nitric Oxide Production in Channels of a
Poly(dimethylsiloxane)-Based Microfluidic Device. Analytical Chemistry 2006, 78.
3193-3197.
16.
C.-J. Ku, T. D'Amico Oblak, D. M. Spence, Interactions between Multiple Cell
Types in Parallel Microfluidic Channels: Monitoring Platelet Adhesion to an
Endothelium in the Presence of an Anti-Adhesion Drug. Anal. Chem. (Washington, DC,
U. S.) 2008, 80. 7543-7548.
17.
E. W. K. Young, A. R. Wheeler, C. A. Simmons, Matrix-dependent adhesion of
vascular and valvular endothelial cells in microfluidic channels. Lab Chip 2007, 7. 17591766.
18.
S. Chung, R. Sudo, P. J. Mack, C.-R. Wan, V. Vickerman, R. D. Kamm, Cell
migration into scaffolds under co-culture conditions in a microfluidic platform. Lab Chip
2009, 9. 269-275.
19.
R. Gomez-Sjoberg, A. A. Leyrat, D. M. Pirone, C. S. Chen, S. R. Quake,
Versatile, fully automated, microfluidic cell culture system. Analytical Chemistry 2007,
79. 8557-63, DOI: 10.1021/ac071311w.
20.
E. W. K. Young, M. W. L. Watson, S. Srigunapalan, A. R. Wheeler, C. A.
Simmons, Technique for Real-Time Measurements of Endothelial Permeability in a
Microfluidic Membrane Chip Using Laser-Induced Fluorescence Detection. Anal. Chem.
(Washington, DC, U. S.) 82. 808-816.
21.
P. A. Vogel, S. T. Halpin, R. S. Martin, D. M. Spence, Microfluidic
Transendothelial Electrical Resistance Measurement Device that Enables Blood Flow and
Postgrowth Experiments. Anal Chem 2011, 83. 4296–4301.
22.
M. W. Li, R. S. Martin, Microchip-based integration of cell immobilization,
electrophoresis, post-column derivatization, and fluorescence detection for monitoring
the release of dopamine from PC 12 cells. Analyst (Cambridge, United Kingdom) 2008,
133. 1358-1366.
76

23.
M. W. Toepke, D. J. Beebe, PDMS absorption of small molecules and
consequences in microfluidic applications. Lab Chip 2006, 6. 1484-1486.
24.
K. J. Regehr, M. Domenech, J. T. Koepsel, K. C. Carver, S. E. Ellison-Zelski, W.
L. Murphy, L. A. Schuler, E. T. Alarid, D. J. Beebe, Biological implications of
polydimethylsiloxane-based microfluidic cell culture. Lab Chip 2009, 9. 2132–2139.
25.
E. Verneuil, A. Buguin, P. Silberzan, Permeation-induced flows: Consequences
for silicone-based microfluidics. Europhys. Lett. 2004, 68. 412-418.
26.
I. Meyvantsson, J. W. Warrick, S. Hayes, A. Skoien, D. J. Beebe, Automated cell
culture in high density tubeless microfluidic device arrays. Lab Chip 2008, 8. 717-724.
27.
J. P. Puccinelli, X. Su, D. J. Beebe, Automated high-throughput microchannel
assays for cell biology: operational optimization and characterization. Jala 15. 25-32.
28.
I. Barbulovic-Nad, H. Yang, P. S. Park, A. R. Wheeler, Digital microfluidics for
cell-based assays. Lab Chip 2008, 8. 519-526.
29.
P. M. van Midwoud, A. Janse, M. T. Merema, G. M. M. Groothuis, E. Verpoorte,
Comparison of Biocompatibility and Adsorption Properties of Different Plastics for
Advanced Microfluidic Cell and Tissue Culture Models. Anal Chem 2012, 84. 3938–
3944.
30.
E. W. Young, E. Berthier, D. J. Guckenberger, E. Sackmann, C. Lamers, I.
eyvantsson, A. Huttenlocher, D. J. Beebe, Rapid prototyping of arrayed microfluidic
systems in polystyrene for cell-based assays. Anal Chem 2011, 83. 1408-1417.
31.
D. A. Detwiler, N. C. Dobes, C. E. Sims, J. N. Kornegay, N. L. Allbritton,
Polystyrene-coated micropallets for culture and separation of primary muscle cells. Anal
Bioanal Chem 2012, 402. 1083-1091.
32.
G. Mehta, J. Lee, W. Cha, Y.-C. Tung, J. J. Linderman, S. Takayama, Hard Top
Soft Bottom Microfluidic Devices for Cell Culture and Chemical Analysis. Anal. Chem.
2009, 81. 3714-3722, DOI: 10.1021/ac802178u.
33.
A. Chen, D. K. Lieu, L. Freschauf, V. Lew, H. Sharma, J. Wang, D. Nguyen, I.
Karakikes, R. J. Hajjar, A. Gopinathan, E. Botvinick, C. C. Fowlkes, R. A. Li, M. Khine,
Shrink-Film Configurable Multiscale Wrinkles for Functional Alignment of Human
Embryonic Stem Cells and their Cardiac Derivatives. Advanced Materials 2011, 23.
5785-5791, DOI: 10.1002/adma.201103463.
34.
V. N. Goral, Y. C. Hsiesh, O. N. Petzold, R. A. Faris, P. K. Yuen, Hot embossing
of plastic microfluidic devices using poly(dimethylsiloxane) molds. J Micromech
Microen 2011, 21. 017002.

77

35.
G. Mehta, J. Lee, W. Cha, Y.-C. Tung, J. J. Linderman, S. Takayama, Hard Top
Soft Bottom Microfluidic Devices for Cell Culture and Chemical Analysis. Anal. Chem.
(Washington, DC, U. S.) 2009, 81. 3714-3722.
36.
Y. Wang, J. Balowski, C. Phillips, R. Phillips, C. E. Sims, N. L. Allbritton,
Benchtop micromolding of polystyrene by soft lithography Lab Chip 2011, 11.
37.
D. C. Duffy, J. C. McDonald, O. J. A. Schueller, G. M. Whitesides, Rapid
Prototyping of Microfluidic Systems in Poly(dimethylsiloxane). Anal. Chem 1998, 70.
4974-4984, DOI: 10.1021/ac980656z.
38.
R. S. Martin, A. J. Gawron, S. M. Lunte, C. S. Henry, Dual-electrode
electrochemical detection for poly(dimethylsiloxane)-fabricated capillary electrophoresis
microchips. Anal. Chem. 2000, 72. 3196-3202.
39.
A. S. Johnson, K. B. Anderson, S. T. Halpin, D. C. Kirkpatrick, D. M. Spence, R.
S. Martin, Integration of multiple components in polystyrene-based microfluidic devices
part I: fabrication and characterization. Analyst (Cambridge, United Kingdom) 138. 129136.

78

CHAPTER 4: ANALYSIS OF ENDOTHELIAL CELLS USING A
POLYSTYRENE BASED MICROFLUIDIC DEVICE AND 96 WELL PLATE
READER DETECTION
4.1 Introduction
As discussed in Chapter 3, microfluidic devices have become popular for studying
cells due to their ability to mimic in vivo dimensions (e.g. blood vessels, which can be
anywhere from 10 µm to 25 mm in diameter) and integrate various detection schemes.

1-4

Polydimethylsiloxane (PDMS) is one of the most widely-used materials in microfluidic
device fabrication because it is easy to work with, inexpensive, transparent and gas
5-6

permeable.

applications,

Although PDMS devices have been successfully used in numerous cell

1, 7-9

there are limitations with this material, including: poor cell adhesion

for some cell lines, partitioning of hydrophobic molecules in the PDMS, and leaching of
un-crosslinked monomers from the bulk PDMS.

10

Thus, making microfluidic devices

out of more biocompatible substrates, such as polystyrene (PS), has been the focus of
several studies.

11-15

Chapter 3 explained how these PS devices were fabricated, while

Chapter 4 will focus on how these devices can be utilized for studying a cell based
system.

4.1.1 Cell Analysis on Polystyrene Microfluidic Devices
Takayama et al. used a device that had a PS top with channels embedded, and
sealed the device with a polyurethane (PU) soft bottom. Human dermal microvascular
cells (HDMECs) were cultured on the device and oxygen tension measurements were
performed via the use of an oxygen sensitive fluorescent dye. The results showed lower
79

evaporation and lower oxygen permeability compared to a device made completely from
PDMS.

16

However, in this construct, the only cellular work that occurred was successful

cell culture on the device, as no other cellular analyses were performed.
Beebe et al. fabricated arrayed polystyrene devices via a simple hot embossing
method that involves PDMS and epoxy molds. Using these arrayed devices, human
umbilical vein endothelial cells (HUVECs) were cultured in the device and after
activation by interleukin 1β, the upregulation of E-selectin was monitored via
fluorescence imaging. Additionally, chemotaxis of neutrophils was monitored across the
length of the channel in the presence of a chemoattractant via imaging.

