Download Visualizing single DNA-bound proteins using DNA as a scanning

ARTICLES
© 2007 Nature Publishing Group http://www.nature.com/naturemethods
Visualizing single DNA-bound proteins using DNA
as a scanning probe
Maarten C Noom1,2, Bram van den Broek1,2, Joost van Mameren1 & Gijs J L Wuite1
Many biological processes involve enzymes moving along DNA.
Such motion might be impeded by DNA-bound proteins or
DNA supercoils. Current techniques are incapable of directly
measuring forces that such ‘roadblocks’ might impose. We
constructed a setup with four independently moveable optical
traps, allowing us to manipulate two DNA molecules held
between beads. By tightly wrapping one DNA around the other,
we created a probe that can be scanned along the contour of the
second DNA. We found that friction between the two polymers
remains below 1 pN. Upon encountering DNA-bound proteins
substantial friction forces are measured, allowing accurate
localization of protein positions. Furthermore, these proteins
remained associated at low probe tensions but could be
driven off using forces greater than 20 pN. Finally, the full
control of the orientation of two DNA molecules opens
a wide range of experiments on proteins interacting with
multiple DNA regions.
Experiments with single DNA molecules have revealed many
intrinsic properties of DNA and associated proteins. In these
experiments, the precise control of a single DNA molecule allows
studying properties of, and interactions between DNA and proteins
difficult to assess in conventional biochemical experiments. However, biological processes that require the involvement of multiple
DNA tracts are still hard to explore with single DNA molecule
experiments. For example, various DNA recombination proteins1–3,
restriction endonucleases4 and bridging nucleoid-associated proteins5 interact simultaneously with two separate DNA regions. The
relative angle between, and tensions on such regions can not be
controlled when handling a single DNA molecule. Here we describe
a method that allows us to manipulate two DNA molecules
independently. This technique permits us to study proteins interacting with multiple DNA binding sites6. To demonstrate the
capabilities of this technique, we used the two DNA molecules as
a scanning probe technique to detect and manipulate DNA-bound
proteins. Many biological processes on DNA involve either ATPdriven or one-dimensional diffusive motion along the DNA contour7–9. Bound proteins or DNA supercoils might act as
roadblocks10, impeding such motion. With this new scanning
technique, one DNA molecule is wrapped around the other
molecule (Fig. 1a) and then used as a mechanical probe to scan
along the contour of the first DNA molecule. With this scanning
technique, we applied forces on single DNA-bound proteins in a
direction along the DNA contour. Consequently, we used this
system to study removal of these roadblocks similar to when they
are encountered in vivo by motor proteins translocating along
the DNA contour.
RESULTS
Dual DNA manipulation
We designed and built an optical-tweezers instrument that allows
manipulation of two DNA molecules in three dimensions simultaneously and independently by trapping micrometer-sized polystyrene beads attached to the ends of the DNA molecules. Four
optical traps are generated by first splitting a laser beam in two
orthogonally polarized beams (Fig. 1b). One of these beams
generates a continuous trap; the other beam is time-shared over
three trap positions using acousto-optic deflectors. Forces acting on
the bead in the continuous trap can be detected with subpicoNewton resolution using back-focal-plane interferometry11,12.
To attach DNA between the four beads held in the optical traps, we
have previously designed and constructed a flow chamber with
multiple laminar flows of solution running parallel to each other13.
By moving the chamber relative to the optical traps, the four
trapped beads can be moved into different solutions. The ends of
two DNA molecules can be attached to the four beads (Fig. 1c).
After ‘catching’ two DNA molecules, the beads are moved into a
channel containing only buffer. Using force-extension analysis we
ensured that every pair of beads only holds one DNA molecule
(Supplementary Fig. 1 online). The four traps can be freely moved
with respect to each other in the sample plane, giving us full control
over the relative orientation of the two DNA molecules as well as
the tension on both molecules. Additionally moving the continuous trap in the third dimension allows us to wind one DNA
molecule around the other.
