Home
Search
Collections
Journals
About
Contact us
My IOPscience
Nanoengineering a single-molecule mechanical switch using DNA self-assembly
This article has been downloaded from IOPscience. Please scroll down to see the full text article.
2011 Nanotechnology 22 494005
(http://iopscience.iop.org/0957-4484/22/49/494005)
View the table of contents for this issue, or go to the journal homepage for more
Download details:
IP Address: 67.186.132.137
The article was downloaded on 15/01/2012 at 04:29
Please note that terms and conditions apply.
IOP PUBLISHING
NANOTECHNOLOGY
Nanotechnology 22 (2011) 494005 (8pp)
doi:10.1088/0957-4484/22/49/494005
Nanoengineering a single-molecule
mechanical switch using DNA
self-assembly
Ken Halvorsen1,2,3 , Diane Schaak1 and Wesley P Wong1,2,3
1
The Rowland Institute at Harvard, Harvard University, Cambridge, MA, USA
The Immune Disease Institute/Program in Cellular and Molecular Medicine, Children’s
Hospital Boston, Boston, MA, USA
3
The Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical
School, Boston, MA, USA
2
E-mail: [email protected]
Received 9 June 2011, in final form 25 July 2011
Published 21 November 2011
Online at stacks.iop.org/Nano/22/494005
Abstract
The ability to manipulate and observe single biological molecules has led to both fundamental
scientific discoveries and new methods in nanoscale engineering. A common challenge in many
single-molecule experiments is reliably linking molecules to surfaces, and identifying their
interactions. We have met this challenge by nanoengineering a novel DNA-based linker that
behaves as a force-activated switch, providing a molecular signature that can eliminate errant
data arising from non-specific and multiple interactions. By integrating a receptor and ligand
into a single piece of DNA using DNA self-assembly, a single tether can be positively identified
by force–extension behavior, and receptor–ligand unbinding easily identified by a sudden
increase in tether length. Additionally, under proper conditions the exact same pair of molecules
can be repeatedly bound and unbound. Our approach is simple, versatile and modular, and can
be easily implemented using standard commercial reagents and laboratory equipment. In
addition to improving the reliability and accuracy of force measurements, this single-molecule
mechanical switch paves the way for high-throughput serial measurements, single-molecule
on-rate studies, and investigations of population heterogeneity.
S Online supplementary data available from stacks.iop.org/Nano/22/494005/mmedia
(Some figures in this article are in colour only in the electronic version)
1. Introduction
activity [6]. In addition, these measurements have yielded
fundamental insights into the dynamical strength of molecular
interactions [7], which have led to the development of creative
new tools for nanoscale assembly [8].
Mechanical forces can be applied to individual molecules
using a broad range of tools, including optical traps, magnetic
tweezers, mechanical cantilevers, and the centrifuge force
microscope [9, 10] (figure 1). Yet a common requirement
of these methods is that single-molecule constructs must be
specifically tethered between two surfaces (e.g. beads, cover
slips or cantilevers) to enable their manipulation and detection.
This leads to one of the major challenges in single-molecule
experimentation—verifying that exactly one molecular tether
is being pulled, and distinguishing this tether from non-specific
The ability to precisely manipulate individual molecules
has led to stunning new discoveries in physics, biology,
and medicine [1, 2], as well as powerful new methods
in nanoscale engineering. For example, single-molecule
force measurements have revealed the basic mechanical
properties of nucleic acids [3], the dynamics and functioning
of molecular motors [4, 5], and the role of hydrodynamic
forces in the circulatory system in regulating enzymatic
Content from this work may be used under the terms of the
Creative Commons Attribution-NonCommercial-ShareAlike
3.0 licence. Any further distribution of this work must maintain attribution to
the author(s) and the title of the work, journal citation and DOI.
0957-4484/11/494005+08$33.00
1
© 2011 IOP Publishing Ltd Printed in the UK & the USA
Nanotechnology 22 (2011) 494005
(a)
K Halvorsen et al
Force
(c)
(a)
(b)
(d)
(b)
Figure 1. Single-molecule manipulation techniques include those of
(a) optical tweezers, (b) magnetic tweezers, (c) the atomic force
microscope, and (d) the centrifuge force microscope.
(c)
and unintended interactions that may occur (e.g. surface–
surface interactions, formation of multiple bonds). The
success and reliability of single-molecule experiments depends
upon the creation of reliable, verifiable and robust linking
techniques. This is particularly important for bond rupture
studies (e.g. characterizing the strength of molecular adhesion
bonds [11], DNA base pairing [12], and cell adhesion and
signaling [13]), as the dissociation between two molecules can
be difficult to positively identify due to the lack of an obvious
mechanical signature.
We have met this experimental challenge with a new
single-molecule attachment technique that facilitates reliable
and accurate single-molecule force measurements. Using DNA
self-assembly techniques, we have nanoengineered a unique
linker that behaves as a force-activated single-molecule switch.
This switch changes conformation under force to signify
bond rupture, providing an identifiable molecular signature
that eliminates the possibility of accidentally measuring nonspecific, multiple and unknown interactions. Furthermore,
this construct enables the same pair of interacting molecules
to be brought back together following rupture, opening the
way to high-throughput serial measurements, single-molecule
on-rate studies, and studies of population heterogeneity.
Our approach is simple, versatile and modular, and can
be easily implemented using standard commercial reagents
and laboratory equipment. We elaborate on this approach
in section 1.1, and provide an overview of standard linker
geometries for single-molecule manipulation experiments.
Figure 2. Single-molecule linking geometries: a cartoon showing
receptor–ligand unbinding using a variety of linking strategies
including (a) direct contact/a short linker, (b) a long linker, and (c) a
looped linker.
materials (e.g. DNA, PEG) [14–18]. Most of these approaches
can be classified into a few basic categories based on their
geometry (figure 2). In this subsection, we review these
categories, and show how appropriately designed linkages
can improve the efficiency and reliability of single-molecule
rupture measurements.
(a) Direct contact/short linkers. First, we consider the
simplest strategy of direct molecular coating onto a surface
(figure 2(a)), which is typically accomplished via adsorption
or conjugate chemistry with short linkers (e.g. SMCC, EDAC).
This approach has been commonly used, particularly in early
single-molecule experiments [19, 9]. Some advantages of this
approach are that it is typically simple, fast and inexpensive,
particularly for the case of physical adsorption of molecules
and cells to a surface. However, the first major problem
that arises from this geometry is the introduction of unwanted
interactions with the surfaces. Since the sample molecules
are typically located within nanometers of the surfaces, nonspecific surface–surface and molecule–surface interactions
(e.g. van der Waals and depletion forces) may significantly
affect the measured force [20]. Furthermore, any impurities
or contamination in the sample may result in unwanted and
unidentifiable adhesion between the surfaces. To deal with
these difficulties, careful characterization and minimization of
1.1. Overview of linker geometries in single-molecule
experiments
A variety of different linker strategies have been employed in
single-molecule experiments. These include different chemical
attachment strategies (e.g. SMCC, click chemistry, EDC, etc),
surface preparations (e.g. silanization methods), and linkage
2
Nanotechnology 22 (2011) 494005
K Halvorsen et al
non-specific interactions are necessary. Reduction of nonspecific adhesion can be accomplished by blocking agents
such as BSA, casein, Pluronics F127, and Tween, but they
are often difficult to eliminate completely. For bond rupture
measurements, the second difficulty with this geometry is
the lack of a specific mechanical signature for the molecule
of interest. Thus, detachment events arising from both
multiple attachments and non-specific attachments are difficult
to distinguish from the single-molecule interaction of interest.
Minimization of multiple attachments can be accomplished
by diluting the active molecules on the surfaces, and by
controlling the distance, touch time and touch force between
the functionalized surfaces during the experiment. While
careful control experiments and the use of statistical techniques
to estimate the number of single bonds can help to mitigate
difficulties arising from direct contact geometry [21], we will
see that the use of longer linkers can improve the reliability,
accuracy, and ease of analysis of the resulting single-molecule
data.