12

This is a good

example of how microscopy can be employed to study cell upregulation and migration on
a polystyrene device.
Another example of a PS device used to study cell behavior is from Allbritton et
al. where PS was molded into arrays of either microwells or microposts for segregation
and tracking of non-adherent and adherent cells. The microwells were used to capture
nonadherent cells and the adherent cells would attach to the microposts and proliferate.
Scanning electron microscopy (SEM) was used to track this cell activity.

17

Allbritton et

al. also created PS coated micropallets for improving cell adhesion and proliferation of
primary muscle cells.

13

Lastly, Midwoud et al. recently compared the use of different thermoplastics,
including polystyrene, for the adherence of human hepatoma cells in a patterned
structure, with the substrates being characterized in terms of surface treatment, adsorption
of hydrophobic compounds, and biocompatibility.
80

11

All of the previously mentioned studies demonstrated that polystyrene is a
suitable substrate for cell studies performed on microfluidic devices. However, there
have not been any examples of using these devices to monitor intra- or extra-cellular
function by integrating the devices with current laboratory equipment.

The device

presented in Chapter 3 has the ability to use a pipet as a pumping mechanism as well as a
96 well plate reader for detection of cellular events.

Furthermore, cell culture is

successful on the device and in this construct, cell uptake, production and release can be
monitored for several analytes.
4.2 Methods
4.2.1 Endothelial Cell Culture in Polystyrene-based Microchannels
In order to facilitate cell adhesion in the channels of the polystyrene device, a 100
g/mL solution of bovine plasma fibronectin (Invitrogen, Carlsbad, CA, USA) cell
adhesion protein was introduced to the channels via the injection block (described in
Chapter 3) and allowed to adsorb for 1 hr in a 37 C and 5% CO2 incubator. The
channels were then dried and exposed to UV light for 15 minutes. Bovine pulmonary
artery endothelial cells (bPAECs) were purchased in frozen cryovials (Lonza,
Walkersville, MD, USA). The vials were thawed to room temperature and added to a T75 tissue culture flask containing 9 mL of endothelial growth media (EGM) that had been
equilibrated to 37 C.

The EGM consists of a low glucose (5.5 mM) Dulbecco’s

Modified Eagles Medium (DMEM, MIDSCI, St. Louis, MO, USA) supplemented with
2.5% v/v adult bovine serum (Sigma-Aldrich, St. Louis, MO, USA), 7.5% fetal bovine
serum (Lonza), penicillin, streptomycin, and amphotericin B (MIDSCI, St. Louis, MO
USA). bPAECs were allowed to grow in a humidified incubator at 37 C and 5% CO2
81

until they were determined as confluent by optical microscopy. Media was changed the
day after plating and then every 2 days thereafter. The bPAECs were subcultured when
the cells reached >80% confluence as visualized by optical microscopy.
In order to seed bPAECs into the channels of the microfluidic device, bPAECs were
washed with 10 mL of HEPES and then treated with 5 mL of 0.25% trypsin/EDTA. This
solution was then removed and the cells suspended in 10 mL of EGM.

The cell

suspension was removed from the flask and centrifuged at 1500 g for 5 min. The
supernatant was removed and the pellet was resuspended in 450 µL of equilibrated
media. This concentrated cell solution was introduced to the channels in the same
manner as the fibronectin and incubated for 1 hr at 37 C and 5% CO2. After an hour of
growth, the ECs were re-seeded in the channel and media was subsequently changed
every 2 hrs. The ECs were allowed to grow to confluence overnight and used the day
after seeding.
To verify cell confluence in the channels, 1 µM 5-chloromethylfluorescein diacetate
(CMFDA) cell tracker (Molecular Probes, Carlsbad, CA, USA) was pumped through the
channels to fluorescently label the bPAECs for monitoring with an optical microscope
(Olympus IX71 Microscope, Olympus America, Melville, NY, USA) fitted with a FITC
filter cube (Chroma Technology Corp, Bellows Falls, VT, USA) containing the excitation
(460-500 nm) and emission (505-560 nm) filters.

4.2.2 Fluorescence Detection of Intracellular Endothelial Nitric Oxide
A 5 mM stock solution of the intracellular nitric oxide (NO) probe DAF-FM-DA (4amino-5-methylamino-2’,7’-difluorofluorescein diacetate, Molecular Probes) was made
82

by dissolving 50 g in 20 L anhydrous dimethyl sulfoxide (DMSO). A 10 M solution
of DAF-FM-DA was prepared by diluting 2 L of stock to 1 mL with physiological salt
solution (PSS). The concentrations of salts (in mM) in the PSS were as follows: 4.7 KCl,
2.0 CaCl2, 1.2 MgSO4, 140.5 NaCl, 21.0 Tris, and 5.5 glucose (pH 7.4). The 10 M
DAF-FM-DA solution was then pumped (via the PDMS injection block) over

the

bPAECs using a 100 L pipet set to 40 L, and then repeated to ensure adequate cell
exposure to probe. The device was then incubated for 1 hr at 37 C and 5% CO2. After
incubation, the channels were rinsed twice with PSS to remove excess probe, followed by
a waiting period of 10 min at 37 C and 5% CO2 to allow for probe de-esterification by
intracellular mechanisms.
Initial fluorescence measurements of the bPAECs were taken by aligning a 96 well
plate, with drilled out wells corresponding to each channel’s serpentine, on top of the PS
device. Fluorescence measurements were performed using a plate reader (Spectramax
M4, Molecular Devices, Sunnyvale, CA, USA) set to an excitation wavelength of 495 nm
and an emission wavelength of 521 nm. These measurements served to represent the
baseline levels of NO in the cells. After measurement, adenosine triphosphate (ATP,
Sigma-Aldrich) was pumped over the bPAECs in the same manner as the probe. ATP
standards were prepared by dissolving ATP in 25 mL PSS to make a 1 mM stock
solution. Next, a 100 M solution was prepared by diluting 100 uL of the ATP stock
solution to 1 mL; this 100 M solution was then diluted further to prepare a 0.5 M ATP
solution that was subsequently pumped over the bPAECs in the device. Following a 30
minute incubation period at 37 C and 5% CO2, the final fluorescence measurements
83

were taken on the plate reader in the same manner as before.

The differences in

fluorescence were obtained and normalized to the channels that had been addressed with
0 M ATP (PSS alone).

4.2.2 Fluorescence Detection of Endothelial Calcium Uptake
2+

A 2.3 mM stock solution of the intracellular calcium (Ca ) specific fluorescence
probe Fluo-4 AM (Invitrogen, Grand Island, NY, USA) was made by dissolving a 50 g
vial in 20 L of anhydrous DMSO. Next, a 5 µL aliquot of the stock solution, along with
5 µL of 200 mg/mL pluronic F-127 surfactant (Invitrogen, Grand Island, NY, USA),
were combined in order to enhance the probe’s penetration capacity. Finally, the probe
solution was diluted to 2 mL in Ca

2+

free PSS to create a 5.7 M solution. The 5.7 M

Fluo-4 AM was then pumped over the bPAECs using a 100 L pipet set to 40 L (via the
PDMS injection block), then repeated to ensure adequate cell exposure to probe. The
device was then incubated for 30 min at 37 C and 5% CO2. After incubation, the
channels were rinsed twice with Ca

2+

free PSS to remove excess probe and incubated for

an additional 10 min at 37 C and 5% CO2 to allow for probe de-esterification by
intracellular mechanisms.

Initial fluorescence measurements of the bPAECs were

obtained using the plate reader as described above for the NO, although the excitation
and emission wavelength settings were modified to 496 nm and 516 nm, respectively.
These measurements served to represent the baseline levels of Ca

84

2+

in the cells.

2+

After measurement, PSS (with Ca ) or 10 µM Ca

2+

ionophore A23187 (Sigma-

Aldrich, St. Louis, MO USA) in PSS were pumped over the bPAECs in the same manner
as the probe. The Ca

2+

ionophore was prepared by dissolving 5 mg in 1.9 mL of

anhydrous DMSO to create a 5 mM stock solution. Next, a 100 µM working solution was
prepared by diluting 20 µL of stock to 1 mL in PSS and finally, a 1:10 dilution in PSS
was performed, resulting in a final concentration of 10 µM Ca

2+

ionophore. Once PSS or

ionophore was flowed over the bPAECs, the device was allowed to incubate for 10 min at
37 C and 5% CO2. Final fluorescence measurements were taken on the plate reader as
described above.

