Natural Computing Series pdf

Part I DNA Nanotechnology – Algorithmic Self-assembly Scaffolded DNA Origami: from Generalized Multicrossovers to Polygonal Networks Paul W.K.. Second, I propose a new method for using sc

Series Editors: G Rozenberg

Th Bäck A.E Eiben J.N Kok H.P Spaink

Leiden Center for Natural Computing

Advisory Board: S Amari G Brassard K.A De Jong

C.C.A.M Gielen T Head L Kari L Landweber T Martinetz

Z Michalewicz M.C Mozer E Oja Gh Paun J Reif H Rubin

A Salomaa M Schoenauer H.-P Schwefel C Torras

D Whitley E Winfree J.M Zurada

°

Natural Computing Series

N

Junghuei Chen · Nataša Jonoska

Grzegorz Rozenberg (Eds.)

123

Nanotechnology: Science and

Computation

With 126 Figures and 10 Tables

Library of Congress Control Number: 2005936799

ACM Computing Classification (1998): F.1, G.2.3, I.1, I.2, I.6, J.3

ISBN-10 3-540-30295-6 Springer Berlin Heidelberg New York

ISBN-13 978-3-540-30295-7 Springer Berlin Heidelberg New York

This work is subject to copyright All rights are reserved, whether the whole or part of the material

is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer Violations are liable for prosecution under the German Copyright Law.

Springer is a part of Springer Science+Business Media

Cover Design: KünkelLopka, Werbeagentur, Heidelberg

Typesetting: by the Editors

Production: LE-TEX Jelonek, Schmidt & Vöckler GbR, Leipzig

Printed on acid-free paper 45/3142/YL – 5 4 3 2 1 0

University of South Florida

4202 E Fowler Av., PHY114

Leiden University Niels Bohrweg 1

2333 CA Leiden, The Netherlands A.E Eiben

Vrije Universiteit Amsterdam The Netherlands

This book is dedicated to Nadrian C Seeman

on the occasion of his 60th birthday

This image was created by DADARA

Nanotechnology is slowly and steadily entering more and more aspects of ourlife It is becoming a base for developing new materials as well as a base fordeveloping novel methods of computing As natural computing is concernedwith information processing taking place in or inspired by nature, the ideascoming from basic interactions between atoms and molecules naturally becomepart of these novel ways of computing

While nanotechnology and nanoengineering have flourished in recent years,the roots of DNA nanotechnology go back to the pioneering work of Nadrian(Ned) C Seeman in the 1980s Many of the original designs and constructions

of nanoscale structures from DNA developed in Ned’s lab provided a pletely new way of looking at this molecule of life Starting with the synthesis

com-of the first immobile Holliday junction, now referred to as J1, through thedouble and triple cross-over molecules, Ned has shown that DNA is a pow-erful and versatile molecule which is ideal for building complex structures atthe nanometer scale

Through the years, Ned has used some of the basic DNA motif tures as ‘tinkertoy’ or ‘lego’ units to build a cube, two-dimensional arrays,and various three-dimensional structures, such as Borromean rings, nanome-chanical devices, nano-walkers (robots), etc All of them were designed anddemonstrated originally in Ned’s lab, but then all these ideas and designs werefollowed up by many other researchers around the world

struc-Adleman’s seminal paper from 1994 provided a proof of principle thatcomputing at a molecular level, with DNA, is possible This led to a realexplosion of research on molecular computing, and very quickly Ned’s ideasconcerning the design and construction of nanoscale structures from DNAhad a profound influence on the development of both the theoretical and theexperimental foundations of this research area

Ned is a scientist and a chemist in the first place Although Ned can

be considered the founder of the DNA nanoengineering field, he has alwaysconsidered himself as a chemist who is interested in basic science Therefore,

he is still very interested in the basic physical properties of DNA and enzymes

VIII Preface

that interact with nucleic acids Ned has been continuously funded by NIH foralmost 30 years and is still providing valuable insights into the DNA and RNAbiophysical and topological properties as well as the mechanism of homologousrecombination between two chromosomal DNAs

Ned’s enormous influence extends also to service to the scientific munity Here one has to mention that Ned is the founding president of theInternational Society for Nanoscale Science, Computation and Engineering(ISNSCE) The respect that Ned enjoys is also manifested through varioushonors and awards that he has received — among others the Feynman Prize

com-in Nanotechnology and the Tulip Award com-in DNA Computcom-ing

Besides science, Ned is very much interested in the world around him, e.g.,

in art Amazingly, some of this interest has also influenced his scientific work:

by studying the work of Escher he got some specific ideas for constructions ofDNA-based nanostructures! Ned is an excellent lecturer and has given talksaround the world, thereby instigating significant interest and research in DNAnanotechnology and computing

With this volume, which presents many aspects of research in basic ence, application, theory and computing with DNA molecules, we celebrate ascientist who has been a source of inspiration to many researchers all over theworld, and to us a mentor, a scientific collaborator, and a dear friend

