Download Red Blood Cell Responses during a Long

micromachines
Article
Red Blood Cell Responses during a Long-Standing
Load in a Microfluidic Constriction
Mitsuhiro Horade, Chia-Hung Dylan Tsai *, Hiroaki Ito and Makoto Kaneko
Department of Mechanical Engineering, Osaka University, Suita 565-0871, Japan;
[email protected] (M.H.); [email protected] (H.I.);
[email protected] (M.K.)
* Correspondence: [email protected]; Tel.: +81-6-6879-7333
Academic Editors: Kwang W. Oh and Nam-Trung Nguyen
Received: 25 January 2017; Accepted: 22 March 2017; Published: 26 March 2017
Abstract: Red blood cell responses during a long-standing load were experimentally investigated.
With a high-speed camera and a high-speed actuator, we were able to manipulate cells staying inside
a microfluidic constriction, and each cell was compressed due to the geometric constraints. During
the load inside the constriction, the color of the cells was found to gradually darken, while the cell
lengths became shorter and shorter. According to the analysis results of a 5 min load, the average
increase of the cell darkness was 60.9 in 8-bit color resolution, and the average shrinkage of the
cell length was 15% of the initial length. The same tendency was consistently observed from cell
to cell. A correlation between the changes of the color and the length were established based on
the experimental results. The changes are believed partially due to the viscoelastic properties of
the cells that the cells’ configurations change with time for adapting to the confined space inside
the constriction.
Keywords: microfluidics; long-standing load; cell manipulation; red blood cell
1. Introduction
There are established relationships between the deformability of red blood cells (RBCs) and
human diseases, such as Malaria, Spherocytosis and Cryohydrocytosis [1–5]. While many works
have focused on RBC deformability through short-duration compressions, such as transit time
through a constriction or micro-slit [6–11], few works have focused on RBC characteristics after
long-standing loads [12,13]. RBC responses under a long-standing load reflect how a RBC changes
while being plugged in microcirculation. For example, Baez et al. found different flow characteristics
of oxygenated and partially deoxygenated red cells plugging in microcirculation [14]. Therefore,
this work is motivated by the idea of investigating cell responses during a long-standing load using
microfluidic constriction.
Figure 1 shows an overview of the proposed idea and sample RBC photos at the beginning and
the end of a long-standing load. A constriction channel is placed in a microfluidic device as illustrated
in Figure 1a. There are two phases of the experiment. First, a target RBC is positioned in front of the
narrow channel for recovering from prior deformation as the “Catching” phase. Afterward, the RBC is
moved into the constriction channel for compression as the “Loading” phase. The cell manipulations
are performed by a vision-feedback system where the real-time cell position is computed for proper
actuator output for moving the cell to a target position. The RBC is maintained inside the constriction
for a specified duration as the long-standing load, and a camera is set for observing the responses of
the cell during the load.
Micromachines 2017, 8, 100; doi:10.3390/mi8040100
www.mdpi.com/journal/micromachines
Micromachines 2017, 8, 100
2 of 10
Figure 1b illustrates a RBC before and after the 5 min load with two sample images from the
experiments. At the beginning of the load, the RBC is slender and almost transparent. After the 5 min
Micromachines 2017, 8, 100
2 of 10
load, the color of the RBC is darkened, and the horizontal length is significantly shortened.
A time-lapse sample video can be found in the Supplementary Materials Video S1. Quantitative
evaluations
of illustrates
the amounta of
change
in and
cell color
length
measured
and images
presented
in the
this
Figure 1b
RBC
before
after and
the 5cell
min
load are
with
two sample
from
paper. The correlation
and interpretation
from
mechanical
aspect transparent.
are discussed
based
the
experiments.
At the beginning
of the load, the
RBCthe
is slender
and almost
After
the on
5 min
experimental
results.
load, the color of the RBC is darkened, and the horizontal length is significantly shortened.
Figure 1.1. An
An overview
overview of
of the
the proposed
proposed method
method and
and example
example results.
results. (a)
(a) An
An illustration
illustration of
of cell
cell
Figure
manipulation
for
the
long-standing
load;
(b)
illustrations
and
experimental
images
of
a
red
blood
cell
manipulation for the long-standing load; (b) illustrations and experimental images of a red blood cell
(RBC)
before
and
after
a
5
min
load.
(RBC) before and after a 5 min load.
In this paper, we particularly focus on the evolution of the RBC color and length during a longA time-lapse sample video can be found in the Supplementary Materials Video S1. Quantitative
standing load. These newly found cell properties could provide new insights or a different evaluation
evaluations of the amount of change in cell color and cell length are measured and presented in this
index for cell evaluation.
paper. The correlation and interpretation from the mechanical aspect are discussed based on the
experimental
results.
2. Related Works
In this paper, we particularly focus on the evolution of the RBC color and length during a
RBC deformability
hasnewly
been found
be correlated
certain
diseases
example,
long-standing
load. These
foundtocell
propertieswith
could
provide
new [5,15–18].
insights orFor
a different
RBCs withindex
reduced
are found in obese patients and in those with diabetic foot diseases
evaluation
for deformability
cell evaluation.
[15,16]. Malaria is a well-known blood-related disease that reduces RBC deformability [17,18].
2.
Relatedmorphology
Works
Unusual
of RBCs has been found from those of the compound heterozygous for
hemoglobin S and hemoglobin C [19]. There are different approaches for investigating RBC
RBC deformability has been found to be correlated with certain diseases [5,15–18]. For example,
deformability [20,21]. For example, Tözeren et al. used micropipette for developing a constitutive
RBCs with reduced deformability are found in obese patients and in those with diabetic foot
equation for the RBC membrane [22]. Brandao et al. employed optical tweezers for measuring RBC
diseases [15,16]. Malaria is a well-known blood-related disease that reduces RBC deformability [17,18].
elasticity [23]. Dulinska et al. applied an atomic force microscope (AFM) for measuring the stiffness
Unusual morphology of RBCs has been found from those of the compound heterozygous for
of erythrocytes [24]. Among different evaluation methods, microfluidic constrictions have become
hemoglobin S and hemoglobin C [19]. There are different approaches for investigating RBC
popular in the last decade because of their precise fabrication and accurate control [25–27]. For
deformability [20,21]. For example, Tözeren et al. used micropipette for developing a constitutive
example, Zheng et al. put RBCs through a constriction for high-throughput biophysical
equation for the RBC membrane [22]. Brandao et al. employed optical tweezers for measuring RBC
measurements [28]. Sakuma et al. evaluated the cell fatigue state by inducing continuous and
elasticity [23]. Dulinska et al. applied an atomic force microscope (AFM) for measuring the stiffness
repetitive deformation [29]. There are also works investigating RBC responses after long-standing
of erythrocytes [24]. Among different evaluation methods, microfluidic constrictions have become
loads. For example, Fischer discovered the shape memory of RBCs based on their recovery from the
popular in the last decade because of their precise fabrication and accurate control [25–27]. For example,
deformation by shear stress up to 4 h [12]. Markle et al. investigated the viscoelastic properties of the
Zheng et al. put RBCs through a constriction for high-throughput biophysical measurements [28].