(b) Long linkers. A variety of linking strategies have been
developed for keeping the molecules of interest far from the
surfaces to which they are attached (figure 2(b)). This can
be accomplished using long polymer linkers such as PEG [17]
or DNA [16], or with more exotic materials such as the M13
filamentous bacteriophage [22]. Long linkers can serve the
double purpose of both eliminating close range non-specific
interactions and facilitating the positive identification of each
tether as a single molecule; the required linker length depends
upon the resolution of the force-probe instrument and the
compliance of the molecule of interest. When used on one
side, a sufficiently long linker can already eliminate surface–
surface non-specific interaction, as well as allowing positive
identification of a single interaction by using the known force–
extension behavior of the tether. When used on both sides as
depicted in figure 2(b), the linkers can additionally eliminate
non-specific surface–molecule interactions. We note that long
linkers between the force probe and the molecule of interest
may complicate the force–loading history and effective energy
landscape, but this can generally be accounted for [23, 15].
(c) Looped linkers.
More recently, single-molecule
experiments have been performed using a looped linker
geometry, in which the interacting molecules/sites of interest
are connected to each other by a molecular tether [24, 25].
This looped linker geometry as shown in figure 2(c) has two
primary advantages. First, it provides an additional signature
for bond rupture events, as this molecular transition will
create a well-defined increase in the tether length. Unlike for
the non-looped geometries, unbinding of the receptor–ligand
pair cannot be confused with accidental unbinding from the
anchoring ends. Second, the looped linker allows for repeated
testing of the exact same pair of molecules, provided the
conditions for reforming the bond are favorable. This opens
up new possibilities for studying population heterogeneity in
a population of molecules—a highly useful, but difficult to
realize, benefit of single-molecule experiments. This looped
linker concept was recently introduced as ReaLiSM (receptor
and ligand in a single molecule) [24], where the A1 domain of
the von Willebrand factor was linked by an amino acid chain
Figure 3. Looped linker construction using DNA origami: circular
single-stranded DNA is enzymatically cleaved at a single site and
mixed with over 100 oligonucleotides to self-assemble into a looped
linker with functional groups. Supplemental figure 2 (available at
stacks.iop.org/Nano/22/494005/mmedia) shows a similar construct
looped with a single ‘bridge’ oligo.
to platelet GPIbα , and the two interacting protein domains
were repeatedly bound and unbound. While highly effective,
the ReaLiSM construct as presented may be difficult for many
labs to create, requiring expertise in protein engineering and
purification.
Here we present a versatile DNA-based alternative
constructed using DNA origami methods [26, 27]. By mixing
a long piece of single-stranded DNA with a carefully designed
soup of short DNA oligomers, we have constructed looped
single-molecule linkers with an integrated receptor–ligand pair
via DNA self-assembly (figure 3).
2. Materials and methods
2.1. Linker design and construction
Two different kinds of linkers were designed and assembled
based on techniques outlined from previous DNA origami
work [26, 27]. Both linkers incorporate functional ‘sticky’ ends
(we used double biotin on both ends) which act as anchors for
single-molecule experiments, as well as two functional sites
near the middle of the linker to form the loop. One set of
constructs forms the loop by hybridizing a single ‘bridge’ oligo
across two distinct locations, while the other set used two
separate oligos each functionalized with digoxigenin or antidigoxigenin, which can bind to each other as a receptor–ligand
pair to form the loop.
To make these linkers, M13mp18 single-stranded DNA
(New England Biolabs) was first linearized by hybridizing
a 40-nucleotide oligo to form a double-stranded region and
then cleaving this region with BtsCI restriction enzyme (New
3
Nanotechnology 22 (2011) 494005
K Halvorsen et al
England Biolabs). The linearized single-stranded DNA was
then mixed with complimentary oligos (Bioneer, Inc.) and
subjected to a temperature ramp from 90 to 20 ◦ C with a
1 ◦ C min−1 ramp in a PCR machine (Bio-Rad) to allow the
oligos to anneal properly. For the linker with the bridge
oligo formed loop, 121 oligos excluding the bridge oligo
were added in tenfold molar excess, with the bridge oligo
added in equimolar concentration to the scaffold strand. For
the receptor–ligand loop construct, 120 oligos excluding the
antibody oligo were in tenfold molar excess, which was added
in equimolar concentration and subjected to a temperature
ramp from 40 to 10 ◦ C with a 0.5 ◦ C min−1 ramp after the other
120 were linked.
Detailed protocols and the full sequences of all oligos are
available in the supplemental information (available at stacks.
iop.org/Nano/22/494005/mmedia). All oligos were purchased
from Bioneer, Inc., with the exception of the 5# double-biotin
oligo (Integrated DNA Technologies), the digoxigenin oligos
(Integrated DNA Technologies), and a few plain oligos ordered
with next-day service (Invitrogen).
the blur-corrected power spectrum fit [29], with additional
calibration information provided by the dsDNA overstretching
transition [30].
Single-molecule force measurements are performed by
bringing linker functionalized beads held in the optical
trap into contact with streptavidin-coated beads held in the
micropipette to form molecular tethers. Tension in each
tether is applied by moving the bead in the micropipette, and
quantified by measuring the displacement of the bead in the
optical trap. The observed distance between the beads gives a
measure of the tether length.
3. Results and discussion
We successfully created looped single-molecule linkers via
DNA self-assembly. Two different kinds of linker constructs
were generated and tested: (i) linkers looped by a short
complementary strand of DNA to study the kinetics of DNA
base pairing, and (ii) linkers looped by a receptor–ligand
pair to study protein–protein interactions. As detailed below,
the proper assembly and functionality of these linkers was
verified using gel-shift assays and optical trap measurements.
We demonstrated their effectiveness for single-molecule force
spectroscopy by measuring the kinetics of bond rupture for
both DNA hybridization and an antibody–antigen interactions,
and by showing how the molecular signature of a looped tether
can be used to improve the accuracy of the data.
2.2. DNA–protein conjugation
A 3# thiol modified oligo was reduced and linked to monoclonal
and polyclonal anti-digoxigenin (Roche Applied Science)
using sulfo-SMCC (Pierce) and the accompanying protocol.
The NHS group on the SMCC was first linked to free amines
on the antibody (at 1 mg ml−1 ) with a 30 min reaction at room
temperature using 20-fold molar excess of SMCC in PBS at
pH 7.4. At the same time, the thiol oligo was deprotected and
reduced by incubating in 50 mM TCEP (Pierce) for 20 min and
then cleaned using a PCR clean up kit (Qiagen). Following the
first SMCC reaction, excess SMCC was removed with a Zeba
desalting column (Pierce), pre-equilibrated with PBS buffer.
The activated protein was then mixed with the reduced thiol
oligo in a 1:1 molar ratio for 30 min at RT.
Conjugation was verified by visualization on a 4–20%
polyacrylamide gel (Bio-Rad) run in 1× TBE buffer at 150 V
for 40 min, where a shift from the protein linkage was readily
apparent (see supplemental figure 3 available at stacks.iop.org/
Nano/22/494005/mmedia). Typically, 5–50% of the oligos
were conjugated to protein, and purification of the protein–
DNA conjugate was accomplished by excising the gel band and
using an electro-elution kit with the accompanying protocol
(Gerard biotech).
3.1. Verification of the linker assembly
We first tested the linkers looped by a single DNA oligo
bridge, as they served as a good model system for testing and
optimizing linker assembly, independent of protein-coupling
efficiency. For these oligo bridge constructs, we made two
different loop lengths: 2580 base pairs and 600 base pairs.
