You are on page 1of 6

Folding dynamics of tethered giant DNA under strong flow

Takuya Saito, Takahiro Sakaue, Daiji Kaneko, Masao Washizu, and Hidehiro Oana

Citation: J. Chem. Phys. 135, 154901 (2011); doi: 10.1063/1.3652957
View online: http://dx.doi.org/10.1063/1.3652957
View Table of Contents: http://jcp.aip.org/resource/1/JCPSA6/v135/i15
Published by the American Institute of Physics.

Related Articles
Folding dynamics of tethered giant DNA under strong flow
JCP: BioChem. Phys. 5, 10B614 (2011)
Rigorous coarse-graining for the dynamics of linear systems with applications to relaxation dynamics in proteins
JCP: BioChem. Phys. 5, 08B605 (2011)
Rigorous coarse-graining for the dynamics of linear systems with applications to relaxation dynamics in proteins
J. Chem. Phys. 135, 054107 (2011)
Comparison of two adaptive temperature-based replica exchange methods applied to a sharp phase transition of
protein unfolding-folding
JCP: BioChem. Phys. 5, 06B614 (2011)
Comparison of two adaptive temperature-based replica exchange methods applied to a sharp phase transition of
protein unfolding-folding
J. Chem. Phys. 134, 244111 (2011)

Additional information on J. Chem. Phys.
Journal Homepage: http://jcp.aip.org/
Journal Information: http://jcp.aip.org/about/about_the_journal
Top downloads: http://jcp.aip.org/features/most_downloaded
Information for Authors: http://jcp.aip.org/authors
Downloaded 02 Nov 2011 to 128.223.222.31. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
THE JOURNAL OF CHEMICAL PHYSICS 135, 154901 (2011)
Folding dynamics of tethered giant DNA under strong flow
Takuya Saito,
1,a)
Takahiro Sakaue,
2,3
Daiji Kaneko,
1
Masao Washizu,
1,4,5
and Hidehiro Oana
1,5,b)
1
Department of Mechanical Engineering, The University of Tokyo, Tokyo 113-8656, Japan
2
Department of Physics, Kyushu University 33, Fukuoka 812-8581, Japan
3
JST, PRESTO, Japan
4
Department of Bioengineering, The University of Tokyo, Tokyo 113-8656, Japan
5
JST, CREST, Japan
(Received 17 May 2011; accepted 28 September 2011; published online 20 October 2011)
Using a microfluidic device, we investigate the folding dynamics of individual linear long DNA,
whose one end is tethered under a strong flow in the presence of a condensing agent. Direct obser-
vations of the folding process of DNA molecules reveal a characteristic dynamics with pronounced
non-monotonic velocity of the folded part at the free end against the flow. We discuss this unique dy-
namics in relation to the inhomogeneous spatial fluctuation and the structure change at the multiple
order levels along the stretched DNA, which is induced by the increasing tension due to the build-up
of the hydrodynamic drag force. © 2011 American Institute of Physics. [doi:10.1063/1.3652957]
I. INTRODUCTION
Genomic DNAs are very long biomacromolecules and as
for the eukaryote, their length range generally from mm to
cm. Nonetheless, they are neatly folded in a narrow cellular
space and function properly with dynamical conformational
changes. The formation of a higher-order structure by DNA
folding is thus expected to be essential for biological func-
tions such as replication and transcription. The equilibrium
aspects of DNA folding have been extensively studied. One
accomplishment is the fact that DNA folding has been clar-
ified to be a discrete transition from a fluctuating coil to a
folded compact state at a single chain level upon the addition
of condensing agents.
1
As compared to the transition manner between stable
structures involved in DNA folding, its dynamics have not
yet been well understood. To obtain further insights, folding
process of individual DNA molecules has been observed uti-
lizing stretched DNA obtained by tethering its one end un-
der a flow of condensing agent’s solution. Previous studies on
DNAchains shorter than ca. 100 μmhave shown that the fold-
ing rate is roughly constant during folding, i.e., the stretched
length from the tethered point to the free end of the chain
roughly monotonically shrinks in this process.
2–5
It is stressed
that this situation is qualitatively different from the folding
transition at equilibrium state in various aspects. Among oth-
ers, the stretching effect caused by the flow is notable. Due
to the building-up of the hydrodynamic drag force along the
chain, the degree of segmental spatial fluctuations should de-
crease in the upstream direction.
6
Thus, if we employ consid-
erably longer chains, the inhomogeneous effect of segmental
spatial fluctuations should be significant.
a)
Present Address: Department of Physics, Kyushu University 33, Fukuoka
812-8581, Japan.
b)
Author to whom correspondence should be addressed. Electronic mail:
oana@mech.t.u-tokyo.ac.jp.
In this paper, we investigate the folding process of long
DNA (more than 200 μm in length) with one end tethered un-
der a strong flow in the presence of a condensing agent. In this
case, it is noteworthy that the describable inhomogeneity is
caused by the deformation of segments (variant from normal
B-form on secondary structure) due to the overstretching of
the double helical backbone
7, 8
as well as spatial fluctuations
of invariant segments. Then, the hierarchical nature of the
DNA molecule becomes explicit, and there may arise a cou-
pling between transitions in different length scales under this
flow condition. In principle, clarifying these aspects should
be important for genome technology from the viewpoint of
understanding living matter, for which the reliable control
of hierarchical structures involving multiple order scales is
required.
II. MATERIALS AND METHODS
In our experiment, we used large DNA molecules which
were obtained from an agarose gel block (Bio-Rad Labora-
tories, Inc.) in which yeast chromosomal DNAs (S. pombe)
are embedded as size markers of linear DNAs with lengths
of 1.2, 1.6, and 1.9 mm. By melting a piece of the gel block
(ca. 10 mm
3
) for 30 min at 80