RBC membrane by characterizing their shape after constant-pressure aspiration up to 90 min [13].
Sakuma et al. evaluated the cell fatigue state by inducing continuous and repetitive deformation [29].
Through the microfluidic platform, it becomes possible to stably monitor cells’ behaviors and
There are also works investigating RBC responses after long-standing loads. For example, Fischer
responses during a fixed-displacement load.
discovered the shape memory of RBCs based on their recovery from the deformation by shear stress up
In our previous works, we succeeded in applying long-standing loads to RBCs by high-speed
to 4 h [12]. Markle et al. investigated the viscoelastic properties of the RBC membrane by characterizing
cell manipulation and discussed how the RBCs recover after the load [30,31]. Preliminary results of
their shape after constant-pressure aspiration up to 90 min [13]. Through the microfluidic platform,
it becomes possible to stably monitor cells’ behaviors and responses during a fixed-displacement load.
Micromachines 2017, 8, 100
3 of 10
Micromachines 2017, 8, 100
3 of 10
In our
previous
works,changes
we succeeded
applying
long-standing
loads observed
to RBCs by
high-speed
cell
the RBC
color
and length
duringinthe
load were
independently
and
presented
at
manipulation
and
discussed
how
the
RBCs
recover
after
the
load
[30,31].
Preliminary
results
of
the
recent conferences [32,33]. In this work, we firstly present a comprehensive analysis of cell changes
RBC
color
length changes
the loadawere
independently
andofpresented
at recent
during
theand
long-standing
load.during
Furthermore,
correlation
between observed
the changes
the RBC color
and
conferences
[32,33].
In
this
work,
we
firstly
present
a
comprehensive
analysis
of
cell
changes
during
length is found according to the experimental results.
the long-standing load. Furthermore, a correlation between the changes of the RBC color and length is
found
according to
the experimental
results.
3. Experimental
System
and Procedure
3.
Experimental
System
and Procedure
3.1.
Cell Manipulation
System
3.1. Cell
System
CellManipulation
manipulation
is the key technique for this work. Without an accurate cell manipulation for
keeping
RBCs
inside
the
constriction,
RBC for
response
under
long-standing
load
cannot
be properly
Cell manipulation is the
key technique
this work.
Without
an accurate
cell
manipulation
for
observed.
Micropipettes
or
robotic
microgrippers
may
also
apply
a
constant
load
to
RBCs,
but
they
keeping RBCs inside the constriction, RBC response under long-standing load cannot be properly
usually take
much longer
to locate
a cell [12,13,21,34].
Thus,a the
microfluidic
systembut
offers
observed.
Micropipettes
ortime
robotic
microgrippers
may also apply
constant
load to RBCs,
theya
more
convenient
platform
for
the
test.
usually take much longer time to locate a cell [12,13,21,34]. Thus, the microfluidic system offers a more
Figure 2 shows the experimental setup of cell manipulation system, which is composed of a polyconvenient platform for the test.
dimethylsiloxane (PDMS) chip, a piezoelectric actuator (MESS-TEK: M-2655s(c), MESS-TEK Ltd.,
Figure 2 shows the experimental setup of cell manipulation system, which is composed of a
Saitama, Japan), a syringe (SGE Analytical Science: 2.5MDF-LL-GT, LGE Japan Ltd., Osaka, Japan), a
poly-dimethylsiloxane (PDMS) chip, a piezoelectric actuator (MESS-TEK: M-2655s(c), MESS-TEK
high-speed camera (Photoron: IDP-Express R2000, Photron Ltd., Tokyo, Japan), a microscope
Ltd., Saitama, Japan), a syringe (SGE Analytical Science: 2.5MDF-LL-GT, LGE Japan Ltd., Osaka,
(OLYMPUS: IX71, Olympus Co., Tokyo, Japan), and a personal computer (PC). The high-speed
Japan), a high-speed camera (Photoron: IDP-Express R2000, Photron Ltd., Tokyo, Japan), a microscope
camera and high-speed actuator are two main components for the cell manipulation inside a
(OLYMPUS: IX71, Olympus Co., Tokyo, Japan), and a personal computer (PC). The high-speed camera
microfluidic device. The camera provides real-time cell position with the frame rate of 1000 frames
and high-speed actuator are two main components for the cell manipulation inside a microfluidic
per second (fps) while the high-speed piezoelectric actuator can rapidly adjust cell position by
device. The camera provides real-time cell position with the frame rate of 1000 frames per second (fps)
generating appropriate fluid flow inside the microfluidic chip. An animation demonstrating the
while the high-speed piezoelectric actuator can rapidly adjust cell position by generating appropriate
manipulation concept can be found in Supplementary Materials Video S2. The captured image frames
fluid flow inside the microfluidic chip. An animation demonstrating the manipulation concept can
are processed by image methods for determining the cell position, and the position is sent to a
be found in Supplementary Materials Video S2. The captured image frames are processed by image
proportional-integral-derivative (PID) controller for computing a suitable output signal to move the
methods for determining the cell position, and the position is sent to a proportional-integral-derivative
actuator. The algorithms are implemented in a C program. The sampling rate in the program is 1000
(PID) controller for computing a suitable output signal to move the actuator. The algorithms are
Hz, which means a slight change of cell position would be immediately compensated by the system
implemented in a C program. The sampling rate in the program is 1000 Hz, which means a slight
within one millisecond [35].
change of cell position would be immediately compensated by the system within one millisecond [35].
Figure 2. The experimental setup. (a) A photo of the setup; (b) a schematic view of the experimental
Figure 2. The experimental setup. (a) A photo of the setup; (b) a schematic view of the experimental
system. The RBC is manipulated by a vision-feedback control system with real-time RBC position and
system. The RBC is manipulated by a vision-feedback control system with real-time RBC position and
microfluidic flow control.
microfluidic flow control.
Micromachines 2017, 8, 100
4 of 10
3.2. Design and Fabrication of PDMS Chip
3.2. Design and Fabrication of PDMS Chip
Figure 3 illustrates the fabrication of the microfluidic chip used in this work. Figure 3a shows
Figure 3 illustrates
the fabrication
of the
microfluidic
chipchannel
used inand
thisthe
work.
Figure 3aare
shows
the dimensions
of the channel
design. The
widths
of the main
constriction
10.0
the
dimensions
of
the
channel
design.
The
widths
of
the
main
channel
and
the
constriction
are
and 3.24 µm, respectively. The channel height is 3.5 µm for the both parts. The dimensions of10.0
the
and
3.24
µm,
respectively.
The
channel
height
is
3.5
µm
for
the
both
parts.
The
dimensions
of
the
channel are designed for introducing deformation to the tested RBCs. Although RBCs are known
channelextraordinary
are designeddeformability
for introducing
to the
RBCs.
