Purification of DNA-origami nanostructures
by rate-zonal centrifugation
Chenxiang Lin1,2,3, Steven D. Perrault1,2,3, Minseok Kwak1,2,3, Franziska Graf1,2,3and
William M. Shih1,2,3,*
1Department of Cancer Biology, Dana-Farber Cancer Institute,2Department of Biological Chemistry and
Molecular Pharmacology and3Wyss Institute for Biologically Inspired Engineering, Harvard University,
Boston, MA 02115, USA
Received September 28, 2012; Revised October 11, 2012; Accepted October 12, 2012
Most previously reported methods for purifying
DNA-origami nanostructures rely on agarose-gel
AGE is routinely used to yield 0.1–1mg purified
DNA nanostructures, obtaining >100mg of purified
DNA-origami structure through AGE is typically
Here, we present a readily scalable purification
approach utilizing rate-zonal centrifugation, which
AGE. The DNA nanostructures remain in aqueous
Therefore, the desired products are easily recovered
with consistently high yield (40–80%) and without
contaminants such as residual agarose gel or DNA
intercalating dyes. Seven distinct three-dimensional
DNA-origami constructs were purified at the scale
of 0.1–100mg (final yield) per centrifuge tube,
showing the versatility of this method. Given the
mixing and fraction collection, this method should
be amenable to automation and further scale up for
quantities) of DNA nanostructures.
Self-assembly of a long, single-stranded circular DNA
(scaffold strand) together with many synthetic oligo-
nucleotides (staple strands)—the method known as
DNA origami (1–7)—has proven an efficient way of
well-defined geometry. In addition, the DNA-origami
offering a powerful technique of organizing material
with up to sub-nanometer precision (8–10). Despite
continued efforts in optimizing the structural design and
folding conditions (11), the assembly yield of DNA
origami is in most cases far <100%. However, well-folded
and contamination-free DNA nanostructures are required
for many applications. Therefore, a purification step after
folding has become a standard procedure for many
DNA-origami studies. Most widely used purification
(AGE), through which the well-folded objects are
resolved as a distinct band and separated from slower
migrating by-products (e.g. misfolded structures and ag-
gregates) as well as the faster migrating non-integrated
staple strands. After electrophoresis, the well-folded struc-
tures are extracted from the gel through homogenization
(3,4,6,11) or electro-elution (12), reconstituted into the
desired buffer and adjusted to desired concentration.
As effective as the AGE-based separation is, the following
two problems remain: the extraction step requires signifi-
cant user input and quickly becomes laborious as the puri-
fication scales up and the DNA nanostructures recovered
in this way usually co-purify with contaminants such as
agarose-gel residues and ethidium bromide (or other
staining reagents). There is therefore a pressing need for
a scalable purification method that reduces labour cost
and contaminants while maintaining the high separation
resolution of AGE.
Here, we report a scalable, cost-effective and con-
nanostructures through rate-zonal centrifugation, which
separates molecular species by subjecting them to high
centrifugal force in a density gradient media (13).
*To whom correspondence should be addressed. Tel: +1 617 632 5143; Fax: +1 617 632 4471; Email: William_Shih@dfci.harvard.edu
Chenxiang Lin, Department of Cell Biology, Yale School of Medicine, New Haven, CT 06520, USA and Nanobiology Institute, Yale University,
West Haven, CT 06516, USA.
Franziska Graf, Insmed Incorporated, Monmouth Junction, NJ 08852, USA.
Published online 15 November 2012 Nucleic Acids Research, 2013, Vol. 41, No. 2e40
? The Author(s) 2012. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nc/3.0/), which
permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact
Similar approaches have been applied to purify macro-
molecules such as proteins (14), protein–RNA complexes
(e.g. ribosomes (15)) and metal colloids (16). In our case,
well-folded DNA-origami structures are separated from
other unwantedspecies (e.g.
strands, misfolded structures and aggregates) due to
their distinct and well-defined mass and shape. A scheme
that depicts the purification workflow is illustrated in
Figure 1. In brief, a folded DNA-origami sample is
loaded to the top of a linear quasi-continuous gradient
of 15–45% glycerol and spun at ?300000g for 1–3h.
Fractions are then collected from the glycerol gradient and
analysed using AGE. Those containing desired products are
then combined, buffer-exchanged and concentrated.
MATERIALS AND METHODS
DNA oligonucleotides were purchased from Bioneer Inc.
(Alameda, CA, USA). Scaffold strands p3024 and p7308
were cloned and amplified in-house following standard
molecular biology protocols (17). Ultracentrifuge tubes
and adaptors were purchased from Beckman-Coulter
Inc. (Miami, FL, USA). Gel-loading tips were purchased
from USA Scientific (Orlando, FL, USA). Rubber
stoppers were purchased from VWR (Radnor, PA,
USA). SYBR Safe DNA gel stain solution was purchased
from Life Technologies (Carlsbad, CA, USA). Amicon
filters were purchased from Millipore (Billerica, MA,
USA). All other reagents were purchased from Sigma-
Aldrich (St. Louis, MO, USA).
The assembly of the DNA-origami nanostructures was
accomplished in a one-pot reaction by mixing 50nM
scaffold strands with a pool of oligodeoxyribonucleotide
staple strands (250 or 500nM of each; reverse-phase cart-
ridge purified) in a folding buffer containing 5mM Tris
(pH 8), 1mM ethylenediaminetetraacetic acid (EDTA)
and 10–20mM MgCl2(exact concentration depends on
the target nanostructure) and subjecting the mixture to a
thermal-annealing ramp that cooled from 80?C to 24?C
over the course of 15–72h (exact annealing time depends
on the target nanostructure).
A linear glycerol gradient (15–45%, v/v) was prepared
in one of the following two ways. In the first way, seven
layers of glycerol solution in 1?TE-Mg buffer (5mM
Tris–HCl, 1mM EDTA and 10mM MgCl2, pH 8), 80
or 400ml per layer, were laid carefully into a 0.8-ml
(Beckman #344090) or 3.5-ml (Beckman #349622) ultra-
centrifuge tube using longneck gel-loading tips with 45%
glycerol solution at the bottom and 5% concentration
decrement per layer (Figure 1a). The tube was then
incubated overnight at 4?C to form a quasi-continuous
gradient. Alternatively, two layers of glycerol solution in
1?TE-Mg buffer, 1.4 ml per layer, were laid carefully
into a 3.5-ml (Beckman #349622) ultracentrifuge tube
with 45% glycerol solution at the bottom and 15%
glycerol solution on top. The tube was capped with a
rubber stopper (00m, VWR #59580-069), laid flat slowly
and incubated at room temperature for 2h. The tube was
then returned to its vertical position (Figure 1b). The gra-
dients prepared in the above two ways have been proven
to yield almost the same separation efficiency for at least
one structure. New or thoroughly cleaned tubes (interior
brushed with mild detergent) should be used when
preparing gradients in the latter way.
Desired amount of folded DNA-origami nanostructures
containing 10% glycerol were loaded on top of the
glycerol gradient (Figure 1c). In a typical experiment,
50ml (or 400ml) of 50nM nanostructure was purified per
0.8ml (or 3.5 ml) tube. The centrifuge tubes were then
placed inside the centrifuge buckets (Beckman #356860
adapters were used to fit the 0.8 ml tubes), suspended on
a swinging-bucket rotor (Beckman SW 55 Ti) and spun at
50000rpm (?300000gmax) for 1 (Figure 2b–g) or 3h
(Figure 2a) at 4?C. The optimal centrifugation time
depends on the exact shape and mass of the nanostructure
and should be experimentally determined. (Excessive cen-
trifugation time may lead to sample pelleting at the tube
bottom and thus negatively impact the purification effi-
ciency.) At this point, the well-folded nanostructures will
have separated from free staple strands and unwanted
multimers due to their different sedimentation rates.
Twelve to sixteen equal-volume fractions were collected
from top to bottom of the centrifuge tube using
longneck gel-loading tips.
Aliquots of each fraction (5–10ml per fraction) were
loaded into separate wells of a non-denaturing, 1.5%
agarose gel containing 1? SYBR Safe gel stain and
separated by electrophoresis at 2.5V/cm in 0.5? TBE
buffer (45mM Tris-borate, 1mM EDTA (disodium salt),
pH 8.3) containing 10mM MgCl2for 2.5h at room tem-
perature (Figure 1d). The gels were then scanned on a
Typhoon FLV 9000 laser scanner. The fractions contain-
ing the desired products were combined and reconstituted
into native folding buffer of the nanostructure or
1?TE-Mg buffer using Amicon Ultra-0.5 ml centrifugal
filters (MWCO 30 or 100kDa) following the manufac-
turer’s manual. Centrifugal force <4500g was used in this
step to minimize sample damage and loss. Typically
50–100ml of solution containing purified DNA nanostruc-
tures were obtained. Alternatively, gel-filtration columns
could be used for buffer exchange if the final DNA con-
centration is not the primary concern, as this process may
dilute the sample slightly. The purified structures were sub-
jected to AGE and transmission electron microscopy
(TEM) to determine their purity and recovery yield.
Sample quality after purification
Seven DNA-origami nanostructures (see Supplementary
Figure S1 for strand diagrams)—a 6-helix-bundle (6-hb)
ring (Figure 2a), a 12-hb ring (Figure 2b), an octahedron
with curved 6-hb edges (Figure 2c), an 18-hb rod bent by
90?(Figure 2d), a 48-hb brick (Figure 2e), a 24-hb rod
(Figure 2g)—were purified using the method described
above. The effectiveness of rate-zonal centrifugation to
separate desired product from non-integrated staple
strands and misfolded structures was evaluated by
resolving post-centrifugation gradient fractions through
e40Nucleic Acids Research, 2013,Vol. 41,No. 2PAGE 2 OF 6
AGE (schematics in Figure 1d; gel images in Figure 2).
Staple strands have the slowest sedimentation rate (i.e.
resides in the top fractions of the gradient) and segregate
far away from most folded materials. Well-folded DNA
nanostructures have distinct mobilities that allow them to
be separatedfrom other
oligomeric nanostructures and aggregates), which usually
travel faster in the gradient. Further quantitative analysis
of gel images revealed that ?85% of the well-folded
nanostructures are enriched in 10–20% volume of the
gradient. The effective separation resulted in final
products with excellent purity, which is supported by
both AGE and TEM study of purified DNA-origami
samples (Figure 2 and Supplementary Figure S2). The
AGE analysis revealed enriched well-folded structures
with greatly reduced misfolded structures and almost com-
impurities could be attributed to disturbance and diffusion
Figure 1. Scheme of the rate-zonal centrifugation purification. (a) Preparation of glycerol gradient through overnight incubation of seven layers of
glycerol solution from 15% to 45% with 5% increment per layer. (b) An alternative way of preparing the same gradient as shown in (a): a capped
tube containing two layers of glycerol solution (15% and 45%) is laid down, incubated for 2h and returned to vertical position. (c) Separation of
different species (free staple strands, well-folded monomers and misfolded multimers) in the DNA-origami folding mixture through centrifugation.
(d) Post-centrifugation process to first identify the fractions containing correctly folded nanostructures and then reconstitute such fractions into
DNA-origami folding buffer.
PAGE 3 OF 6 Nucleic AcidsResearch, 2013, Vol.41,No. 2 e40
of the gradient during sample handling (e.g. pipetting frac-
tions or transferring tubes in and out of the centrifuge)
and slight structural deformation of purified DNA-
origami molecules. The already low level of remaining
free staple strands can be further reduced to gel-
undetectable levels by post-processing through Amicon
filtration (Figure 2, all gel slices to the left of the noted
percentage yields). For comparison, we purified the 6-hb
rings (Figure 2a) from the same folding batch through
either electrophoresis (band electro-eluted) or this centri-
fugation method. Similar DNA nanostructure purity was
observed by resolving the final products side-by-side on an
Figure 2. Purification result of a series of 3D DNA-origami structures: (a) 6-helix-bundle (6-hb) ring, (b) 12-hb ring, (c) octahedron with curved 6-hb
edges, (d) 18-hb rod bent by 90?, (e) 48-hb brick, (f) 24-hb rod and (g) 24-hb rod with two cavities. Computer-rendered 3D models of the structures
are shown in the leftmost column, where each straight or curved DNA helix is shown as a white or red cylinder, respectively. Cross-sections of 6-hb
and 12-hb rings are shown on the upper right corner of the 3D models in (a) and (b). AGE analyses of glycerol-gradient fractions collected after
centrifugation are shown in the second column from the left. M: 0.1–10.0kb DNA ladder (New England Biolabs); R: Raw DNA-origami assemblies
before purification; consecutive Arabic numbers denote the fractions collected from top to bottom of the gradient, with fraction 1 the lightest.
Underscored fractions are those enriched for well-folded DNA-origami structures. AGE characterizations of the purified products are shown in the
third column from the left. Lanes (from left to right) are loaded with 0.1–10.0kb DNA ladder, unpurified DNA-origami folding mixture and
DNA-origami structures purified through centrifugation, respectively. Recovery yield of each structure (calculated from band intensities measured
using ImageJ) is presented to the right of the corresponding gel image. Representative TEM images of the purified structures are shown in the
rightmost column. Scale bars: 50nm.
e40 Nucleic Acids Research, 2013,Vol. 41,No. 2PAGE 4 OF 6
agarose gel (Supplementary Figure S3). In addition, the
DNA-origami structures purified through the centrifuga-
tion method were never in contact with agarose gel or
DNA intercalating reagent (e.g. ethidium bromide or
SYBR Safe), which benefits downstream applications
involving TEM or fluorescence study by providing
cleaner background or baseline signal.
Scalability and recovery yield
The purification was performed on 0.5pmol or 20pmol
folded scaffold strand for each structure (Figure 2) using
one 0.8 or 3.5 ml centrifuge tube. It is important to
note that these numbers are far below the maximum
purification capacity of each tube. The 0.8 and 3.5 ml
tubes have been successfully used to purify 8 and
120pmol of raw DNA-origami structures (i.e. starting
material), respectively, without compromising separation
resolution or recovery yield. A purification trial using
one centrifuge tube per sample generated 0.05–50pmol
(0.1–100mg) enriched correctly folded product, depending
on the purification scale, DNA-origami folding efficiency
and recovery yield. One centrifuge rotor (Beckman SW55)
accommodates six centrifuge tubes and therefore can be
used to purify six different DNA-origami species or larger
amount of a single species in one purification trial.
Another valuable feature of the centrifugation-based puri-
fication method is the reproducible, high recovery yield. In
contrast to the AGE-based purification, the DNA always
stays in the aqueous phase and there is no need to extract
them from the gel matrix. Therefore, losses of material are
greatly reduced. In addition, there is less batch-to-batch
variation in the recovery yield, which is a common
problem associated with the AGE-based purification
methoddueto the experimental
(e.g. electro-elution time, homogenization temperature,
etc.). As shown in Figure 2, ?40–80% properly folded
DNA-origami objects were recovered after centrifugation
separation and post-processing (measured by ImageJ).
Subsequent TEM analyses also confirmed the strong
enrichment of nanostructures with designated geometry
DNA nanotechnology is leading to various applications
in structural biology (17), biophysics (18–20), biosensing
(21–24) and therapeutic delivery (25–28), many of which
entail the preparation of micrograms to milligrams of
enriched high-quality DNA nanostructures. Traditional
electrophoresis-based methods excel at separation reso-
lution but often fall short in scalability. The rate-zonal
centrifugation method presented here is an effective com-
plement to the current state-of-the-art in DNA-origami
purification—properly folded DNA structures are well
separated in the glycerol gradient with similar reso-
lution provided by AGE and can be recovered in large
quantity, with high efficiency and ease. Furthermore, we
demonstrated that purified 6-hb rings carrying staple ex-
tensions (handles)were able
nanoparticles through handle/anti-handle hybridization
[Supplementary Figure S4, with design and protocol
adapted from (29,30)], confirming the well-preserved
activity of the single-stranded handles after purification.
We note that the reported method may be improved
further in the future to achieve better consistency,
greater end-product purity and higher throughput, par-
ticularly through the automation of the gradient-mixing
and fraction-collection processes using commercially
available liquid-handling equipment. It is worth pointing
out that in this work a universal gradient (15–45% linear
glycerol gradient) was used for all structures. Therefore,
when applied to a variety of DNA-origami structures,
such a gradient may be more effective in purifying
certain ones versus others. Users should be able to
fine-tune the gradient to find the optimal conditions for
each DNA structure of interest. Finally, higher capacity
centrifuge tubes and rotor sets could be used to increase
the purification scale by another order of magnitude (e.g.
Beckman SW32 rotor accommodates up to 38.5-ml tubes).
The rate-zonal centrifugation method, as an effective and
easy-to-adapt purification approach, could help address-
ing the technical challenge
nanostructure preparation and in turn, enable the devel-
opment of many applications in the field of structural
to host5-nm gold
of large-scale DNA
Supplementary Data are available at NAR Online:
Supplementary Figures 1–4.
The authors thank the Imaging Core and General
Equipment Facility of Wyss Institute for Biologically
Inspired Engineering for the use of TEM and ultracentri-
fuge. They thank Dr Weiming Xu at Yale School of
Medicine for inspiring discussion at the early stage of
National Institutes of Health (NIH) [1DP2OD004641,
[N000141010241, N000014091118]; Wyss Institute for
Biologically Inspired Engineering Faculty Award (to
(to S.D.P.); Netherlands Organization for Scientific
Research (NWO-Rubicon) (to M.K.). Funding for open
access charge: NIH [1DP2OD004641].
of Naval Research
for Health Research
Conflict of interest statement. None declared.
1. Rothemund,P. (2006) Folding DNA to create nanoscale shapes
and patterns. Nature, 440, 297–302.
PAGE 5 OF 6 Nucleic AcidsResearch, 2013, Vol.41,No. 2e40
2. Andersen,E., Dong,M., Nielsen,M., Jahn,K., Subramani,R., Download full-text
Mamdouh,W., Golas,M., Sander,B., Stark,H., Oliveira,C. et al.
(2009) Self-assembly of a nanoscale DNA box with a controllable
lid. Nature, 459, 73–76.
3. Douglas,S., Dietz,H., Liedl,T., Ho ¨ gberg,B., Graf,F. and
Shih,W.M. (2009) Self-assembly of DNA into nanoscale
three-dimensional shapes. Nature, 459, 414–418.
4. Dietz,H., Douglas,S. and Shih,W.M. (2009) Folding DNA into
twisted and curved nanoscale shapes. Science, 325, 725–730.
5. Han,D., Pal,S., Liu,Y. and Yan,H. (2010) Folding and cutting
DNA into reconfigurable topological nanostructures. Nat.
Nanotech., 5, 712–717.
6. Liedl,T., Ho ¨ gberg,B., Tytell,J., Ingber,D. and Shih,W.M. (2010)
Self-assembly of three-dimensional prestressed tensegrity structures
from DNA. Nat. Nanotech., 5, 520–524.
7. Han,D., Pal,S., Nangreave,J., Deng,Z., Liu,Y. and Yan,H. (2011)
DNA origami with complex curvatures in three-dimensional
space. Science, 332, 342–346.
8. Nangreave,J., Han,D., Liu,Y. and Yan,H. (2010) DNA origami:
a history and current perspective. Curr. Opin. Chem. Biol., 14,
9. Shih,W.M. and Lin,C. (2010) Knitting complex weaves with
DNA origami. Curr. Opin. Struct. Biol., 20, 276–282.
10. Tørring,T., Voigt,N., Nangreave,J., Yan,H. and Gothelf,K. (2011)
DNA origami: a quantum leap for self-assembly of complex
structures. Chem. Soc. Rev., 40, 5636–5646.
11. Ke,Y., Bellot,G., Voigt,N.V., Fradkov,E. and Shih,W.M. (2012)
Two design strategies for enhancement of multilayer–
DNA-origami folding: underwinding for specific intercalator
rescue and staple-break positioning. Chem. Sci., 3, 2587–2597.
12. Bellot,G., McClintock,M.A., Lin,C. and Shih,W.M. (2011)
Recovery of intact DNA nanostructures after agarose gel-based
separation. Nat. Methods., 8, 192–194.
13. Churchill,L., Banker,G. and Cotman,C. (1973) Gradient design
to optimize rate zonal separations. Anal. Biochem., 56, 370–382.
14. Patsch,J., Sailer,S., Kostner,G., Sandhofer,F., Holasek,A. and
Braunsteiner,H. (1974) Separation of the main lipoprotein density
classes from human plasma by rate-zonal ultracentrifugation.
J. Lipid Res., 15, 356–366.
15. Eikenberry,E., Bickle,T., Traut,R. and Price,C. (1970) Separation
of large quantities of ribosomal subunits by zonal
ultracentrifugation. Eur. J. Biochem., 12, 113–116.
16. Steinigeweg,D., Schu ¨ tz,M., Salehi,M. and Schlu ¨ cker,S. (2011) Fast
and cost-effective purification of gold nanoparticles in the 20-250
nm size range by continuous density gradient centrifugation.
Small, 7, 2443–2448.
17. Douglas,S., Chou,J. and Shih,W.M. (2007)
DNA-nanotube-induced alignment of membrane proteins for
NMR structure determination. Proc. Natl Acad. Sci. USA, 104,
18. Fu,J., Liu,M., Liu,Y., Woodbury,N. and Yan,H. (2012)
Interenzyme substrate diffusion for an enzyme cascade organized
on spatially addressable DNA nanostructures. J. Am. Chem. Soc.,
19. Wilner,O., Weizmann,Y., Gill,R., Lioubashevski,O., Freeman,R.
and Willner,I. (2009) Enzyme cascades activated on topologically
programmed DNA scaffolds. Nat. Nanotech., 4, 249–254.
20. Liu,C., Kim,E., Demple,B. and Seeman,N. (2012) A DNA-based
nanomechanical device used to characterize the distortion of
DNA by Apo-SoxR protein. Biochemistry, 51, 937–943.
21. Ke,Y., Lindsay,S., Chang,Y., Liu,Y. and Yan,H. (2008)
Self-assembled water-soluble nucleic acid probe tiles for label-free
RNA hybridization assays. Science, 319, 180–183.
22. Lin,C., Liu,Y. and Yan,H. (2007) Self-assembled combinatorial
encoding nanoarrays for multiplexed biosensing. Nano Lett., 7,
23. Bell,N., Engst,C., Ablay,M., Divitini,G., Ducati,C., Liedl,T. and
Keyser,U. (2012) DNA origami nanopores. Nano Lett., 12,
24. Wei,R., Martin,T., Rant,U. and Dietz,H. (2012) DNA origami
gatekeepers for solid-state nanopores. Angew. Chem. Int. Ed., 51,
25. Douglas,S., Bachelet,I. and Church,G. (2012) A logic-gated
nanorobot for targeted transport of molecular payloads. Science,
26. Jiang,Q., Song,C., Nangreave,J., Liu,X., Lin,L., Qiu,D., Wang,Z.-
G., Zou,G., Liang,X., Yan,H. et al. (2012) DNA origami as a
carrier for circumvention of drug resistance. J. Am. Chem. Soc.,
27. Walsh,A., Yin,H., Erben,C., Wood,M. and Turberfield,A. (2011)
DNA cage delivery to mammalian cells. ACS Nano, 5,
28. Liu,X., Xu,Y., Yu,T., Clifford,C., Liu,Y., Yan,H. and Chang,Y.
(2012) A DNA nanostructure platform for directed assembly of
synthetic vaccines. Nano Lett., 12, 4254–4259.
29. Ding,B., Deng,Z., Yan,H., Cabrini,S., Zuckermann,R. and
Bokor,J. (2010) Gold nanoparticle self-similar chain structure
organized by DNA origami. J. Am. Chem. Soc., 132, 3248–3249.
30. Zhao,Z., Jacovetty,E., Liu,Y. and Yan,H. (2011) Encapsulation
of gold nanoparticles in a DNA origami cage. Angew. Chem. Int.
Ed., 50, 2041–2044.
e40Nucleic Acids Research, 2013,Vol. 41,No. 2PAGE 6 OF 6