Nataˇsa JonoskaGrzegorz Rozenberg

Part I DNA Nanotechnology – Algorithmic Self-assembly

Scaffolded DNA Origami: from Generalized Multicrossovers

to Polygonal Networks

Paul W.K Rothemund 3

A Fresh Look at DNA Nanotechnology

Zhaoxiang Deng, Yi Chen, Ye Tian, Chengde Mao 23

DNA Nanotechnology: an Evolving Field

Hao Yan, Yan Liu 35

Self-healing Tile Sets

Erik Winfree 55

Compact Error-Resilient Computational DNA Tilings

John H Reif, Sudheer Sahu, Peng Yin 79

Graphs

Giuditta Franco, Nataˇsa Jonoska 105

Part II Codes for DNA Nanotechnology

Finding MFE Structures Formed by Nucleic Acid Strands in

a Combinatorial Set

Mirela Andronescu, Anne Condon 121

Involution Solid Codes

Lila Kari, Kalpana Mahalingam 137

Part III DNA Nanodevices

DNA-Based Motor Work at Bell Laboratories

Bernard Yurke 165

Jong-Shik Shin, Niles A Pierce 175

Part IV Electronics, Nanowire and DNA

A Supramolecular Approach to Metal Array Programming

Using Artificial DNA

Mitsuhiko Shionoya 191

Multicomponent Assemblies Including Long DNA and

Nanoparticles – An Answer for the Integration Problem?

Andreas Wolff, Andrea Csaki, Wolfgang Fritzsche 199

Molecular Electronics: from Physics to Computing

Yongqiang Xue, Mark A Ratner 215

Part V Other Bio-molecules in Self-assembly

Towards an Increase of the Hierarchy in the Construction

of DNA-Based Nanostructures Through the Integration of

Inorganic Materials

Bruno Samor`ı, Giampaolo Zuccheri, Anita Scipioni, Pasquale De Santis 249

Adding Functionality to DNA Arrays: the Development of

Semisynthetic DNA–Protein Conjugates

Christof M Niemeyer 261

Bacterial Surface Layer Proteins: a Simple but Versatile

Biological Self-assembly System in Nature

1 Adapted with permission (Table 1, Figs 1–3, and associated text) from J Am.

Chem Soc 2004, 126, 10834–10835 Copyright 2004 American Chemical Society.

Contents XI

Part VI Biomolecular Computational Models

Computing with Hairpins and Secondary Structures of DNA

Masami Hagiya, Satsuki Yaegashi, Keiichiro Takahashi 293

Bottom-up Approach to Complex Molecular Behavior

Milan N Stojanovic 309

Aqueous Computing: Writing on Molecules Dissolved in

Water

Tom Head, Susannah Gal 321

Part VII Computations Inspired by Cells

Turing Machines with Cells on the Tape

Francesco Bernardini, Marian Gheorghe, Natalio Krasnogor, GheorgheP˘aun 335

Insights into a Biological Computer: Detangling Scrambled

Genes in Ciliates

Andre R.O Cavalcanti, Laura F Landweber 349

Modelling Simple Operations for Gene Assembly

Tero Harju, Ion Petre, Grzegorz Rozenberg 361

Part VIII Appendix

Publications by Nadrian C Seeman

377

Part I

DNA Nanotechnology – Algorithmic

Self-assembly

Scaffolded DNA Origami: from Generalized Multicrossovers to Polygonal Networks

Ned’s DNA sculptures did turn out to have a relationship to tion In 1994, Len Adleman’s creation of a DNA computer [1] showed thatlinear DNA self-assembly, together with operations such as PCR, could tackleNP-complete computational problems Excited by this result, Erik Winfreequickly forged an amazing link that showed how the self-assembly of geo-metrical DNA objects, alone, can perform universal computation [21] Thedemonstration and exploration of this link have kept a small gaggle of com-puter scientists and mathematicians tangled up with Ned and his academicchildren for the last decade At an intellectual level, the technical achieve-ments of the resulting collaborations and interactions have been significant,among them the first two-dimensional DNA crystals [22] and algorithmic self-assembly of both linear [7] and two-dimensional [10] arrays By various otherpaths, a number of physicists have joined the party, mixing their own ideaswith Ned’s paradigm of “DNA as Tinkertoys” to create nanomechanical sys-tems such as DNA tweezers [26] and walkers [25, 17, 20] DNA nanotechnologyhas taken on a life of its own since Ned’s original vision of DNA fish flying

computa-in an extended Escherian lattice [14], and we look forward to a new “DNAworld” in which an all-DNA “bacterium” wriggles, reproduces, and computes

On a personal level, I and many others have gotten to find out exactlywhat kind of twisted genius Ned is Ned is a singular character He is atonce gruff and caring, vulgar and articulate, stubborn and visionary Ned

is generous both with his knowledge of DNA and his knowledge of life His

4 P.W.K Rothemund

life’s philosophy includes a strong tension between the abysmally negative(the general state of the world) and the just tolerably positive (that whichone can, with great effort, hope to achieve) To paraphrase and to whitewash,