Using DNA as scanning probe
To scan one DNA duplex (the scanned DNA) using the second one
as probe (the probing DNA), we positioned the beads such that the
two DNA molecules are in a ‘crossed’ configuration. Next we
1Physics
2These
of Complex Systems, Department of Physics and Astronomy, Faculty of Sciences, Vrije Universiteit, De Boelelaan 1081, 1081 HV Amsterdam, The Netherlands.
authors contributed equally to this work. Correspondence should be addressed to G.J.L.W. (gwuite@nat.vu.nl).
RECEIVED 31 JULY; ACCEPTED 10 OCTOBER; PUBLISHED ONLINE 11 NOVEMBER 2007; DOI:10.1038/NMETH1126
NATURE METHODS | ADVANCE ONLINE PUBLICATION | 1
ARTICLES
© 2007 Nature Publishing Group http://www.nature.com/naturemethods
wrapped the probing DNA around the scanned DNA once or
multiple times (Fig. 1a and Supplementary Movie 1 online). We
stretched both DNA molecules to a preset tension, (typically 5–20
pN) to create a tight DNA loop. For a circular DNA loop, the
diameter D depends on the tension S and the persistence length P as
(see ref. 14):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
D 2P kB T=S ðref : 14Þ;
(s.d., 0.2 pN) and even in DNA-condensing conditions16 (200 mM
spermine; s.d., 0.4 pN). Scans at various speeds (B200–2,000 bp/s)
and with various tensions on both DNA molecules, with single and
multiple DNA windings yielded similar results as well. Thus, even
though in DNA-condensing conditions the repulsion between the
negatively charged DNA molecules is practically neutralized, very
little frictional interaction was present when the two tightly
wrapped DNA molecules were moved past each other.
To further demonstrate the localization and manipulation capabilities of our technique, we incubated the scanned l DNA with
Type IIP restriction enzymes in noncleaving conditions. These
restriction enzymes bind specifically to their recognition sequence.
Because the locations of these recognition sequences are known,
these enzymes function as a convenient site-specific marker in these
approximately 5–10 nm for the tensions mentioned above. This is
an upper limit of the actual probe size because in our experiments
the DNA loop does not have a circular but a twisted or supercoillike structure (Fig. 1a). By moving both beads that hold the probing
DNA simultaneously parallel to the scanned DNA, we used the
DNA loop to probe the contour of the
scanned DNA (Fig. 2a). If this DNA loop
a
stalls because of an obstacle or friction on
the scanned DNA, we measured an increase
in force on the bead in the continuous trap.
We recorded a scanning trace of l DNA
in the presence of 5 mM Ca2+ (Fig. 2b). The
feature on the left of the measured reference
scan (Fig. 2b) is due to the collinear alignment of the time-shared optical traps 3 and
4 with the (continuous) trap 1, used for
force detection15. We corrected for this in all
subsequent traces (see Supplementary
Fig. 2 online for details). The left edge of
a scanning trace is marked by the probing
DNA pushing against bead 1, resulting in a
b
negative force. Similarly, at the right edge of
the trace, the probing DNA pushes against
bead 2 resulting in a positive force meaBeam steering
sured at bead 1. The difference in slopes
Polarizing
originates from additional stretching of the
beam splitter
scanned DNA.
Trapping
Beam
laser
Notably, the friction force between the
expander
1,064
tightly pulled DNA loop and the scanned
nm
DNA remains well below 1 pN (s.d. of
friction signal was 0.3 pN around the baseλ /2
line). We obtained similar results when
Polarizer
scanning in the presence of 150 mM Na+
Figure 1 | Dual DNA manipulation assay. (a) Two
l DNA molecules suspended between polystyrene
beads held with optical tweezers. The probing DNA
molecule is wound around the scanned DNA
molecule. (b) Schematic representation of the
experimental setup. Inset, an impression of the
multi-channel flow cell. (c) The dual-DNA
experiment is conducted in a four-channel flow
cell with nonmixing laminar flows. Four beads are
trapped with four optical traps in the bead
channel (i). Two DNA molecules (l DNA molecules
in experiments presented here) are caught
between the beads in the DNA channel (ii).
DNA windings are imposed in the buffer channel
(iii). In the last protein channel the DNA is
incubated in protein solution (iv). Scanning
can be performed either in the protein or
buffer channel.