C in 1 ml of 2 M sodium
chloride solution, a solution of chromosomal DNA was ob-
tained. With this procedure, we obtained a dilute DNA solu-
tion which contains DNA molecules longer than 200 μm al-
though large DNA molecules in random coil state are easy
to be fragmented during pipette-base handling due to hy-
drodynamic shear force. The direct observation of the single
chain folding was performed using a microfluidic channel de-
vice fabricated by standard soft lithography with PDMS. As
schematically depicted in Fig. 1, this main channel (0.5 mm
in width) was equipped with a micro-pillar array to hook and
stretch linear DNA chains. The hooked DNA can be regarded
as a chain with one end tethered at the pillar. In addition, these
pillars stand the stall force produced by the DNA chain. The
0021-9606/2011/135(15)/154901/5/$30.00 © 2011 American Institute of Physics 135, 154901-1
Downloaded 02 Nov 2011 to 128.223.222.31. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
154901-2 Saito et al. J. Chem. Phys. 135, 154901 (2011)
Top view
micro pillar region
2
,
0
0
0
μ
m
22 mm
0.5 mm
Cover slip
PDMS
Micro pillar
Inlets Outlet
2000 μm
20 μm
9
0

μ
m
6
0

μ
m
2
5

μ
m
Cross-sectional view
FIG. 1. The schematic representation of the micro-flow channel on the in-
verted microscope for the observation of DNA folding process. The in-
set represents arrangement of micro-pillars for the immobilization of DNA
molecules. (Reproduced and modified from Ref. 9, T. Saito et al., 2008 Inter-
national Symposium on Micro-NanoMechatronics and Human Science, with
permission from c [2008] IEEE.)
inlets and outlet were on the upstream and downstream sides
of the pillars, respectively. Each reagent was fed from dif-
ferent inlets, which were connected with sample solutions
through silicone tubes, respectively. The flow of each solution
was independently controlled by the microfluidics flow con-
trol system (MFCS; FLUIGENT). First, the genomic DNA
solution was fed, hooked to the pillar by chance, and stretched
by a flow. Then, to rinse off sodium chloride and unhooked
DNA from the main channel, purified water (by Milli-Q Gra-
dient; Millipore) was introduced. After the immobilization of
a single DNA molecules on the micro pillars, i.e., obtain-
ing tethered DNA molecules, the condensing agent (1 mM
spermidine (trivalent cation)) was fed, and we observed the
dynamics of DNA folding under inverted fluorescence mi-
croscopy (IX-71; Olympus Corp.). In all solutions, 1 μM YO-
PRO-1 as a fluorescent dye for DNA and 1 mM dithiothre-
itol as a deoxidizer were dissolved. The fluorescence image
was acquired using a high-sensitivity video camera (EB-CCD
camera, C7190-43; Hamamatsu Photonics K.K.).
III. RESULTS AND DISCUSSION
Figure 2 shows a time series of fluorescence images of
DNAfolding induced by the addition of the condensing agent,
spermidine. In this article, we set the term “stretched length”,
which is the distance from tethered point to the downstream
free end of the DNA chain. Typically, before folding, the
stretched length is greater than the width of the field of view