“In a world full of execrable excrescences, there is always a fetid coprostasis

of an idea to make your own.” Once one is correctly calibrated to Ned, thissuperficially gloomy counsel becomes positively bright and Ned’s success withDNA nanotechnology serves as an example for the young scientist In fact,Ned’s education of young scientists reveals a latent optimism As an advisorNed plots a strategic course, giving graduate students projects with risks andpayoffs calculated to help them succeed at every stage — from confidencebuilders in their first years to high-risk/high-gain projects in later years.Ned’s own relationship with science is equally telling of his character He ishealthily (and vocally) paranoid about Nature’s determination to screw up hisexperiments To combat this, he practices a capricious paganism, frequentlyswitching between gods in the hope that one will answer his prayers for ahighly-ordered three-dimensional DNA crystal (A habit which he attemptedunsuccessfully to break when he abandoned crystallography.) Such supersti-tion is tongue-in-cheek, however, and Ned is one of the most careful scientiststhat I know He is ever-mindful that, as Peter Medawar wrote, “research issurely the art of the soluble” and, while his highly imaginative research isconstructive and nonreductionist in its goals, Ned makes sure that it rests onfalsifiable Popperian bedrock

In celebration of Ned the character, as well as the box of Tinkertoys andLegos that he has created, I cover two topics First, I review the recent gener-alization of Ned’s geometry of parallel crossovers to the creation of arbitraryshapes and patterns via a method called scaffolded DNA origami I give anexample pattern with roughly 200 pixels spaced 6 nm apart Second, I propose

a new method for using scaffolded DNA origami to make arbitrary nal networks, both two-dimensional planar stick figures and three-dimensionalpolyhedra

polygo-1 Scaffolded DNA Origami for Parallel Multicrossovers

Fig 1a,b show one of the most successful of Ned’s noncanonical DNA motifs,

a “double-crossover” molecule [4] fashioned from two parallel double helicaldomains that comprise four distinct strands of DNA Each DNA strand windsalong one helix for a number of bases before switching to the other helix bypassing through a structure called a “crossover” (small black triangles) Be-cause strands reverse direction at the crossovers, the crossovers are termed

“antiparallel” It is the juxtaposition of two crossovers that holds the helices

in their parallel arrangement (isolated crossovers assume an equilibrium

rigidly together (isolated crossovers are floppy) These properties allow doublecrossovers to assemble into large extended lattices [22], and nanotubes [12]

Scaffolded DNA Origami 5

C T G A

C A G

C G C C

T

T T T

T T T T

single-stranded origami

c

G A C T G

C T G A C

C A G T

C C G C C C

A G T

G G C T T

C C G A A

C C

G G C G

G G C

T A

A T

2<GATGGCGT CCGTTTAC AGTCGAGG ACGGATCG>3

1>TCACTCTACCGCA GGCAAATG TCAGCTCC TGCCTAGCTCACT<4

1<TAGAGGTAAGACC TGCGGTAT AGATAGCA GGCTACTGGAGAT>4

2>CATTCTGG ACGCCATA TCTATCGT CCGATGAC<3

1

4

Fig 1 Double-crossover molecules, and flavors of DNA design.

The idea of holding helical domains in a parallel arrangement via thejuxtaposition of antiparallel crossovers has become a general principle in DNAnanotechnology, used in at least a dozen constructions For example, it hasbeen extended to molecules with three parallel helices [6], and it has beenused to attach triangles rigidly to a nanomechanical device [23]

A key question is how to create generalized multicrossover molecules withparallel helices To answer this question, it is necessary to understand theadvantages and disadvantages of different approaches Within the DNA nan-otechnology paradigm, designs may be classified by how they are built upfrom component strands, being (1) composed entirely of short oligonucleotidestrands as in Fig 1c, (2) composed of one long “scaffold strand” (black) andnumerous short “helper strands” (colored) as in Fig 1d, or (3) composed

of one long strand and few or no helpers as in Fig 1e Here these designapproaches are termed “multistranded”, “scaffolded”, and “single-stranded”,respectively The last two are termed “DNA origami” because a single longstrand is folded, whether by many helpers or by self-interactions

Multistranded designs (such as Ned’s original cube) suffer from the ficulty of getting the ratios of the component short strands exactly equal Ifthere are not equal proportions of the various component strands, then in-complete structures form and purification may be required Because, for largeand complex designs, a structure missing one strand is not very differentfrom a complete structure, purification can be difficult and may have to beperformed in multiple steps Single-stranded origami such as William Shih’soctahedron [19] cannot, by definition, suffer from this problem Scaffoldedorigami sidesteps the problem of equalizing strand ratios by allowing an ex-cess of helpers to be used As long as each scaffold strand gets one of each

dif-6 P.W.K Rothemund

c

b a

x y

Fill the shape

with helices and

a periodic array

of crossovers.

Raster fill helices with a single long scaffold strand.