Additionally, we varied the length of the bridge oligo on one
side to be 30 bp, 20 bp, 15 bp, and 10 bp, while the other side
was maintained at 30 bp. The formations of both long and
short loops were easily distinguishable from those of unlooped
products by a gel shift due to slower migration on a 0.7%
agarose gel (figure 4), for both the 30 bp and 20 bp bridge
constructs. We were unable to observe any looped construct in
the gel when using the 15 bp or 10 bp bridge oligo, presumably
due to the harsh conditions of electrophoresis (e.g. low salt,
high temperature, high voltage). Confirmation that the shifted
gel band was indeed the looped construct was accomplished
by cutting the construct with a single-cut enzyme in the loop
region (figure 4)—the shifted band (looped DNA) was largely
unaffected by the enzyme, while the lower band (unlooped
DNA) was completely digested into two separate pieces.
Next, these products (with double-biotin ends) were
verified directly in the optical trap by pulling them end
to end with linear ramps of force (figure 4). Unlooped
linkers show characteristic DNA force–extension behavior
with typical contour lengths of 2000–2300 nm, consistent with
the number of DNA bases within the construct. The looped
linkers initially start with a shorter contour length, then exhibit
a sudden increase to this full contour length when the DNA
2.3. Single-molecule force spectroscopy
The final unpurified linkers with double-biotin ends were
incubated with streptavidin polystyrene beads (Corpuscular)
for 15 min, and then injected into a chamber with PBS
buffer for use in the optical trap. The optical trap setup
consists of a single stationary trap and a piezo-controlled
micropipette integrated into an inverted light microscope
(Nikon). The setup is functionally identical to previously
described instruments [6, 28], but with 160× overall
magnification instead of 400×. High-speed video microscopy
is used to measure bead positions in 1D with a resolution
of ∼4 nm at ∼2 kHz. The optical trap is calibrated using
4
Nanotechnology 22 (2011) 494005
K Halvorsen et al
its antibody were assembled to form linkers with a loop
length of 600 base pairs. The verification of these constructs
was conducted in the same way as for the DNA bridge
looped linkers, using both gel electrophoresis shift assays
and single-molecule pulling experiments. We note that while
the polyclonal looped constructs could be readily seen in a
gel as a distinct shifted band (identical to the DNA bridge
looped construct in figure 4), the monoclonal constructs could
not. This is likely due to the much lower affinity between
digoxigenin and its monoclonal antibody, and is consistent
with the lack of a band for the 10 bp and 15 bp DNA bridge
constructs. However, both the monoclonal and polyclonal
constructs could be observed in the optical trap, and exhibited
force–extension curves that matched those of the 600 bp oligo
bridge construct.
3.2. Demonstration of single-molecule force spectroscopy
We tested the dynamic strength of DNA hybridization with
the optical trap by repeatedly applying linear force ramps to
the DNA bridge constructs to determine the distribution of the
rupture force. The molecular signature of the looped linker
serves as a powerful filtering method to distinguish the rupture
of the DNA bridge from non-specific, unknown and multiple
interactions. This is illustrated in figure 5 (left) for the rupture
force of the 20 bp bridge, where positive identification of the
correct rupture transition (using the change in tether length,
overall tether length, and overstretching of the linker) enabled
the removal of erroneous data that accounted for 57% (34/60)
of the measured events. In the resulting data, we measured
a mean rupture force of 52 pN with a standard deviation of
6 pN at a nominal loading rate of 100 pN nm−1 (this was
a combination of experiments with a mean loading rate of
98 pN nm−1 and a standard deviation of 35 pN nm−1 ). This
agrees within error with the expected force of 39 ± 15 pN for
the mechanical shearing of DNA (based upon their empirical
formula) [12].
When testing the other DNA bridge lengths, we observed
fewer rupture events for the 30 bp bridge, as the biotin–
streptavidin bonds anchoring the linker would often rupture
first. In addition, we also saw evidence of the 15 bp and
10 bp bridges in single-molecule pulling experiments, despite
not seeing these constructs with the gel-shift assay. While
the formation of these loops should be energetically favorable
even with the additional entropic cost of closing the loop [32],
it is possible that the conditions of electrophoresis lower the
stability of these constructs leading to their absence in the gel
assays.
As another demonstration, we measured the forcedependent unbinding kinetics of digoxigenin with its antibody
in the optical trap (figure 5 (right)). By making repeated
measurements of bond rupture under a constant force, we
found a characteristic lifetime of 1.3 s (with a 95% confidence
band of 0.9–2.0 s) at a force of 49 ± 2 pN for the polyclonal
antibody, using maximum likelihood estimation with an
exponential decay model. We found this interaction to be
relatively strong, in agreement with other single-molecule
measurements that used it as a molecular anchor [22]. We
Figure 4. Verification of DNA bridge looped linker. Top: gel
electrophoresis on a 0.7% agarose gel at 5 V cm−1 for 2 h, stained
with Sybr gold. Bottom: single-molecule pulling trajectories
demonstrating the sudden increase in tether length that signifies bond
rupture events. The inset shows a histogram of the change in tether
length for the 2580 bp (red) and 600 bp (black) looped linkers.
bridge ruptures under the application of mechanical stress. We
measured an average increase in contour length of 884 nm and
208 nm for the long and short loops, respectively (figure 4
inset), which is within a few nanometers of the expected length
changes of 877 and 204 nm predicted from the worm-like
chain polymer model using a contour length of 0.34 nm per
base pair [31, 3]. As can be seen in figure 4, after bridge
rupture both curves roughly follow the curve for the unlooped
linker. As the linker was stressed above 65 pN, the DNA
overstretching transition could be observed, which served
as an additional mechanical signature for identifying singlemolecule tethers. However, pulling the molecule through this
transition always resulted in detachment of the linker from
the functionalized beads, presumably due to the force-induced
melting of the biotinylated anchor oligos off of the ssDNA
scaffold. While this effectively limits the use of this linker
to measurements below about 65 pN, this could likely be
overcome by covalently cross-linking the DNA linker or by
using much longer anchoring oligos. Regarding the observed
length of the linkers, we note that a distribution of lengths
is expected from multiple-tether measurements, even if every
tether is identical on a molecular level. Since the bead in
the pipette is rotationally constrained, tethers may be held at
different angles, causing the measured distance between the
beads to differ from the molecular tether length.
Receptor–ligand looped linkers were also created in order
to measure the force-dependent kinetics of an antibody–
antigen interaction. Oligos coupled to digoxigenin and to
5
Nanotechnology 22 (2011) 494005
K Halvorsen et al
Figure 5. Single-molecule force spectroscopy results for the rupture of (left) DNA hybridization and (right) antibody–antigen interactions.
Left: rupture force histogram for shearing a 20 bp DNA segment, demonstrating the filtering of erroneous data via the looped linker molecular
signature. Right: survival trajectory for digoxigenin against its antibody under constant force, with results of maximum likelihood estimation
superimposed.
Figure 6. Trajectory demonstrating repeated rupture and formation of a single receptor–ligand pair (digoxigenin with its monoclonal
antibody): (left) force versus time and (right) force versus extension traces for repeated cycles of force application and release. Bond rupture
events are observable by a sudden drop in force and an increase in tether length, as demarcated by red arrows. Rebinding/bond formation
during a low force clamp can be observed by subsequent bond rupture under the application of force.
note that without the looped linker the high bond strength of
this interaction can make rupture measurements difficult, as
it can be difficult to distinguish the rupture of digoxigenin–
antibody from the failure of molecular anchors in the absence
of an additional molecular signature.
Not only were we able to measure bond rupture, but
single-molecule bond formation could also be observed. In
many cases with the digoxigenin–antibody construct, we were
able to reform the complex after dissociation by bringing the
beads closer together and waiting for a short time (figure 6).
However, we were not able to reform the oligo bridge after
rupture under similar conditions, suggesting that the formation
of secondary structure or the extra time to diffusively align the
two strands slowed the rebinding kinetics.
increase the accuracy and reliability of single-molecule force
measurements. The method is versatile enough to be useful
for a wide range of molecular interactions, and simple enough
to be made by researchers of diverse backgrounds without
significant investment of time or money (see the appendix).
We demonstrated this functionality by constructing and testing
two different looped linkers, designed for studying the dynamic
strength of DNA base pairing and receptor–ligand interactions.
In addition, we showed how the molecular signature provided
by this ‘DNA mechanical switch’ enables the removal of
erroneous data that can arise from non-specific, unknown, and
multiple interactions. Not only is this construct useful for
traditional bond rupture measurements and force spectroscopy,
but it also enables the same pair of interacting molecules to
be brought back together following rupture, opening the way
toward high-throughput serial measurements, single-molecule
on-rate studies, and studies of population heterogeneity.
4. Conclusion
We have presented a simple and effective method for making
functional looped linkers using DNA self-assembly, which can
6
Nanotechnology 22 (2011) 494005
K Halvorsen et al
Acknowledgments
To generate linkers of different lengths, alternative ssDNA
scaffolds could be used. For example, ssDNA can be
generated using asymmetric PCR, or obtained by strandseparating standard PCR products. Product from rolling-circle
amplification could also be used for the generation of periodic
linkers. An alternative method for constructing looped linkers
that we have been exploring is ligating together the products of
multiple PCR reactions.
Lastly, we will discuss the time, equipment, and expertise
requirements for making these linkers, as these were primary
considerations for us in starting this project. By following the
protocols in the methods section (described in detail in the
supplemental information available at stacks.iop.org/Nano/22/
494005/mmedia), the linkers can be completely constructed
and verified in a single working day, and this time could
probably be reduced with further optimization. The protein
conjugation procedure takes about 2–3 h, with an additional
2 h for purification. The oligo annealing process (excluding
protein-conjugated oligos) takes 70 min, and can be done
in parallel with the protein conjugation step. Annealing
the protein-conjugated oligo(s) takes another 1 h, and final
verification of the looped linker on a gel takes an additional
1–2 h. Overall, it can be done in 6–8 h from start to finish.
The equipment requirements are minimal, and the protocols
can be carried out with only modest pipetting skills. The only
major equipment needs are for a centrifuge, and horizontal
and vertical electrophoresis systems (the vertical one was used
for protein–oligo purification, but we have seen gel shifts in
agarose as well). We used a PCR machine for the annealing
protocol, but this is not strictly necessary and could be replaced
with a hot water bath cooling slowly to room temperature. All
reactants and kits used are commercially available.
These linkers are intended to be versatile for studying a
variety of molecules, and in fact much of the protocols will
remain the same even as specific molecules on the linker are
changed. The modularity of the linker makes it simple to use
different proteins by just swapping one functionalized oligo
for another. If the linker is constructed with only the proteinconjugated oligos omitted, this stock solution of partially
constructed linker can then be mixed with the specific protein
oligos to get the final construct. The only additional step
required will be to repeat the chemical conjugation of oligo
to protein for each specific case. In our example linkers, we
used SMCC, but there are a variety of other bioconjugation
techniques that could be used instead. In the case of antibodies,
it is also possible to use antibody binding proteins such as
protein G to simplify the process of swapping in different
antibodies.
The authors would like to thank Teresa Kao and Daniel Cheng
for laboratory support, and William Shih and Ted Feldman for
helpful discussions. This work was supported by the Rowland
Junior Fellows program.
Appendix. Practical considerations for the
construction of looped linkers
The goal of this work was to develop a looped single-molecule
linker, with an easily identifiable molecular ‘signature’, that is
both versatile and easy to implement. While the DNA origami
construction of the linker may sound intimidating due to the
sheer quantity of oligos used, it is rather simple in practice.
We recommend dividing the ssDNA scaffold into fixed and
variable regions. The variable regions are those that may be
changed in the future to have different labels, while the fixed
regions will only have unlabeled oligos. Then all of the ‘fixed’
oligos can be ordered pre-mixed to save time and eliminate
the possibility of mixing error (Bioneer was accommodating in
this regard). In our case, we designated 12 variable oligos each
spaced apart by 9–10 fixed oligos, and ordered the 109 fixed
oligos pre-mixed in a single tube. The variable oligos gave us
the flexibility to choose different oligo labels and to choose
different loop lengths for our construct, without significant
protocol modification.
The choice of oligo length should also be considered.
We chose 60 nucleotide oligos to minimize the number of
oligos needed (companies vary in maximum oligo size), but in
hindsight the pre-mixing of the fixed oligos makes the number
of oligos less important. The main differences to consider
with shorter oligos will be differences in annealing behavior,
requisite protocol modifications, and price.
Regarding
annealing, longer oligos have higher melting temperatures,
which should increase the annealing speed in a decreasing
temperature ramp. On the other hand, longer oligos tend
to have more secondary structure which can sometimes
impede proper annealing, especially at lower temperatures.
Regarding protocol modification, oligos that are 30 bp or
shorter can be washed out with PCR clean up kits, which would
necessitate the use of a different cleaning method in our SMCC
conjugation. Regarding price, the yield of shorter oligos can be
significantly higher, making them more cost effective.
It is worth considering cost in more detail, as it could
be a deciding factor in the use of these linkers. The
cost of oligos continues to decrease, and many companies
offer discounts for large oligo orders or those ordered in
plates instead of individual tubes. As an example, our
construct cost under $1000 with unmodified oligos. Modified
oligos will increase the cost depending on the modification,
from tens to hundreds of dollars per oligo. However, this
initial investment can produce several liters of product at
the nanomolar concentrations that are typically required for
single-molecule experiments. Surprisingly, the most expensive
ingredient on a molar basis when using short unpurified and
unmodified oligos is the M13 scaffold [26].
References
[1] Deniz A A, Mukhopadhyay S and Lemke E A 2008
Single-molecule biophysics: at the interface of biology,
physics and chemistry J. R. Soc. Interface 5 15
[2] Ritort F 2006 Single-molecule experiments in biological
physics: methods and applications J. Phys.: Condens.
Matter 18 R531
[3] Bustamante C, Bryant Z and Smith S B 2003 Ten years of
tension: single-molecule DNA mechanics Nature 421 423–7
7
Nanotechnology 22 (2011) 494005
K Halvorsen et al
[4] Svoboda K, Schmidt C F, Schnapp B J and Block S M 1993
Direct observation of kinesin stepping by optical trapping
interferometry Nature 365 721–7
[5] Greenleaf W J, Woodside M T and Block S M 2007
High-resolution, single-molecule measurements of
biomolecular motion Annu. Rev. Biophys. Biomol. Struct.
36 171
[6] Zhang X, Halvorsen K, Zhang C Z, Wong W P and
Springer T A 2009 Mechanoenzymatic cleavage of the
ultralarge vascular protein von Willebrand factor Science
324 1330
[7] Evans E 2001 Probing the relation between force–lifetime–and
chemistry in single molecular bonds Annu. Rev. Biophys.
Biomol. Struct. 30 105–28
[8] Kufer S K, Puchner E M, Gumpp H, Liedl T and Gaub H E
2008 Single-molecule cut-and-paste surface assembly
Science 319 594
[9] Neuman K C and Nagy A 2008 Single-molecule force
spectroscopy: optical tweezers, magnetic tweezers and
atomic force microscopy Nature Methods 5 491–505
[10] Halvorsen K and Wong W P 2010 Massively parallel
single-molecule manipulation using centrifugal force
Biophys. J. 98 L53–5
[11] Evans E and Ritchie K 1997 Dynamic strength of molecular
adhesion bonds Biophys. J. 72 1541–55
[12] Strunz T, Oroszlan K, Schäfer R and Güntherodt H J 1999
Dynamic force spectroscopy of single DNA molecules Proc.