2 | ADVANCE ONLINE PUBLICATION | NATURE METHODS
LED
Quadrant
photodiodes
Detection
laser
980
nm
Dichroic 3
Dichroic 4
Microscope
x,y
x,y,z
Dichroic 1
Polarizing
beam splitter
Dichroic 2
AOD
VCO
CCD cameras
c
(i)
Beads
(ii)
DNA
(iii)
Buffer
(iv)
Proteins
ARTICLES
Figure 2 | DNA scanning scheme and reference scan. (a) To detect
DNA-bound proteins, the probing DNA is moved along the scanned DNA.
Upon encountering a protein bound to the scanned DNA, the measured force
on bead 1 scales with distance Dy. (b) Typical scanning trace of l DNA
without proteins (gray) using a probing force of 10 pN. The black trace
shows the same data, corrected for interference between time-shared traps
holding beads 3 and 4 and the continuous trap holding bead 1
(Supplementary Fig. 2).
y
Forward
3
k,
L0,
S0
Scanned DNA
2
Pro
bin
at specific and nonspecific sites imply that the nature of the events
is different. The fact that peak forces at nonspecific sites are lower
may indicate that the probe triggers such a protein to dissociate,
whereas a specifically bound protein may either dissociate
(at a higher peak force) or stay bound with the probe slipping
over it. Below we demonstrate that for low probe tensions the latter
is the case.
gD
L0 = initial length of probing DNA
S0 = initial tension on probing DNA
4
S 0= protein
∆y
b
Friction force (pN)
8
Reference scan
Corrected reference scan
4
0
–4
0
2
4
6
8
10
12
14
16
Lateral position of beads holding probing DNA (µm)
experiments. The plot in Figure 3a displays three consecutive scans
in forward and backward directions along the same l DNA
molecule with EcoRI restriction enzymes bound it (Fig. 3a). An
additional friction force is measured when the probing DNA loop
stalls behind DNA-bound EcoRI proteins. We detected force peaks
at four locations that match with the expected specific target sites of
EcoRI, immediately revealing the orientation of the DNA. Besides
peaks at these specific sites, we occasionally observed additional
(smaller) peaks at other locations. Presumably this is a signature of
nonspecific or noncognate binding of EcoRI to sites that are similar
(for example, 1-base-pair mismatch) to the specific binding
sequence (over 250 of such sites are present on l DNA). At every
bound protein that is encountered by the DNA probe the measured
force on bead 1 increases with similar inclination, after which it
‘snaps back’ to a zero friction force. The distribution of measured
peak forces for both nonspecific and specific events from the scan
displayed in Figure 3b indicates that forces measured at nonspecific
sites, 2 ± 1 pN (s.e.m.; n ¼ 7) with a typical duration of 3 ± 1 s, are
substantially lower than at specific sites, 5 ± 1 pN (s.e.m.; n ¼ 11)
with a duration of 7 ± 2 s. The distinctly different forces measured
Resolution and accuracy
To determine the accuracy and resolution of our scanning technique, we corrected our data to compensate for lateral stretching of
the scanning DNA molecule and bead displacement out of the
optical traps (Supplementary Note online). Figure 3b displays a
corrected force-distance trace for the EcoRI forward curve (Fig. 3a),
where the x axis represents the actual position of the probing loop
and the force increases vertically at every encountered protein. We
assessed the accuracy of localization of the DNA-bound proteins by
a
8
4
*
*
**
*
0
Forward scan 1
Backward scan 1
Forward scan 2
EcoRI sites
Nonspecific sites
*
–4
–8
0
2
4
6
8
10
12
14
16
18
Lateral position of beads holding probing DNA (µm)
b
Measured force
Calculated force on probing DNA loop
EcoRI sites
8
6
4
2
0
0
2
4
6
8
10
12
14
16
18
Scanning loop (probe) position along DNA (µm)
Figure 3 | Detection of individual DNA-bound restriction enzymes. (a) Three
consecutive scans along l DNA in forward and backward directions, with 0.1
units/ml (B1 nM) EcoRI bound to the DNA using a probe tension of 10 pN.
(b) During each event, the traveled distance by the probing DNA, Dy, can be
corrected by modeling the lateral stretching of the probing DNA and the bead
displacement out of the optical traps. Corrected data of the probing DNA
scanning loop was calculated from the EcoRI forward scan (measured force).