due to the optical setup (ca. 200 μm), and the free down-
stream end is not observed in the flame. Upon the addition
of condensing agents, the stretched length starts decreasing.
Shortly after the condensing agent is introduced, the down-
stream free end enters the visual field owing to the shrinkage
of the stretched length. Note that the free downstream end
was observed as the first bright spot that corresponds to the
folded part. Here, we set this time as t = 0. This downstream
folded part (#1) moves toward the tethered point at the pillar
with time. In many samples, plural bright folded parts (#2, #3,
. . . ) were newly observed along the stretched chain, especially
0 s
0.17 s
1.0 s
10.0 s
59.0 s
60.0 s
60.4 s
60.8 s
61.2 s
61.6 s
61.8 s
62.0 s
100 μm
t

(
s
)
Flow direction
#1
#3
#4
#1
#2
Tethered point at micro pillar
DNA
FIG. 2. The typical folding process of long DNA at a flow velocity of ca.
300 μm/s. The flow is directed rightward. The bright spots indicate folded
parts, which are labeled by #1 ∼ #4 from the downstream end. (Reproduced
and modified from Ref. 9, T. Saito et al., 2008 International Symposium
on Micro-NanoMechatronics and Human Science, with permission from c
[2008] IEEE.)
near the downstream free end through the folding process, as
seen after 60 s. These move against the flow while fusing with
the downstream end or other folded parts.
To clearly characterize the folding dynamics using the
sample shown in Fig. 2, we established the time evolution
of the stretched length from the tethered point to the folded
part at the downstream end, together with the positions of
other folded parts, as shown in Fig. 3. Hereafter, the veloc-
ity at the downstream end is called the folding velocity, and
it corresponds to the shrinkage rate of the stretched length.
Figure 3(a) shows the overall folding process. This result
indicates that folding proceeds through three stages: [I] the
folded part at the free end rapidly moves toward the teth-
ered point, but slows down at a length of around 170 μm;
[II] then, the folding almost plateaus in around a dozen sec-
onds; and [III] finally, the chain suddenly and rapidly folds
completely within a few seconds. This non-monotonic fold-
ing process was observed under flow velocity of more than
ca. 200 μm/s (we examined up to ca. 600 μm/s). On the other
hand, for the relatively shorter DNA(<200 μm), the plateau
stage was not observed during the folding under flow velocity
of less than ca. 300 μm/s (not shown). This rather monotonic
folding for the shorter DNA under the weak flow is in agree-
ment with previous reports.
2–5
Figures 3(b) and 3(c) show
magnified views of stages [I] and [III], respectively. Also,
the plural folded parts have been often observed. Similarly,
Downloaded 02 Nov 2011 to 128.223.222.31. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
154901-3 DNA folding under flow J. Chem. Phys. 135, 154901 (2011)
I II III
I III
II
200
150
100
50
0
30 20 10 50 0 40 60 70
t (s)
180
170
t (s)
0 1 2 3
200
100
0
59 60 61 62
t (s)
(a)
(b)
(c)
S
t
r
e
t
c
h
e
d