4.2.3 Cell Viability Study
Saponin (Sigma Aldrich, St. Louis, MO, USA) was used to permeabilize bPAEC
membranes in order to induce cell death. A 0.25% solution of Saponin was made by
dissolving 13.2 mg into 5.28 mL of water. A 1 mL aliquot of the 0.25% solution was
then diluted to 10 mL with PSS resulting in a 0.025% solution. This solution, or PSS,
was then flowed over bPAECs cultured in the polystyrene device.
The nucleic acid stain, Sytox Green (Invitrogen), which only labels cell with
compromised membranes, was used to determine whether or not saponin had killed the
immobilized bPAECs.

Its excitation and emission maxima are 504 and 523 nm,

respectively when bound to DNA. After Saponin or PSS was flowed over the cells, 1 µM
Sytox Green was flowed over the bPAECs. The probe was allowed to incubate at 37 ˚C

85

for 15 min. and then imaged using fluorescence microscopy and fluorescence intensity
recorded using a plate reader.
4.3 Results
A limitation of PDMS devices is poor cell adhesion of some cell lines for example,
PC 12 cells. Adherent proteins, such as collagen, or fibronectin, are commonly used to
facilitate cell adherence can be coated onto PDMS surfaces. However, despite the use of
adhesion proteins, cells are still not as adherent on PDMS as they are in conventional cell
culture substrates such as polystyrene.

Previous studies have investigated the

biocompatibility of cells on various thermoplastics including polystyrene, with the
surfaces being rendered hydrophilic by means of UV-ozone or plasma treatment.

11, 17

In

the case of PC 12 cells, it was found that they were more adherent on a polystyrene
surface than a PDMS device with 8% and 75% of the cells lost after rinsing,
respectively.

18

4.3.1 Monitoring Nitric Oxide Production, Calcium Uptake and Viability in
Endothelial Cells
The pumping mechanism using the PDMS injection block to interface with a pipet (as
described in Chapter 3) was used to introduce bPAECs into to the polystyrene-based
microchannels. A side view of one channel is shown in Figure 4.1, where a layer of
fibronectin (cell adhesion protein) was first introduced to the channel (via the injection
block). After allowing the fibronectin to absorb for 1 hr, a concentrated solution of

86

Figure 4.1: Endothelial cell culture on a polystyrene device. To culture endothelial cells, the device was first coated with fibronectin.
Next, a suspension of endothelial cells was pumped into the device. The device was then incubated at 37 °C, 5% CO 2 for 2 hours.
Media was changed every two hours until cell confluence was observed. A bright field image is shown on the top right, while the
image on the bottom right was obtained by incubating the cells with CellTracker CMFDA fluorescenct probe.

87

bPAECs was introduced into the device and allowed to adhere and grow in the
microchannel (in an incubator), with frequent media changes (every 2 hrs). The bPAECs
on the polystyrene grew to confluence overnight and were visualized by optical
microscopy as shown in the bright field image in Figure 4.1. To further ensure that the
cells in the serpentine channels were confluent, they were also labeled with the cell
tracker CMFDA and visualized under the FITC filter of the microscope (as pictured in
the fluorescent image, Figure 4.1). Before any biological studies were performed on the
bPAECs cultured in the polystyrene device, they were first visualized with optical
microscopy to confirm the presence of healthy, confluent cells.
The in-channel cultured bPAECs were incubated with the intracellular NO probe
DAF-FM DA and then exposed to ATP, a known stimulus of NO production in
endothelial cells;

7, 19

therefore, it was anticipated that ATP would result in increased

intracellular bPAEC NO production. Upon application of 0.5 µM ATP to the cells in the
serpentine channel, a significant increase in NO production was observed when compared
to applying PSS alone (Figure 4.2). This increase was measured by aligning the device in
the plate reader and performing top-read fluorescence measurements in a high-throughput
manner.
2+

Ca

The device was also utilized to measure cellular uptake by monitoring

influx into the bPAECs after stimulation by a Ca
2+

treated with Ca

2+

ionophore. Cells in the device

ionophore exhibited a 42 ± 13% increase in fluorescence compared to

cells treated with PSS alone, which only exhibited a 17 ± 2% increase in fluorescence
(N=3, error: SEM, p-value < 0.05) as shown in Figure 4.3. These studies suggest that the
polystyrene devices with imprinted channels can be integrated with existing infrastructure

88

Figure 4.2: Intracellular measurement of NO. These cells were incubated with 10 µM
DAF-FM-DA, an intracellular NO probe, at 37 C for 30 min; excess probe was rinsed
off the cells by pumping buffer without probe through the channels. After the incubation
period, a basal fluorescence measurement was taken using a standard plate reader. Next,
0.5 µM ATP was pumped over the cells and allowed to incubate for an additional 30 min
in order to stimulate NO production in the endothelial cells. A final fluorescence
measurement was taken using the plate reader and the difference in fluorescence
calculated. This value was normalized to the difference in fluorescence for a 0 µM ATP
standard. The average changes in fluorescence for n=3 separate devices is shown in,
along with error bars representing the standard error of the mean. The changes are
significant for p<0.001.

89

Figure 4.3 Intracellular Measurement of Ca

2+

Uptake: These cells were incubated with

2+

5.7 µM Fluo 4AM, an intracellular Ca probe, at 37 C for 30 min; excess probe was
rinsed off the cells by pumping buffer without probe through the channels. After the
incubation period, a basal fluorescence measurement was taken using a standard plate
2+
reader. Next, 10 µM Ca ionophore was pumped over the cells and allowed to incubate
2+

for an additional 10 min in order to stimulate Ca uptake in the endothelial cells. A
final fluorescence measurement was taken using the plate reader and the difference in
fluorescence calculated. This value was normalized to the difference in fluorescence for
a 0 µM ionophore. The average changes in fluorescence for n=3 separate devices is
shown, along with error bars representing the standard error of the mean. The changes
are significant for p<0.05.

90

(pipets for pumping and plate readers for detection) and can be utilized to study both an
intracellular process (NO production) and cellular uptake (Ca

2+

influx).

Another study was performed to see if this device could be used to monitor cell
viability. Saponin was flowed over cells cultured in the PS device in order to induce cell
death. The bPAECs were then incubated with Sytox Green, which is a fluorescent
nucleic acid stain that does not cross the membrane of live cells. The increase in
fluorescence for cells treated with saponin, versus the control treated with PSS, is shown
in Figure 4.4 and is statistically different (p<0.001).
4.4 Conclusions
The polystyrene device whose fabrication was described in Chapter 3 was used to
study endothelial cells. Endothelial cells, those that comprise the inner wall of blood
vessels, were immobilized into a polystyrene device. These cells were stimulated to
produce nitric oxide, which was determined intracellularly with a fluorescence probe.
Furthermore, these cells’ uptake of Ca

2+

was also determined with an intracellular

fluorescence probe. In a separate set of experiments done by the Martin group, though
+

using a similar polystyrene device integrated with electrodes the stimulated (K ) release
18

of catecholamines from a layer of immobilized PC 12 cells was also measured.

Finally, the ability to induce and study cell death was also accomplished through the use
of this device. These results demonstrate that devices fabricated in polystyrene are
capable of being used in many of the same types of assays as those devices made from
other materials, such as PDMS. The use of polystyrene is advantageous in fabrication

91

Figure 4.4: Cell Viability. Saponin, a detergent that permeabilizes cell membranes, was used to induce endothelial cell death as
visualized by the nucleic acid stain Sytox Green. Sytox Green does not label live cells thus, the image on the right indicates the
presence of dead cells after saponin administration. The bar graph shows an increase in fluorescence, indicative of cell death, for cells
treated with saponin as compared to cells treated with PSS. The average changes in fluorescence for n=5 separate devices is shown,
along with error bars representing the standard error of the mean. The changes are significant for p<0.001.

92

strategies that seek to employ integration with electrodes, injection blocks for pipet
pumping, and plate reader detection schemes. This allows for endothelial cell behaviour
(NO production, Ca

2+

uptake, viability) to be studied in a high throughput manner.

Furthermore, the rigidity of polystyrene and its ability to be integrated with a 96 well
reader show promise in its ability to be further utilized with established, automated lab
equipment.
Though in this construct the PS device was only used to study endothelial cell
behaviour, it has the potential to be a powerful tool used to study multiple cell types. In
Chapter 2, I discussed an aspect of platelet biology that could play a larger role in the
vasculature. The Spence group hypothesizes that the ATP activated P2X1 receptor on the
platelet surface is influenced by RBC and endothelial cell behaviour, specifically by their
release of ATP, NO, and other platelet inhibitors or activators. The PS device can be
used to study interaction between these three cell types.

