-
S1
A rapid and practical technique for real-time monitoring of
biomolecular interactions using
mechanical responses of macromolecules Mehmet C. Tarhan1,2,*,
Nicolas Lafitte2, Yannick Tauran2,3, Laurent Jalabert2, Momoko
Kumemura1, Grgoire Perret2,4, Beomjoon Kim1, Anthony W. Coleman1,3,
Hiroyuki Fujita1 and Dominique Collard1,2,*
1CIRMM, Institute of Industrial Science, The University of
Tokyo, Tokyo, Japan.
2LIMMS/CNRS-IIS UMI2820, Institute of Industrial Science, The
University of Tokyo, Tokyo, Japan.
3LMI UMR5615, University of Lyon 1, Lyon, France.
4IEMN UMR8520, Lille, France.
-
S2
Supplementary Figures
Supplementary Figure 1: Tweezers insertion protocol was
performed in three steps. i) At first the PDMS rim was used to
detect the top-level of the channel (1,2) and tweezers were
positioned close to the channel
vicinity (3). ii) The second step was to detect the precise
location of the channel. The top and bottom level of the channel
were detected (4,5) and positioned at the required height (6). iii)
The last step was to detect the air-liquid meniscus (7). After
detection, a safe position was saved before DNA capturing (8).
Assays were
performed inside channel at the desired immersion depth.
2
7
Placement for safe insertion2
7 Detecting the meniscus location
In-channel position:
Safe position: In air (50 m away from interface)
1
1 Detecting the PDMS location
3
3 Placing channel opening to the tweezers vicinity
45
6
4 5
6
Detecting the channel height & position &
Locating device at the desired height
8
8 Moving to the safe position
PDMS
Channel
GlassTweezers
-
S3
Supplementary Figure 2: The position of the air-liquid interface
changes slightly closer to the glass surface due to the hydrophilic
surface. 30 m above the glass surface resulted in 5 for all
cases).
Supplementary Figure 3: An example of tweezers frequency
response in air and in liquid without any macromolecules between
the tips. Inserting the SNT tips in the liquid resulted in 1.6 Hz
increase in fR and
0.5% decrease in Amax due to the air-liquid interface.
0
20
40
60
80
100
120
0 1 2 3 4
Dis
tanc
e fro
m g
lass
sur
face
(m
)
Interface location (m)
No access due totweezers geometry (30m)
Glas
ssu
rface
PDMS
surfa
ce
0
2
4
6
8
10
12
14
1150 1200 1250 1300 1350 1400 1450 1500
Ampl
itude
(mV)
Frequency (Hz)
in liquidin air
4
6
8
10
12
1280 1300 1320 1340
Ampl
itude
(mV)
Frequency (Hz)
-
S4
Supplementary Figure 4: Tweezers were inserted into the channel
(from an opening of h= 115, w= 108 m) at different immersion depth.
Between each step, tweezers were taken out to a position 3 m away
from the
air-liquid interface. The position of each step with respect to
the air-liquid interface is shown in the inset. The error bars
correspond to standard deviations.
Supplementary Figure 5: a) Tweezers were inserted into channel
(t=0 min) at different insertion height
(varying from 10 m to 70 m above the glass surface for a device
with h=115 m, w=108 m). b) Zoom out view of the results shown in
(a). The initial immersion depth was 5 m for all the cases.
However, insertion at 10-m height caused solution to go out under
the tweezers because of the capillary effect between the glass
surface and the bottom of the tweezers. c) A graph showing the
average resonance
frequency shift. Z-position corresponds to the distance from the
glass surface. The error bars correspond to standard
deviations.
Supplementary Figure 6: Another example of devices with larger
openings (h=171 m, w= 134 m)
showing similar characteristics as supplementary figure 4.
Tweezers insertion at 10-m height resulted in an unstable response.
Height > 30 m showed stable results. Z-position corresponds to
the distance from the
glass surface. The error bars correspond to standard
deviations.
0
1.0
2.0
3.0
4.0
0 5 10 15 20 25 30 35 40Time (min)
f R s
hift
(Hz) 0
1
2
3
-5 0 5 10 15 20x-position (m)
Avg.
f R s
hift
(Hz)
a)
0
1
2
3
4
5
70mz-position:
50m
40m30m10m
f R s
hift
(Hz) f R s
hift
(Hz)
Time (min)0
10
20
30
0 2 4 6 8 10
0 2 4 6 8 10Time (min)
0
10
20
10 20 30 40 50 60 70 80z-position (m)
b)
c)
Avg.
f R s
hift
(Hz)
0
1
2
3
4
5
6
7
0 20 40 60 80 100 120 140
Avg.
f R s
hift
(Hz)
z-position (m)
-
S5
Supplementary Figure 7: Dimensions of the side opening affected
the stability. Tweezers were inserted into
the channel (at t = 0 min) and monitored for 20 min for
stability. Some of the examples are shown above. Lower openings
caused instability due to capillary effect between the bottom of
the tweezers and the glass surface as seen in the given standard
deviations. Similarly, narrow openings caused low stability due to
the
capillary effect between the tweezers arms and the PDMS
walls.