Add helper strands

to bind the scaffold

Fig 2 Design of DNA origami.

helper, all scaffolds may fold correctly (some might get trapped in ings) Because origami are easily differentiable from the helpers, separatingthem is not difficult (e.g large origami stick much more strongly to micasurfaces than do tiny helpers and so excess helpers can be washed away).Single-stranded origami and scaffolded origami thus seem the best can-didates for the creation of large, complex structures As Shih has observed(personal communication), the geometry used for the octahedron should gen-eralize and allow the creation of arbitrary polygonal networks However, the

misfold-Scaffolded DNA Origami 7

use of single-stranded origami to create parallel multi-crossover designs seemsdifficult (but perhaps only to me)

Generalization of the parallel helical geometry introduced by crossover molecules is simple using scaffolded DNA origami; I have recentlydemonstrated a technique for the creation of six arbitrary shapes and sixarbitrary patterns (including the one shown here); the design method andexperiments showing its generality are described in [11] To get a feeling forthe method, look at Fig 2 Shapes are approximated by laying down a series

double-of parallel helical domains inside the shape (Fig 2a) Helices are cut to fitthe shape, in a series of sequential pairs from top to bottom, so that the re-

in the y-direction To make a molecular design, a scaffold is run exactly oncethrough each helix; performed in a raster-fill manner, this creates a “foldingpath” (Fig 2b) To hold the scaffold in this shape, helper strands are added

to create a regular pattern of antiparallel crossovers (Fig 2c)

a

Fig 3 Several folding paths (top) drawn without helper strands, and predicted

structures (bottom) that use an ∼7000-base-long scaffold Colors indicate the baseposition on the scaffold, from 1 (red–orange) to 7000 (purple) Arrows indicateseams, which are bridged by helper strands for mechanical stability Scale bar,

8 P.W.K Rothemund

200 DNA strands for a final molecular weight of 15,000 nucleotides Thus theseDNA origami have a molecular weight 100 times that of the original double-crossover and almost 6 times larger than Ned’s largest geometric construction,

a truncated octahedron [27] Further, such scaffolded origami are created in a

Given a shape, such as the rectangle in Fig 4a,b, it is simple to decorate itwith an arbitrary pattern of binary pixels The position of each helper strand(of which there are roughly 200) is considered to be a pixel The original set

of helper strands is taken to represent binary ‘0’s To represent binary ‘1’s

a new set of labeled helper strands is constructed; so far, they have beenlabeled with extra DNA hairpins To create a desired pattern (say Fig 4c),the appropriate complementary sets of strands are drawn from the originalhelper strands and the labeled helper strands Everywhere the pattern has a

‘0’, an original helper strand is used; everywhere the pattern has a ‘1’, a newhelper strand is used Creating the mixture of strands for a desired patternrequires about 1.5 hours of pipetting

e d

Fig 4 An arbitrary pattern The white features are DNA hairpins The black scale

bar in (a) applies to (b,c) and (e) as well Scale bars, both black and white, 100

Scaffolded DNA Origami 9

tall (letters half this height are shown in [11]) Roughly 50 billion copies ofthe pattern were made; copies stick to each other along their vertical edgesvia blunt-end stacking Note that the pattern clearly shows the influence ofNed on DNA nanotechnology

Because scaffolded DNA origami makes the creation of arbitrary shapesand patterns so simple, and because it provides the ability to pattern at the

6 nm length scale, scaffolded origami has the potential to play an importantrole in future lithographic techniques for nanocircuits and other nanodevices

2 DNA Origami for Polygonal Networks

Given the ease with which scaffolded origami generalizes parallel crossovers,the question becomes, “what other general methods of creating shapes mightthere be?” The first thing that would probably spring to a geometer’s mind isthe use of polygons Indeed an attempt to create polygonal networks – DNAstick figures – was where Ned began his quest for 3D structure [14, 15] Hisoriginal vision was to “trash the symmetry” of DNA branch junctions to cre-ate immobile motifs, which could then be assembled into polygonal networksvia sticky ends (Fig 5a,b) Unfortunately, it wasn’t that easy; single-branchedjunctions resisted crystallization into 2D lattices for many years In general,branched junctions formed from single helices are floppy and tend to cyclizeinto families of trimers, tetramers, and higher macrocycles In particular, four-armed branch junctions vacillate between one of two different “stacked-X”

symmetries, one can use specific sticky ends that force a particular tivity, such as the DNA cube [2], but, because of uncertainty in the junctiongeometry, it is still unknown whether the DNA cube was a cube or some otherparallelopiped

connec-It was out of such frustrations that the parallel helical geometry used

by Ned to create the double crossovers was born [4], giving us DNA “Lego”bricks rather than the “Tinkertoy” spools and sticks originally envisioned.DNA lattices were eventually formed from unconstrained four-arm junctionseither by letting the junctions have their way, to create rhomboidal lattices

the junctions to crystallize into lattices of parallel helices [13] None of theseexperiments, however, gets us any closer to Tinkertoys

Recently, in an attempt to create DNA motifs with a square 1:1 aspect

(Fig 5c) By using two DNA helices rather than one for each arm of theirfour-arm motif, and connecting these arms with apparently floppy junctions,Yan and LaBean have created a motif that crystallizes into rectilinear domains

three-arm motifs (Fig 5d), which he calls “3-point stars”, that crystallize beautifully

Fig 5 Ned’s original vision for branch junction lattices, and the motifs that have

succeeded them The sticky-end placement and arm lengths in (c) and (d) are not

accurate; refer to [24, 5] for the actual structures

into 30-micron hexagonal lattices [5] It is amazing that the combination ofsingle covalent bonds and poly-T linkers at the centers of these motifs yieldsstructures rigid enough to form large lattices These successes hint that theprinciple may be generalized to other numbers of arms — and may provide

us with the sticks and spools for DNA Tinkertoys

three-point stars in that it uses two helical domains per arm, that may beused in the context of scaffolded DNA origami to create arbitrary polygonal

Scaffolded DNA Origami 11

networks I begin by describing its use to create arbitrary pseudohexagonalnetworks

.