Although
are
known
having
anddeformation
can go through
1–3tested
µm pores
without
lysis, RBCs
the 3.24
µm
wide
having
extraordinary
deformability
and
can
go
through
1–3
µm
pores
without
lysis,
the
3.24
µm
wide
channel is designed to let the tested RBCs deform around 50% while the length of 15 µm is to make
channel
is designed
to let
tested
RBCs
deform around
while thethrough
length ofthe
15different
µm is to width
make
sure
all deformed
RBCs
arethe
kept
inside
the constriction.
RBC50%
deformation
sure
all
deformed
RBCs
are
kept
inside
the
constriction.
RBC
deformation
through
the
different
width
of the channel has been previously investigated [36]. An inlet and an outlet are on the two sides of
of the
channel
has for
been
previously
[36]. An
inletconnection.
and an outlet
are on
theand
twoFigure
sides of
the
the
main
channel
RBCs
feedinginvestigated
and the syringe
pump
Figure
3b–f
3g–k
main channel
for RBCs feeding
theand
syringe
pump connection.
Figures 3b–f
andmold
3g–kfabrication,
illustrate the
illustrate
the fabrication
of the and
mold
microfluidic
chip, respectively.
In the
a
fabrication
of
the
mold
and
microfluidic
chip,
respectively.
In
the
mold
fabrication,
a
silicon
substrate
silicon substrate is spin coated with a layer of photoresist (SU-8 3005, Microchem Corp., Newton,
MA,
is spinbefore
coatedbeing
with exposed
a layer oftophotoresist
Microchem
Corp.,
Newton,
USA) before
USA)
ultraviolet (SU-8
(UV) 3005,
light for
lithography.
The
mold isMA,
developed
using
being
exposed
to
ultraviolet
(UV)
light
for
lithography.
The
mold
is
developed
using
propylene
glycol
propylene glycol monomethylether acetate (PM) thinner and isopropyl alcohol (IPA). The mixture
of
monomethylether
(PM) thinner
(IPA). TheMI,
mixture
and
curing
® 184, and
PDMS
and curing acetate
agent (Sylgard
Dowisopropyl
Corning alcohol
Corp., Midland,
USA) of
arePDMS
poured
over
the
® 184, Dow Corning Corp., Midland, MI, USA) are poured over the fabricated mold
agent (Sylgard
fabricated
mold in
a container for molding. After the PDMS is cured, the PDMS chip is peeled off
in a container
molding.
After
PDMS is
the and
PDMS
chip Finally,
is peeledthe
offmicrofluidic
from the mold
and
from
the moldfor
and
two holes
arethe
punched
forcured,
the inlet
outlet.
chip
is
two
holes
are
punched
for
the
inlet
and
outlet.
Finally,
the
microfluidic
chip
is
bonded
to
a
slide
glass
bonded to a slide glass using a plasma device (Cute-MP, Femto Science Inc., Gyeonggi-Do, Korea).
using a plasma device (Cute-MP, Femto Science Inc., Gyeonggi-Do, Korea).
Figure 3.
3. Design
Design and
and fabrication
fabrication of
of the
the microfluidic
microfluidic chip.
chip. (a)
(a) Schematic
Schematic view
view of
of the
the microfluidic
microfluidic chip.
chip.
Figure
The
device
consists
of
an
inlet,
an
outlet,
a
microchannel
(3.5
µm
in
height
and
10.0
µm
in
width),
and
The device consists of an inlet, an outlet, a microchannel (3.5 µm in height and 10.0 µm in width), and
narrow channel
channel (3.5
(3.5 µm
µm in
in height,
15.0 µm
µm in
narrow
height, 15.0
in length
length and
and 3.24
3.24 µm
µm in
in width);
width); (b–f)
(b–f) the
the step-by-step
step-by-step
photolithography
for
mold
fabrication;
(g–k)
the
step-by-step
procedure
for
poly-dimethylsiloxane
photolithography for mold fabrication; (g–k) the step-by-step procedure for poly-dimethylsiloxane
(PDMS) chip fabrication from the mold.
(PDMS) chip fabrication from the mold.
3.3. RBC
RBC Sample
Sample Preparation
Preparation and
3.3.
and Control
Control Sequence
Sequence
Blood for
for the
the experiment
was provided
provided by
by aa volunteer
subject who
who has
has read
read and
Blood
experiment was
volunteer subject
and agreed
agreed with
with
the
consent
of
the
test.
About
5
µL
of
blood
was
taken
by
a
sterilized
lancet
from
a
fingertip
of
the consent of the test. About 5 µL of blood was taken by a sterilized lancet from a fingertip of the
the
subject. The
The blood
blood is
is diluted
diluted 100
100 times
times by
by standard
standard saline
saline for
for reducing
reducing the
the number
number density
density of
of the
subject.
the
RBCs in
in the
the test
test sample.
sample. The
The single
single cell
cell test
test cannot
cannot be
be performed
performed without
without the
the dilution
dilution due
due to
to the
the high
high
RBCs
density of
of RBCs
RBCs in
in the
The sample
sample solution
solution is
is injected
injected into
into the
the channel
channel from
from the
the chip
chip
density
the whole
whole blood.
blood. The
inlet, and
inlet,
and then
then the
the syringe
syringe pump
pump is
is connected
connected to
to the
the outlet
outlet for
for the
the RBC
RBC manipulation.
manipulation. Because
Because the
the
changes
of
RBC
color
and
length
are
faster
at
the
beginning
and
gradually
approaching
to
asymptotic
changes of RBC color and length are faster at the beginning and gradually approaching to asymptotic
values,
rates
are set
0.1, 1 and
for the
t = 0–1 s, of
t = t1–20
s and
values, the
theimage
imagesampling
sampling
rates
areevery
set every
0.1, 10
1 sand
10 intervals
s for theofintervals
= 0–1
s,
20–290
s, respectively.
tt == 1–20
s and
t = 20–290 s, respectively.
4. Experimental Results
Figure 4 shows an example RBC in the test. The RBC was firstly caught on the left of the
constriction from a constant pressure-driven flow, as shown in Figure 4a. During the catching phase,
Micromachines 2017, 8, 100
5 of 10
4. Experimental Results
Micromachines 2017, 8, 100
Figure 2017,
4 shows
Micromachines
8, 100 an
5 of 10
example RBC in the test. The RBC was firstly caught on the left of
5 ofthe
10
constriction
from
a
constant
pressure-driven
flow,
as
shown
in
Figure
4a.
During
the
catching
phase,
the RBC was held steady for 2 s to recover from prior deformations, such as deformation caused by
the RBC was held steady for 2 s to recover from prior deformations, such as deformation caused by
the shear stress from the flow. Afterward, a new target position was given as the right end of the
the shear stress from the flow.
flow. Afterward, a new target position was given as the right end of the
constriction, and as a result, the RBC was moved into the constriction and was compressed. Figure
constriction,
andas
asaaresult,
result,the
the
RBC
was
moved into
the
constriction
and
was
compressed.
Figure
constriction,
and
RBC
was
moved
constriction
and
wasand
compressed.