l
e
n
g
t
h

L

(
μ
m
)
S
t
r
e
t
c
h
e
d

l
e
n
g
t
h

L

(
μ
m
)
FIG. 3. Time evolution of the length from the tethered point to the folded
parts along the folding. Data points are obtained from the sample shown
in Fig. 2. (), (), (), and () correspond to folded parts, #1 ∼ #4,
from the downstream end, respectively. (a) Overall process. Magnified views:
(b) [I] first stage and (c) [III] last stage. Red lines serve as a visual guide for
the stretched length. (Reproduced and modified from Ref. 9, T. Saito et al.,
2008 International Symposiumon Micro-NanoMechatronics and Human Sci-
ence, with permission from c [2008] IEEE.)
this phenomenon has been observed in other experiments by
using shorter DNA chain.
2
Note that, in our experiment, the
plural folded parts show the explicit tendency to emerge near
the downstream end, which should reflect the great difference
of the spatial fluctuation along the chain in relation to the nu-
cleus formation. The non-monotonic spatial fluctuation along
the chain will be discussed.
How can we describe the non-monotonic folding process
of the tethered long DNA under strong flow? To begin with,
let us describe the DNA as “an inextensible worm-like chain”
(inextensible WLC), the contour length of which is invariant.
As shown in Fig. 4(a), its one end is fixed at the origin x = 0.
Suppose that the DNA is folded against a uniform flow with
velocity V
s
where single folded part at the downstream end is
considered for simplicity. This WLC polymer consists of M
0
segments, which are labeled from the tethered point (n = 0).
M(t)-th segment is the boundary between the folded and un-
folded parts at time t, and the unfolded and folded segments
are assumed to be in the range 0 ≤ n ≤ M(t) and M(t) ≤ n
≤ M
0
, respectively. The distance from the origin to n-th seg-
ment is given by
x[n] =
n

k=1
l[k], (1)
where l[k] is the projected length of the k-th segment to the
flow direction. Here we obtain the stretched length of the un-
folded DNA as L = x[M(t )] (=

M(t )
k=1
l[k]). Note that the arc
length of each segment is the Kuhn length b
0
, and the total
contour length is given by L
c
= b
0
M
0
.
The tension at n-th segment is given by the building-
up of the hydrodynamic force from downstream end to
(c)
0
Vs Flow velocity:
Tethered point
L x
(a)
Folded part
Inextensible WLC
~100 nm
(b)
0 L
b
0
l(x)
b
0
l(x)
0
x
t / τ
0
2000 1000 0
L
/
L
0
1
0.5
0
εb
0
/k
B
T:
4.0
5.0
6.0
FIG. 4. (a) Schematic representation of long DNA conformation under flow.
x-axis indicates the direction of flow. (b) Plot of the projected length per
segment l(x) in the case of the “inextensible WLC” model under steady strong
flow. Note that the arc length per segment is constant b
0
. (c) Theoretical time
evolution of the stretched length L for the inextensible WLC.
n-th segment.
10
Assuming that the extension of each seg-
ment could be characterized by the uniformtension-stretching
relation,
11, 12
we find the local force balance as follows:
2πη(M(t ) −n)b
0
V
s

k
B
T
2b
0
(1 −l[n]/b
0
)
2
, (2)
where k
B
T is the thermal energy, and η is the viscosity of the
solvent, and the left-hand side corresponds to the tension of
n-th segment. Note that the drag force acting on the
folded part is comparable with that for the single unfolded
segment.
13
This balance implies that an upstream transverse
fluctuation decreases because of the increasing tension due to
the build-up of the drag force along the stretched chain.
10
We
demonstrate the projected length per segment l(x) in the case
of the inextensible WLC under the steady flow V
s
as shown
in Figure 4(b). The projected length per segment should prac-
tically approach its arc length b
0
as x goes from downstream
end to 0.
The free energy of the folding chain may be written as F
= − b
0
[M
0
− M(t)], where a phenomenological parameter
Downloaded 02 Nov 2011 to 128.223.222.31. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
154901-4 Saito et al. J. Chem. Phys. 135, 154901 (2011)
( > 0) represents a free energy gain upon folding per length.
Balancing the rate of free energy change with the dissipation
rate associated with the folding nucleus dF/dt = −dQ/dt, we
obtain
b
0
dM(t )
dt
=
_
V
s