Whole blood, or RBCs

combined with platelets, could be flowed over a layer of immobilized endothelial cells
and multiple detection schemes can be incorporated in the device. The 96 well plate
format would lend itself to fluorescence measurements or chemiluminescence,
furthermore, cells can be collected at the end of the channel for off chip analysis such as
flow cytometry for platelet activation. In this manner, P2X1 receptor function would be
probed at the RBC and endothelial cell level.
As discussed previously, RBCs release ATP and these levels of ATP may
influence platelet activation. To investigate this relationship, studies previously done in
the Spence lab where platelets incubated with RBCs and various ATP release
activators/inhibitors adhering to an uninjured endothelium
93

20

can be repeated on the PS

platform. In this manner, it would be possible to measure ATP release from RBCs,
endothelial cell NO production and platelet activation.

To measure ATP release a

membrane could be incorporated into the device so that ATP released from RBCs or
platelets could diffuse above the membrane to be measured via the luciferin/luciferase
chemiluminescence reaction, these analytical wells would be fabricated in a 96-well
format. The endothelial cell NO (platelet inhibitor) could be measured as discussed in
this chapter, but the RBCs, due to their red color, would have to be washed out of the
channel before a fluorescence measurement was taken. Finally, samples can be collected
for off chip analysis via flow cytometry to investigate the levels of platelet activation.
Flow cytometry would separate the different cell types and allow for the detection of
fluorescently labelled antibodies indicative of platelet activation.

The integration

possibilities for this PS device would be useful for studying the interactions between
different cell types though more work still needs to be done on integrating detection
2+

schemes for analytes besides NO or Ca .

94

REFERENCES

95

REFERENCES

1.
A. L. Bowen, R. S. Martin, Integration of On-Chip Peristaltic Pumps and
Injection Valves with Microchip Electrophoresis and Electrochemical Detection.
Electrophoresis 2010, 31. 2534–2540.
2.

J. El-Ali, P. K. Sorger, K. F. Jensen, Cells on chips. Nature 2006, 442. 403-411.

3.
R. S. Martin, P. D. Root, D. M. Spence, Microfluidic technologies as platforms
for performing quantitative cellular analyses in an in vitro environment. Analyst 2006,
131. 1197-1206.
4.
D. M. Spence, N. J. Torrence, M. L. Kovarik, R. S. Martin, Amperometric
Determination of Nitric Oxide in An Endothelial Cell-coated Microchip Channel. Analyst
2004, 129. 995-1000.
5.
D. C. Duffy, J. C. McDonald, O. J. A. Schueller, G. M. Whitesides, Rapid
Prototyping of Microfluidic Systems in Poly(dimethylsiloxane). Anal. Chem 1998, 70.
4974-4984, DOI: 10.1021/ac980656z.
6.
M. L. Kovarik, P. C. Gach, D. M. Ornoff, Y. Wang, J. Balowski, L. Farrag, N. L.
Allbritton, Micro Total Analysis Systems for Cell Biology and Biochemical Assays. Anal
Chem 2012, 84. 516-540.
7.
L. I. Genes, N. V. Tolan, M. K. Hulvey, R. S. Martin, D. M. Spence, Addressing a
vascular endothelium array with blood components using underlying microfluidic
channels. Lab Chip 2007, 7. 1256-1259.
8.
S. T. Halpin, D. M. Spence, Direct Plate-Reader Measurement of Nitric Oxide
Released from Hypoxic Erythrocytes Flowing through a Microfluidic Device. Anal.
Chem. 2010, 82. 7492-7497, DOI: 10.1021/ac101130s.
9.
M. W. Li, R. S. Martin, Microchip-based integration of cell immobilization,
electrophoresis, post-column derivatization, and fluorescence detection for monitoring
the release of dopamine from PC 12 cells. Analyst 2008, 133. 1358-1366.
10.
K. J. Regehr, M. Domenech, J. T. Koepsel, K. C. Carver, S. E. Ellison-Zelski, W.
L. Murphy, L. A. Schuler, E. T. Alarid, D. J. Beebe, Biological implications of
polydimethylsiloxane-based microfluidic cell culture. Lab Chip 2009, 9. 2132–2139.
11.
P. M. van Midwoud, A. Janse, M. T. Merema, G. M. M. Groothuis, E. Verpoorte,
Comparison of Biocompatibility and Adsorption Properties of Different Plastics for
Advanced Microfluidic Cell and Tissue Culture Models. Anal Chem 2012, 84. 3938–
3944.
96

12.
E. W. Young, E. Berthier, D. J. Guckenberger, E. Sackmann, C. Lamers, I.
eyvantsson, A. Huttenlocher, D. J. Beebe, Rapid prototyping of arrayed microfluidic
systems in polystyrene for cell-based assays. Anal Chem 2011, 83. 1408-1417.
13.
D. A. Detwiler, N. C. Dobes, C. E. Sims, J. N. Kornegay, N. L. Allbritton,
Polystyrene-coated micropallets for culture and separation of primary muscle cells. Anal
Bioanal Chem 2012, 402. 1083-1091.
14.
G. Mehta, J. Lee, W. Cha, Y.-C. Tung, J. J. Linderman, S. Takayama, Hard Top
Soft Bottom Microfluidic Devices for Cell Culture and Chemical Analysis. Anal. Chem.
2009, 81. 3714-3722, DOI: 10.1021/ac802178u.
15.
A. Chen, D. K. Lieu, L. Freschauf, V. Lew, H. Sharma, J. Wang, D. Nguyen, I.
Karakikes, R. J. Hajjar, A. Gopinathan, E. Botvinick, C. C. Fowlkes, R. A. Li, M. Khine,
Shrink-Film Configurable Multiscale Wrinkles for Functional Alignment of Human
Embryonic Stem Cells and their Cardiac Derivatives. Advanced Materials 2011, 23.
5785-5791, DOI: 10.1002/adma.201103463.
16.
G. Mehta, J. Lee, W. Cha, Y.-C. Tung, J. J. Linderman, S. Takayama, Hard Top
Soft Bottom Microfluidic Devices for Cell Culture and Chemical Analysis. Anal. Chem.
(Washington, DC, U. S.) 2009, 81. 3714-3722.
17.
Y. Wang, J. Balowski, C. Phillips, R. Phillips, C. E. Sims, N. L. Allbritton,
Benchtop micromolding of polystyrene by soft lithography Lab Chip 2011, 11. 30893097.
18.
K. B. Anderson, S. T. Halpin, A. S. Johnson, R. S. Martin, D. M. Spence,
Integration of multiple components in polystyrene-based microfluidic devices part II:
cellular analysis. Analyst (Cambridge, United Kingdom) 138. 137-143.
19.
R. S. Sprague, M. L. Ellsworth, A. H. Stephenson, A. J. Lonigro, ATP: the red
blood cell link to NO and local control of the pulmonary circulation. Am J Physiol 1996,
271. H2717-22.
20.
W. Karunarathne, C. J. Ku, D. M. Spence, The dual nature of extracellular ATP as
a concentration-dependent platelet P2X1 agonist and antagonist. Integr Biol 2009, 1. 655663, DOI: Doi 10.1039/B909873a.

97

CHAPTER 5: CONCLUSIONS AND FUTURE DIRECTIONS
5.1 Platelet P2X1 Receptor Function
Circulating platelets are in an inactive state, but upon blood vessel injury, platelets
become activated and aggregate at the injury site forming a plug, or clot, that prevents
further blood loss. Purinergic receptors on the platelet surface play a role in this clotting
process and include the adenosine diphosphate (ADP) activated P2Y1 and P2Y12, and
the adenosine triphosphate (ATP) activated receptor P2X1. The P2X1 receptor is an
2+

ATP-gated calcium (Ca ) channel.

The platelet P2X1 receptor’s role in platelet

activation and aggregation is not well understood, but there is evidence in the literature
that it contributes to platelet activation.

1-3

The Spence group believes that RBC-derived

ATP release has an effect on platelet activation via this P2X1 receptor. Disease states
with increased (multiple sclerosis
9

4-5

, sickle cell anemia

10-11

, diabetes

, pulmonary hypertension

12-13

6-7

) or decreased (cystic fibrosis

8-

) RBC ATP release all exhibit platelet

hyperactivity.
NF449, commonly used as a P2X1 inhibitor, was found to be a platelet sensitizer
that can be used to study platelet Ca
presence of added apyrase.