Figure
4b–e
4b–e are the images of the RBC
staying
inside into
the the
channel
for 0, 30,
100,
300 s, respectively.
In
4b–e
areimages
the images
ofRBC
the RBC
staying
inside
the channel
for
0,100,
30, and
100, 300
and s,300
s, respectively.
In
are
the
of
the
staying
inside
the
channel
for
0,
30,
respectively.
In
this
this example, we can directly observe the changes of the RBC color and length during the loading
this
example,
we directly
can directly
observe
the changes
the color
RBC and
colorlength
and length
during
the loading
example,
welength
can
observe
the changes
theofRBC
during
the loading
phase. The
of the RBC
became
shorter,of
while
the color
got darker.
A
time-lapse
samplephase.
video
phase.
The of
length
of the
RBC became
shorter,
while
thegot
color
got darker.
A time-lapse
sample
video
The
length
the
RBC
became
shorter,
while
the
color
darker.
A
time-lapse
sample
video
can
be
can be found in Supplementary Materials Video S1.
can
be in
found
in Supplementary
Materials
Video S1.
found
Supplementary
Materials
Video S1.
Figure 4. An example of a RBC in the microfluidic channel before and during loading. (a) The RBC at
Figure
4.
of
in the
channel
before
loading.
(a)(a)
The
RBC
at
Figure
4. An
An example
example
ofaaRBC
RBC
themicrofluidic
microfluidic
channel
beforeand
andduring
during
loading.
The
RBC
the catching
phase; (b–e)
The in
RBC
response
to the
long-standing
load
at 0,
30, 100,
and
300
s,
the
catching
phase;
(b–e)
The
RBC
response
to
the
long-standing
load
at
0,
30,
100,
and
300
atrespectively.
the catching
phase;
response
to the long-standing
at 0, 30,time.
100, and 300 s,
s,
The
RBC’s(b–e)
colorThe
andRBC
length
were changed
with respect toload
the loading
respectively.
The RBC’s
RBC’s color
color and
and length
length were
were changed
changed with
with respect
respect to
to the
the loading
loadingtime.
time.
respectively. The
Figure 5a–c and Figure 5d-f show the color distribution of two different RBCs during the same
Figure 5a–c and Figure 5d-f show the color distribution of two different RBCs during the same
long-standing
load.
y and
axes distribution
in the figureof represent
the x,
y positions,
and
the
Figures 5a–c
andThe
5d–fx,show
thez color
two different
RBCs
during the
same
long-standing load. The x, y and z axes in the figure represent the x, y positions, and the
corresponding
color
value,
respectively.
The
color
is
in
standard
8-bit
grayscale,
where
the
color
of
long-standing load. The x, y and z axes in the figure represent the x, y positions, and the corresponding
corresponding color value, respectively. The color is in standard 8-bit grayscale, where the color of
each value,
pixel ranges
from 0 The
(white)
tois255
durations
of loading
shown
5a,d,ranges
Figure
color
respectively.
color
in (black).
standardThe
8-bit
grayscale,
where the
colorin
ofFigure
each pixel
each pixel ranges from 0 (white) to 255 (black). The durations of loading shown in Figure 5a,d, Figure
5b,e
and
Figure
5c,f
are
1,
15
and
290,
respectively.
One
consistent
tendency
for
the
two
different
from 0 (white) to 255 (black). The durations of loading shown in Figure 5a,d, Figure 5b,e and Figure 5c,f
5b,e and Figure 5c,f are 1, 15 and 290, respectively. One consistent tendency for the two different
RBCs,
well290,
as all
the tested RBCs,
was that the
overall for
pixel
color
the cellsRBCs,
got darker
with
are
1, 15asand
respectively.
One consistent
tendency
the
twoof
different
as well
as respect
all the
RBCs, as well as all the tested RBCs, was that the overall pixel color of the cells got darker with respect
to the RBCs,
loading
time.
thepixel
colorcolor
patterns
were
Thewith
pattern
around
theloading
center of
the
tested
was
thatHowever,
the overall
of the
cellsdifferent.
got darker
respect
to the
time.
to the loading time. However, the color patterns were different. The pattern around the center of the
RBC
in
Figure
5c
is
flat
whereas
the
one
in
Figure
5f
is
like
a
valley
where
the
color
of
the
central
However, the color patterns were different. The pattern around the center of the RBC in Figure 5c is
RBC in Figure 5c is flat whereas the one in Figure 5f is like a valley where the color of the central
pixels
is lighter
the
surrounding
Thewhere
shrinkages
of the
RBCcentral
lengths
during
the load
can
flat
whereas
the than
one in
Figure
5f is likepixels.
a valley
the color
of the
pixels
is lighter
than
pixels is lighter than the surrounding pixels. The shrinkages of the RBC lengths during the load can
alsosurrounding
be observedpixels.
in Figure
The lengthofchanges
RBC Sample
andload
2 can
bealso
seenbeinobserved
Figure 5a–c
the
The5.shrinkages
the RBCoflengths
during1the
can
in
also be observed in Figure 5. The length changes of RBC Sample 1 and 2 can be seen in Figure 5a–c
and
Figure
5d–f,
respectively.
The
shrinkages
of
the
RBCs
on
the
top
and
bottom
of
Figure
5
were
Figure 5. The length changes of RBC Sample 1 and 2 can be seen in Figures 5a–c and 5d–f, respectively.
and Figure 5d–f, respectively. The shrinkages of the RBCs on the top and bottom of Figure 5 were
about
5% and 14%
with
respect
to top
theirand
initial
lengths
at t = 05s.were
According
to the
two
samples
shown
The
shrinkages
of the
RBCs
on the
bottom
of Figure
about 5%
and
14%
with respect
about 5% and 14% with respect to their initial lengths at t = 0 s. According to the two samples shown
in
Figure
5,
we
find
that
while
different
RBCs
share
a
similar
tendency
of
getting
darker
and
shorter,
to their initial lengths at t = 0 s. According to the two samples shown in Figure 5, we find that while
in Figure 5, we find that while different RBCs share a similar tendency of getting darker and shorter,
they alsoRBCs
haveshare
different
responses
in terms
of the
colorand
patterns
andthey
thealso
amount
shrinkage
during
different
a similar
tendency
of getting
darker
shorter,
haveof
different
responses
they also have different responses in terms of the color patterns and the amount of shrinkage during
the
long-standing
in
terms
of the colorload.
patterns and the amount of shrinkage during the long-standing load.
the long-standing load.
Figure
Figure5.5.The
Thecolor
colordistributions
distributionsof
oftwo
twodifferent
differentRBC
RBCsamples
samplesatatthe
theloading
loadingtimes
timesofoft t==11s,s,tt==15
15s,s,
Figure
5.