dL
dt
_
dL
dt
. (3)
where it is assumed that the free energy change in the folded
part is dissipated owing to the drag force against the folded
part, and is the relevant drag coefficient. Introducing con-
tinuous variable s( ≡b
0
k), we replace the summation with the
integral in Eq. (1), and then putting Eq. (2) into the resultant
integral leads to
L b
0
_
M −
_
2Mb
0
τ
0
V
s
_
, (4)
where τ
0
= 2πηb
3
0
/k
B
T . Its derivative with respect to t is
dL
dt
b
0
_
1 −
_
b
0
2Mτ
0
V
s
_
dM
dt
. (5)
The combination of this result with Eq. (3) gives the following
time evolution for M(t):
dM
dt




V
s
b
0

b
0
/k
B
T

0
_
1 −
_
b
0
2Mτ
0
V
s
_



__
1 −
_
b
0
2Mτ
0
V
s
_
.
(6)
where the coefficient c = /2πηb
0
is assumed to be a con-
stant of order unity, which might be a reasonable approxi-
mation from the observation that the hydrodynamic radius
of the compactly folded DNA is of the order of the Kuhn
segment.
13
Figure 4(c) shows the theoretical time evolution
of the stretched length L(t) obtained from Eqs. (4), (6), which
well captures the qualitative feature of the folding dynamics
in stage [III].
However, this analysis does not interpret a longer plateau
after initial rapid folding at stage [I]. Here, as the poten-
tial mechanism, we discuss the secondary structure change,
i.e., overstretching beyond its natural contour length in B-
form. Caron’s group
7
and Bustamante’s group
8
independently
found that DNA shows abrupt overstretching when it is pulled
beyond the critical tension ∼70 pN. Caron’s group termed this
overstretched form of the secondary structure as S-form. In
our experimental condition (V
s
∼ 100 − 600 μm/s), the drag
force per segment is more than 2πηb
0
V
s
∼ 0.1 − 1 pN.
14
Therefore, as for the long tethered DNA (more than ∼200
μm in contour length), its segmental number greater than 2
×10
3
should result in a large upstream tension ∼10
2
pN. Tak-
ing account of this estimation, the upstream secondary struc-
ture of DNA in our experiments should be remarkably de-
formed and might become S-form as shown in Fig. 5(a) due
to the extreme building-up of the hydrodynamic drag force
from its downstream part.
To discuss how S-DNA is involved in the non-monotonic
folding process as shown in Fig. 3(a), we introduced an exten-
sible WLC model, the contour length of which is extensible.
In this model, when the tension is small, the DNA chain is
~100 nm
Folded part
S-DNA B-DNA
x
Extensible WLC
0
b
0
l(x)
0 L
Secondary structure
(i) S-form DNA (ii) B-form DNA
(S-DNA) (B-DNA)
(a)
(b)
x
Flow
0 L
(c)
T
h
r
e
s
h
o
l
d

l
e
n
g
t
h

L
c
Flow velocity Vs
Monotonic
Folding
Non-monotonic Folding
Vs
*
FIG. 5. (a) Schematic representation of long tethered DNA under strong flow
as “extensible WLC” model. The upstream segments are remarkably over-
stretched accompanied with the secondary structure change (closeup (i)).
This overstretched part (S-form) of the DNA is indicated by a red dashed
rectangle. Other downstream part of the DNA takes normal B-form DNA
(closeup (ii)) as the secondary structure, and the downstream end part is
folded. (b) Plot of the projected length per segment l(x), corresponding to
Fig. 5(a). The value l(x) at the upstream segment (S-DNA) is greater than its
arc length b
0
in the absence of the tension. The red dashed rectangle indi-
cates the overstretched part of the DNA. (c) Diagram separating the mono-
tonic folding and non-monotonic folding regions in (V
s
− M) plane with the
boundary given by Eq. (7). From our experiments, the order of the critical
velocity is V