.

.

.

.

.

.

.

Fig 6 A pseudohexagonal network composed of geometrical 3-stars, and the DNA

3-stars used to build a molecular approximation

Fig 6a shows what is meant by pseudohexagonal networks: planar figurescomposed from the two three-armed components at the left (which I call 3-stars) without rotation or bending I propose that such structures can becreated from scaffolded DNA origami by replacing each geometrical 3-star

12 P.W.K Rothemund

the black strand is intended to be the scaffold strand of a DNA origami, andthe colored strands are helper strands, each 32 nucleotides long DNA 3-starsare classified by the number of “open ends” that they have, i.e the number ofbreaks in the scaffold strand as it travels around the circumference of the DNA3-star Thus DNA 3-stars can be “type-0”, “type-1”, “type-2”, or “type-3”.The type-0 DNA 3-star is the simplest pseudohexagonal network; each arm isclosed at the end by the scaffold as it crosses from one helix of the arm to theother Note that these DNA 3-stars differ from Mao’s 3-point stars (as well as

rather than in the middle of each arm – and thus it is uncertain how DNA3-stars will behave in the laboratory Let us assume for now that they willform well

When two DNA 3-stars abut in a pseudohexagonal network, they can

be joined in one of two ways: either two closed ends meet (Fig 6c, left) ortwo open ends meet (Fig 6c, right) If two closed ends meet then they aremechanically joined by modified helper strands that cross the ends closed by

two open ends meet then they are joined by the scaffold strand – the scaffoldstrand passes along the top helix from right to left, and returns along thebottom helix from left to right Call this structure a “scaffold join” Fig 6dshows the helical representation of both helper and scaffold joins

Given an arbitrary pseudohexagonal network of N 3-stars, a simple gorithm allows a molecular design M to be built up from N DNA 3-stars.Fig 7a shows an example network; Fig 7b shows simplified diagrams of DNA3-stars that show only the scaffold strand and are colored according to theirtype The algorithm begins by placing a type-0 DNA 3-star over a randomlychosen 3-star in the network; Fig 7c,d show one particular choice, and Fig 7eshows another The algorithm proceeds by adding type-1 DNA 3-stars one at

al-a time, until the entire network is covered (Fig 7c–e, step 2 through step 7).Each time a type-1 DNA 3-star is added, it is positioned next to an already-placed DNA 3-star (which such a position may be chosen randomly) and it isfastened to the already-placed DNA 3-star by a scaffold join Thus the type ofthe already-placed 3-star is incremented by 1 (visualized in Fig 7 as a colorchange) If the type-1 DNA 3-star is placed next to two or more already-placedDNA 3-stars (Fig 7d,e, step 7), then it is fastened to one of the DNA 3-stars(chosen randomly) by a scaffold join and to the remaining DNA 3-stars byhelper joins (arrows, Fig 7c–e) Before each addition of a type-1 DNA star,the scaffold is a single closed loop At the end of each addition, the scaffold

1 Technically, this motif should be called a 1.5-turn DNA 3-star; any odd number

of half-turns may be used in the arm

2 Here each helper strand is drawn as binding to 24 bases in one DNA 3-star, and

to eight bases in the other This is by analogy with similar joints in previouslycreated scaffolded origami; what lengths may work the best are unknown

Scaffolded DNA Origami 13

is still a single closed loop Thus the algorithm always generates a design Mthat has a single continuous scaffold strand

Fig 7 A pseudohexagonal network, converted to a molecular design in three

dif-ferent ways Arrows point to helper joins

14 P.W.K Rothemund

As described, the algorithm is nondeterministic and can generate differentfolding paths; the positions of helper and scaffold joins in M depend on the

as in Fig 7, the pattern of scaffold and helper joins seems irrelevant In largedesigns, however, such as those in Fig 8, it is easy to imagine that the pattern

of joins may have a bearing on whether the structures fold correctly or ontheir mechanical stability For example, perhaps local folds may form fasterthan long-distance ones, causing short, wiggly paths to fold more reliably thanlong, straight ones; if this is true then the tree-like folding path of the design inFig 8c might fold more robustly into a triangular figure (Fig 8a) than wouldthe comb-like folding path of the design in Fig 8b Or we might expect thatthe folding path of Fig 8e (for which every radius of the hexagon intersects

at least two covalent scaffold bonds) would yield a more mechanically stableversion of Fig 8d than would the folding path of Fig 8f (for which one radius

of the hexagon – the dotted line – intersects only helper joins) If it is learnedthat the pattern of scaffold and helper joins matters, such information can beincorporated into the design algorithm

Technically, large designs such as those in Fig 8 seem within easy reach (atleast to try) The triangular network (Fig 8a) would require a 5856-base-longscaffold, and the hexagonal ring (Fig 8b) a scaffold 6912 bases long (renderedusing 1.5-turn DNA 3-stars)

While polygonal networks are planar graphs, the objects created with themneed not be planar Fig 9 (top left) reproduces Ned’s proposal for a single-

dodecahedron In this scheme, the single blue strand that winds around thedodecahedron must leave the dodecahedron once per face, and jump to anadjacent face (Fig 9, bottom right, makes this path clear) Ned’s plan was

to cut off these exocyclic arms with restriction endonucleases after the

do-3 Note that the number of scaffold and helper joins in M remains the same,

inde-pendent of the order in which M is built By construction, the number of scaffoldjoins, S, equals N− 1, where N is the number of 3-stars The number of helperjoins, H, is obviously J− S, where J is the total number of joins (determined bythe network geometry) More fun (and perhaps more useful) than counting J or

H is to observe that H is the number of “holes” in the network If the network

is embedded in a plane, the number of holes is the number of unconnected gions that the network divides the plane into, disregarding the region outside ofthe network For example, the network in Fig 8a has 21 holes (small hexagons),and the molecular designs in Fig 8b,c both have 21 helper joins The network inFig 8d has 19 holes (18 small hexagons and 1 large interior hexagonal void) andthe designs in Fig 8e,f both have 19 helper joins The relationship J = S + H =

re-N −1+H is just a restatement of Euler’s theorem for planar graphs V −E+F = 2,where the number of vertices V is equal to N , the number of edges E is equal to

J, and the number of faces F is equal to H + 1 (the number of faces of a graphincludes all the holes, plus the region of the plane outside the graph.)

4 A Schlegel diagram for a polyhedron is just the planar graph associated with that

polyhedron

Scaffolded DNA Origami 15

Fig 8 Given a particular network, folding paths in molecular designs are not

unique Vertically oriented scale bar, 100 nm

decahedron had folded More inconvenient than the surplus arms is that thisstructure is a formal knot – in order for it to fold, the single strand would have

16 P.W.K Rothemund

to be cut (say at the black arrow) and threaded through itself many times (atleast twice per edge as drawn)

Fig 9 Ned’s vision of a single-stranded dodecahedron (Top left: figure credit, Ned

Seeman.) Eleven faces of the dodecahedron are represented as interior pentagons ofthe Schlegel diagram; the twelfth face is the pentagon formed by the outer edges

approach (Fig 10a,b) would allow the dodecahedron to be created without any

5 Shih’s single-stranded approach would also eliminate such knots.

Scaffolded DNA Origami 17

contour and spiraling back in along the black contour More tree-like foldingpaths similar to that of Fig 8c are obviously possible, but it is my intuition

The dodecahedron uses only 12 DNA 3-stars – using the standard 7000-basescaffold would thus allow the use of larger DNA 3-stars with longer arm lengths(and requiring more than two helper strands per arm) Using 5.5-turn DNA3-stars, the edge lengths would be 11 turns (116 bases) and the total scaffoldwould be 6960 bases long Each edge would be 39.4 nm and the diameter of

a sphere enclosing the dodecahedron would be 110 nm

Ned has described his work on geometrical DNA constructs as “pure minster Fuller” Scaffolded origami may now allow the simple construction of

Buck-a “DNA buckybBuck-all” (Fig 10c,d show the Schlegel diBuck-agrBuck-am Buck-and moleculBuck-ar

DNA 3-stars, such an analog would require only a 5760-base scaffold and wouldthus be a little smaller and less complex than current scaffolded designs Car-bon buckyballs are 0.7 nm in diameter – a DNA buckyball would be 50 nm

in diameter and have over 300,000 times the volume Probably too floppy toimage well with atomic force microscopy, DNA buckyballs (and dodecahedra)would have to be characterized by an electron microscopy technique such assingle-particle analysis or electron tomography

While I have so far presented structures created from DNA 3-stars, it

is possible that scaffolded polygonal origami can be created from other

4 molecules are so well-behaved DNA 5-stars tolerant of the appropriateangles would make scaffolded icosahedra possible (5.5-turn DNA 5-stars wouldyield icosohedra with a 75 nm enclosing sphere and a 6960-base scaffold).Eventually, as k increases, a star’s central section is likely to become so floppythat it collapses and admits blunt-ended stacking between pairs of helices inopposing arms My intuition is that this is the major obstacle to high k-starsrendered in DNA Figures made of stars of mixed valence may also be possible.Note that the algorithm for constructing a molecular design (adding type-1stars) is the same for k > 3 and mixed-valence designs; also, the number of

here are all planar graphs, the number of helper joins remains equal to thenumber of holes

It will be interesting to see whether polygonal origami works as well asparallel multicrossover origami in the lab – if so, it will be another example

of a system for creating a general class of DNA shapes With a wealth of

6 This intuition is in opposition to my previous suggestion for why tree-like folding

paths might fold better My imagination is that the more long branches there arefloating about, the higher the probability of unintended catenation, for examplethat two faces of a polyhedron might form in an interlocking manner Lots of

“imaginations” are possible I hope that someday some new technique will allow

us to make movies of the process and give us a real intuition about folding

18 P.W.K Rothemund

Fig 10 A dodecahedron and buckyball designed as scaffolded origami DNA

3-stars are asymmetric and have a distinct ‘top’ and ‘bottom’ face It is unclear if thiswill result in one or two forms (each inside-out of the other) for each polyhedron

structural experience under its belt, the DNA nanotechnology community isexploring such generalized approaches for a variety of motifs For example,William Sherman has proposed a neat framework [18] for the creation ofDNA nanotubes of arbitrary cross section In another example, as discussedabove, William Shih has observed that single-stranded origami may be used to

7 To see this, replace helper joins with paranemic cohesion motifs and scaffold joins

with Shih’s double-crossover struts in all the diagrams of this section

Scaffolded DNA Origami 19

4-stars

5-stars

6-stars

3,4,5,6-stars

Fig 11 Figures constructed using 4-stars, 5-stars and 6-stars.

we would have such a scheme for composing the motif into larger, arbitrarystructures In our attempt to do this, some motifs present stimulating anddifficult challenges Ned’s surprising paranemic crossover DNA [16] might begeneralized to form large sheets with the interesting property that, althoughthey were made from DNA “helices”, no strands would cross from one surface

of the sheet to the other!