The
color
distributions
of
two
different
RBC
samples
at
the
loading
times
of
t
=
1
s,
t
=
15
s,
and
RBCsample
sample22at
atthe
theloading
loadingtimes.
times.
andt t==290
290s.s.(a–c)
(a–c)RBC
RBCsample
sample11at
atthe
theloading
loadingtimes.
times.(d–f)
(d–f)RBC
and t = 290 s. (a–c) RBC sample 1 at the loading times. (d–f) RBC sample 2 at the loading times.
Figure 6 explains the image processing method for the representative cell color and length.
Figure 6 explains the image processing method for the representative cell color and length.
Figure 6a shows a captured image from the camera. A background image was subtracted from each
Figure 6a shows a captured image from the camera. A background image was subtracted from each
captured image for background removal. Figure 6b shows the cell image after background removal.
captured image for background removal. Figure 6b shows the cell image after background removal.
Although the color on a RBC is not uniform and may have different patterns, such as the two
Although the color on a RBC is not uniform and may have different patterns, such as the two
Micromachines 2017, 8, 100
6 of 10
examples shown in Figure 5, for the convenience of analysis, the gray level was defined as a
representative
color
Micromachines 2017,
8, 100value for a RBC and was the average of the color values over the RBC. The 6RBC
of 10
length is the distance between the left-most and right-most points of the detected cell area image as
the green arrow indicates in Figure 6.
Figure7a,b
6 explains
thechange
imageofprocessing
for the
representative
cellwith
color
and length.
Figure
show the
gray levelsmethod
and lengths
of all
the tested RBCs
respect
to the
Figure
6a
shows
a
captured
image
from
the
camera.
A
background
image
was
subtracted
from
each
elapsed time, while Figure 7c shows the correlation between the gray levels and lengths at different
Micromachines
2017,
8,
100
6
of 10
captured image
fordifferent
background
removal.
Figure 6binitial
shows
the levels,
cell image
instances.
Because
RBCs
have different
gray
the after
gray background
level at t = removal.
0 s was
Although the
RBC is not uniform
and may have
different
patterns,
as bar
the represent
two examples
normalized
tocolor
0 foron
thea convenience
of comparison.
The data
points
and thesuch
error
the
examples
shown 5,infor
Figure
5, for the convenience
of analysis,
the
gray
levelaswas
defined as a
shown
in
Figure
the
convenience
of
analysis,
the
gray
level
was
defined
a
representative
averages and standard deviations of all the tested RBCs. The results in Figure 7a show that the gray
representative
value for
RBCaverage
and was
the
of theover
colorthe
values
over
the
RBC.
TheisRBC
color
for color
a RBC
wasato
the
oftime
theaverage
color
values
RBC.
The
RBC
length
the
level
isvalue
increasing
withand
respect
the elapsed
of loading.
The curves
of the
gray
level
show
an
length
is
the
distance
between
the
left-most
and
right-most
points
of
the
detected
cell
area
image
as
distance
between
the
left-most
and
right-most
points
of
the
detected
cell
area
image
as
the
green
arrow
asymptotic approach to converging values for the RBCs.
the greenin
arrow
indicates
in Figure 6.
indicates
Figure
6.
Figure 7a,b show the change of gray levels and lengths of all the tested RBCs with respect to the
elapsed time, while Figure 7c shows the correlation between the gray levels and lengths at different
instances. Because different RBCs have different initial gray levels, the gray level at t = 0 s was
normalized to 0 for the convenience of comparison. The data points and the error bar represent the
averages and standard deviations of all the tested RBCs. The results in Figure 7a show that the gray
level is increasing with respect to the elapsed time of loading. The curves of the gray level show an
asymptotic approach to converging values for the RBCs.
Figure 6. Determination of RBC gray level and length from a captured image. (a) A raw image frame;
Figure 6. Determination of RBC gray level and length from a captured image. (a) A raw image frame;
(b) the enhanced image by removing the background.
(b) the enhanced image by removing the background.
Figure 7a,b show the change of gray levels and lengths of all the tested RBCs with respect to the
elapsed time, while Figure 7c shows the correlation between the gray levels and lengths at different
instances. Because different RBCs have different initial gray levels, the gray level at t = 0 s was
normalized to 0 for the convenience of comparison. The data points and the error bar represent the
averages and standard deviations of all the tested RBCs. The results in Figure 7a show that the gray
6. Determination
of RBC
level andtime
length
a captured
image. of
(a) the
A raw
image
frame;
level Figure
is increasing
with respect
to gray
the elapsed
offrom
loading.
The curves
gray
level
show an
(b)
the
enhanced
image
by
removing
the
background.
asymptotic approach to converging values for the RBCs.
Figure 7. The experimental results of the gray level and cell length of the RBCs with respect to the
loading time. (a) The change of the gray level with respect to the loading time; (b) the change of the
cell length with respect to the loading time; (c) the change of the gray level with respect to the cell
length at different instances.
The normalized cell length in Figure 7b was calculated as the ratio between every measured cell
length and its initial length. The normalization was done for the convenience of comparing RBCs
with different initial lengths in the narrow channel. The curves of the length were also logarithmically
Figure
experimental results
results of the
the gray
gray level
cell
the
Figure 7.
7. The
level and
cell length
length of
the RBCs
RBCs with
with respect
respect to
to the
the
converging
inThe
theexperimental
5 min load. Figureof 7c
indicates
aand
series
of dataofsets
at different
elapsed
times
loading
time.
(a)
The
change
of
the
gray
level
with
respect
to
the
loading
time;
(b)
the
change
of
the
loading time. (a) The change of the gray level with respect to the loading time; (b) the change of cell
the
between
the
gray
level
and
length.
Negative
correlations
between
the
gray
level
and
cell
length
are
length
with with
respect
to thetoloading
time; (c)
the (c)
change
of the gray
level
with
respect
the celltolength
at
cell length
respect
the loading
time;
the change
of the
gray
level
withtorespect
the cell
observed.
Aninstances.
interesting observation from Figure 7c is that during the first 15 s of the loading, the
different
length at different instances.
The normalized cell length in Figure 7b was calculated as the ratio between every measured cell
length and its initial length. The normalization was done for the convenience of comparing RBCs
with different initial lengths in the narrow channel. The curves of the length were also logarithmically
converging in the 5 min load. Figure 7c indicates a series of data sets at different elapsed times
between the gray level and length. Negative correlations between the gray level and cell length are
Micromachines 2017, 8, 100
7 of 10
The normalized cell length in Figure 7b was calculated as the ratio between every measured cell
length and its initial length. The normalization was done for the convenience of comparing RBCs
with different initial lengths in the narrow channel. The curves of the length were also logarithmically
converging in the 5 min load. Figure 7c indicates a series of data sets at different elapsed times between
Micromachines 2017, 8, 100
7 of 10
the gray level and length. Negative correlations between the gray level and cell length are observed.
An interesting
observation
from Figure
is gray
that during
the first 15
s of the
loading,
the 15
cells length
cell
length continuously
decreased
while7cthe
level remained
almost
constant.
From
to the
continuously
decreased
while
the
gray
level
remained
almost
constant.
From
15
s
to
the
end of
the
end of the loading, the gray level increased, corresponding to the decrease of the cell length.