s
∼ 200 μ m/s in the present spermidine concentration.
almost inextensible with keeping its structure as B-form, and
when the tension is larger than a certain threshold value, it
becomes extensible with changing its structure from B-form
to S-form. Based on this model, we established the conjec-
tured plot of the projected length per segment, l(x), to the flow
direction as shown in Fig. 5(b). Here, this plot is obtained
by utilizing the reported experimental results of “overstretch-
ing” at uniform tension.
7, 8
Figure 5(b) represents that in the
downstream part, l(x) approaches the arc interval b
0
toward
the upstream with keeping its secondary structure in the B-
form. Furthermore, it represents that in the upstream part of
the tethered DNA, the abrupt secondary structure change, i.e.,
overstretching occurs, and l(x) in the S-form becomes greater
than b
0
due to suffering a large tension which exceeds the
critical tension (∼70 pN
7, 8
).
While the upstream part moving against the flow must
pull the downstream part, the secondary structure change
in the extremely overstretched part should provide a much
greater gain in the free energy. Hence, during the early stage
Downloaded 02 Nov 2011 to 128.223.222.31. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions
154901-5 DNA folding under flow J. Chem. Phys. 135, 154901 (2011)
in the folding, the overstretched upstream part could be re-
tracted, leading to greater shrinkage of the stretched length
together with free energy gain. However, with a decrease in
the unfolded segments, the upstream structure approaches the
normal B-form, and the gain in the free energy due to the sec-
ondary structure change decreases. In this situation, the fold-
ing velocity should decrease. This process should correspond
to the period stage [I] in Fig. 3.
In addition, dM/dt = 0 in the equation of time evolution
(Eq. (6)) suggests the presence of the threshold in the un-
folded contour length L
c
(, V
s
) (=b
0
M
c
(, V
s
)) with the given
attraction strength and flow velocity V
s
, implying that DNA
longer than that threshold is not folded within the framework
of the inextensible WLC model. From Eq. (6), the following
relation is satisfied at the threshold condition;
b
0
k
B
T

τ
0
V
s
b
0
_
1 −
_
b
0
2Mτ
0
V
s
_
. (7)
In Fig. 5(c), we show a diagram separating the monotonic
folding and non-monotonic folding regions in (V
s
−M) plane
with the boundary curve given by Eq. (7). Below the flow ve-
locity V