Natl Acad. Sci. 96 11277
[13] Evans E A and Calderwood D A 2007 Forces and bond
dynamics in cell adhesion Science 316 1148
[14] Hermanson G T 2008 Bioconjugate Techniques 2nd edn
(New York: Academic)
[15] Friedsam C, Wehle A K, Kühner F and Gaub H E 2003
Dynamic single-molecule force spectroscopy: bond rupture
analysis with variable spacer length J. Phys.: Condens.
Matter 15 S1709
[16] Hegner M 2000 DNA handles for single molecule experiments
Single Molecules 1 139–44
[17] Hinterdorfer P, Gruber H J, Kienberger F, Kada G, Riener C,
Borken C and Schindler H 2002 Surface attachment of
ligands and receptors for molecular recognition force
microscopy Colloids Surf. B 23 115–23
[18] Perret E, Leung A, Morel A, Feracci H and Nassoy P 2002
Versatile decoration of glass surfaces to probe individual
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
8
protein–protein interactions and cellular adhesion Langmuir
18 846–54
Lee G U, Kidwell D A and Colton R J 1994 Sensing discrete
streptavidin–biotin interactions with atomic force
microscopy Langmuir 10 354–7
Israelachvili J N 1992 Intermolecular and Surface Forces: With
Applications to Colloidal and Biological Systems
(New York: Academic)
Evans E, Halvorsen K, Kinoshita K and Wong W P 2009 A new
approach to analysis of single-molecule force measurements
Handbook of Single-Molecule Biophysics (Dordrecht:
Springer) pp 571–89
Khalil A S, Ferrer J M, Brau R R, Kottmann S T, Noren C J,
Lang M J and Belcher A M 2007 Single M13 bacteriophage
tethering and stretching Proc. Natl Acad. Sci. 104 4892
Evans E and Ritchie K 1999 Strength of a weak bond
connecting flexible polymer chains Biophys. J. 76 2439–47
Kim J, Zhang C Z, Zhang X and Springer T A 2010 A
mechanically stabilized receptor–ligand flex-bond important
in the vasculature Nature 466 992–5
Wiita A P, Ainavarapu S R K, Huang H H and Fernandez J M
2006 Force-dependent chemical kinetics of disulfide bond
reduction observed with single-molecule techniques Proc.
Natl Acad. Sci. 103 7222–7
Rothemund P W K 2006 Folding DNA to create nanoscale
shapes and patterns Nature 440 297–302
Douglas S M, Dietz H, Liedl T, Högberg B, Graf F and
Shih W M 2009 Self-assembly of DNA into nanoscale
three-dimensional shapes Nature 459 414–8
Halvorsen K 2007 Probing weak single-molecule interactions:
development and demonstration of a new instrument PhD
Thesis Boston University, Massachusetts
Wong W P and Halvorsen K 2006 The effect of integration time
on fluctuation measurements: calibrating an optical trap in
the presence of motion blur Opt. Express 14 12517–31
Smith S B, Cui Y and Bustamante C 1996 Overstretching
b-DNA: the elastic response of individual double-stranded
and single-stranded DNA molecules Science 271 795
Bustamante C, Marko J F, Siggia E D and Smith S 1994
Entropic elasticity of lambda-phage DNA Science
265 1599–600
Hanke A and Metzler R 2003 Entropy loss in long-distance
DNA looping Biophys. J. 85 167–73
Nanoengineering a single-molecule mechanical
switch using DNA self-assembly: Supplemental
Ken Halvorsen1 , Diane Schaak1 and Wesley P. Wong1,2,3
The Rowland Institute at Harvard, Harvard University, Cambridge, MA
The Immune Disease Institute/Program in Cellular and Molecular Medicine,
Children’s Hospital Boston, Boston, MA
3
The Department of Biological Chemistry and Molecular Pharmacology, Harvard
Medical School, Boston, MA
1
2
E-mail: [email protected]
This supplemental section provides detailed information for the construction of
looped single-molecule linkers using DNA self-assembly. Following a description of the
DNA oligos that we used to construct this linker, a protocol for linker construction is
presented.
1. Oligo details
The full sequence of all the oligos used are shown in Figure 1 and are based on the M13
sequence given by New England Biolabs and used for previous DNA origami work [1].
The oligos are stored at 100 µM at −20 ◦ C, and all mixing of oligos is done at these stock
concentrations unless noted. We will refer to the numbering throughout this section as
we explain the construction of the various linkers.
2. Detailed protocol for looped linker construction
2.1. Step 1: ssDNA linearization
(i) Mix the following in a clean PCR tube:
•
•
•
•
5 µL M13mp18 ssDNA (NEB product N4040S 0.25mg/mL or 100nM)
2.5 µL 10x buffer 4 (NEB)
0.5 µL 100 µM cut-site oligo (oligo 000A)
16.5 µL water
(ii) Briefly bring to 95 ◦ C ( 30 seconds), lower to 50 ◦ C, and add 1µL BtsCI enzyme.
(iii) Incubate for 1 hour at 50 ◦ C.
(iv) Heat deactivate the BtsCI enzyme by incubating at 95 ◦ C for 1 minute.
Note: To assay the linearization efficiency, add 1.21 µL of a mixture of all numbered
oligos to 5 µL of linearized ssDNA and construct the dsDNA piece by heating to 90 ◦ C
Nanoengineering a single-molecule mechanical switch using DNA self-assembly: Supplemental 2
121 Plain oligos spanning the entire M13 template
001
002
003
004
005
006
007
008
009
010
011
012
013
014
015
016
017
018
019
020
021
022
023
024
025
026
027
028
029
030
031
032
033
034
035
036
037
038
039
040
041
042
043
044
045
046
047
048
049
050
051
052
053
054
055
056
057
058
059
060
AACATCCAATAAATCATACAGGCAAGGCAAAGAATTAGCAAAATTAAGCAATAAAGCCTC
AGAGCATAAAGCTAAATCGGTTGTACCAAAAACATTATGACCCTGTAATACTTTTGCGGG
AGAAGCCTTTATTTCAACGCAAGGATAAAAATTTTTAGAACCCTCATATATTTTAAATGC
AATGCCTGAGTAATGTGTAGGTAAAGATTCAAAAGGGTGAGAAAGGCCGGAGACAGTCAA
ATCACCATCAATATGATATTCAACCGTTCTAGCTGATAAATTAATGCCGGAGAGGGTAGC
TATTTTTGAGAGATCTACAAAGGCTATCAGGTCATTGCCTGAGAGTCTGGAGCAAACAAG
AGAATCGATGAACGGTAATCGTAAAACTAGCATGTCAATCATATGTACCCCGGTTGATAA
TCAGAAAAGCCCCAAAAACAGGAAGATTGTATAAGCAAATATTTAAATTGTAAACGTTAA
TATTTTGTTAAAATTCGCATTAAATTTTTGTTAAATCAGCTCATTTTTTAACCAATAGGA
ACGCCATCAAAAATAATTCGCGTCTGGCCTTCCTGTAGCCAGCTTTCATCAACATTAAAT
GTGAGCGAGTAACAACCCGTCGGATTCTCCGTGGGAACAAACGGCGGATTGACCGTAATG
GGATAGGTCACGTTGGTGTAGATGGGCGCATCGTAACCGTGCATCTGCCAGTTTGAGGGG
ACGACGACAGTATCGGCCTCAGGAAGATCGCACTCCAGCCAGCTTTCCGGCACCGCTTCT
GGTGCCGGAAACCAGGCAAAGCGCCATTCGCCATTCAGGCTGCGCAACTGTTGGGAAGGG
CGATCGGTGCGGGCCTCTTCGCTATTACGCCAGCTGGCGAAAGGGGGATGTGCTGCAAGG
CGATTAAGTTGGGTAACGCCAGGGTTTTCCCAGTCACGACGTTGTAAAACGACGGCCAGT
GCCAAGCTTGCATGCCTGCAGGTCGACTCTAGAGGATCCCCGGGTACCGAGCTCGAATTC
GTAATCATGGTCATAGCTGTTTCCTGTGTGAAATTGTTATCCGCTCACAATTCCACACAA
CATACGAGCCGGAAGCATAAAGTGTAAAGCCTGGGGTGCCTAATGAGTGAGCTAACTCAC
ATTAATTGCGTTGCGCTCACTGCCCGCTTTCCAGTCGGGAAACCTGTCGTGCCAGCTGCA
TTAATGAATCGGCCAACGCGCGGGGAGAGGCGGTTTGCGTATTGGGCGCCAGGGTGGTTT
TTCTTTTCACCAGTGAGACGGGCAACAGCTGATTGCCCTTCACCGCCTGGCCCTGAGAGA
GTTGCAGCAAGCGGTCCACGCTGGTTTGCCCCAGCAGGCGAAAATCCTGTTTGATGGTGG
TTCCGAAATCGGCAAAATCCCTTATAAATCAAAAGAATAGCCCGAGATAGGGTTGAGTGT
TGTTCCAGTTTGGAACAAGAGTCCACTATTAAAGAACGTGGACTCCAACGTCAAAGGGCG
AAAAACCGTCTATCAGGGCGATGGCCCACTACGTGAACCATCACCCAAATCAAGTTTTTT
GGGGTCGAGGTGCCGTAAAGCACTAAATCGGAACCCTAAAGGGAGCCCCCGATTTAGAGC
TTGACGGGGAAAGCCGGCGAACGTGGCGAGAAAGGAAGGGAAGAAAGCGAAAGGAGCGGG
CGCTAGGGCGCTGGCAAGTGTAGCGGTCACGCTGCGCGTAACCACCACACCCGCCGCGCT
TAATGCGCCGCTACAGGGCGCGTACTATGGTTGCTTTGACGAGCACGTATAACGTGCTTT
CCTCGTTAGAATCAGAGCGGGAGCTAAACAGGAGGCCGATTAAAGGGATTTTAGACAGGA
ACGGTACGCCAGAATCCTGAGAAGTGTTTTTATAATCAGTGAGGCCACCGAGTAAAAGAG
TCTGTCCATCACGCAAATTAACCGTTGTAGCAATACTTCTTTGATTAGTAATAACATCAC
TTGCCTGAGTAGAAGAACTCAAACTATCGGCCTTGCTGGTAATATCCAGAACAATATTAC
CGCCAGCCATTGCAACAGGAAAAACGCTCATGGAAATACCTACATTTTGACGCTCAATCG
TCTGAAATGGATTATTTACATTGGCAGATTCACCAGTCACACGACCAGTAATAAAAGGGA
CATTCTGGCCAACAGAGATAGAACCCTTCTGACCTGAAAGCGTAAGAATACGTGGCACAG
ACAATATTTTTGAATGGCTATTAGTCTTTAATGCGCGAACTGATAGCCCTAAAACATCGC
CATTAAAAATACCGAACGAACCACCAGCAGAAGATAAAACAGAGGTGAGGCGGTCAGTAT
TAACACCGCCTGCAACAGTGCCACGCTGAGAGCCAGCAGCAAATGAAAAATCTAAAGCAT
CACCTTGCTGAACCTCAAATATCAAACCCTCAATCAATATCTGGTCAGTTGGCAAATCAA
CAGTTGAAAGGAATTGAGGAAGGTTATCTAAAATATCTTTAGGAGCACTAACAACTAATA
GATTAGAGCCGTCAATAGATAATACATTTGAGGATTTAGAAGTATTAGACTTTACAAACA
ATTCGACAACTCGTATTAAATCCTTTGCCCGAACGTTATTAATTTTAAAAGTTTGAGTAA
CATTATCATTTTGCGGAACAAAGAAACCACCAGAAGGAGCGGAATTATCATCATATTCCT
GATTATCAGATGATGGCAATTCATCAATATAATCCTGATTGTTTGGATTATACTTCTGAA
TAATGGAAGGGTTAGAACCTACCATATCAAAATTATTTGCACGTAAAACAGAAATAAAGA
AATTGCGTAGATTTTCAGGTTTAACGTCAGATGAATATACAGTAACAGTACCTTTTACAT
CGGGAGAAACAATAACGGATTCGCCTGATTGCTTTGAATACCAAGTTACAAAATCGCGCA
GAGGCGAATTATTCATTTCAATTACCTGAGCAAAAGAAGATGATGAAACAAACATCAAGA
AAACAAAATTAATTACATTTAACAATTTCATTTGAATTACCTTTTTTAATGGAAACAGTA
CATAAATCAATATATGTGAGTGAATAACCTTGCTTCTGTAAATCGTCGCTATTAATTAAT
TTTCCCTTAGAATCCTTGAAAACATAGCGATAGCTTAGATTAAGACGCTGAGAAGAGTCA
ATAGTGAATTTATCAAAATCATAGGTCTGAGAGACTACCTTTTTAACCTCCGGCTTAGGT
TGGGTTATATAACTATATGTAAATGCTGATGCAAATCCAATCGCAAGACAAAGAACGCGA
GAAAACTTTTTCAAATATATTTTAGTTAATTTCATCTTCTGACCTAAATTTAATGGTTTG
AAATACCGACCGTGTGATAAATAAGGCGTTAAATAAGAATAAACACCGGAATCATAATTA
CTAGAAAAAGCCTGTTTAGTATCATATGCGTTATACAAATTCTTACCAGTATAAAGCCAA
CGCTCAACAGTAGGGCTTAATTGAGAATCGCCATATTTAACAACGCCAACATGTAATTTA
GGCAGAGGCATTTTCGAGCCAGTAATAAGAGAATATAAAGTACCGACAAAAGGTAAAGTA
061
062
063
064
065
066
067
068
069
070
071
072
073
074
075
076
077
078
079
080
081
082
083
084
085
086
087
088
089
090
091
092
093
094
095
096
097
098
099
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
ATTCTGTCCAGACGACGACAATAAACAACATGTTCAGCTAATGCAGAACGCGCCTGTTTA
TCAACAATAGATAAGTCCTGAACAAGAAAAATAATATCCCATCCTAATTTACGAGCATGT
AGAAACCAATCAATAATCGGCTGTCTTTCCTTATCATTCCAAGAACGGGTATTAAACCAA
GTACCGCACTCATCGAGAACAAGCAAGCCGTTTTTATTTTCATCGTAGGAATCATTACCG
CGCCCAATAGCAAGCAAATCAGATATAGAAGGCTTATCCGGTATTCTAAGAACGCGAGGC
GTTTTAGCGAACCTCCCGACTTGCGGGAGGTTTTGAAGCCTTAAATCAAGATTAGTTGCT
ATTTTGCACCCAGCTACAATTTTATCCTGAATCTTACCAACGCTAACGAGCGTCTTTCCA
GAGCCTAATTTGCCAGTTACAAAATAAACAGCCATATTATTTATCCCAATCCAAATAAGA
AACGATTTTTTGTTTAACGTCAAAAATGAAAATAGCAGCCTTTACAGAGAGAATAACATA