14

2+

influx induced by ATP or ADP without the

This allows for a new way of studying ATP’s effect on the

platelet without the use of apyrase (which breaks down ATP) or a non-metabolizable
form of ATP. By using NF449 in this manner, the effect of native ATP on the P2X1
receptor can be investigated. Additionally, the effects of ATP can be distinguished from

98

2+

those of ADP, with regard to platelet Ca . Experiments on washed platelets are useful
for isolating the platelet’s response to agonists, while the platelet’s interaction with other
cell types in the vasculature (RBCs, endothelial cells) is more relevant to in vivo
conditions.
In Chapter 1, various microfluidic devices used to study platelet activation were
described. The microfluidic platelet adhesion device developed by the Spence group has
an advantage over many other devices as it incorporates 3 blood cell types (RBCs,
platelets, and endothelial cells) to study platelet activity, whereas other systems only
employ collagen and platelets. However, the adhesion device still has some drawbacks.
For example, in order to observe platelet adhesion, a low flow rate (0.2 µL/min) was used
during experiments and, after the samples were pumped, the device was further incubated
before rinsing and detecting platelets sticking to the endothelium. This does not represent
well in vivo conditions as blood is not continuously flowing throughout the experiment.
Furthermore, while this may indicate a baseline of platelet activation, there is no injury to
the endothelial cells so, it does not truly represent the formation of a thrombus. Lastly,
platelets adhering were counted individually, which is a time consuming process. I
propose an in vitro microfluidic platform for studying thrombus formation by controlled
3

cell lysis. Endothelial cells can be cultured in the device as described previously, but
with collagen as the adhesion protein to better represent the environment in vivo. An
15

16

endothelial cell injury can be induced via laser , chemical , or applied potential

17

thus,

exposing collagen. Next, either whole blood or platelets alone or some combination of
RBCs and platelets could be flowed over the injury and thrombus formation monitored.
99

Ideally, the platelets would be tagged with a fluorescent probe and visualized using
microscopy and the fluorescence intensity of the thrombus monitored rather than
individual platelets adhering to an undisturbed endothelium.
This device would have an advantage over current in vitro microfluidic models as
it would not only incorporate cells of the vasculature, but also a vascular injury. This
would be a useful tool in probing P2X1 function in thrombus formation as each
component of the blood could be manipulated, for example, RBCs separated from
platelets could be treated with an ATP release inhibitor, then recombined with platelets
and the effects of this inhibition observed. Additionally, whole blood samples from
healthy controls versus disease states with altered ATP release/platelet activity could also
be flowed and studied in this controlled environment.
5.2 Polystyrene Microfluidic Devices
The polystyrene (PS) device from chapters 3 and 4 was successfully integrated
with standard lab equipment (pipet and 96 well plate reader) and used to study
endothelial cells cultured in the serpentine channels.

18-19

PS is a desirable microfluidic

device material because it is commonly used for cell culture and the rigidity allows for
easier integration as compared to the flexible polydimethylsiloxane (PDMS) polymer.
Furthermore, one of the drawbacks of PDMS devices is absorption of compounds into the
PDMS. This was demonstrated in a static system where the drug, clopidogrel, absorbed
significantly into PDMS, but not PS, when compared to standards made up in
polyethyelene microcentrifuge tubes.

19

100

The study of drug transport across a membrane is useful in creating in vitro
pharmacokinetic profiles for new drug candidates. This is commonly performed using a
hollow fiber bioreactor cartridge employing cell culture. These cartridges contain fibers
that retain bacteria or cells, but allow drugs and nutrients to flow freely. Media propelled
through the cartridge provides oxygen and nutrients for the cells.

20-22

However, the

cartridges use large volumes of media ~ 20 mL, are low throughput, and do not allow for
any post analysis of cells. Our group has developed a PDMS based microfluidic device
with a polycarbonate membrane capable of performing these measurements (unpublished
data), however, as previously mentioned, PDMS may not be the best material as certain
drugs may absorb into the device material. Thus, using the PS technology developed
herein, a membrane was incorporated during fabrication in order to study drug transport
across the membrane.
A device made of PS and PDMS that incorporated a polyester membrane and
fittings to integrate pumps was developed and an image of the device is shown in figure
5.1. This device was made in a similar manner to the devices in chapter 3, but before the
top PDMS layer was plasma sealed to the PS, a polyester membrane dipped in 25%
toluene 75% methanol was laid across the channel and immediately dried. Though drug
transport was able to be measured using this device, no improvements in drug absorption
were observed when compared to a PDMS device (results not shown) even though 3
sides of the channel were made of PS and one side PDMS.
The ultimate goal would be to create a device that is made solely from
polystyrene, but still incorporate a membrane. A PDMS negative mold of channels was

101

Figure 5.1: Membrane based polystyrene device. The bottom layer of this device is PS
containing embedded channels, while the top layer is a PDMS slab containing wells for
sample introduction (adapters), sampling (above membrane) and waste. Between these
two layers is a polyester membrane that allows for diffusion from the channel to above
the well. The luer adapters are attached to the device with a quick dry epoxy and this
allows for syringes to be used for sample pumping.

102

fabricated, whereby pouring and curing epoxy over this PDMS mold, an epoxy mold with
positive features was made. In brief, a 10:1 ratio EC 415 epoxy to curing agent (Adtech
Plastic Systems) was poured over the PDMS mold with negative features and allowed to
spread over the features. Manual spreading should be avoided as it adds bubbles to the
epoxy. To reduce bubble formation the PDMS mold can be desiccated prior to pouring
the epoxy or the mold with the epoxy can be put under vacuum for a very short time, the
epoxy will expand rapidly under vacuum. The epoxy mold was then cured for 24 hrs on a
40 ˚C hot plate and any bubbles that surface are manually popped with a Pasteur pipet.
After 24 hrs, the epoxy was carefully demolded from the PDMS mold and the following
post cure was carried out on a hot plate: 93 ˚C for 2 hrs, 121 ˚C for 2 hrs, 149 ˚C for 2
hrs, 177 ˚C for 2 hrs. The result was a solid gray positive-relief epoxy mold that
replicates the original SU-8 master.
This epoxy can withstand higher temperatures and pressures therefore, it can be
used in a heated hydraulic press to transfer channels to sheets of polystyrene. Finally, the
polystyrene embedded channels can be sealed to another sheet of polystyrene with wells
to address sample and waste.

The initial prototypes contained channels that were

approximately 200 µm wide, although these channels would become deformed or
completely blocked when sealed to a PS sheet. Thus, the channels were scaled up to 1.5
mm wide in order to optimize the sealing procedure. The final hydraulic press schedule
used to embed channels in the PS is described below
The press was set to 115 ˚C (~ 120 ˚C) and allowed to equilibrate at that
temperature for 20 min. Then the following layers were inserted into the press in order
from bottom to top: mirror finished stainless steel plate, featureless PDMS, PS blank
103

sheet, epoxy mold, rubber sheet, Teflon sheet, mirror finished stainless steel plate. The
press was closed with 2000 lbs of pressure and after 7 min. the temperature was turned
down, the water turned on, and the press allowed to cool to 65 ˚C before removing the
sample. The PS was carefully demolded from the epoxy with the PS now containing
embedded channels.

Any thicker edges were removed before moving on to the next

step.
In order to seal the PS sheet with channels to a blank PS sheet with wells, the
hydraulic press was set to 75 ˚C (~ 80 ˚C) and allowed to equilibrate for 10 min. The
following layers were then inserted into the press in order from bottom to top: mirror
finished stainless steel plate, featureless PDMS, PS blank with wells for inlets and
outlets, PS with channels, featureless PDMS, rubber sheet, Teflon sheet, mirror finished
stainless steel plate. The press was closed with 3000 lbs of pressure and after 10 min. the
temperature was turned down, the water turned on, and the press was allowed to cool to
65 ˚C before removing the sample. The final product is a microfluidic device made
completely out of PS with channels and inlets and outlets.

These channels were

successfully sealed with no blockages as shown in figure 5.2 where fluorescein was
flowed and visualized using fluorescence microscopy and no leaking occurred. Further
development is still needed on sealing a membrane between the two PS slabs. One of the
downsides to these lithographically based fabrication strategies is the time it takes to go
from device idea to functional prototype. There are multiple intermediate steps before
the device is finished and this is one reason why 3 dimensional (3D) printing was
investigated

as

a

new

method

104

for

creating

these

devices.

Figure 5.2: Polystyrene sealed to polystyrene device. The image on the left shows a two
layer device where the bottom layer has channels embedded in PS and the top blank layer
containing wells, also PS. These to pieces of PS was thermally sealed together using a
heated hydraulic press. To ensure that no channels were obstructed, fluorescein was
flowed though the channels, and the representative fluorescent image on the right shoes
that the channel is sealed, not leaking.

105

5.3 Introduction to 3 Dimensional Printing
In this dissertation, PDMS and polystyrene have been discussed as materials for
microfluidic devices and the processes by which those devices are photolithography and
soft lithography.