s
b
2
0
/(τ
0
k
B
T ), DNA chain shows the monotonic
folding irrespective of the chain length, and then the boundary
line of the threshold length L
c
appears beyond that. Indeed, in
our experiments, the non-monotonic folding was not observed
for the relatively shorter DNA (<200 μm) and slower flow
velocity (<ca. 300 μm/s), and the non-monotonic folding ap-
peared for longer DNA (>200 μm) and faster flow velocity
(>ca. 200 μm/s). These correspond to our theoretical trend.
Given the simplicity in our theoretical model, this should be
a rather reasonable agreement, which qualifies the present
argument.
What is the mechanism inducing the transition to stage
[III]? The observed plateau time in experiments differs each
time even under the same conditions, suggesting that the tran-
sition to stage [III] in Fig. 3 is an activation process, although
the precise mechanism is not known.
Here, let us compare our experimental results with previ-
ous ones.
2–5
These experiments have been performed in a sim-
ilar situation. The folding dynamics, however, differs consid-
erably, i.e., previous works reported a nearly constant speed
on folding, whereas our observed result indicates more in-
volved dynamics with a highly non-monotonic folding veloc-
ity. This difference between them should be attributable to
the initial chain length. In other words, the folding velocity
for DNA smaller than 100 μm should correspond to that at
the late stage [III] for longer DNA, because initially large ten-
sion variation in long chain becomes less noticeable with the
growth of the folded part.
Our scenario for three-stage-folding [I]–[III] is summa-
rized as follows:
[I]: The upstream segments are initially overstretched with
the deformation of the secondary structure. The decrease
in the unfolded downstream segments reduces the ten-
sion at the upstream. Hence, the overstretched segments
are retracted, accompanied by the secondary structure
change.
[II]: Then, the folding becomes stagnant because the free en-
ergy gain by the overstretching is exhausted. A long
plateau continues until the length of the unfolded part
becomes lesser than the threshold by some sort of
fluctuation.
[III]: Eventually, the rapid folding is caused, because the up-
stream part of the tensed B-form DNA is slacked due
to the lowering of the tension accompanied with the nu-
cleus growth.
IV. CONCLUDING REMARKS
In summary, we have studied the folding process of long
DNA under a strong flow at a velocity ca. 300 μm/s. Direct
observations revealed the following non-monotonic change
in the folding velocity: [I] an initial fast folding; [II] subse-
quent slowing down, leading to the long plateau; and [III]
the final sudden acceleration and the completion of the fold-
ing. We consider that this notable feature in the folding
dynamics should arise from the inhomogeneous conforma-
tion change involving various order levels on the hierarchi-
cal structure together with the non-uniform spatial fluctuation
that is inherent to a much longer tethered chain under a strong
flow.
ACKNOWLEDGMENTS
T.S. and T.S. thank H. Nakanishi at Kyushu University
for useful discussions. This work was supported in part by an
Industrial Technology Research Grant Program from NEDO,
Japan, and KAKENHI (No. 20034008, No. 20840027, No.
21114507) from the MEXT, Japan. Photography masks were
fabricated using the EB lithography apparatus available at the
VLSI Design and Education Centre (VDEC), the University
of Tokyo.
1
K. Yoshikawa, M. Takahashi, V. V. Vasilevskaya, and A. R. Khokhlov,
Phys. Rev. Lett. 76, 3029 (1996).
2
L. R. Brewer, M. Corzett, and R. Balhorn, Science 286, 120 (1999).
3
T. J. Su, E. Theofanidou, J. Arlt, D. T. F. Dryden, and J. Crain, J. Fluoresc.
14, 65 (2004).
4
G. Wagner, A. Bancaud, J.-P. Jean-Pierre Quivy, C. Clapier, G. Almouzni,
and J.-L. Viovy, Biophysical J. 89, 3647 (2005).
5
K. Hirano, M. Ichikawa, T. Ishino, M. Ishikawa, Y. Baba, and
K. Yoshikawa, “How environmental solution conditions determine the
compaction velocity of single DNA molecules,” Nuc. Acids Res. (in press).
6
T. T. Perkins, D. E. Smith, R. G. Larson, and S. Chu, Science 268, 83
(1995).
7
P. Cluzel, A. Lebrun, C. Heller, R. Lavery, J.-L. Viovy, D. Chatenay, and
F. Caron, Science 271, 792 (1996).
8
S. B. Smith, Y. Cui, and C. Bustamante, Science 271, 795 (1996).
9
T. Saito, T. Sakaue, D. Kaneko, M. Washizu, and H. Oana, Dynamics in the
Folding of Long DNA Chain under Strong Flow, IEEE International Sym-
posiumon Micro-NanoMechatronics and Human Science in 2008, 118–123
(2008).
10
F. Brochard-Wyart, A. Buguin, and P. G. de Gennes, Europhys. Lett. 47,
171 (1999).
11
J. F. Marko and E. D. Siggia, Macromolecules 28, 8579 (1995).
12
R. G. Larson, T. T. Perkins, D. E. Smith, and S. Chu, Phys. Rev. E 55, 1794
(1997).
13
H. Oana, K. Tsumoto, Y. Yoshikawa, and K. Yoshikawa, FEBS. Lett. 530,
143 (2002).
14
M. Doi and S. F. Edwards, The Theory of Polymer Dynamics (Oxford
Science, Oxford, 1986).
Downloaded 02 Nov 2011 to 128.223.222.31. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions