Yossi Weizmann et al- A polycatenated DNA scaffold for the one-step assembly of hierarchical...

8/3/2019 Yossi Weizmann et al- A polycatenated DNA scaffold for the one-step assembly of hierarchical nanostructures

1/6

A polycatenated DNA scaffold for the one-stepassembly of hierarchical nanostructuresYossi Weizmann, Adam B. Braunschweig, Ofer I. Wilner, Zoya Cheglakov, and Itamar Willner*

Institute of Chemistry and the Center for Nanoscience and Nanotechnology, The Hebrew University of Jerusalem, Jerusalem 91904, Israel

Communicated by Raphael D. Levine, The Hebrew University of Jerusalem, Jerusalem, Israel, February 6, 2008 (received for review December 31, 2007)

A unique DNA scaffold was prepared for the one-step self-assembly

of hierarchical nanostructures onto which multiple proteins ornanoparticles are positioned on a single template with precise

relative spatial orientation. The architecture is a topologically

complex ladder-shaped polycatenane in which the rungs of theladder are used to bring together the individual rings of the

mechanically interlocked structure, and the rails are available forhierarchical assembly,whose effectiveness has beendemonstrated

with proteins, complementary DNA, and gold nanoparticles. The

ability of this template to form from linear monomers and simul-taneously bind two proteins was demonstrated by chemical force

microscopy, transmission electron microscopy, and confocal fluo-

rescence microscopy. Finally, fluorescence resonance energy trans-fer between adjacent fluorophores confirmed the programmed

spatial arrangement between two different nanomaterials. DNAtemplates thatbring together multiple nanostructures withprecise

spatial control have applications in catalysis, biosensing, and nano-

materials design.

chemical force microscopy proteins wires nanoparticle catenane

Versatile scaffolds for the immobilization of proteins andother nanosized objects are targets of intensive research inthe broader field of nanotechnology because of their potentialapplications (1) in the areas of catalysis, biosensing, and nano-materials. In nature, complex catalytic cascades, such as theKrebs cycle (2), photosynthesis (3), or glycolysis (4), involvemultiple proteins assembled with exact relative spatial orienta-

tions to facilitate cooperative binding, to facilitate electron orenergy transfer, and to optimize substrate conversion. Enzy-matic cascades that are used industrially (57) c ould also benefitfrom such closely arranged proteins if a template with the abilityto fix precisely various proteins were available. Although func-tional synthetic polymers (8) are potential matrices for forminguseful scaffolds, this approach is limited by polymer compati-bility with biomolecules and the f lexibility of the polymerizationprotocol to synthesize macromolecules that controllably bindmultiple proteins. DNA, however, is a robust biopolymer that,through WatsonCrick base pairing and the multitude of se-quences that can be formulated, might yield self-assembledprotein scaffolds. Indeed, DNA has been used for the organi-zation of proteins (9, 10) and elegant, topologically complex 3Darchitectures, such as Borromean rings (11), nanotubes (12),

[2]catenanes (13), 2D tilings (1418),and chains(19),onto whichnanoparticles (2028) and proteins (19, 20, 2932) have beentethered by using complementary single-stranded DNA(ssDNA), biotinstreptavidin interactions, or gold-thiol bonds.These studies, however, have been limited to the immobilizationof a single biomolecule or nanoparticle onto the scaffold orrequire multistep protocols to organize more than one nano-material onto the template.

Results and Discussions

In this report, we describe a modular DNAscaffold that wasusedto orient nanoparticles, a protein, or several different fluoro-phores onto a single scaffold with precise relative spatial orien-tation in a one-step, robust, programmable fashion. The scaffold

is a DNA ABAB copolymer whose linear monomers are me-chanically interlocked with ligase to form a polycatenated ladderin which each ring contains two ssDNA domains available forfurther hierarchical self-assembly. These ssDNA sections havebeen used to bind fluorophores or proteins attached to com-plementary ssDNA sequences, or we have modified the code ofthe sequence to the aptamer (33, 34) for thrombin and subse-quently bound fluorescent-dye-labeled thrombin. The inter-locked nature of the chain was c onfirmed with gel electrophore-sis, dynamic force spectroscopy (DFS), and transmissionelectron microscopy (TEM). The ability of this DNA sequenceto bind two f luorophores by using specific, complementary DNAinteractions that bind either monomer 1 or 2 was confirmed byusing confocal fluorescence microscopy (CFM) and atomic force

microscopy (AFM). The close proximity necessary for fluores-cence resonance energy transfer (FR ET) to occur c onfirmed theperiodic arrangement of the fluorophores. This mechanicallyinterlocked scaffolda modular polycatenated DNA poly-meris unique in itsability to bring together different nanosizedobjects, be they enzymes or nanoparticles, with precise relativeorientations in a programmable fashion by self-assembly.

The assembly of two linear ssDNA strands into a mechanicallyinterlocked polycatenated scaffold is shown in Fig. 1A. The twocomplementary ssDNA monomers, 1 and 2, were designed suchthat upon assembly the enzyme ligase joins the adjacent 3 and5 ends of the monomer in the hybridized, complementary DNA,thereby mechanically interlocking 1 and 2 to form the links thatgenerate the ladder-shaped polycatenane, 3. Chain extension ofthe AB copolymer continues by a polycondensation-type mech-anism, in which, after each catenation occurs, active sites forfurther and subsequent catenation remain at each end of thepolymer. Polymerization continues as a consequence of eitheraddition of a single link or by chain combination. The result ofthis process is a ladder-shaped DNA polymer in which the rungsare composed of hybridized DNA and the rails are ssDNA on

which other objects can be appended to form hierarchicalnanostructures.

The ability of this process to form high-molecular-weight,interlocked nanowires was confirmed by gel electrophoresis andatomic force microscopy (Fig. 1). To assess the ability of themonomers to self-assemble into a chain, an agarose gel w ith themonomers alone, and mixed together, was run against linearDNA molecular weight standards. No bands appear in lanes 4

and5

after staining because the SYBR Green I only binds todouble-stranded DNA. However, when the two monomers werepremixed in the buffer with ligase, the result was a broad bandat the baseline whose intensity increased with reaction time,signaling the creation of a species with higher molecular weight

Author contributions: Y.W., A.B.B., O.I.W., Z.C., and I.W. designed research; Y.W., A.B.B.,

O.I.W.,and Z.C.performedresearch; Y.W.,A.B.B., O.I.W., andZ.C. analyzeddata;and Y.W.,

A.B.B., and I.W. wrote the paper.

The authors declare no conflict of interest.

*To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/cgi/content/full/

0800723105/DCSupplemental.

2008 by The National Academy of Sciences of the USA

www.pnas.orgcgidoi10.1073pnas.0800723105 PNAS April 8, 2008 vol. 105 no. 14 52895294
http://www.pnas.org/cgi/content/full/0800723105/DCSupplementalhttp://www.pnas.org/cgi/content/full/0800723105/DCSupplementalhttp://www.pnas.org/cgi/content/full/0800723105/DCSupplementalhttp://www.pnas.org/cgi/content/full/0800723105/DCSupplementalhttp://www.pnas.org/cgi/content/full/0800723105/DCSupplemental


2/6

than that corresponding to the 1-kb marker of 10,000 bp.Although the linear standards do not give an accurate repre-sentation of the molecular weight of the double-stranded, cir-cular chains, the new band does demonstrate that the two

monomers hybridize to form aggregates of significantly highermolecular weight. A second gel was run under denaturingconditions to explore mechanical bonding in the polycatenane[supporting information (SI) Fig. S1]. The polycatenane, 3, wasassembled and exposed to ligase, and this mixture was run in thedenaturing polyacrylamide gel against the same mixture beforeligase exposure and the monomers A and B. The mixture thathad not been exposed to ligase resulted in a single band that ranidentically to the monomers. However, themixture that had beenexposed to ligase stayed near the baseline with a broad bandreflecting both high molecular weight and the polydispersityreflective of a polymeric material. This result only occursbecause the chains are held together by bonds that are unaf-fected by denaturing conditions, that is, mechanical bonds.

Topographical AFM images (Fig. S2) showed flexible chainsmany micrometers in length, and the height profile of thesechains of 1.6 nm (Fig. 1C) is consistent with double-strandedDNA. AFM is a more accurate measure of chain length than the

linear molecular weight standards in the gel, and it is more likelythat the chains imaged by the AFM consist of several thousandbase pairs that are observed in the AFM based on an estimateof 250 links per micrometer.

By using DFS (35), the mechanical bonding within the cat-enane was further investigated. To study the ability of our systemto catenate, the circular oligonucleotide 4 was tethered to a

Au-coated AFM tip, and the complementary semicircular oli-gonucleotide 5 was immobilized on a Au-coated glass slide. Inthe presence of ligase, these sequences should form the mechan-ically bonded [2]catenane that yields the monomer of thepolycatenane. The ligation should result in increased ruptureforce because the surface and the tip are covalently linkedthrough the mechanical bond, and rupture could only occur by

Fig. 1. Preparation and characterization of DNA polycatenanes and [2]catenanes.(A) The formation of the polycatenatedDNA by polymerization and ligation

ofthe twossDNAmonomers, 1 and 2, resultsin themechanicallyinterlocked ABABcopolymer,3. Thecolorsrepresent thecomplementary sequences, greenbeing

complementarywith orange andblue withyellow.Gray areas areamenableto further modification.(B) Gelelectrophoresisshowingformation of polycatenane,

3. Lane M contains a 1-kb marker. Lanes 1, 2, and3 contain 1:1mixtures of monomers 1 and 2 after 90 min, 60 min, and30 minself-assembly times,respectively.

Lane4 contains monomer 1, andlane5 containsmonomer 2. (C) TheAFM image of thepolycatenated DNAsequence, 3. (D) Representative schematic of dynamic

force microscopy (DFS) measurementsbetweenoligonucleotides 4 and 5. Forcesweremeasured both in thepresence andabsenceof ligase. (E) The dimerization

of 1.4-nmgold particlesby theinterlocking of Au nanoparticles appended to thessDNA, 7 and 8. (F) TEMimageof thesingle oligonucleotide, 7, functionalized

with 1.4-nm Au nanoparticles. (G) TEM image of Au-nanoparticle-functionalized [2]catenanes, 9.

5290 www.pnas.orgcgidoi10.1073pnas.0800723105 Weizmann et al.
http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF1http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF1http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF1http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF2http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF2http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF2http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF1


3/6

the destruction of the weakest link in the new system, most likelythe rupture of a AuAu bond. The forces on retraction of the tipfrom the surface were measured both before and after exposureto ligase. The forces measured from each pull-off event wererecorded and used to form histograms (Fig. S3) from which theaverage forces were extracted. The rupture of two hybridized30-bp oligonucleotides on separation with an AFM tip resultedin pull-off forces of 20200 pN (3638), with an accepted valueof 2050 pN (39), whereas the force required to detach a Au

atom from a bulk Au surface is

1 nN, which is slightly lowerthan the force required to rupture a AuS bond (40). Beforeligation, forces in multiples of 31 26 pN were measured aftererror analysis (Fig. S4 and Fig. S5), which is within the accepted

value of 2050 pN between 30 bp of DNA. In the presence ofligase, both the shape of the energy landscape contour and themagnitude of the forces changed dramatically (Fig. S6). In eachrecorded force curve, a strong pull-off event of 1.3 0.82 nN inmagnitude occurred, with several smaller events leading up tothis rupture. The smaller events could be caused by interactions

with ligase present in the system, whereas the large rupturecorresponds to the cleavage of the Au-thiolated bond or theremoval of the Au atoms from the bulk surface (see SI Text forfurther discussion). Control experiments such as force measure-ments between a linear analog of 5 and 4 (Fig. S7), and

measurements of the effects of ligase for the dihydroxylatedanalog of 5, which could not be ligated, was measured in theabsence (Fig. S8) and in the presence (Fig. S9) of ligase, withoutany significant change in the magnitude of force compared withthe value obtained between 4 and 5 in the absence of ligase.These experiments provide further evidence that a mechanicalbond is formed in the polycatenane.

The same nucleic acid sequences used in the DFS experimentswere used to bring together pairs of gold nanoparticles as partof functionalized [2]catenanes to show that catenane formationis not inhibited by appending large objects to the rails. The aminesequences, 4 and 5, were immersed in a solution of 1.4-nm goldnanoparticles to form gold nanoparticle-immobilized oligonu-cleotides 7 and 8 (Fig. 1E). These oligonucleotides were hybrid-ized, ligated, and imaged by TEM. As a c ontrol, TEM images of

the circular oligonucleotide, 7, were taken, and nanoparticlesthat were randomly distributed along the TEM grid were ob-served (Fig. 1F). When both rings, 7 and 8, were added to thesolution, the grids were composed predominantly of twinnednanoparticles (Fig. 1G). The paired nanoparticles are the resultof the [2]catenane, 9, formed between the complementary Aunanoparticle-functionalized rings. Close examination reveals anincrease in the percentage of twinned nanoparticles separated by8 nm (Fig. S10), which is the maximum possible separationbetween the thiols on fully extended [2]catenanes. Subsequently,the solution that resulted in the twinned nanoparticles wasexposed to BsaAI, a restriction enzyme that cleaves the hybrid-ized DNA. Examination of the TEM image revealed that mostof the particles existed as single nanoparticles, which suggeststhat the scission of the [2]catenanes in solution occurred. Pop-

ulation histograms (Fig. S10) show a demonstrable, quantitativechange in the distribution of nanoparticles on adding comple-mentary pairs that is partly reversed on exposure to the restric-tion enzyme. These experiments demonstrate the ability to usethis system to self-assemble Au nanoparticles and separate themenzymatically, a benefit of using the bioactive DNA scaffold.

Although the TEM is not stand-alone evidence for programmedassembly, it is an important demonstration that catenation is notprohibited by the presence of bulky nanoparticles. Together withthe gels and DFS, TEM provides evidence for a catenanestructure that acts as a scaffold for hierarchical nanostructures.

After demonstration of mechanical bonding in the polycat-enane, we proceed to demonstrate the ability of this material toserve as a scaffold for hierarchical assembly by using AFM,

TEM, CFM, and FRET measurements. The ability to formcomplex nanostructures was shown by using either proteins orssDNA that contained fluorescent dyes. A tetramethylrhod-amine (TAMRA)-labeled ssDNA sequence, 10, that was com-plementary to the rails of the polycatenated DNA, 3, formed thefunctionalized polycatenane 11 when mixed with 10 (Fig. 2A). Byusing confocal f luorescent microscopy, a solution of 11 that wasexcited at 543 nm revealed micrometer-long wires with a strong

emission at 570 nm, the emission wavelength characteristic ofTAMRA (Fig. 2B). As a result,we conclude that thessDNArailsare accessible for modification by complementary ssDNA. Toextend the means by which the scaffold could be modified, themonomer sequence of the polycatenane was changed such thatit contained the aptamer for the protein thrombin on the rails,and the functional polycatenane 12 was formed (Fig. 2C). Onexposing this polycatenane to TAMRA-labeled thrombin, 13,the nanowires of 14, several micrometers in length, were ob-served by confocal f luorescence microscopy (Fig. 2D). Thethrombin-functionalized DNA polycatenane, 15, was also exam-ined by AFM, and circular nanostructures were detected (Fig.2F). Nanowires with regularly spaced bumps with heights of2nm were detected (Fig. 2F). These regularly spaced bumps are

Fig. 2. Modification of thepolycatenated DNAwith complementary DNAor

proteins. (A) Functionalization of the polymer rails of 3 with the fluorescent-

labeled ssDNA, 10. (B) Confocal fluorescence microscopy image of 11. The

polymers are excited at 543 nm, and emission is recorded at 570 nm. (C)The functionalization of polycatenane 12 with TAMRA-labeledthrombinby the

aptamer-induced association of the protein to form 14. (D) Confocal fluores-

cence microscopy image of 14 (ex 543 nm, em 570 nm). (E) Functional-

ization of 12 polycatenane with thrombin to form 15. AFM images of 15.

Weizmann et al. PNAS April 8, 2008 vol. 105 no. 14 5291
http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF3http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF3http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF3http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF4http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF4http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF4http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF4http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF4http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF6http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF6http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF6http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=STXThttp://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=STXThttp://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF7http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF7http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF7http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF8http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF8http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF8http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF9http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF8http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF8http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF9http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF10http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF9http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF8http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF7http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=STXThttp://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF6http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF4http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF4http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF3


4/6

most likely the thrombin on the DNA backbone (41), and crosssections of these bumps show a small valley between two peaks.These valleys are likely the distance between two thrombin unitson the same link, and the spaces between bumps are the resultof the empty links that were programmed between thrombin-aptamer-containing rings. Interestingly, all thrombin-labeled

polycatenanes form rings with diameters of0.5 m, and thismay be a result of drying effects (42), because the identicalstructure does not form rings in the confocal microscopy imagethat is carried out in solution. These two techniques to modifythe polycatenated scaffoldusing complementary DNA andconverting the rails to an aptamer demonstrate the modularityof this template onto which proteins, dyes, and nanoparticles canbe immobilized to form ordered, hierarchical nanostructures.

The preprogrammed assembly of two dyes, sequentially and within nanometers of each other, was confirmed by confocalfluorescence microscopy and FRET measurements. The fluo-rescent dyes TAMR A and f luorescein were tethered to the endsof two different oligonucleotide sequences that were comple-mentary to the rails of A andB, respectively (Fig. 3). Themixture

of2, 16, 17, 18, and ligase in solution resulted in polycatenane 19that included both dyes in an alternating repeat sequence. Thepresence of both dyes on the DNA chains was confirmed byCFM. Excitation of TAMRA at 543 nm resulted in strongred fluorescence, 570 nm, from the observed wires, andexcitation of fluorescein at 488 nm also resulted in redfluorescence, 570 nm, as a result of energy transfer fromfluorescein to TAMRA. However, after the photobleaching ofTAMRA, excitation at 488 nm resulted in the green

fluorescence characteristic of f luorescein (Fig. 3E

). Because theacceptor was bleached and the f luorescence intensity decreased,the fluorescence of the donor increased (Fig. S11). This exper-iment confirmed that both dyes are present on the scaffold, andthe energy transfer only occurs if the dyes are in close proximity(5 nm). Quantitative FRET experiments demonstrate theincrease in fluorescein fluorescence as TAMRA is photo-bleached (Fig. S11), and the increase in FRET as the self-assembly time increases, which demonstrate that FRET onlyoccurs as the scaffold becomes populated with dyes. Theseexperiments could not be used to establish the degree ofsubstitution, and some binding sites may remain unoccupied,however, FRET originates from species on the nanowires thatare close together and suggests that there is a significantpopulation of adjacent periodic dyes that communicate and

enable FRET. The fluorescence experiments demonstrate theability of the self-assembly protocol to simultaneously bringtogether multiple components in a single reaction step; in thisexperiment, six individual components come together to formeach repeat unit.

Conclusions

To conclude, we have reported the preparation of a modularDNA scaffold onto which proteins, nanoparticles, and dyes werefixed with precise control. A topologically unique ladder-shapedpolycatenane was prepared by two different monomers followedby ligation to form the mechanically interlocked macromolecule.The mechanically bonded nature was confirmed by a series ofexperiments including electrophoresis, DFS, and TEM. Thedenaturing gel showed a clear increase in molecular weight after

ligation, and by modifying the monomers such that only the[2]catenane forms, we were able to examine their mechanicalstrength and bring together gold nanoparticles. Although therungs of this ladder are hybridized into double-stranded DNA toform the polymer, the rails remain available as sites for furtherhierarchical assembly that was shown to serve as a scaffold with

AFM, CFM, and FRET. We have used these sites to create longpolycatenane chains modified w ith proteins by self-assembly. Bymodifying the DNA sequence of these polycatenanes, the com-ponents and structure of the assembly can be changed. Theability to complex different structures lies in the modularity ofthe DNA sequence in the rails of the DNA ladders. Two dyes

were assembled in an alternating sequence by using an efficientone-step protocol that simultaneously brings together six com-ponents in each polymeric unit. The versatility of the polycat-enated template motif enables the formation of more elaboratestructures by the design of an ABC copolymer or by differen-tiating the rails within each ring, thereby resulting in nanostruc-tures with many interacting components within a well pro-grammed spatial arrangement. The potential of hierarchicalnanoassemblies in biosensing, catalyzing enzymatic cascades,and forming complex nanostructures are just part of the promiseof mechanically interlocked, polymeric DNA scaffolds.

Materials and MethodsMethods Summary. Thrombin was labeled with TAMRA and DNA with fluo-

rescein and TAMRA according to standard protocols. Oligonucleotide se-

quences 1 and 2 (1 109 M) were polymerized by using a Quick Ligation Kit

and analyzedby bothgel electrophoresis and AFM. The interlocked nature of

Fig. 3. One-step self-assembly and FRET analysis of bifunctional polycat-

enated DNA structures. (A) Monomers 2 and 16 were combined with fluoro-

phores 17 and 18 in a 1:1:2:2 ratio to form 19. (B and C) Confocalfluorescence

microscopy image of the multicomponent complex, 19, exposed to ex 488

nm,em520and ex543nm, em570nm, respectively.(D and E) Confocal

fluorescence microscopy image of 19 during photobleaching at 0 s and 7 s,

respectively. (F) Time-dependent fluorescence resonance energy transfer

(FRET) spectra after the self-assembly of 19 after (1) t 0 min,(2) 6 min,(3) 12

min, (4) 18 min, and (5) 30 min.

http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF11http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF11http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF11http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF11http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF11http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=SF11


5/6

the rings was confirmed by chemical force microscopy and denaturing poly-

acrylamide gel measurements between 4 and 5, which were immobilized on

gold-coated AFM cantilevers and gold slides, respectively. Forces were mea-

sured both in the absence and in the presence of T4 Ligase (50 units/l). The

ability of the [2]catenanes to assemble nanoparticles was determined by

modifying7 and 8 with mono-sulfo-NHS gold nanoparticles(Nanoprobes) and

imaging the resulting assemblies with TEM (FEI Company). The ability of 3 to

bind TAMRA-labeled ssDNA or 11 to bind TAMRA-labeled thrombin was

confirmed by confocal fluorescence microscopy (Olympus FluoView FV300

Confocal Laser Scanning Microscope). One-step self-assembly of 19 was ac-

complished by mixing 2, 16, 17, 18 (1:1:2:2 mixture) for 4 h at 37C.

Materials. Oligonucleotides 1, 2, 10, 16, 17, and 18 were purchased from

Genosys (Sigma),and 4 and 5 werepurchased fromDNA Technologies.Chem-

icals were purchased from Sigma unless otherwise noted. The deoxynucle-

otide (dNTP) solution mixture, Exonuclease I, Exonuclease III, BsaAI endonu-

clease, and Quick Ligation Kit were purchased from New England Biolabs.

Mono-Sulfo-NHS gold nanoparticles were purchased from Nanoprobes, Inc.

Thrombin from human plasma was purchased from Sigma. Glass slides with a

250-nm gold coating were purchased from Analytical -Systems. Centricon

separation devices were purchased from Millipore. Ultrapure water from a

NANOpure Diamond (Barnstead) source was used throughout all of the

experiments. The following were purchased from commercial sources: 8%

denaturing polyacrylamide gel (Biological Industries), SYBR gold nucleic acid

stain (Molecular Probes), 50-bp DNA ladder (New England Biolabs). 0.6%

Agarose gel (SeaKem LE Agarose from Cambrex BioScience Rockland), 1-kb

DNA ladder (Promega), and SYBR Green I (Sigma).

General Methods. Transmission electron microscopy (TEM)images weretaken

on a FEI Tecnai F20 G2 instrument with 0.24-nm resolution operating in

bright-field mode on 300-mesh copper grids (Electron Microscopy Sciences).

All atomic force microscopy (AFM) imaging and dynamic force microscopy

(DFS) measurements were performed at room temperature by using a Multi-

mode scanning probe microscope with a Nanoscope 3A controller (Digital

Instruments/Veeco Probes). Gel electrophoresis was carried out with 0.6%

agarose gel and SYBRGreen I (MolecularProbes) stain. Confocalfluorescence

microscopy (CFM) was carried out on a LSM 410 Zeiss Confocal Laser Micro-

scope with DlanApochromat 64 /1.4 oil lens, an Olympus FluoView FV300

Confocal Laser Scanning Microscpe with a UIS PLAPO 60/1.4 oil lens with a

488 nm Argon laser, and a 543 nm He-Ne Laser. TAMRA fluorescence was

observed at 560570 nm, and fluorescein fluorescence was observed at 505

530 nm. FRET confocal microscopy was carried out on an Olympus FV-1000

confocal microscope (60 /1.35 oil-immersion objective). Bleaching (100%)

was accomplished at 543 nm for 7 sec. Fluorescein (ex 488 nm, em 505530) nm and TAMRA (ex 543 nm, em 560570 nm). Real-time FRET

measurements carried out in photon-counting spectrometer (Edinburgh In-

struments, FLS920) equippedwith a cooled photomultiplier detectionsystem,

connected to a computer (F900 v.6.3 software; Edinburgh Instruments). UV/

Ozonecleaningwas carried outin a T1O 10/OES/E UV/ozone chamber from

UVOCS. AFM topographical images were taken on samples deposited on

freshly cleaved micasurfaces (Structure Probe) thatwere first passivated with

a 5 mM MgCl2 solution for 1 min followed by dropcasting the solution of

interest. Images were taken with Ultrasharp SiN AFM tips (Mikromasch) in

tapping mode at their resonant frequency, and these images were analyzed

with WsXM SPIP software (Nanotec) (43).

Preparation of Polycatenated DNA 3. Linear DNA, 1 (5-GTAGTACAGA-

CGCTCAAAAAAAAAAAAAAACAGTATTAGCACGTGCTTCACAGTCTCACA-

AAAAAAAAAAAAAAACAGACGATCCTAGAC-3 ), 10 109 M, was reacted

with linear DNA, 2 (5-CACGTGCTAATACTGAAAAAAAAAAAAAAAGA-

GCGTCTGTACTACGTCTAGGATCGTCTGAAAAAAAAAAAAAAATGTGAGACT-

GTGAAG-3), 10 109 M, in Quick Ligation Kit buffer. The solution was first

heated to 90C for 10 min and fast cooled to 50C at which it was held for 30

min. The solution was then fast cooled to 25C, and 40 units/l ligase was

added for 30 min. The enzyme was denatured by heating at 65C for 10 min.

DNA 3 was washed with ultrapure water and separated with a Centricon

filtration device (30,000 cutoff).

Gel Electrophoresis of 3. Two micromolar 1 wasreactedwith2 M 2 in ligation

buffer consisting of 66 mM TrisHCl, 10 mM MgCl2, 1 mM ATP, 15% polyeth-

ylene glycol (PEG 6,000), 1 mM DTT, pH 7.6. The mixture was first heated to

90C for10 minthencooled to 50C fordifferent time intervals(30,60, and90

min, respectively), fast cooled to 25C, and exposed to ligase 200 units/l for

30 min. The resulting mixture was loaded in 0.6% agarose gel and exposed

with SYBR Green I.

Denaturing Gel Electrophoresis of 3. Two micromolar 1 was reacted with 2 M

2 in ligation buffer. The mixture was first heated to 90C for 10 min then

cooled to 50C for 90 min, fast cooled to 25C, and exposed to ligase 200

units/l, for 30 min. The resulting mixture was loaded in 8% polyacrylamide

gel in the presence of 8 M urea.

Preparation of Circular DNA 4. The phosphorylated linear DNA

(5-GATCCTAAT-(NH 2)AATAGTACACATGCTCAAAAAAAAAAAAAAACAGT-

ATTAGCACGTGCTTCACAGTCTCACAAAAAAAAAAAAAAAAAACCACAC-3),

0.8 106 M, was reacted with the ligation template (5-TTAGGATCGTGTG-

GTT-3), 9

10

6

M, in the Quick Ligation Kit buffer, in the presence of T4Ligase(400 units/l),at25Cfor30 min. Theenzyme wasdenaturedbyheating

at 65C for 10 min, and the ligated circular DNA, 4, was treated with Exonu-

clease I (5 units/l) for 40 min at 37C to degrade the unligated primer that

remained in solution. Exonuclease III (5 units/l) was added to degrade the

ligated primerthatremainedattached to thecircular DNAfor 40 minat 37C.

The enzymes were denatured by heating at 80C for 20 min. DNA 4 was

washed withultrapure water andseparated witha Centriconfiltration device

(30,000 cutoff).

Immobilization of 5 on Au Slide. The gold surface was cleaned by a 15-min

immersion in piranha solution [70% (vol/vol) concentrated H2SO4 and 30% (vol/

vol)30% hydrogenperoxide] and rinsed by water. The surface was subsequently

soaked in concentrated nitric acid, rinsed again with water, and dried with

nitrogen. The clean, dry Au surface was immersed in a 10 mM 3,3-dithiodipro-

pionic acid bis(N-hydroxysuccinimide ester) in dry DMSO for 1 h to obtain the

active ester monolayer. The resulting surface was washed with dry DMSO andHepesbuffer (0.1 M, pH7.4).The activatedsurfacewasimmersedin a solutionof

linear DNA, 5 (5-CACGTGCTAATACTGAAAAAAAAAAAAAAAAAAAAA-

AAAAT-(NH2)AAAAAAAAAAAAAAAAAAAAAAAAATGTGAGACTGTGAAG-3),

1 106 M, for 30 min. The DNA-modified surface was immersed in Tris buffer

(0.2 M, pH 7.4) for 10 min to passify active ester that had not reacted with 5.

Cantilever Modification with 4. Gold-coated silicon cantilevers (Mikromasch)

were cleaned in a UV/ozone chamber for 20 min and immediately immersed

in a 10 mM 3,3-dithiodipropionic acid bis(N-hydroxysuccinimide ester) in dry

DMSO for 1 h to obtain an active ester monolayer. The resulting cantilevers

were washedwithDMSOand Hepes buffer(0.1M, pH 7.4). TheNHS-activated

cantilevers were immersed in a solution of the ligated circular DNA, 4, 20

109 M in Hepes buffer (0.1 M, pH 7.4) for 30 min. The resulting cantilevers

were reacted with Tris buffer (0.2 M, pH 7.4) for 10 min to eliminate any

unreacted active ester.

Measuring Forces between 4 and 5. All measurementswere carried out at room

temperature in ligation buffer consisting of 66 mM Tris HCl, 10 mM MgCl2, 1

mM ATP (pH 7.6) in a total volume of 300 l. Measurements were carried out

in a liquid cell on a PicoForce module (Digital Instruments/Veeco Probes).

Spring constants of the cantilevers were determined in air by using the

thermal noise method(44) to give an average springconstantof 0.06 Nm.To

measure the forces, the modified cantilevers were lowered to the surface for

30 s and retracted at a rate of 0.1 m/s. Data points were analyzed with their

associated spring constants. Histograms were prepared by using OriginPro

(OriginLab). Peaks were fitted by using autocorrelation analysis (45) or a

Lorentzian function (see SI Text for further discussion). Each histogram was

the result of at least 100 separateforce measurements. Forces weremeasured

inthe absence,then in thepresence ofT4 Ligase(20,000units).For thesystem

that did not include ligase, 300 retraction experiments were performed,

resulting in 150 force curves giving rise to a success rate of 50%. For the

system that included ligase out of 300 pulls, 66 data points were identified,

giving rise to a success rate of 20%.

Preparation of Gold Nanoparticle-Labeled DNA 7 and 8. DNA oligonucleotide 4

and 5 (1.2 109 mol) were each reacted with 1.4-nm Mono-Sulfo-NHS gold

nanoparticles, 6 109 mol for 1 h in a buffer consisting in Hepes buffer, 0.1

M (pH 7.4), for 40 min at room temperature. The reaction mix was then

purified and separatedfrom the excess Au-nanoparticlesby using a Centricon

filtration device (30,000 cutoff).

Preparation of Gold Nanoparticle-Labeled DNA [2]Catenane 9. To create the

[2]catenane, 9, ligated oligonucleotide, 7, labeled Au nanoparticles, 2 107

M, were reacted with oligonucleotide, 8, labeled Au nanoparticles, 2 107

M,for 1 h ina bufferconsistingof 50mM TrisHCl, 10mM MgCl2, 100mM NaCl,

and0.05mM DTT. Theligation reactionwas completedby adding2.5mM ATP

and ligase (4,000 units) in a total volume of 50 l for 30 min at 25C. The

Weizmann et al. PNAS April 8, 2008 vol. 105 no. 14 5293
http://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=STXThttp://www.pnas.org/cgi/data/0800723105/DCSupplemental/Supplemental_PDF#nameddest=STXT


6/6

mixture washeatedto55Cfor 20minand fast cooledin ice. Fortherestriction

assay, endonuclease BsaA I (5 units per reaction) was added, for 1 h, at 37C.

Preparation of TAMRA-oligonucleotide 10. Oligonucleotide (5-NH 2-

TTTTTTTTTTTTTTT-3),10 106 M, wasreactedwith TAMRA-NHS, 20 106

M, inHepesbuffer (0.1 M, pH7.4) for1 h at room temperature.The excessdye

was separated with a Centricon filtration device (3,000 cutoff) and was

washed with water.

TAMRA-Labeled Polycatenane 11. Polycatenated DNA 3 (50 nM), DNA 2 (50

nM), and 10 (200 nM) were mixed in hybridization buffer consisting of 0.3 MNaCl and 10 mM phosphate buffer (pH 7.4) for 4 h.

TAMRA-Labeled Trombin 13. Thrombin, 5 M, was reacted with TAMRA-NHS,

10 M, to form 13 in Hepes buffer, 0.1 M, pH 7.4, for 50 min at room

temperature. The reaction mixtures were then purified and separated from

the excess dyes by the aptamer-binding proteins buffer consisting of 20 mM

TrisHCl (pH 7.4), 140 mM NaCl, 5 mM KCl, 5 mM CaCl 2, and 1 mM MgCl2, by

using a Centricon filtration device (10,000 cutoff).

TAMRA-Containing Thrombin-Labeled Polycatenane14. Linear DNA 2, 10 109

M, was reacted with aptamer-containing linear DNA, 16 (5-CACGTG-

CTAATACTGAAAAAGGTTGGTGTGGTTGGAAAAAGAGCGTCTGTACTACGTCT-

AGGATCGTCTGAAAAAGGTTGGTGTGGTTGGAAAAATGTGAGACTGTGAAG-

3), 10 109 M, in an aptamer-binding buffer consisting of 20 mM TrisHCl

(pH 7.4), 140 mM NaCl, 5 mM KCl, 5 mM CaCl2, and 1 mM MgCl2 to create the

aptamer-containing polycatenane 12. The mixture was heated to 90C for 10

min and fast cooled to 37C, which was followed by immediate addition of

TAMRA-labeledthrombin 13 (100 109 M), ina finalvolume of50lfo r2 h .

14 was examined with no further purification.

Thrombin-Labeled Polycatenane 15. 15 was made identically to 14, although

with thrombin instead of 13.

FRET Bifunctional Polycatenane 19. Monomers 2, 1 109

M; 16, 1 109

M;17 (5-fluorescein-TTTTTTTTTTTTTTT-3), 10 109 M; and 18 (5-TTTTTC-

CAACCACACCAACCTTTTT-TAMRA-3), 10 109 M were combined in a final

volumeof 200l inthehybridizationbufferconsistingof0.3 M NaCl and1 mM

phosphate buffer (pH 7.4).Time-dependent changes in the FRETfluorescence

emission were measured at ex 488 nm, em 550600 nm.

Photobleaching of 19. Photoleaching (100%) was carried out by exposure at

543 nm for 7 s.

ACKNOWLEDGMENTS. We thank Dr. Mark N. Berman for assistance withstatistical analysis. This work was supported by the BioInfoNanoCognaArtsFund, The Hebrew University of Jerusalem, and by a fellowship from TheJohanna Friedlander Memorial Fund (to Y.W.).

1. Niemeyer CM (2001) Nanoparticles, proteins, and nucleic acids: Biotechnology meets

materials science. Angew Chem Int Ed 40:41284158.2. Winkel BS (2004) Metabolic channeling in plants. J Annu Rev Plant Biol55:85107.

3. Nelson N, Ben-Shem A (2004) The complex architecture of oxygenic photosynthesis.

Nat Rev Mol Cell Biol 5:112.

4. Plaxton WC (1996) Theorganizationand regulation of plant glycolysis.AnnuRev Plant

Physiol Plant Mol Biol 47:185214.

5. Schmid A, et al. (2001) Industrial biocatalysis today and tomorrow. Nature 409:258

268.

6. Parmar A, Kumar H, Marwaha SS, Kennedy JF (1998) Recent trends in enzymatic

conversion of Cephalosporin C to 7-Aminocephalosporanic Acid (7-ACA). Crit Rev

Biotech 18:118.

7. Bhattacharya S, Schiavone M, Gomes J Bhattacharya SK (2004) Cascade of bioreactors

in series for conversion of 3-phospho-D-gycerate into D-ribulose-1,5-bisphosphate:

Kinetic parameters of enzymes and operation variables. J Biotech 111:203217.

8. Patil AO, Schulz DN, Novak BM, eds (1998) Functional Polymers: Modern Synthetic

Methods andNovel Structures, ACSSymposium Series704 (AmericanChemicalSociety,

Washington, DC).

9. Feldkamp U, Niemeyer CM (2006) DNA nanoarchitectures. Angew Chem Int Ed

45:18561876.

10. Seeman NC (1998) DNA nanotechnology: NovelDNA constructions.AnnuRev Biophys

Biomol Struct 27:225248.

11. Mao C, Sun W, Seeman NC (1997) Assembly of Borromean rings from DNA. Nature

386:137138.

12. Kuzuya A, Wang R, Sha R, Seeman NC (2007) Six-helix and eight-helix DNA nanotubes

assembled from half-tubes. Nano Lett7:17571763.

13. Kumar R, et al. (2007) Template-directed oligonucleotide strand ligation, covalent

intramolecular DNA circularization using click chemistry. J Am Chem Soc 129:6859

6865.

14. Winfree E, Liu F, Wenzler LA, Seeman NC (1998) Design and self-assembly of two-

dimensional DNA crystals. Nature 394:539544.

15. Garibotti AV, Knudsen SM, Ellington AD, Seeman NC (2006) Functional DNAzymes

organized into two-dimensional arrays. Nano Lett6:15051507.