At four peaks the distribution is fit with Gaussians (bottom) to determine the
location of specifically bound proteins with B120 bp resolution. (c) Detail of
a scan along a l DNA molecule incubated with EcoRV using a probe tension of
25 pN. The black line is drawn to guide the eye. Vertical gray bars indicate the
spatial resolution of the localization.
c
12
Measured force
EcoRV sites
10
Friction force (pN)
© 2007 Nature Publishing Group http://www.nature.com/naturemethods
NA
k0 = DNA spring constant
Friction force (pN)
Force
1
Friction force (pN)
a
8
6
4
2
0
–2
2.6
2.8
3.0
3.2
3.4
Probe position along DNA (µm)
NATURE METHODS | ADVANCE ONLINE PUBLICATION | 3
ARTICLES
© 2007 Nature Publishing Group http://www.nature.com/naturemethods
20
15
10
Forward scan 1
Forward scan 2
Forward scan 3
Sbf I site
b
Friction force (pN)
Friction force (pN)
a
40
30
Forward scan 1
Backward scan 1
Forward scan 2
EcoRI site
absence of proteins at specific binding locations is limited by the biochemical binding
probability of proteins18 or irregularities on
the DNA.
20
(Non-)destructive imaging
Do the specifically bound proteins remain
bound when the scanning loop encounters
0
0
them, or are they pulled off their binding
0
2
4
6
8
10 12 14 16
3.0
3.0
2.5
2.5
2.0
1.5
1.0
site and subsequently replaced by others
Probe position along DNA (µm)
Probe position along DNA (µm)
present in the surrounding solution? To
examine this, we first ‘loaded’ the dual
Figure 4 | Nondestructive and destructive imaging. (a) Three consecutive scans (in forward direction and
DNA construct with SbfI restriction enzyme
corrected for lateral stretching of the probing DNA) acquired in a buffer solution without proteins, after
loading of the DNA with SbfI. All detected proteins remain bound when scanned with a low probe tension
(under conditions where dissociation gen(5–10 pN). (b) Destructive imaging. A larger tension on the probing DNA (35–40 pN) yields a smaller
erally takes B1 h)19 by briefly moving it
loop-size. Consecutive traces show that this results in a DNA molecule with no proteins associated to it.
into a flow channel containing the proteins.
The two peaks right after the third specific site are attributed to nonspecifically bound proteins, briefly
We then moved the DNA into a channel
interacting (B3 s) with the scanning probe before dissociating.
with identical buffer but without proteins
to ensure that no more proteins could bind
evaluating a dwell time histogram of the corrected data (Fig. 3b). to the DNA. Then we scanned the DNA repeatedly at a probe
From the standard deviation values of Gaussian fits to the observed tension of B10 pN. We measured the friction force of a part of a l
peaks, we determined the spatial resolution to be B120 bp. DNA molecule with a protein bound to the single SbfI site present
Comparison with the known locations of the specific binding in this region, by scanning it three times in succession (Fig. 4a). SbfI
sequences on l DNA yielding a relative position accuracy of remained bound, demonstrating that protein binding is undisB50 bp. Closer analysis of the DNA that contains four EcoRV turbed by the DNA scanning. Apparently, the DNA loop that exerts
restriction sites allowed us to unambiguously distinguish three of lateral force on the bound protein slips over it when some critical
the sites (Fig. 3c). We could not distinguish the fourth site because
force is reached (in this case B10–20 pN).
it is only separated 35 base pairs from the adjacent site, demonIncreasing the tension on the probing DNA should result in a
strating the resolution of the technique.
tighter DNA loop, potentially preventing it from slipping over the
The binding and unbinding of DNA-associated proteins can be
protein. When we increased this probe tension to 25–40 pN, the
probed with a repetition rate equal to the scan time, typically 100 s observed force peaks were in fact higher: 20–40 pN (Fig. 4b). In
for the complete l DNA at a scanning speed of 500 bp/s. The
subsequent scans no more force peaks appeared, indicating that protemporal resolution can be greatly improved by scanning only part teins were indeed pushed off by the application of high lateral forces.