That
loading,
graywere
levelshrinking
increased,
corresponding
to the
decrease
of the
That means the
means
thethe
RBCs
first
before changing
color
in the first
15 cell
s of length.
the loading.
RBCs were shrinking first before changing color in the first 15 s of the loading.
5. Discussion
5. Discussion
According to the experimental results in Figure 7, there is a correlation between the change of
According to the experimental results in Figure 7, there is a correlation between the change of the
the gray level and the cell length during the long-standing loads. The decrease of the cell length
gray level and the cell length during the long-standing loads. The decrease of the cell length happened
happened earlier than the increase of the gray level. This fact says that the increase of the gray level
earlier than the increase of the gray level. This fact says that the increase of the gray level may be
may be brought about by the decrease of the cell length. For interpreting the phenomenon, we
brought about by the decrease of the cell length. For interpreting the phenomenon, we separately
separately consider the cell behaviors in the entrance phase and the loading phase as follows:
consider the cell behaviors in the entrance phase and the loading phase as follows:
Figure 8a–d show a RBC entering a constriction. The left column of Figure 8 shows a series of
Figure 8a–d show a RBC entering a constriction. The left column of Figure 8 shows a series of
actual photos indicating the cell behavior during the entrance phase. A video clip of a RBC entering
actual photos indicating the cell behavior during the entrance phase. A video clip of a RBC entering
and exiting the constriction can be found in Supplementary Materials Video S3. A RBC is usually
and exiting the constriction can be found in Supplementary Materials Video S3. A RBC is usually
assumed being able to deform freely as long as the total surface area remains constant [37]. Based on
assumed being able to deform freely as long as the total surface area remains constant [37]. Based on
the assumption, the possible shape changes of the RBC during the entrance phase are illustrated in
the assumption, the possible shape changes of the RBC during the entrance phase are illustrated in the
the middle and right columns of Figure 8, which are the top view and the cross-sectional view,
middle and right columns of Figure 8, which are the top view and the cross-sectional view, respectively.
respectively. According to the photos, the RBC seems not only to be compressed but also folded in
According to the photos, the RBC seems not only to be compressed but also folded in half for entering
half for entering into the constriction. The cross-sectional area of the channel is gradually filled with
into the constriction. The cross-sectional area of the channel is gradually filled with the deformed
the deformed RBC during the entrance. However, there are still vacancies around the deformed RBCs
RBC during the entrance. However, there are still vacancies around the deformed RBCs after the full
after the full entrance of the RBC.
entrance of the RBC.
Figure 8. Interpretations of RBC deformation while squeezing into a constriction channel. (a) RBC
Figure 8. Interpretations of RBC deformation while squeezing into a constriction channel. (a) RBC
before the deformation; (b) the tip of the RBC is compressed and folded for squeezing into the channel;
before the deformation; (b) the tip of the RBC is compressed and folded for squeezing into the channel;
(c) a greater area of the cross-section is filled by the RBC as it moves into the channel; (d) the RBC is
(c) a greater area of the cross-section is filled by the RBC as it moves into the channel; (d) the RBC is
fully squeezed into the narrow channel.
fully squeezed into the narrow channel.
Figure 99illustrates
a possible
shape
change
of the RBC
during
long-duration
load. Illustrations
Figure
illustrates
a possible
shape
change
of the
RBCthe
during
the long-duration
load.
of
another
cross-sectional
view
from
the
bottom
are
additionally
included
in
Figure
9.
The in
decreases
of
Illustrations of another cross-sectional view from the bottom are additionally included
Figure 9.
the cell
lengthof
may
as the
cell
gradually
adapts
to the confined
the space
constriction,
The
decreases
thehappen
cell length
may
happen
as the
cell gradually
adaptsspace
to theinside
confined
inside
and
tries
to
fill
the
vacancies.
This
kind
of
shape
adaption
is
generally
known
as
strain
creep
for
typical
the constriction, and tries to fill the vacancies. This kind of shape adaption is generally known as
viscoelastic
materials,
and it means
that the and
deformation
of a compressed
RBCofisaacompressed
function of RBC
time
strain
creep for
typical viscoelastic
materials,
it means that
the deformation
when
the load
constraint
is fixed.
Based
on the assumption
of the
surface of
area,
shape
is
a function
ofor
time
when the
load or
constraint
is fixed. Based
on constant
the assumption
thethis
constant
surface area, this shape change makes the cell length shorten in the narrow channel. Furthermore, the
adaption results in an equivalently thicker RBC from the top view as shown on the right of Figure 9c.
The increase of the RBC thickness would reduce the transparency of the cell, and as a result, the cell
is darkened and the gray levels increase as the experimental results show in Figure 7. However,
further fluorescence or confocal microscopic images are needed to confirm the morphology changes
Micromachines 2017, 8, 100
8 of 10
change makes the cell length shorten in the narrow channel. Furthermore, the adaption results in an
equivalently thicker RBC from the top view as shown on the right of Figure 9c. The increase of the RBC
thickness would reduce the transparency of the cell, and as a result, the cell is darkened and the gray
levels increase as the experimental results show in Figure 7. However, further fluorescence or confocal
microscopic images are needed to confirm the morphology changes of RBCs during the long-standing
Micromachines 2017, 8, 100
8 of 10
load and are the future direction of this work.
Figures 88 and
and 99 provide
provide aa possible
possible interpretation
interpretation for
for the
the correlation
correlation between
between the
the changes
changes of
of the
the
Figures
RBC color
color and
and length
length from
from the
the perspectives
perspectives of
of mechanical
mechanical deformation.
deformation. We
We would
would like
like to
to emphasize
emphasize
RBC
that
the
interpretation
does
not
exclude
other
possible
changes
inside
the
RBC,
such
as
reformation
of
that the interpretation does not exclude other possible changes inside the RBC, such as reformation
the
cytoskeleton
or
the
exhaust
of
adenosine
triphosphate
inside
the
RBCs.
In
fact,
it
is
very
possible
of the cytoskeleton or the exhaust of adenosine triphosphate inside the RBCs. In fact, it is very
that there
were
fundamental
changes changes
of the RBCs
based
onbased
the findings
from the from
previous
study on
possible
that
there
were fundamental
of the
RBCs
on the findings
the previous
cell
fatigue
[24].
The
other
possible
effects
of
RBCs
under
long-standing
loads
are
worth
investigating
study on cell fatigue [24]. The other possible effects of RBCs under long-standing loads are worth
further from further
different
perspectives,
such as chemistry
biology,
forbiology,
RBC deformability.
investigating
from
different perspectives,
such asand
chemistry
and
for RBC deformability.
Figure
InterpretationsofofRBC
RBC
changes
during
the long-standing
in the constriction.
(a)
Figure 9. Interpretations
changes
during
the long-standing
load inload
the constriction.
(a) Loading
Loading
time
t = 60
s; (b) time
loading
times;t (c)
= 180
s; (c) loading
time t = 60
s; (b)
loading
t = 180
loading
time t = time
300 s.t = 300 s.