AAAACAGGGAAGCGCATTAGACGGGAGAATTAACTGAACACCCTGAACAAAGTCAGAGGG
TAATTGAGCGCTAATATCAGAGAGATAACCCACAAGAATTGAGTTAAGCCCAATAATAAG
AGCAAGAAACAATGAAATAGCAATAGCTATCTTACCGAAGCCCTTTTTAAGAAAAGTAAG
CAGATAGCCGAACAAAGTTACCAGAAGGAAACCGAGGAAACGCAATAATAACGGAATACC
CAAAAGAACTGGCATGATTAAGACTCCTTATTACGCAGTATGTTAGCAAACGTAGAAAAT
ACATACATAAAGGTGGCAACATATAAAAGAAACGCAAAGACACCACGGAATAAGTTTATT
TTGTCACAATCAATAGAAAATTCATATGGTTTACCAGCGCCAAAGACAAAAGGGCGACAT
TCAACCGATTGAGGGAGGGAAGGTAAATATTGACGGAAATTATTCATTAAAGGTGAATTA
TCACCGTCACCGACTTGAGCCATTTGGGAATTAGAGCCAGCAAAATCACCAGTAGCACCA
TTACCATTAGCAAGGCCGGAAACGTCACCAATGAAACCATCGATAGCAGCACCGTAATCA
GTAGCGACAGAATCAAGTTTGCCTTTAGCGTCAGACTGTAGCGCGTTTTCATCGGCATTT
TCGGTCATAGCCCCCTTATTAGCGTTTGCCATCTTTTCATAATCAAAATCACCGGAACCA
GAGCCACCACCGGAACCGCCTCCCTCAGAGCCGCCACCCTCAGAACCGCCACCCTCAGAG
CCACCACCCTCAGAGCCGCCACCAGAACCACCACCAGAGCCGCCGCCAGCATTGACAGGA
GGTTGAGGCAGGTCAGACGATTGGCCTTGATATTCACAAACAAATAAATCCTCATTAAAG
CCAGAATGGAAAGCGCAGTCTCTGAATTTACCGTTCCAGTAAGCGTCATACATGGCTTTT
GATGATACAGGAGTGTACTGGTAATAAGTTTTAACGGGGTCAGTGCCTTGAGTAACAGTG
CCCGTATAAACAGTTAATGCCCCCTGCCTATTTCGGAACCTATTATTCTGAAACATGAAA
GTATTAAGAGGCTGAGACTCCTCAAGAGAAGGATTAGGATTAGCGGGGTTTTGCTCAGTA
CCAGGCGGATAAGTGCCGTCGAGAGGGTTGATATAAGTATAGCCCGGAATAGGTGTATCA
CCGTACTCAGGAGGTTTAGTACCGCCACCCTCAGAACCGCCACCCTCAGAACCGCCACCC
TCAGAGCCACCACCCTCATTTTCAGGGATAGCAAGCCCAATAGGAACCCATGTACCGTAA
CACTGAGTTTCGTCACCAGTACAAACTACAACGCCTGTAGCATTCCACAGACAGCCCTCA
TAGTTAGCGTAACGATCTAAAGTTTTGTCGTCTTTCCAGACGTTAGTAAATGAATTTTCT
GTATGGGATTTTGCTAAACAACTTTCAACAGTTTCAGCGGAGTGAGAATAGAAAGGAACA
ACTAAAGGAATTGCGAATAATAATTTTTTCACGTTGAAAATCTCCAAAAAAAAGGCTCCA
AAAGGAGCCTTTAATTGTATCGGTTTATCAGCTTGCTTTCGAGGTGAATTTCTTAAACAG
CTTGATACCGATAGTTGCGCCGACAATGACAACAACCATCGCCCACGCATAACCGATATA
TTCGGTCGCTGAGGCTTGCAGGGAGTTAAAGGCCGCTTTTGCGGGATCGTCACCCTCAGC
AGCGAAAGACAGCATCGGAACGAGGGTAGCAACGGCTACAGAGGCTTTGAGGACTAAAGA
CTTTTTCATGAGGAAGTTTCCATTAAACGGGTAAAATACGTAATGCCACTACGAAGGCAC
CAACCTAAAACGAAAGAGGCAAAAGAATACACTAAAACACTCATCTTTGACCCCCAGCGA
TTATACCAAGCGCGAAACAAAGTACAACGGAGATTTGTATCATCGCCTGATAAATTGTGT
CGAAATCCGCGACCTGCTCCATGTTACTTAGCCGGAACGAGGCGCAGACGGTCAATCATA
AGGGAACCGAACTGACCAACTTTGAAAGAGGACAGATGAACGGTGTACAGACCAGGCGCA
TAGGCTGGCTGACCTTCATCAAGAGTAATCTTGACAAGAACCGGATATTCATTACCCAAA
TCAACGTAACAAAGCTGCTCATTCAGTGAATAAGGCTTGCCCTGACGAGAAACACCAGAA
CGAGTAGTAAATTGGGCTTGAGATGGTTTAATTTCAACTTTAATCATTGTGAATTACCTT
ATGCGATTTTAAGAACTGGCTCATTATACCAGTCAGGACGTTGGGAAGAAAAATCTACGT
TAATAAAACGAACTAACGGAACAACATTATTACAGGTAGAAAGATTCATCAGTTGAGATT
TAGGAATACCACATTCAACTAATGCAGATACATAACGCCAAAAGGAATTACGAGGCATAG
TAAGAGCAACACTATCATAACCCTCGTTTACCAGACGACGATAAAAACCAAAATAGCGAG
AGGCTTTTGCAAAAGAAGTTTTGCCAGAGGGGGTAATAGTAAAATGTTTAGACTGGATAG
CGTCCAATACTGCGGAATCGTCATAAATATTCATTGAATCCCCCTCAAATGCTTTAAACA
GTTCAGAAAACGAGAATGACCATAAATCAAAAATCAGGTCTTTACCCTGACTATTATAGT
CAGAAGCAAAGCGGATTGCATCAAAAAGATTAAGAGGAAGCCCGAAAGACTTCAAATATC
GCGTTTTAATTCGAGCTTCAAAGCGAACCAGACCGGAAGCAAACTCCAACAGGTCAGGAT
TAGAGAGTACCTTTAATTGCTCCTTTTGATAAGAGGTCATTTTTGCGGATGGCTTAGAGC
TTAATTGCTGAATATAATGCTGTAGCTCAACATGTTTTAAATATGCAACTAAAGTACGGT
GTCTGGAAGTTTCATTCCATATAACAGTTGATTCCCAATTCTGCGAACGAGTAGATTTAG
TTTGACCATTAGATACATTTCGCAAATGGTCAATAACCTGTTTAGCTATATTTTCATTTG
GGGCGCGAGCTGAAAAGGTGGCATCAATTCTACTAATAGTAGTAGCATT
Oligo for M13 linearization
000A
CTACTAATAGTAGTAGCATTAACATCCAATAAATCATACA
BtsCI cut site oligo
Labeled oligos
001A
033A
044A
121A
5’ dual biotin - AACATCCAATAAATCATACAGGCAAGGCAAAGAATTAGCAAAATTAAGCAATAAAGCCTC
TCTGTCCATCACGCAAATTAACCGTTGTAGCAATACTTCTTTGATTAGTAATAACATCAC - 3’thiol
5‘ dig - ATTCGACAACTCGTATTAAATCCTTTGCCCGAACGTTATTAATTTTAAAAGTTTGAGTAA
GGGCGCGAGCTGAAAAGGTGGCATCAATTCTACTAATAGTAGTAGCATT - dT internal biotin 3’ biotin
Bridge oligos
Binds M13 from
bases 1950-1980
033-044A
033-044B
033-044C
033-044D
033-077A
033-077B
033-077C
033-077D
Binds M13 starting at:
base 2580 (short loop)
base 4560 (long loop)
CAATACTTCTTTGATTAGTAATAACATCACATTCGACAACTCGTATTAAATCCTTTGCCC
CAATACTTCTTTGATTAGTAATAACATCACATTCGACAACTCGTATTAAA
CAATACTTCTTTGATTAGTAATAACATCACATTCGACAACTCGTA
CAATACTTCTTTGATTAGTAATAACATCACATTCGACAAC
CAATACTTCTTTGATTAGTAATAACATCACTCAACCGATTGAGGGAGGGAAGGTAAATAT
CAATACTTCTTTGATTAGTAATAACATCACTCAACCGATTGAGGGAGGGA
CAATACTTCTTTGATTAGTAATAACATCACTCAACCGATTGAGGG
CAATACTTCTTTGATTAGTAATAACATCACTCAACCGATT
(60bp
(50bp
(45bp
(40bp
(60bp
(50bp
(45bp
(40bp
short loop bridge oligo)
short loop bridge oligo)
short loop bridge oligo)
short loop bridge oligo)
long loop bridge oligo)
long loop bridge oligo)
long loop bridge oligo)
long loop bridge oligo)
Figure 1. Sequences of all oligos used. The numbered oligos cover the entire sequence
of the M13 scaffold strand, and the lettered strands are functionalized or special
purpose oligos. The bridge oligos are labeled red for regions that bind M13 bases
1950-1980, green for regions that bind starting at M13 base 2580, and blue for regions
that bind starting at M13 base 4560.