of the DNA or by increasing the scanning speed. However, high
scanning speeds will result in larger loading rates acting on the DISCUSSION
enzyme-DNA bond, possibly activating dissociation. Moreover, the An alternative technique capable of manipulating multiple DNA
stretch correction fails where the DNA elasticity starts to deviate
molecules is to attach two DNA molecules between a surface and a
from the worm-like chain (WLC), at B50 pN total force. Finally, single paramagnetic bead manipulated by magnetic tweezers20.
the nature of the DNA itself sets an upper limit to the force that can Braids can be induced by rotating the magnetic bead. Unlike our
be applied because at 65 pN the DNA double helix starts to approach, this technique yields very limited control over the
relative orientation of the DNA molecules and lacks the ability to
overstretch, making it difficult to interpret the signal.
exert different forces on them.
An established scanning probe technique is atomic force
Detection efficiency
We determined the detection efficiency using Type IIP restriction microscopy (AFM)21. With AFM, deviations of a scanning tip are
enzymes that have well-known (and well-separated) locations of used to generate a topographic image. DNA can be visualized
their binding sites on l DNA. This allows to distinguish false posi- and associated proteins may be observed22 and manipulated23.
tives (probe sticking at locations where no protein is bound) from In contrast to AFM experiments, the technique introduced here is
performed far away from any (charged) surfaces that potentially
specifically bound proteins. We did not observe such events when
affect protein-DNA interactions24,25. Additionally, we exert and
scanning DNA in the absence of proteins. As restriction enzymes can
17
measure forces on the proteins in the direction of the DNA contour,
bind DNA in a nonspecific manner , we did not regard events in the
presence of proteins at nonspecific sites as false positives.
as is expected for a roadblock encounter in vivo.
On l DNA there are five recognition sites for EcoRI but we
Recently, the development of a technique combining a scanning
detected four specifically bound proteins (Fig. 3a). The fact that we
probe and optical tweezers was reported26. A micropipette is used
as a probe to scan along the contour of a DNA molecule suspended
neither detected a protein at this particular location in the second
between optical tweezers. The technique, however, has not yet been
and third scan (while the other four are consequently detected)
makes it unlikely that there is actually a protein bound at this shown to enable detection of proteins bound to DNA. As the
micropipette scans the DNA on one side only and does not enclose
location. When we repeatedly scanned other constructs we
it as in our case, it might miss bound proteins owing to the finite
observed similar results and therefore we argue that hardly any
bound proteins are missed using this method. Apparently, the torsional compliance of the DNA.
5
4 | ADVANCE ONLINE PUBLICATION | NATURE METHODS
10
© 2007 Nature Publishing Group http://www.nature.com/naturemethods
ARTICLES
Finally, a technique in which mechanical separation of the two
strands of the DNA double helix provides information about the
location of associated proteins and binding strengths has been
described27. Contrary to our capability to visualize individual
DNA-associated proteins repeatedly without affecting the binding,
unzipping the double helix leads to destruction of the protein
recognition site.
In conclusion, the realization of four moveable optical traps in
combination with the laminar flow system provides a powerful
means to study biological processes governed by proteins interacting with multiple DNA sites4,28–30. The involvement of multiple
DNA-binding domains complicates the analysis of their interactions with conventional single-molecule approaches. With our
dual-DNA manipulation technique, such interactions can now be
explored in detail6. The scanning probe technique introduced here
is indicative of the topological freedom this dual DNA assay offers.
METHODS
DNA and proteins. To allow specific binding to streptavidincoated beads (1.87 mm diameter, Spherotech Inc.), we biotinylated
l phage dsDNA (Roche) on both ends, as described previously31.
EcoRI and Sbf I were purchased from New England Biolabs and
used without further purification. We recovered EcoRV from
ammonium sulfate precipitates, dialyzed it into (10% (vol/vol)
glycerol, 20 mM Tris-HCl (pH 7.5), 1 M NaCl, 10 mM 2mercaptoethanol, 1 mM EDTA and filtered it through 0.2 mm
syringe filters. Aliquots (59 mM) were flash-frozen and stored at
–80 1C. We performed all protein scanning experiments in 10 mM
Tris-HCl (pH 7.5), 100 mM NaCl, 5 mM CaCl2 and 1 mM DTT,
with protein concentrations of B1–5 nM.