16. Feng L, Park SH, Reif JH, Yan H (2003) A two-state DNA lattice switched by DNA

nanoactuator. Angew Chem Int Ed 42:43424346.

17. Ding B, Seeman NC (2006) Operation of a DNA robot arm inserted into a 2D DNA

crystalline substrate. Science 314:15831585.

18. Mao C, Sun W, Shen Z, Seeman NC (1999) A nanomechanical device based on the B-Z

transition of DNA. Nature 397:144146.19. YanH, ParkSH, Finkelstein G, ReifJH, LaBean TH (2003) DNA-templated self-assembly

of protein arrays and highly conductive nanowires. Science 301:18821884.

20. Park SH, et al. (2005) Three-helix bundle DNA tiles self-assemble into 2D lattice or 1D

templates for silver nanowires. Nano Lett5:693696.

21. Yamada F, SachoY, Matsumoto T, Tanaka H, Kawai T (2004) DNA-templated assembly

of Au nanoparticles via step-by-step binding reaction. J Surf Sci Nanotech 2:222225.

22. Deng Z, Tian Y, Lee SH, Ribbe AE, Mao C (2005) DNA-encoded self-assembly of gold

nanoparticles into one-dimensional arrays. Angew Chem Int Ed 44:35823585.

23. LeJD, etal. (2004) DNA-templated self-assemblyof metallicnanocomponent arrays on

a surface. Nano Lett4:23432347.

24. Lin C, Liu Y, Rinker S, Yan H (2006) DNA tile based self-assembly: Building complex

nanoarchitectures. ChemPhysChem 7:16411647.25. Sharma J, Chhabra R, Liu Y, Ke Y, Yan H (2006) DNA-templated self-assembly of

two-dimensionaland periodicalgold nanoparticlearrays.AngewChemIntEd45:730

735.

26. Zhao W, Gao Y, Kandadai SA, Brook MA, Li Y (2006) DNA polymerization on gold

nanoparticles through rolling circle amplification: Towards novel scaffolds for three-

dimensional periodic nanoassemblies. Angew Chem Int Ed45:24092413.

27. Zheng J, et al. (2006) Two-dimensional nanoparticle arrays show the organization

power of robust DNA motifs. Nano Lett6:15021504.

28. Aldaye FA,SeimanHF (2006) Sequentialself-assemblyof a DNA hexagonas a template

for the organization of gold nanoparticles. Angew Chem Int Ed45:22042209.

29. ParkSH, etal. (2005) ProgrammableDNA self-assemblies fornanoscaleorganizationof

ligands and proteins. Nano Lett5:729733.

30. Niemeyer CM, Sano T, Smith CL, Cantor CR (1994) Oligonucleotide-directed self-

assemblyof proteins:Semisynthetic DNA- streptavidinhybrid moleculesas connectors

for the generation of macroscopic arrays and the construction of supramolecular

bioconjugates. Nucleic Acids Res 22:55305539.

31. LiH, ParkSH,Reif JH,LaBeanTH, YanH (2004)DNA-templatedself-assemblyof protein

and nanoparticle linear arrays. J Am Chem Soc 126:418419.

32. Liu Y, Lin C, Li H, Yan H (2005A) Aptamer-directed self-assembly of protein arrays on

a DNA nanostructure. Angew Chem Int Ed 44:43334338.

33. CoxJC, EllingtonAD (2001)Automated selectionof anti-proteinaptamers. Bioorg Med

Chem 9:25252531.

34. Xu W, EllingtonDA (1996) Anti-peptide aptamersrecognize amino acidsequence and

bind a protein epitope. Proc Natl Acad Sci USA 93:74757480.

35. Janshoff A, Neitzert M, Oberdorfer Y, Fuchs H (2000) Force spectroscopy of molecular

systems- Singlemoleculespectroscopyof polymersand biomolecules.AngewChemInt

Ed39:32123237.

36. Lee GU, Chrisey LA, Colton RJ (1994) Direct measurement of the forces between

complementary strands of DNA. Science 266:771773.

37. Sattin BD,PellingAE, GohMC (2004) DNAbase pairresolutionby single moleculeforce

spectroscopy. Nucleic Acids Res 32:48764883.

38. Strunz T, Oroszlan K, Schafer R, Guntherodt HJ (1999) Dynamic force spectroscopy of

single DNA molecules. Proc Natl Acad Sci USA 96:1127711282.

39. Basnar B, Elnathan R, Willner I (2006) Following aptamer-thrombin binding by force

measurements. Anal Chem 78:36383642.

40. Brough B, et al. (2006) Evaluation of synthetic linear motor-molecule actuation ener-

getics. Proc Natl Acad Sci USA 103:85838588.41. Chhabra R, et al. (2007) Spatially addressable multiprotein nanoarrays templated by

aptamer-tagged DNA nanoarchitectures. J Am Chem Soc 129:1030410305.

42. Tripp SL,PusztaySV, RibbeAE, WeiA (2002) Self-assemblyof cobalt nanoparticlerings.

J Am Chem Soc 124:79147915.

43. Horcas I, et al. (2007) WSxM: A software for scanning probe microscopy and a tool for

nanotechnology. Rev Sci Instr 78:013705.

44. Hutter JL, Bechhoefer J (1993) Calibration of atomic-force microscope tips. Rev Sci

Instrum 64:18681878.

45. Schonherr H etal. (2000) Individual supramolecularhost-guest interactionsstudied by

dynamic single molecule force spectroscopy. J Am Chem Soc 122:49634967.


Yossi Weizmann et al- A polycatenated DNA scaffold for the one-step assembly of hierarchical...

Documents

Transcript of Yossi Weizmann et al- A polycatenated DNA scaffold for the one-step assembly of hierarchical...