6.
6. Conclusions
Conclusions
The
The responses
responses of
of RBCs
RBCs during
during aa long-standing
long-standing load
load in
in aa microfluidic
microfluidic constriction
constriction are
are
experimentally
experimentally investigated.
investigated.The
Thecolor
colorof
ofRBCs
RBCsin
in the
the constriction
constriction becomes
becomes darker
darker while
while the
the length
length
of
respect
to to
thethe
loading
time.
The The
spatial
and temporal
variations
of theof
color
of the
theRBCs
RBCsshrinks
shrinkswith
with
respect
loading
time.
spatial
and temporal
variations
the
change
are
analyzed.
The
results
show
both
logarithmic
changes
of
the
color
and
the
length
of
color change are analyzed. The results show both logarithmic changes of the color and the lengththe
of
RBCs.
We
also
find
an
unanticipated
negative
correlation
between
the
color
and
the
RBC
length
the RBCs. We also find an unanticipated negative correlation between the color and the RBC length
according
according to
to the
the experimental
experimental results.
results. Interpretation
Interpretation from
from the
the perspective
perspective of
ofmechanical
mechanicaldeformation
deformation
is
is discussed.
discussed. RBC
RBCresponses
responsesunder
undersuch
suchaa long-standing
long-standingload
loadare
aresimilar
similarto
to aa RBC
RBC being
being plugged
plugged in
in
microcirculation
inside
a
body,
and
the
results
could
provide
new
insights
into
the
time-dependent
microcirculation inside a body, and the results could provide new insights into the time-dependent
quality
quality of
of plugged
plugged RBCs.
RBCs.
Supplementary
areare
available
online
at www.mdpi.com/2072-666X/8/4/100/s1;
Video
Supplementary Materials:
Materials: The
Thefollowing
following
available
online
at www.mdpi.com/2072-666X/8/4/100/s1;
Video
S1: A time-lapse
theresponse
cell response
during
a long-standing
Video
S2: The
conceptual
animation
S1:
A time-lapse
video video
of the of
cell
during
a long-standing
load;load;
Video
S2: The
conceptual
animation
of
of
cell
manipulation;
Video
S3:
A
slow-motion
video
of
a
RBC
passing
through
a
constriction
channel
showing
the
cell manipulation; Video S3: A slow-motion video of a RBC passing through a constriction channel showing the
dynamic response of cell folding and unfolding.
dynamic response of cell folding and unfolding.
Acknowledgments: This work was partially supported by JSPS KAKENHI Grant Nos. JP15H05761, JP16K14197,
Acknowledgments:
This work
was partially
supported
by JSPSInstitute,
KAKENHI
Grant
Nos. JP15H05761,
JP16K14197,
JP16K18051, JP16H06933,
Foundation
Advanced
Technology
and
“Nanotechnology
Platform
Project
JP16K18051,
JP16H06933,
Foundation
Advanced
Technology
Institute,
“Nanotechnology
Platform Project
(Nanotechnology
Open Facilities
in Osaka
University)”
of MEXT
Japan and
[F-16-OS-0006,
S-16-OS-0005].
(Nanotechnology Open Facilities in Osaka University)” of MEXT Japan [F-16-OS-0006, S-16-OS-0005].
Author Contributions: All authors conceived and designed the experiments; M.H., C.-H.D.T. and H.I.
performed the experiments; M.H., C.-H.D.T. and H.I. contributed to the data analysis and interpretations; M.H.,
C.-H.D.T. and M.K. wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
Micromachines 2017, 8, 100
9 of 10
Author Contributions: All authors conceived and designed the experiments; M.H., C.-H.D.T. and H.I. performed
the experiments; M.H., C.-H.D.T. and H.I. contributed to the data analysis and interpretations; M.H., C.-H.D.T.
and M.K. wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Glenister, F.; Coppel, R.; Cowman, A.F.; Mohandas, N.; Cooke, B.M. Contribution of parasite proteins to
altered mechanical properties of malaria-infected red blood cells. Blood 2002, 99, 1060–1063. [CrossRef]
[PubMed]
Lim, C.T. Single cell mechanics study of the human disease malaria. J. Biomech. Sci. Eng. 2006, 1, 82–92.
[CrossRef]
Tomaiuolo, G. Biomechanical properties of red blood cells in health and disease towards microfluidics.
Biomicrofluidics 2014, 8, 51501. [CrossRef] [PubMed]
Mokken, F.; Kedaria, M.; Henny, C. The clinical importance of erythrocyte deformability, a hemorrheological
parameter. Ann. Hematol. 1992, 64, 113–122. [CrossRef] [PubMed]
Li, X.; Vlahovska, P.; Karniadakis, G. Continuum- and particle-based modeling of shapes and dynamics of
red blood cells in health and disease. Soft Matter 2013, 9, 28–37. [CrossRef] [PubMed]
Roth, K.B.; Eggleton, C.D.; Neeves, K.B.; Marr, D.W.M. Measuring cell mechanics by optical alignment
compression cytometry. Lab Chip 2013, 13, 1571–1577. [CrossRef] [PubMed]
Gossett, D.R.; Henry, T.K.; Lee, S.A.; Ying, Y.; Lindgren, A.G.; Yang, O.O.; Rao, J.; Clark, A.T.; Di Carlo, D.
Hydrodynamic stretching of single cells for large population mechanical phenotyping. Proc. Natl. Acad.
Sci. USA 2012, 109, 7630–7635. [CrossRef] [PubMed]
Otto, O.; Rosendahl, P.; Mietke, A.; Golfier, S.; Herold, C.; Klaue, D.; Girardo, S.; Pagliara, S.; Ekpenyong, A.;
Jacobi, A.; et al. Real-time deformability cytometry: On-the-fly cell mechanical phenotyping. Nat. Methods
2015, 12, 199–202. [CrossRef] [PubMed]
Tsai, C.D.; Sakuma, S.; Arai, F.; Taniguchi, T.; Ohtani, T.; Sakata, Y.; Kaneko, M. Geometrical alignment for
improving cell evaluation in a microchannel with application on multiple myeloma red blood cells. RSC Adv.
2014, 4, 45050–45058. [CrossRef]
Salehyar, S.; Zhu, Q. Deformation and internal stress in a red blood cell as it is driven through a slit by an
incoming flow. Soft Matter 2016, 12, 3156–3164. [CrossRef] [PubMed]
Park, S.Y.; Dimitrakopoulos, P. Transient dynamics of an elastic capsule in a microfluidic constriction.