Experimental setup. We performed the experiments using a
custom-built inverted microscope (Fig. 1b). To generate the
optical traps we used a Nd:YVO4 laser (1,064 nm 10 W cw,
Millennia IR; Spectra Physics), isolated against back-reflections by
a Faraday isolator (IO-3-l-VHP; Optics For Research) and
expanded by a beam expander (2–8; Linos Photonics GmbH).
We split this laser beam into two beams by a polarizing beam
splitter cube (PBS-1064; CVI). In both beam paths, we implemented a 1:1 telescope system (f ¼ 150 mm) allowing beam
steering in the sample32. In one path (which we refer to as the
‘continuous’ path), the first telescope lens could be displaced
laterally using two computer-controlled actuators (T-LA28; Zaber
Technologies Inc.). In the other ‘time-shared’ path, we placed two
orthogonal acousto-optic deflectors (AODs, DTD 276HD6;
IntraAction) directly in front of the telescope. We then coupled
first-order deflected beam via a dichroic mirror (1,020 dichroic
longpass; Chroma Tech Corp.) into a 60 water-immersion
objective (Plan Apochromat 60, numerical aperture (NA) ¼
1.20; Nikon) to form the other laser traps.
A third computer-controlled actuator moved the first telescope
lens in the continuous path in the direction of the laser light,
thereby changing the depth of the laser focus with respect to the
traps from the time-shared path. Doing so, we could wind two
DNA molecules around one another.
For displacement detection of the continuous trap, we imaged
the intensity profile in the back focal plane of the condenser
(Achromat/Aplanat, NA ¼ 1.4; Nikon) onto a quadrant photodiode (SPOT-9DMI; UDT Sensors)12. We likewise measured
displacements in one of the AOD-generated traps using a separate
(nontrapping) detection laser (980 nm IQ2C140/6018; Power
Technology Inc.), overlaid on the trap and imaged onto a separate
quadrant photodiode. To generate multiple traps, the AODs were
driven by voltage-controlled oscillators (VCOs; DE-272H Deflector Driver; IntraAction) as radio frequency (RF) synthesizers (see
below). A bright-field image of the trapped beads, illuminated by a
blue LED (LXHL-NB98 Luxeon Star/O; LumiLeds) was imaged
onto a charge-coupled device (CCD) camera (902B; Watec).
Quadruple trap implementation. We used AODs to generate
three independent, time-shared traps by modulating the VCOs
that synthesize the RF signal with analog voltages generated by a
multifunction data acquisition printed circuit board (PCI-6221;
National Instruments). Using two orthogonally placed AODs,
traps can be generated and steered in both directions in the
sample plane. The voltage that is input to the VCO determines
the frequency deviation of the synthesized RF signals (typically
27 MHz) that drive the AODs. Therefore, VCO-input signals
consisting of repeated patterns with a number of voltage levels
yield a corresponding number of successively scanned, independent laser deviations (Supplementary Fig. 3 online).
Microfluidic flow cell. To enable swift exchange of buffers and to
have fine control over the process of catching the DNA molecules
between beads, we used a custom-built microfluidic flow chamber
(Figs. 1b,c). The central part of this flow chamber consists of four
channels cut manually out of a spacer (Parafilm) sandwiched
between a 24 60 mm #1 cover slip and a 50 75 1 mm
microscope slide. The slide contains 1-mm diameter holes that
connect to the input channels. By using a pattern with merging
channels, a region exists in the flow chamber where juxtaposed
buffers exhibit laminar flow. At the locus of the experiment in
the flow chamber the channels are well separated, facilitating
rigorous sub-second buffer exchange by simply moving the
microscope stage in a direction perpendicular to the flow. A
custom-made, sealed pressure chamber holds a reservoir for
different solutions. Contents of each channel are input to the
channels through polyetheretherketone (PEEK Upchurch
Scientific Inc.) tubing and can be altered using selection valves
(V-241; Upchurch Scientific Inc.). We controlled the flow speed
through adjustment of the air pressure in the pressure chamber,
thereby pushing the buffers through the flow chamber33. Fine
control of pressure was attained by using solenoid valves
(ES-2T-6; Clippard Europe S.A.) to in- or decrease the pressure,
while monitoring the pressure using a differential pressure meter
(CTE8005GY0; Sensortechnics GmbH). This approach yielded
smoother flow (transitions) than when using a stepper motor
syringe pump31. Typical working flow speeds were on the order of
100 mm/s, achieved at 50–100 mbar overpressure.