Nanoengineering a single-molecule mechanical switch using DNA self-assembly: Supplemental 3
cut site
enzyme
oligos
circular
ssDNA
unbound
bound
Figure 2. Looped linker construction using DNA origami: Circular single-stranded
DNA is enzymatically cleaved at a single site and mixed with over 100 oligonucleotides
to self assemble into a looped linker held together by a single “bridge” or “staple”
oligo.
and cooling to 20 ◦ C at 1 ◦ C/minute or slower. The product can either be run on a 0.7%
agarose gel to separate circular from linear, or the strand can be cut with a single cut
enzyme (we added 1µ AfeI) to make sure most or all of the ssDNA was linearized.
2.2. Step 2: Conjugating protein to oligo
The protocol we used for conjugating protein to oligo was loosely based on the Pierce
protocol provided with the sulfo-SMCC.
(i) Deprotect the SH oligos by mixing the following in a clean PCR tube
• 5µL 0.5M TCEP (Pierce)
• 40µL water
• 5µL oligo 033A at 100 µM
Let the mixture sit for at least 30 minutes at room temperature
(ii) Make 10mM Sulfo-SMCC solution
• add 20µL DMSO to 2 mg Sulfo-SMCC and pipette up and down to mix well
• dilute into 450µL of PBS
• Use immediately
(iii) Activate the protein
• suspend protein in PBS at 1mg/ml concentration (use desalting column or
centrifuge filter if necessary to concentrate or change buffer)
Nanoengineering a single-molecule mechanical switch using DNA self-assembly: Supplemental 4
• use 20x molar excess of Sulfo-SMCC for 1mg/ml protein (e.g. 1mg/ml
Antibody - 20x: 20µL protein @ 6.7µM + 0.27µL Sulfo-SMCC @ 10mM).
• React for 30 minutes at room temperature.
(iv) Wash TCEP from oligos just before Sulfo-SMCC reaction is finished
• use Qiagen PCR clean up kit, following the protocol except eluting into 50µL
PBS for 10µM final concentration (note: using a Tris buffer will interfere due
to the presence of amines)
(v) Wash the Sulfo-SMCC from the activated protein
• pre-equilibrate a Zeba desalt column (Pierce) with PBS
• do one or two passes of the protein through the column to eliminate SulfoSMCC
• resuspend at 5µM concentration
(vi) Mix the activated protein with the deprotected oligos
• Use 1:1 molar ratio or excess of protein (e.g. 1:1 - 2µL oligo @ 10µM + 4µL
antibody @ 5µM)
• React for 30 minutes at room temperature
2.3. Step 3: Purifying protein conjugated oligo
Once the protein has been conjugated to the oligos, it may be necessary to purify the
conjugated oligo from unreacted byproducts depending on the yield of the reaction. We
were typically able to get 5-50% yields (see Figure 2), and even with a 50% yield we
had trouble getting the loop to form with the digoxigenin antibody complex without
purification. This is probably due to unconjugated oligos competing with the conjugated
ones and due to excess protein reacting with dig-labeled oligos on the DNA construct.
To purify, we ran the same product in a 4-20% polyacrylamide gel (Bio-rad) enough
to separate the conjugated and unconjugated oligos (see Figure S2), typically 150V for
40 minutes in 1x TBE buffer. Then we stained the gel for 10-15 minutes in a 1x solution
of Sybr Gold (Invitrogen) and used a razor blade to cut out the relevant band(s). Once
the gel slices were cut, we used an electroelution kit and supplied protocol to extract the
conjugated oligos from the gel. Briefly, we put the slices in a midi sized electroelution
tube (Gerard biotech) with 600µL of buffer, and ran them in a horizontal electrophoresis
at 150V for an additional hour. We estimated the final concentration based on the known
amount of oligo put into each gel lane, the conjugation yield, and the amount of dilution
in the electroelution step.
2.4. Step 4: Assembly of DNA linkers
We have made several different linkers over the course of this study. All of these utilize
the same basic protocol which was adapted from other DNA origami work [1] without
further optimization. This involves mixing the oligos with the M13 ssDNA (typically in
od
y
An
tib
+
o
Ol
ig
Ol
ig
ch o +
em po
ica lyc
lly lon
co al
nj an
Ol
ug t i
ig
at bo
o
ed dy
ch +
em m
ica ono
lly clo
co na
nj l a
ug n
at tib
ed od
y
Re
fe
(1 ren
00 ce
bp la
in dde
cr r
em
en
ts
)
Nanoengineering a single-molecule mechanical switch using DNA self-assembly: Supplemental 5
Antibody bound
60 bp oligo
60 bp oligo
Figure 3. Gel verification of antibody-oligo conjugation. The left hand lanes
show chemical conjugation of the oligo with polyclonal and monoclonal antibody,
respectively. The right hand lane is a mixture of digoxigenin labelled oligo and antidigoxigenin without chemical conjugation.
a 10x molar excess) and subjecting the mixture to a temperature ramp from 90 ◦ C to
20 ◦ C at 1 ◦ C/minute. This protocol may be slower than necessary, as it was based on
folding complex 2D shapes rather than simply making a linear piece of dsDNA as we
are doing here. We used a thermal cycler to apply the temperature ramp, but heating a
water bath and letting it cool to room temperature over the same time should provide
the similar results.
The following mixtures of oligos will be used in the construction of various linkers:
plain linear: mix of all numbered oligos
linear with double biotin ends: mix of all numbered oligos except substitute 001A
for 001 and 121A for 121
short loop with double biotin ends: same mixture as linear with double biotin
ends, but do not include 033 or 044 (or use truncated versions of 033 and 044
to ensure all bases of M13 are paired).
Nanoengineering a single-molecule mechanical switch using DNA self-assembly: Supplemental 6
long loop with double biotin ends: same mixture as linear with double biotin ends,
but do not include 033 or 077 (or use truncated versions of 033 and 077 to ensure
all bases of M13 are paired).
To make the desired construct, mix the following in a clean PCR tube and apply
the temperature ramp to anneal oligos:
• 5µL of linearized ssDNA from section 2.1
• 1.2µL of one of the above mixtures (for 10:1 oligo:DNA ratio)
• For looped constructs, additionally add:
(i) 1µL of 1000x dilute digoxigenin oligo 044A for dig-antibody construct
(ii) 1µL 1000x dilute bridge oligo 033-044B for short loop with 50bp oligo bridge
(iii) 1µL 1000x dilute bridge oligo 033-077B for long loop with 50bp oligo bridge
The temperature ramp completes the protocol, except for the dig-antibody
construct. For the dig-antibody looped construct, add an additional 1µL of purified
antibody conjugated oligo at an approximate concentration of 100nM once the first
temperature ramp is complete. Note that it if a larger volume of lower concentration
oligo is used, it is important to maintain the proper buffer conditions for hybridization.
Additional concentrated buffer may need to be added in this case. To anneal this oligo,
we used a temperature ramp from 40 ◦ C to 10 ◦ C at 0.5 ◦ C/minute, though we did not
exhaustively test simpler or shorter protocols. For the looped constructs, we typically
got a 50% looping yield following this protocol. This looped product can be excised and
purified from a gel as we did with the antibody conjugated oligo, but we did not find
this purification step to be necessary for this study.
References
[1] P.W.K. Rothemund. Folding DNA to create nanoscale shapes and patterns. Nature, 440(7082):297–
302, 2006.