One of the downsides of conventional photolithography, soft

lithography, and hot embossing is the time it takes to go from device design to a final
usable device. Additionally, any modifications to the device have to be made at the
master level, thereby starting the whole process anew. This, in particular, is a drawback
when developing novel devices, where multiple prototypes may be necessary before a
final design is decided upon. 3D printing technologies allow for fast device production,
and also for designs to be modified and optimized quickly based on the ongoing data
collection in the lab.
3D printing has commonly been used in manufacturing industries to produce
design prototypes, but recently it has been utilized in the areas of scaffolds for tissue
23-26

growth,

prototyping,

27-28

29

electronics , microfluidics

30-31

, and pneumatics.

32

3D

printers build a device layer by layer based on “sliced” 3D models that are drawn using a
computer program, such as Computer Aided design (CAD).

33-34

In work here, the Objet

Connex 350 is used to create all 3D printed devices. The Objet 3D printer constructs the
device layer by layer by inkjet deposition of photopolymerizable materials that are
hardened by subsequent UV irradiation. The Objet Connex 350 is able to print polymer
droplets that are 42 µm in diameter, which allows for features ranging in size from 100 –
1000 µm. Though fluidic devices that allow for mixing
reactions

31

30

and chambers for organic
25

have been fabricated as well as scaffolds allowing for cell growth , to date,
106

no device has been directly printed that incorporates flow, drug transport and cell based
techniques together on one 3D printed device. Here we fabricate a 3D printed millifluidic device that directly integrates with syringe pumps, allows for drug transport across
a membrane, and can be used to study cell viability using commercially available cell
inserts. These studies were executed with the help of my fellow graduate student, Sarah
Lockwood, who performed all the mass spectrometry studies.
5.3.1 3D Printed Device Fabrication
All 3D printed devices were printed on an Objet Connex 350 printer housed in
Department of Electrical and Computer Engineering at Michigan State University. All
devices were printed using Objet Vero White Plus material whose composition is
proprietary, but contains: isobornyl acrylate (15-30%), acrylic monomer (15-30%),
urethane acrylate (15-50%), epoxy acrylate (5-10; 10-15%), acrylic monomer (5-10; 1015%), acrylic oligomer (5-10; 10-15%) and photoinitiator (0.1-1; 1-2%). Once printed,
the devices are opaque, white and rigid. Initial prototypes contained only one channel,
but as many as 8 channels on one device have been printed.

Using commercially

available cell culture inserts (0.4 um pore diameter, Corning Incorporated, Corning, NY,
USA), polycarbonate membranes were clicked into place above the channel.

The

channels are 3.0 mm wide and 1.5 mm deep under the membrane. To integrate the
devices with syringes for pumping, the inlet was printed so that a female luer adapter (P629, IDEX Health and Science, Oak Harbor, WA, USA) could be clicked into place and
sealed with quick drying epoxy. In this manner, syringes were hooked up to the device
via this adapter and a dual syringe pump was used to flow sample through the channels.

107

5.3.2 Sample Preparation for Drug Transport
To study drug transport in the 3D printed device, three established drugs,
clopidogrel, levofloxacin (Sigma Aldrich, St. Louis, MO, USA)and linezolid (TOCRIS
Bioscience, Bristol, United Kingdom), were prepared in standard solutions ranging from
100 to 2000 nM.

Clopidogrel was solubilized in 0.1 M HCl, and linezolid and

levofloxacin were dissolved in water to prepare 1 mg/mL stock solutions. Using a
physiological salt solution (PSS) comprised of 2.0 mM CaCl2, 4.7 mM KCl, 11.1 mM
dextrose, 12 mM MgSO4, 21.0 mM tris(hydroxymethyl)aminomethane, and 140.5 mM
NaCl (final pH 7.4)(Sigma Aldrich) as a buffer, the stock solutions were diluted to the
desired standard concentrations. Standards were then pumped at 1 µL/min through the
channels of the 3D device for one hour.

In this case, the cell culture inserts that were

clicked in above the channel contained 150 µL of PSS and from that aliquots of sample
(25 µL) were taken after an hour of pumping. The 25 µL samples were added to 96 µL
of cold acetonitrile (Mallinckrodt Chemicals, Phillipsburg, NJ, USA) containing the
internal standard, 750 nM ciprofloxacin (Sigma Aldrich). To remove any precipitate,
samples were centrifuged (500 g for 10 min) and the supernatant removed.

After

centrifugation, the samples were added to a polycarbonate PCR plate (Denville,
Metuchen, NJ, USA) and sealed using a RapidEPS seal (Bio Chromato, Fujisawa, Japan)
5.3.3 Mass Spectrometric Analysis of Drug Transport
Detection of the drugs was achieved by using a Shimadzu high performance
liquid chromatograph

(HPLC) coupled with multiple reaction monitoring mass

spectrometry (MRM-MS/MS) with electrospray ionization (ESI) using a Waters Quattro
108

Micro triple quadrupole mass spectrometer. The reverse phase HPLC separation was
performed through the use of a Supelco Ascentis precolumn, followed by a packed
Supelco Ascentis Express C18 column (length: 3 cm, ID: 2.1 mm, particles: 2.7
µm)(Supelco, Bellefonte, PA, USA). A 10 µL sample from the PCR plate was injected
onto the column where it underwent separation at 0.45 mL/min with the corresponding
gradient, 95% A, 0% B for 0.25 min; 2% A, 0% B at 1.00 min; 2% A, 0% B at 1.25 min;
30% A, 15% B at 1.75 min; 30% A, 15% B at 1.90 min; 95% A, 0% B at 2.00 min.
Solvent A is methanol, solvent B is acetonitrile and solvent C is 1% formic acid in water.
Retention times of ciprofloxacin, clopidogrel, levofloxacin and linezolid are 1.51, 2.08,
1.48, and 1.71 minutes, respectively. Upon introduction into the Quattro Micro mass
spectrometer, analytes were ionized by ESI and further analyzed using the MRMMS/MS.

The method for detection of ciprofloxacin, clopidogrel, levofloxacin and

linezolid involved a cone voltage of 25 V for the internal standard and 30 V for the
remaining drugs. Collisional energies for ciprofloxacin, clopidogrel, levofloxacin, and
linezolid were 40, 35, 30, and 35 eV, respectively. The MRM transitions monitored were
m/z 332.11>231.06 (ciprofloxacin), 322.08>155.03 (clopidogrel), 362.11>234.05
(levofloxacin), and 338.13>296.1 (linezolid). The collision gas, argon, was 1.96 x 10

-3

Torr. Quantitative data processing was obtained through Waters MassLynx software
(Milford, MA, USA). Calibration curves were generated by taking the ratio of sample
peak area to the internal standard peak area.
5.3.4 Endothelial Cell Culture onto Inserts

109

Bovine pulmonary artery endothelial cells (ECs) were used in all cell studies. The
vials were thawed to room temperature and added to a T-75 tissue culture flask
containing 9 mL of endothelial growth media (EGM) that had been equilibrated to 37 C.
The EGM consists of a low glucose (5.5 mM) Dulbecco’s Modified Eagles Medium
(DMEM, MIDSCI, St. Louis, MO, USA) supplemented with 2.5% v/v adult bovine
serum (Sigma-Aldrich, St. Louis, MO, USA), 7.5% fetal bovine serum (Lonza,
Walkersville, MD, USA), penicillin, streptomycin, and amphotericin B (MIDSCI, St.
Louis, MO USA). bPAECs were allowed to grow in a humidified incubator at 37 C and
5% CO2 until they were determined as confluent by optical microscopy. Media was
changed the day after plating and then every 2 days thereafter. The ECs were subcultured
when the cells reached >80% confluence as visualized by optical microscopy. ECs were
then seeded onto 6.5 mm cell inserts that contain a permeable polycarbonate membrane
(0.4 um pore diameter, Corning Incorporated, Corning, NY, USA). In short, T-75 flasks
of ECs were washed with 10 mL of HEPES and then treated with 5 mL of 0.25%
trypsin/EDTA, which was then removed and the cells suspended in 10 mL of media. The
cell suspension was removed from the flask and centrifuged at 1500 g for 5 min. The
supernatant was removed and the pellet was resuspended in approximately 2 mL of
equilibrated media. Then 100 µL of concentrated cell solution was pipette onto the
inserts with 600 µL of media in the lower chamber. The cells were allowed to grow to
confluence for 24 hours as visualized by optical microscopy. Cell inserts were then
clicked into place above the channels on the 3D printed device and 100 µL of HBSS was
pipetted onto the insert.