To diminish any effects of flow on the measured force, we
turned off the buffer flow during measurements. To ensure that the
solution in which the experiments are performed remains uncontaminated, the measurements were typically done in a channel
before it merged with the others in the laminar flow chamber.
Data acquision and analysis. For two traps, we recorded bead
displacements within the traps in x and y directions using
two quadrant photodiodes and a data acquisition board (AD16
NATURE METHODS | ADVANCE ONLINE PUBLICATION | 5
ARTICLES
module on a ChicoPlus analog-to-digital computer interface
board, maximum sampling rate 195 kHz; Innovative Integration)12,15. We calibrated voltages to forces using power spectrum
analysis34. For concurrent force and extension recordings of the
captured DNA molecules, we measured the distances between
pairs of beads on-line using pattern matching on a digitized
microscope image (IMAQ PCI-1409; National Instruments).
© 2007 Nature Publishing Group http://www.nature.com/naturemethods
Note: Supplementary information is available on the Nature Methods website.
ACKNOWLEDGMENTS
We thank E.J.G. Peterman and R.T. Dame for useful discussions. We acknowledge
D.A. Hiller (Yale) and J.J. Perona (University of California Santa Barbara) for the
kind gift of EcoRV. This work is part of the research program of the ’Stichting
voor Fundamenteel Onderzoek der Materie (FOM)’, which is financially supported
by the ’Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO)’ and was
supported by a NWO-Vernieuwingsimpuls grant.
Published online at http://www.nature.com/naturemethods/
Reprints and permissions information is available online at
http://npg.nature.com/reprintsandpermissions
1. West, S.C. Molecular views of recombination proteins and their control. Nat. Rev.
Mol. Cell Biol. 4, 435–445 (2003).
2. van den Bosch, M., Lohman, P.H. & Pastink, A. DNA double-strand break repair by
homologous recombination. Biol. Chem. 383, 873–892 (2002).
3. van Mameren, J. et al. Dissecting elastic heterogeneity along DNA molecules
coated partly with Rad51 using concurrent fluorescence microscopy and optical
tweezers. Biophys. J. 91, L78–L80 (2006).
4. van den Broek, B., Vanzi, F., Normanno, D., Pavone, F.S. & Wuite, G.J. Real-time
observation of DNA looping dynamics of Type IIE restriction enzymes NaeI and
NarI. Nucleic Acids Res. 34, 167–174 (2006).
5. Luijsterburg, M.S., Noom, M.C., Wuite, G.J. & Dame, R.T. The architectural role of
nucleoid-associated proteins in the organization of bacterial chromatin: A
molecular perspective. J. Struct. Biol. 156, 262–272 (2006).
6. Dame, R.T., Noom, M.C. & Wuite, G.J. Bacterial chromatin organization by
H-NS protein unravelled using dual DNA manipulation. Nature 444, 387–390
(2006).
7. Berg, O.G., Winter, R.B. & von Hippel, P.H. Diffusion-driven mechanisms of
protein translocation on nucleic acids. 1. Models and theory. Biochemistry 20,
6929–6948 (1981).
8. Gowers, D.M. & Halford, S.E. Protein motion from non-specific to specific DNA
by three-dimensional routes aided by supercoiling. EMBO J. 22, 1410–1418
(2003).
9. von Hippel, P.H. & Berg, O.G. Facilitated target location in biological systems.
J. Biol. Chem. 264, 675–678 (1989).
10. Epshtein, V., Toulme, F., Rahmouni, A.R., Borukhov, S. & Nudler, E. Transcription
through the roadblocks: the role of RNA polymerase cooperation. EMBO J. 22,
4719–4727 (2003).
11. Allersma, M.W., Gittes, F., deCastro, M.J., Stewart, R.J. & Schmidt, C.F.
Two-dimensional tracking of ncd motility by back focal plane interferometry.
Biophys. J. 74, 1074–1085 (1998).
6 | ADVANCE ONLINE PUBLICATION | NATURE METHODS
12. Gittes, F. & Schmidt, C.F. Interference model for back-focal-plane displacement
detection in optical tweezers. Opt. Lett. 23, 7–9 (1998).