Soft Matter 2013, 19, 8844–8855. [CrossRef] [PubMed]
Fischer, T.M. Shape memory of human red blood cells. Biophys. J. 2004, 86, 3304–3313. [CrossRef]
Markle, D.R.; Evans, E.A.; Hochmuth, R.M. Force relaxation and permanent deformation of erythrocyte
membrane. Biophys. J. 1983, 42, 91–98. [CrossRef]
Baez, S.D.; Kaul, M.E.; Nagel, R.L. Microvascular determinants of blood flow behavior and HbSS erythrocyte
plugging in microcirculation. Blood Cells 1982, 8, 127–137. [PubMed]
Wiewiora, M.; Sosada, K.; Wylezol, M.; Slowinska, L.; Zurawinski, W. Red blood cell aggregation and
deformability among patients qualified for bariatric surgery. Obes. Surg. 2007, 17, 365–371. [CrossRef]
[PubMed]
Cahn, A.; Livshits, L.; Srulevich, A.; Raz, I.; Yedgar, S.; Barshtein, G. Diabetic foot disease is associated with
reduced erythrocyte deformability. Int. Wound J. 2015, 13, 500–504. [CrossRef] [PubMed]
Dondorp, A.M.; Kager, P.A.; Vreeken, J.; White, N.J. Abnormal blood flow and red blood cell deformability
in severe malaria. Parasitol. Today 2000, 16, 228–232. [CrossRef]
Pivkin, I.V.; Peng, Z.; Karniadakis, G.E.; Buffet, P.A.; Dao, M.; Suresh, S. Biomechanics of red blood cells in
human spleen and consequences for physiology and disease. Proc. Natl. Acad. Sci. USA 2016, 113, 7804–7809.
[CrossRef] [PubMed]
Lawrence, C.; Fabry, M.E.; Nagel, R.L. The unique red cell heterogeneity of SC disease: Crystal formation,
dense reticulocytes, and unusual morphology. Blood 1991, 78, 2104–2112. [PubMed]
Zheng, Y.; Nguyen, J.; Wei, Y.; Sun, Y. Recent advances in microfluidic techniques for single-cell biophysical
characterization. Lab Chip 2013, 13, 2464–2483. [CrossRef] [PubMed]
Micromachines 2017, 8, 100
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
10 of 10
Reich, E.; Goldberg, I.H. Actinomycin and nucleic acid function. Prog. Nucleic Acid Res. Mol. Boil. 1964, 3,
183–234.
Tözeren, A.; Skalak, R.; Sung, K.L.; Chien, S. Influence of physicochemical factors on rheology of human
neutrophils. Biophys. J. 1982, 39, 23–32. [CrossRef]
Brandao, M.M.; Fontes, A.; Barjas-Castro, M.L.; Barbosa, L.C.; Costa, F.F.; Cesar, C.L.; Saad, S.T.O. Optical
tweezers for measuring red blood cell elasticity: Application to the study of drug response in sickle cell
disease. Eur. J. Haematol. 2003, 70, 207–211. [CrossRef] [PubMed]
Dulińska, I.; Targosz, M.; Strojny, W.; Lekka, M.; Czuba, P.; Balwierz, W.; Szymoński, M. Stiffness of normal
and pathological erythrocytes studied by means of atomic force microscopy. J. Biochem. Biophys. Methods
2006, 66, 1–11. [CrossRef] [PubMed]
Hirose, Y.; Kaneko, M.; Kawahara, T.; Yamanishi, Y.; Arai, F. High speed cell stiffness evaluation toward 100%
reliability. In Proceedings of the 2011 IEEE SENSORS, Limerick, Ireland, 28–31 October 2011; pp. 962–965.
Hirose, Y.; Tadakuma, K.; Higashimori, M.; Arai, T.; Kaneko, M.; Iitsuka, R.; Yamanishi, Y.; Arai, F. A new
evaluation toward high speed cell sorter. In Proceedings of the 2010 IEEE International Conference Robotics
and Automation (ICRA), Anchorage, MD, USA, 3–7 May 2010; pp. 4113–4118.
Tsai, C.D.; Sakuma, S.; Arai, F.; Kaneko, M. A new dimensionless index for evaluating cell stiffness-based
deformability in microchannel. IEEE Trans. Biomed. Eng. 2014, 61, 1187–1195. [CrossRef] [PubMed]
Zheng, Y.; Shojaei-Baghini, E.; Azad, A.; Wang, C.; Sun, Y. High-throughput biophysical measurement of
human red blood cells. Lab Chip 2012, 13, 2560–2567. [CrossRef] [PubMed]
Sakuma, S.; Kuroda, K.; Tsai, C.D.; Fukui, W.; Arai, F.; Kaneko, M. Red blood cell fatigue evaluation based on
close-encountering point between extensibility and recoverability. Lab Chip 2014, 14, 1135–1141. [CrossRef]
[PubMed]
Ito, H.; Murakami, R.; Sakuma, S.; Tsai, C.D.; Gutsmann, T.; Brandenburg, K.; Ito, H.; Murakami, R.;
Sakuma, S.; Tsai, C.D.; et al. Mechanical diagnosis of human erythrocytes by ultra-high speed manipulation
unraveled critical time window for global cytoskeletal remodeling. Sci. Rep. 2017, 7, 43134. [CrossRef]
[PubMed]
Murakami, R.; Tsai, C.D.; Ito, H.; Tanaka, M.; Sakuma, S.; Arai, F.; Kaneko, M. Catch, Load and Launch
toward On-Chip Active Cell Evaluation. In Proceedings of the 2016 IEEE International Conference Robotics
and Automation (ICRA), Stockholm, Sweden, 16–21 May 2016; pp. 1713–1718.
Horade, M.; Tsai, C.D.; Ito, H.; Tanaka, M.; Kaneko, M. Chameleon effect of RBC under loading in microfluidic
channel. In Proceedings of the 19th International Conference on Miniaturized Systems for Chemistry and
Life Sciences (MicroTAS16), Dublin, Ireland, 9–13 October 2016; pp. 311–312.
Tsai, C.D.; Horade, M.; Ito, H.; Tanaka, M.; Kaneko, M. High-resolution cell manipulation for longstanding
load on red blood cells. In Proceedings of the IEEE International Conference on Mechatronics and
Automation (ICMA2016), Harbin, China, 7–10 August 2016; pp. 914–919.
Horade, M.; Kojima, M.; Kamiyama, K.; Kurata, T.; Mae, Y.; Arai, T. Development of an optimum end-effector
with a nano-scale uneven surface for non-adhesion cell manipulation using a micromanipulator. J. Micromech.
Microeng. 2015, 25, 5002. [CrossRef]
Monzawa, T.; Kaneko, M.; Tsai, C.D.; Sakuma, S.; Arai, F. On-chip actuation transmitter for enhancing the
dynamic response of cell manipulation using a macro-scale pump. Biomicrofluidics 2015, 9, 014114. [CrossRef]
[PubMed]
Tsai, C.D.; Tanaka, J.; Kaneko, M.; Horade, M.; Ito, H.; Taniguchi, T.; Ohtani, T.; Sakata, Y. An on-chip
RBC deformability checker significantly improves velocity-deformation correlation. Micromachines 2016, 7,
176–185. [CrossRef]
Evans, E.A.; Waugh, R.; Melnik, L. Elastic area compressibility modulus of red cell membrane. Biophys. J.
1976, 16, 585–595. [CrossRef]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).