Simply proposing a scheme for a general architecture, as this paper hasdone, is not enough A complete generalized approach would have three parts:(1) the definition of an infinite family of DNA shapes, (2) the experimentaldemonstration of a convincing and representative set of examples, and (3) thecreation of automated design tools for that family of shapes The last of theseparts, while seeming a simple matter of software engineering, is of equal impor-tance to the first two It will allow the community of DNA nanotechnologists

to reproduce and extend each other’s work but, of more importance perhaps, itwill allow scientists outside of the community – physicists, chemists, materialsscientists, and biologists – to make and explore DNA nanostructures of their

20 P.W.K Rothemund

own design As we create architectures and tools that put DNA ogy into the hands of the research community at large, it will be exciting tosee the legacy of Ned’s flying DNA fish continue to grow

4 T.-J Fu and N.C Seeman DNA double-crossover molecules Biochemistry,32:3211–3220, 1993

5 Y He, Y Chen, H Liu, A.E Ribbe, and C Mao Self-assembly of hexagonalDNA two-dimensional (2D) arrays Journal of the American Chemical Society,10:1021, 2005

6 T.H LaBean, H Yan, J Kopatsch, F Liu, E Winfree, J.H Reif, and N.C man Construction, analysis, ligation, and self-assembly of DNA triple crossovercomplexes Journal of the American Chemical Society, 122:1848–1860, 2000

using algorithmic self-assembly of DNA triple-crossover molecules Nature,407(6803):493–496, 2000

8 C.D Mao, W.Q Sun, and N.C Seeman Designed two-dimensional DNA liday junction arrays visualized by atomic force microscopy Journal of theAmerican Chemical Society, 121:5437–5443, 1999

Hol-9 A.I.H Murchie, R.M Clegg, E von Kitzing, D.R Duckett, S Diekmann, andD.M.J Lilley Fluorescence energy transfer shows that the four-way DNA junc-tion is a right-handed cross of antiparallel molecules Nature, 341:763–766, 1989

10 P.W.K Rothemund, N Papadakis, and E Winfree Algorithmic self-assembly

of DNA Sierpinski triangles PLoS Biology, 2(12):e424, 2004

11 P.W.K Rothemund Generation of arbitrary nanoscale shapes and patterns byscaffolded DNA origami (submitted), 2005

12 P W K Rothemund, A Ekani-Nkodo, N Papadakis, A Kumar, D.K son, E Winfree Design and characterization of programmable DNA nanotubes.Journal of the American Chemical Society, 26(50):16344–16353, 2004

Fygen-13 P.W.K Rothemund DNA self-assembly with floppy motifs – single crossoverlattices Foundations of Nanoscience, Self-Assembled Architectures and Devices,Proceedings of FNANO’05 (J.H Reif eds.) 185–186, 2005

14 N.C Seeman Nucleic-acid junctions and lattices Journal of Theoretical Biology,99:237–247, 1982

15 N.C Seeman Construction of three-dimensional stick figures from branchedDNA DNA and Cell Biology, 7(10):475–486, 1991

Scaffolded DNA Origami 21

16 Z.Y Shen, H Yan, T Wang, and N.C Seeman Paranemic crossover DNA: Ageneralized Holliday structure with applications in nanotechnology Journal ofthe American Chemical Society, 126:1666–1674, 2004

17 W.B Sherman and N.C Seeman A precisely controlled DNA biped walkingdevice Nanoletters, 4(7):1203–1207, 2004

18 W.B Sherman and N.C Seeman The design of nucleic acid nanotubes Journal

of Biomolecular Structure and Dynamics, 20(6):930–931, 2003

19 W.M Shih, J.D Quispe, and G.F Joyce A 1.7-kilobase single-stranded DNAthat folds into a nanoscale octahedron Nature, 427(6453):618–621, 2004

20 J.S Shin and N.A Pierce A synthetic DNA walker for molecular transport.Journal of the American Chemical Society, 126(35):10834–10835, 2004

21 E Winfree On the computational power of DNA annealing and ligation InR.J Lipton and E.B Baum, editors, DNA Based Computers, DIMACS, AMSPress, Providence, RI, 27:199–221, 1996

22 E Winfree, F Liu, L.A Wenzler, and N.C Seeman Design and self-assembly

of two-dimensional DNA crystals Nature, 394:539–544, 1998

23 H Yan, X Zhang, Z Shen, and N.C Seeman A robust DNA mechanical devicecontrolled by hybridization topology Nature, 415:62–65, 2002