110

5.3.5 Sample Preparation and Experimental Procedure for Cell Viability Study
Saponin (Sigma Aldrich) was used to permeabilize EC membranes in order to
induce death. A 0.25% solution of saponin was made by dissolving 13.2 mg into 5.28
mL of water. A portion of the 0.25% solution was then diluted to 0.025% in Hank’s
balanced salt solution (HBSS, Sigma Aldrich). This solution, or HBSS, was then loaded
into syringes and flowed through the 3D printed device underneath the cell insert
containing cultured ECs, for 30 min. at 1 µl/min using a dual syringe pump. The inserts
were then removed from the device and placed into a 24 well plate containing 600 µL
HBSS in the lower chamber.
The nucleic acid stain Sytox Green (Invitrogen, Grand Island, NY, USA), which
only labels cell with compromised membranes, was used to determine whether or not
saponin had killed ECs. Its excitation excitation and emission maxima are 504 and 523
nm, respectively when bound to DNA. After saponin or HBSS was flowed under the
cells, 4 µL of 1 µM Sytox Green was added to the cell insert, resulting in a concentration
of 40 nM above the cells. The probe was allowed to incubate at 37 ˚C for 30 min. and
then imaged using fluorescence microscopy.

An Olympus MVX10 macroscope

(Olympus America, Melville, NY, USA) was used to collect fluorescence images for
each insert and using the MicroSuite Biological Suite software (Olympus America)
average pixel intensities of the fluorescence image were obtained.

The data from

corresponding to the saponin treated cell insert was normalized to the data from the
control cell insert with HBSS flowing through the channel.
5.4 Preliminary Results with 3D Printed Device

111

The device design used in these studies consisted of 8 channels with inlets
allowing for sample introduction/syringe pump integration, analytical wells for
membrane containing cell culture inserts and outlets for waste as shown in figure 5.3.
The side profile of one channel (figure 5.4) shows how each channel will address the
membrane allowing for diffusion of analytes from the channel, through the membrane, to
the area above the insert. Once printed, the device was used to study drug transport
across the membrane.
5.4.1 Drug Transport across a Membrane
Standard samples containing linezolid and levofloxacin were prepared in PSS no
BSA buffer. Concentrations of 1.1 and 2 µM standard solutions were introduced to the
3D device through syringes and soft Teflon tubing. The 3D device included a 3 mm
channel which lay directly under an open well that was fitted to the cell insert. Upon
insertion of the cell insert, buffer was added to the insert and the standard solutions were
flowed through the channels for an hour at 1 µL/min. After an hour, liquid was sampled
from the insert and added to a vial containing a solution of acetonitrile and the internal
standard, ciprofloxacin. The samples were then analyzed using LC/MS/MS to monitor
the diffusion of linezolid and levofloxacin from the channel, across the porous membrane
into the cell insert. After an hour of flow, the 1.1 µM and 2 µM samples had between

112

Figure 5.3: 3D printed device. The final 3D printed device contains adapters for syringe
based pumps, channels, membrane insertion port and outlets. The top image shows a side
view of the device where sample is pumped though the channel via tubing where it flows
under the membrane where molecules can diffuse. The bottom image shows the device
from the angle of sample entry where all 8 channels of the device can be used
simultaneously to address 8 membranes.

113

Figure 5.4: 3D printed device side view. The side view of the device shows how the inlet addresses the channel and allows fluid to
flow under the membrane. The membrane is part of a commercially available cell insert that is manually inserted into the port on top
of the device. Finally, there is an outlet to allow fluid to leave the device.

114

18.4 % and 20.5% drug transport across the polycarbonate membrane (figure 5.5).
Moreover, results yielded reproducible drug transport concentrations between runs
furthering the reusability of the device.
In other polymer based devices reusability is not an option due to issues with
maintaining seals or the need for resealing devices thought the experiment.
Contamination is also a concern due to challenges cleaning devices or absorption of
materials into the polymer base. The use of a new device for each experiment can lead to
high variability between runs. Also the incorporation of a membrane into the polymer
based device, typically reversibly sealed between two pieces, can be easily compromised
due to the flexibility of the support material. Cleaning and extended use can also weaken
the fragile membrane.

The aforementioned strains on conventional fluidic membranes

lend virtue to the cell culture insert, which can easily be replaced for additional runs on
the same device. The cell insert has a rugged base which supports the membrane and can
be easily discarded after use. The rigidity of the 3D device when used simultaneously
with the disposable cell insert offers a supportive platform for a reusable fluidic device.
5.4.2 Cell Viability Assessment
Commercially available cell culture inserts were used to integrate the 3D printed
device with cultured cells. In this design, the insert clicks into place above the channel
and the membrane would contain a layer of cultured cells as visualized in figure 5.6.
Furthermore, the ECs were stained with Hoechst dye (ex. 350 nm, em. 460 nm), a simple
nucleic acid stain to confirm the presence of a confluent layer of cells on top of the

115

Figure 5.5: Drug transport across a membrane. Standards of the drugs linezolid (N = 4)
and levofloxacin (N = 5) were flowed through the channels of the device; samples were
collected above the membrane and analyzed via MS. As concentration of the drug
increased so, did transport across the membrane with each concentration being
statistically different from the previous. (p value < 0.001).

116

membrane. The image in figure 5.6 was taken from a fluorescence microscope with a
DAPI filter and the stained cells are visualized on top of the membrane.
In order to determine whether or not this 3D printed device could be used to study
cell activity, a simple viability study was performed. Either HBSS or saponin (kills cells
by forming a complex with cholesterol and forming pores in the membrane) was
delivered through the channel under the membrane that contained cultured ECs. It was
expected that saponin would diffuse through the polycarbonate membrane and come into
contact with the ECs. As shown in figure 5.7 A, the ECs that were treated with saponin
had almost 5 times more fluorescence than the cells treated with HBSS. The Sytox Green
does not stain live cells so; the increased fluorescence indicates a higher population of
dead cells. The images from the fluorescence microscope (figure 5.7 B) confirm this
fact, as the image of the saponin treated ECs show more fluorescence than the image of
ECs treated with HBSS alone.
5.5 Conclusions on 3D Printing
The 3D printed device was capable of studying drug transport and cell viability in
a high throughput manner. Furthermore, the device was rugged and reusable, which
gives it an advantage over devices made out of soft polymers. Additionally, the ability to
rapidly modify and print prototypes is advantageous when optimizing a device design.
Experiments on these 3D printed devices are ongoing in the Spence lab with the latest
prototype being in 24-well format with threads directly printed on the device in order to
integrate pumping mechanisms as shown in figure 5.8.

117

Figure 5.6: Incorporation of endothelial cells on 3D printed device. Endothelial cells were
incorporated onto the 3D printed device by culturing them onto the membrane insert. These are
commercially available cell inserts well suited for cell culture. Cells were cultured on the inserts
and stored in an incubator (37 ˚C, 5% CO2) until experiments were performed. Cells were
labeled with Hoechst dye, a nucleic acid stain, to confirm their presence on the membrane as
evidenced in the fluorescent image. Cell inserts were then clicked into place on the device and
samples flowed through the channel under the membrane containing cells.

118

Figure 5.7: Cell viability experiment. Saponin, a known cell detergent, was flowed under
endothelial cells to induce cell death. Sytox Green, which does not label live cells, was used in
order to determine the extent of cell death. The bar graph in part A shows an increase in Sytox
Green fluorescence when cells were treated with saponin compared to buffer alone. The
fluorescence images in part B represent cells treated with HBSS (left) and cells treated with
saponin (right) with the Sytox Green label.

119

Figure 5.8: Single channel 3D printed device prototype. The prototype shown in the top image
has the port situated in “24-well plate” format to enable future handling by automated systems.
Additionally, the bottom image shows the ability to print threads directly into the device so
adapters can be screwed in without the need of an additional adapter or to seal with epoxy.