Supplementary Figure 8: Devices with different size opening were
monitored for stability based on a) the width and b) the height of
the opening. Some examples of the monitored devices were shown
in
supplementary figure 1. Wider (w>110 m) and higher openings
(h>90 m) provided more stable results.
0
1
2
0 10 20
0
1
2
3
4
0
1
2
0
5
10
0
5
10
0
1
2
3
0
1
2
0 10 20
0
1
2
4
f R s
hift
(Hz)
Time (min)
71 m 40 m
width:height:
58 m 110 m
width:height:
Std. dev.:0.53 Hz
Std. dev.:1.45 Hz
128 m 117 m
width:height:
195 m 190 m
width:height:
114 m 43 m
width:height:
74 m 94 m
width:height:
117 m 89 m
width:height:
205 m 270 m
width:height:
Std. dev.:0.07 Hz
Std. dev.:0.08 Hz
Std. dev.:0.24 Hz
Std. dev.:1.10 Hz
Std. dev.:0.08 Hz
Std. dev.:0.08 Hz
0
1
2
3
4
5
6
7
8
9
50 100 150 200 250
Width (m)
Avg.
f R s
hift
(Hz)
0
1
2
3
4
5
6
7
8
9
0 50 100 150 200 250 300
Height (m)
a) b)
Avg.
f R s
hift
(Hz)
-
S6
Supplementary Figure 9: Pump working at the withdrawn mode
generated a flow in the channel. Applying different pressure did
not compromise stability. In the experiments, the necessary
pressure level was applied
(for 20 s) to induce a flow rate of ~50 l min-1.
Supplementary Figure 10: DNA capturing was performed using
DEP-assisted lateral combing. a) Tweezers were inserted into DNA
solution (0.175 g ml-1). Then an AC voltage (1 MHz, 3.2-4.8 Vp-p)
was
applied between tweezers tips. b) The stage was moved laterally
to remove the tips of the tweezers consequently for the capturing
process. c) Molecular combing steps: (i) Parts of DNA molecules
were
attached on the Al-coated tips after inserted into the solution.
(ii) Lateral motion of the droplet allowed one of the tips to be
removed from the droplet while performing molecular combing.
Regions of DNA drawn out of
the solution were stretched and aligned. (iii) The bridge was
formed with stretched and aligned DNA molecules when the second tip
was removed from the droplet.
-0.20
-0.15
-0.10
-0.05
0.00
0.05
0 2 4 6 8 10
f R s
hift
(Hz)
Time (min)
Pressure level (mbar):
Pressure level
0 mbar
0 00-5 -5-10 -10 -10 -10-15 -15-20 -20
-5 mbar
-10 mbar
-15 mbar
-20 mbar
a) b)
CapturedDNA bundle
CapturedDNA bundle
DNAsolution
Tweezers
DNAsolution
Motiondirection
Motiondirection
Tweezerstips
c)
i)
ii)
iii)
-
S7
Supplementary Figure 11: No-DNA control experiments did not show
any increase in the stiffness or the viscous losses for a)
decreasing pH (for 5 min of buffer/acid cycles with a pH decreasing
from 4.8 to 2.1) and
b) increasing Ag concentration (from 10 M to 100 mM). For better
comparison both axis are having the same scales as the DNA
experiments shown in Fig. 5.
Supplementary Figure 12: Different DNA bundles resulted in
similar characteristics when exposed to
varying pH levels. Decreasing pH resulted in increasing
stiffness and viscous losses.
Supplementary Figure 13: Different DNA bundles resulted in
similar characteristics when exposed to
increasing Ag+ concentration. Unlike the pH case, increasing Ag+
concentration affected only the stiffness. Viscous losses stayed
constant even though stiffness increased with the increasing Ag+
concentration.