13. Brewer, L.R., Corzett, M. & Balhorn, R. Protamine-induced condensation and
decondensation of the same DNA molecule. Science 286, 120–123 (1999).
14. Arai, Y. et al. Tying a molecular knot with optical tweezers. Nature 399, 446–448
(1999).
15. Vermeulen, K.C. et al. Calibrating bead displacements in optical tweezers using
acousto-optic deflectors. Rev. Sci. Instrum. 77, 013704 (2006).
16. Gosule, L.C. & Schellman, J.A. Compact form of DNA induced by spermidine.
Nature 259, 333–335 (1976).
17. Shimamoto, N. One-dimensional diffusion of proteins along DNA - Its biological
and chemical significance revealed by single-molecule measurements. J. Biol.
Chem. 274, 15293–15296 (1999).
18. Trabesinger, W., Schutz, G.J., Gruber, H.J., Schindler, H. & Schmidt, T. Detection
of individual oligonucleotide pairing by single-molecule microscopy. Anal. Chem.
71, 279–283 (1999).
19. Hiller, D.A. et al. Simultaneous DNA binding and bending by EcoRV endonuclease observed by real-time fluorescence. Biochemistry 42, 14375–14385
(2003).
20. Charvin, G., Vologodskii, A., Bensimon, D. & Croquette, V. Braiding DNA:
experiments, simulations, and models. Biophys. J. 88, 4124–4136
(2005).
21. Binnig, G., Quate, C.F. & Gerber, C. Atomic force microscope. Phys. Rev. Lett. 56,
930–933 (1986).
22. Lindsay, S.M. et al. STM and AFM images of nucleosome DNA under water.
J. Biomol. Struct. Dyn. 7, 279–287 (1989).
23. Fritzsche, W., Takac, L. & Henderson, E. Application of atomic force microscopy to
visualization of DNA, chromatin, and chromosomes. Crit. Rev. Eukaryot. Gene Expr.
7, 231–240 (1997).
24. Kasas, S. et al. Escherichia coli RNA polymerase activity observed using atomic
force microscopy. Biochemistry 36, 461–468 (1997).
25. van Noort, S.J. et al. Direct visualization of dynamic protein-DNA interactions
with a dedicated atomic force microscope. Biophys. J. 74, 2840–2849 (1998).
26. Huisstede, J.H.G., van der Werf, K.O., Bennink, M.L. & Subramaniam, V. Force
detection in optical tweezers using backscattered light. Opt. Express 13,
1113–1123 (2005).
27. Koch, S.J., Shundrovsky, A., Jantzen, B.C. & Wang, M.D. Probing protein-DNA
interactions by unzipping a single DNA double helix. Biophys. J. 83, 1098–1105
(2002).
28. Kowalczykowski, S.C., Dixon, D.A., Eggleston, A.K., Lauder, S.D. & Rehrauer, W.M.
Biochemistry of homologous recombination in Escherichia coli. Microbiol. Rev. 58,
401–465 (1994).
29. Dame, R.T., Wyman, C. & Goosen, N. Structural basis for preferential binding of
H-NS to curved DNA. Biochimie 83, 231–234 (2001).
30. Halford, S.E., Welsh, A.J. & Szczelkun, M.D. Enzyme-mediated DNA looping. Annu.
Rev. Biophys. Biomol. Struct. 33, 1–24 (2004).
31. van den Broek, B., Noom, M.C. & Wuite, G.J.L. DNA-tension dependence of
restriction enzyme activity reveals mechanochemical properties of the reaction
pathway. Nucleic Acids Res. 33, 2676–2684 (2005).
32. Svoboda, K. & Block, S.M. Biological applications of optical forces. Annu. Rev.
Biophys. Biomol. Struct. 23, 247–285 (1994).
33. Wuite, G.J.L., Davenport, R.J., Rappaport, A. & Bustamante, C. An integrated
laser trap/flow control video microscope for the study of single biomolecules.
Biophys. J. 79, 1155–1167 (2000).
34. Gittes, F. & Schmidt, C.F. Signals and noise in micromechanical measurements.
Methods Cell Biol. 55, 129–156 (1998).