120

REFERENCES

121

REFERENCES

1.
C. Y. Fung, C. A. Brearley, R. W. Farndale, M. P. Mahaut-Smith, A major role
for P2X1 receptors in the early collagen-evoked intracellular Ca2+ responses of human
platelets. Thromb Haemost 2005, 94. 37-40.
2.
C. Y. E. Fung, A. J. Marcus, M. J. Broekman, M. P. Mahaut-Smith, P2X1
Receptor Inhibition and Soluble CD39 Administration as Novel Approaches to Widen the
Cardiovascular Therapeutic Window. Trends Cardiovasc. Med. 2009, 19. 1-5.
3.
W. Karunarathne, C. J. Ku, D. M. Spence, The dual nature of extracellular ATP as
a concentration-dependent platelet P2X1 agonist and antagonist. Integr Biol 2009, 1. 655663, DOI: Doi 10.1039/B909873a.
4.
S. Letourneau, L. Hernandez, A. N. Faris, D. M. Spence, Evaluating the effects of
estradiol on endothelial nitric oxide stimulated by erythrocyte-derived ATP using a
microfluidic approach. Anal. Bioanal. Chem. 2010, 397. 3369-75, DOI: 10.1007/s00216010-3687-7.
5.
H. P. Wright, R. H. Thompson, K. J. Zilkha, Platelet adhesiveness in multiple
sclerosis. Lancet 1965, 2. 1109-10.
6.
K. I. Ataga, E. P. Orringer, Hypercoagulability in sickle cell disease: a curious
paradox. Am. J. Med. 2003, 115. 721-728.
7.
R. B. Francis, Jr., Platelets, coagulation, and fibrinolysis in sickle cell disease:
their possible role in vascular occlusion. Blood Coagul Fibrinolysis 1991, 2. 341-53.
8.
B. P. O'Sullivan, M. D. Linden, A. L. Frelinger, III, M. R. Barnard, M. SpencerManzon, J. E. Morris, R. O. Salem, M. Laposata, A. D. Michelson, Platelet activation in
cystic fibrosis. Blood 2005, 105. 4635-4641.
9.
R. S. Sprague, M. L. Ellsworth, A. H. Stephenson, M. E. Kleinhenz, A. J.
Lonigro, Deformation-induced ATP release from red blood cells requires CFTR activity.
American Journal of Physiology 1998, 275. H1726-H1732.
10.
A. I. Vinik, T. Erbas, T. S. Park, R. Nolan, G. L. Pittenger, Platelet dysfunction in
type 2 diabetes. Diabetes Care 2001, 24. 1476-1485.
11.
J. Carroll, M. Raththagala, W. Subasinghe, S. Baguzis, T. D. a. Oblak, P. Root, D.
Spence, An altered oxidant defense system in red blood cells affects their ability to
release nitric oxide-stimulating ATP. Mol. BioSyst. 2006, 2. 305-311.

122

12.
L. Somova, J. Mufunda, Platelet activity and salt sensitivity in the pathogenesis of
systemic (essential) hypertension in black Africans. Clin Exp Hypertens 1993, 15. 78196.
13.
R. S. Sprague, A. H. Stephenson, M. L. Ellsworth, C. Keller, A. J. Lonigro,
Impaired release of ATP from red blood cells of humans with primary pulmonary
hypertension. Exp. Biol. Med. (Maywood, NJ, U. S.) 2001, 226. 434-439.
14.
K. B. Anderson, W. Karunarathne, D. M. Spence, Measuring P2X1 receptor
activity in washed platelets in the absence of exogenous apyrase. Anal Methods-Uk 2012,
4. 101-105, DOI: Doi 10.1039/C1ay05530e.
15.
E. D. Rosen, S. Raymond, A. Zollman, F. Noria, M. Sandoval-Cooper, A.
Shulman, J. L. Merz, F. J. Castellino, Laser-induced noninvasive vascular injury models
in mice generate platelet- and coagulation-dependent thrombi. Am. J. Pathol. 2001, 158.
1613-1622.
16.
H. Matsuno, T. Uematsu, S. Nagashima, M. Nakashima, Photochemically
Induced Thrombosis Model in Rat Femoral-Artery and Evaluation of Effects of Heparin
and Tissue-Type Plasminogen-Activator with Use of This Model. J Pharmacol Methods
1991, 25. 303-317, DOI: Doi 10.1016/0160-5402(91)90030-9.
17.
P. J. Marc, C. E. Sims, M. Bachman, G. P. Li, N. L. Allbritton, Fast-lysis cell
traps for chemical cytometry. Lab on a Chip 2008, 8. 710-716, DOI: Doi
10.1039/B719301g.
18.
K. B. Anderson, S. T. Halpin, A. S. Johnson, R. S. Martin, D. M. Spence,
Integration of multiple components in polystyrene-based microfluidic devices part II:
cellular analysis. Analyst (Cambridge, United Kingdom) 138. 137-143.
19.
A. S. Johnson, K. B. Anderson, S. T. Halpin, D. C. Kirkpatrick, D. M. Spence, R.
S. Martin, Integration of multiple components in polystyrene-based microfluidic devices
part I: fabrication and characterization. Analyst (Cambridge, United Kingdom) 138. 129136.
20.
J. A. Bilello, G. Bauer, M. N. Dudley, G. A. Cole, G. L. Drusano, Effect of 2',3'Didehydro-3'-Deoxythymidine in an in-Vitro Hollow-Fiber Pharmacodynamic Model
System Correlates with Results of Dose-Ranging Clinical-Studies. Antimicrob Agents Ch
1994, 38. 1386-1391.
21.
S. L. Preston, P. J. Piliero, J. A. Bilello, D. S. Stein, W. T. Symonds, G. L.
Drusano, In vitro-in vivo model for evaluating the antiviral activity of amprenavir in
combination with ritonavir administered at 600 and 100 milligrams, respectively, every
12 hours. Antimicrob Agents Ch 2003, 47. 3393-3399, DOI: Doi
10.1128/Aac.47.11.3393-3399.2003.

123

22.
G. L. Drusano, P. A. Bilello, W. T. Symonds, D. S. Stein, J. McDowell, A. Bye, J.
A. Bilello, Pharmacodynamics of abacavir in an in vitro hollow-fiber model system.
Antimicrob Agents Ch 2002, 46. 464-470, DOI: Doi 10.1128/Aac.46.2.464-470.2002.
23.
J. N. H. Shepherd, S. T. Parker, R. F. Shepherd, M. U. Gillette, J. A. Lewis, R. G.
Nuzzo, 3D Microperiodic Hydrogel Scaffolds for Robust Neuronal Cultures. Adv. Funct.
Mater. 2011, 21. 47-54, DOI: DOI 10.1002/adfm.201001746.
24.
Y. B. Lee, S. Polio, W. Lee, G. H. Dai, L. Menon, R. S. Carroll, S. S. Yoo, Bioprinting of collagen and VEGF-releasing fibrin gel scaffolds for neural stem cell culture.
Exp Neurol 2010, 223. 645-652, DOI: DOI 10.1016/j.expneurol.2010.02.014.
25.
W. Lee, V. Lee, S. Polio, P. Keegan, J. H. Lee, K. Fischer, J. K. Park, S. S. Yoo,
On-Demand Three-Dimensional Freeform Fabrication of Multi-Layered Hydrogel
Scaffold With Fluidic Channels. Biotechnol. Bioeng. 2010, 105. 1178-1186, DOI: Doi
10.1002/Bit.22613.
26.
A. Butscher, M. Bohner, S. Hofmann, L. Gauckler, R. Muller, Structural and
material approaches to bone tissue engineering in powder-based three-dimensional
printing. Acta Biomater 2011, 7. 907-920, DOI: DOI 10.1016/j.actbio.2010.09.039.
27.
J. Stampfl, R. Liska, New materials for rapid prototyping applications. Macromol
Chem Physic 2005, 206. 1253-1256, DOI: DOI 10.1002/macp.200500199.
28.
D. S. Rangel, Claire; Bielschowsky, Mayra; Farris, Katie; Falconer, Ruth;
Baveye, Philippe Rapid Prototyping and 3-D Printing of Experimental Equipment in Soil
Science Research Soil Sci Soc Am J 2012, 77. 54-59.
29.
B. Y. Ahn, E. B. Duoss, M. J. Motala, X. Y. Guo, S. I. Park, Y. J. Xiong, J. Yoon,
R. G. Nuzzo, J. A. Rogers, J. A. Lewis, Omnidirectional Printing of Flexible, Stretchable,
and Spanning Silver Microelectrodes. Science 2009, 323. 1590-1593, DOI: DOI
10.1126/science.1168375.
30.
D. Therriault, S. R. White, J. A. Lewis, Chaotic mixing in three-dimensional
microvascular networks fabricated by direct-write assembly. Nat. Mater. 2003, 2. 265271, DOI: Doi 10.1038/Nmat863.
31.
P. J. Kitson, M. H. Rosnes, V. Sans, V. Dragone, L. Cronin, Configurable 3DPrinted millifluidic and microfluidic 'lab on a chip' reactionware devices. Lab on a Chip
2012, 12. 3267-3271, DOI: Doi 10.1039/C2lc40761b.
32.
F. Ilievski, A. D. Mazzeo, R. E. Shepherd, X. Chen, G. M. Whitesides, Soft
Robotics for Chemists. Angew Chem Int Edit 2011, 50. 1890-1895, DOI: DOI
10.1002/anie.201006464.

124

33.
E. Sachs, M. Cima, P. Williams, D. Brancazio, J. Cornie, 3-Dimensional Printing
- Rapid Tooling and Prototypes Directly from a Cad Model. J Eng Ind-T Asme 1992, 114.
481-488.
34.
W. Sun, B. Starly, J. Nam, A. Darling, Bio-CAD modeling and its applications in
computer-aided tissue engineering. Comput Aided Design 2005, 37. 1097-1114, DOI:
DOI 10.1016/j.cad.2005.02.002.

125