0.1
1
10
100
10-7
10-6
10-5
10-4
22.533.544.55pH
Stiff
ness
(N m
-1)
Viscous Losses (N s m
-1)
Stiffness
Viscous Losses
0
0.1
1
10
10-7
10-6
10-5
10-6 10-5 10-4 10-3 10-2 10-1
Stiffness
Viscous Losses
Stiff
ness
(N m
-1)
Viscous Losses (N s m
-1)
Ag+ concentration (M)
-
S8
Supplementary Figure 14: Ag+ selectively and specifically
affected the DNA bundle. a) A DNA bundle was exposed to 1 mM Ag+
solution right after a metal ion mixture (Na+, K+, Mg2+, Ca2+ and
Co2+, 1 mM each). The stiffness showed a big increase even though
the pH of solutions were almost identical (4.85 and 4.80
respectively). b) A similar result was obtained when the metal
mixture was followed by an Ag-metal-mixture (Ag+, Na+, K+, Mg2+,
Ca2+ and Co2+, 1 mM each). Ag+ selectively showed an effect on the
DNA bundle in
presence of other metal ions.
Supplementary Figure 15: Effect of acid on the mechanical
properties of the DNA bundle was reversible even at very low pH.
Tris buffer (pH 6.8) followed HNO3 solution (pH 2.1) injection in 5
min cycles. HNO3 solution increased fR and decreased Amax of the
tweezers and buffer returned them back to the initial values.
a) b)
0
0.2
0.4
0.6
0.8
1
0
0.05
0.10
0.15
0.20
Mixture Ag &Mixture
Mixture(1mM each)
Ag(1mM)
Stiff
ness
(N m
-1)
Stiff
ness
(N m
-1)
Mixture: Na+, K+, Mg2+,Ca2+ & Co2+
-
S9
Supplementary Figure 16: Images of DNA bundle at different pH
levels monitored with visual light. At low
pH level, (-) charges on DNA molecules possibly decreased due to
protonation and thus, DNA bundle became tighter causing an increase
in the viscous losses (image on the left). Increasing the pH
resulted in deprotonation and thus DNA molecules in the bundle
started repelling each other (image in the middle).
Therefore, the bundle was difficult to visualize. 3 minutes
after acid injection, the bundle was visible again (image on the
right). The differential interference contrast (DIC) images were
taken with a photometrics
camera (Cascade II) on an inverted microscope stage (Olympus
IX71).
Supplementary Figure 17: Longer experiments (> 3h) were also
performed on the same DNA bundle. The example shown in this figure
examined nickel ion effect on DNA (until t = 1 h). Then, on the
same DNA,
zinc ion effect at low pH values were tested for demonstration
purposes.
Supplementary Figure 18: Microtubules were captured between the
tweezers tips to demonstrate possibility
of capturing other macromolecules using the proposed method. The
gap between the tips is 16 m.
pH 3.1
DNAbundle
20 m
pH 3.1pH 6.8
1300
1350
1400
1450
1500
1550
10
0 1 2 30
5
15
20
25
30
35
Res. Freq. (Hz)
Ampl. (V)
Time (h)
Ampl. (m
V)R
es. F
req.
(Hz)
CapturedMTs
20 m
-
S10
Supplementary Figure 19: a) Frequency response (amplitude and
phase) of tweezers with and without DNA (in air). Blue dots
correspond to the measurements and red lines correspond to the
performed fitting by least square method to extract the fR and Amax
values. b) A damped harmonic oscillator model was used to
derive
the molecular stiffness and viscous losses values of the
molecular bundle.
Supplementary Figure 20: A portion of the raw data presented in
figure 2a. After smoothing the correlation
between the fR and temperature could be seen much more clearly.
Temperature increase of 0.2C inside the box caused a decrease of
~0.02 Hz in fR
Supplementary Movies Supplementary Movie 1: DNA capturing
Supplementary Movie 2: Automated detection of air-liquid
interface
Supplementary Movie 3: Liquid exchange in the microfluidic
device
0
2
4
6
8
10
12
1100 1200 1300 1400 1500 1600
with DNA
w/o DNA
Ampl
. (m
V)
Freq. (Hz)
with DNAw/o DNA
-100
-50
0
50
100
Phas
e (d
eg)
a) b)
R=0.99996 R=0.99991
fR,w/o DNA= 1199Hz
Amax,w/o DNA= 10.9mV
fR,with DNAAmax,with DNA
Qw/o DNA= 49.2
Molecularbundle
Fixedtip
kmb
mb
MTweezers Actuation
k
Actuatingtip
0.50
0.51
0.52
0.53
0.54
0.55
28.90
28.95
29.00
29.05
29.10
29.15
0 0.5 1 1.5 2 2.5
Time (hour)
f R s
hift
(Hz)
Temperature (C
)