24 H Yan, S.H Park, G Finkelstein, J.H Reif, and T.H LaBean DNA-templatedself-assembly of protein arrays and highly conductive nanowires Science,301:1882–1884, 2003

25 P Yin, H Yan, X.G Daniell, A.J Turberfield, and J.H Reif A unidirectionalDNA walker that moves autonomously along a track Angewandte Chemie In-ternational Edition, 43(37):4906–4911, 2004

26 B Yurke, A.J Turberfield, A.P Mills, Jr., F.C Simmel, and J.L Neumann ADNA-fuelled molecular machine made of DNA Nature, 406:605–608, 2000

27 Y Zhang and N.C Seeman The construction of a DNA truncated octahedron.Journal of the American Chemical Society, 116:1661–1669, 1994

A Fresh Look at DNA Nanotechnology

Zhaoxiang Deng, Yi Chen, Ye Tian, and Chengde Mao

Department of Chemistry, Purdue University, West Lafayette, Indiana 47907, USAmao@purdue.edu

Synthetic DNA structures for nanotechnological applications have experiencedsubstantial success during the past decades benefiting from Seeman and hiscoworkers’ pioneering work In the last few years, some new branches havebeen emerging in this field This review will summarize some recent progress

in the authors’ group

1 Two-Dimensional DNA Triangle Arrays Designed with

a Tensegrity Strategy

Forming crystalline DNA lattices in one, two and even three dimensions haslong been a hot topic of DNA nanotechnology These artificially designedlattices are the basis for a variety of applications The first success with a 2DDNA lattice was achieved in [22] with building blocks of double-crossover (DX)DNA molecules Following that, rhombus motifs, triple-crossover molecules,and a cross motif were also constructed from branched four-arm Hollidayjunctions [19] Here we present a tensegrity strategy for the construction ofwell-structured DNA triangle molecules [14] A DNA triangle consists of threevertices (DNA four-arm junctions) and three sides (DNA duplexes) Althoughindividual four-arm junctions are flexible, the rigidity of the three duplex edgesrestricts the freedom of the component four-arm junctions and only trianglescan form The shape of such a triangle is fully defined by the lengths ofthe three edges By rational use of sticky-end cohesion, we have successfullyassembled triangle arrays in one and two dimensions Fig 1 shows the design

of some such triangle arrays and some atomic force microscope (AFM) images

of such triangle arrays

24 Z Deng, Y Chen, Y Tian, C Mao

Fig 1 DNA triangle arrays Left panel : Schematic representation of one- and

two-dimensional arrays (a) A DNA triangle contains three DNA duplexes, shown as rods

of different colors (b) Strand structure of a DNA triangle Each thin line represents

detailed structure of a triangle vortex (d, e) 1D and (f ) 2D self-assembly of DNA triangles Right panel : AFM images of one-dimensional (a, b) and two-dimensional (c, d) DNA triangle arrays (Reproduced from [14] with permission).

2 DNA Molecular Motors

Molecular motors are a very attractive topic in many scientific fields, becausethey are expected to be mechanical parts of future nanorobots Complemen-tarily to other molecular systems, DNA motors afford rational design, easyconstruction, and, most importantly, good control of the motions that theygenerate Earlier models of DNA motors include a nanomechanical devicebased on B–Z transition upon a change of the ionic strength of a solution[17], and molecular tweezers with their opening and closing controlled by se-quential addition of DNA strands [23] Inspired by those early successes, theauthors’ group has worked intensively in this field Comparison between cel-lular protein motors and macroscale man-made machines leads us to ask fourquestions, as listed below Answering these questions is fundamental for thefurther development of molecular motors

2.1 Can DNA Motors Perform Complicated Motions? —

Modeling Gear Motion at Molecular Scale

Gears have many useful functions such as changing the direction and speed

of movement, and are important parts in real machines It is reasonable toexpect that gears might play similar roles in small motor systems such nanomotors and molecular motors This notion has motivated us to model gears

A Fresh Look at DNA Nanotechnology 25

with circular DNA molecules, which can roll controllably against each other byuse of a strand displacement strategy (Fig 2) [21] A DNA gear has a central,circular single strand of DNA, which base-pairs with three linear DNA strands,leaving three unpaired tails as cogs for the gears When a linker strand L1 isadded to the sample, the two gears are bridged together and become readyfor rolling Upon addition of another linker strand L2, two linkages are built

strand displacement method, a removal strand R1 is then added, which forms

a duplex with strand L1 and strips off L1 from the gear pair This step creates

a continuous rotation of the two gears is realized

Fig 2 Design and rolling mechanism of a pair of molecular gears (a) Structures

of the individual gears C and P indicate DNA strands, and T indicates teeth (b)

Operation of the gears L and R represent linker and removal strands, respectively

L1 and R1 are complementary to each other Both circles remain intact during therolling process The only changed strands are the linker (L) and removal (R) strands.Note that no twisting motion is generated in the central strands during the rollingprocess (Reproduced from [21] with permission)

26 Z Deng, Y Chen, Y Tian, C Mao

2.2 Can DNA Motors Work Autonomously?

A DNA Machine Contains a DNAzyme Domain

Working autonomously is an essential feature for cellular protein motors andman-made macroscale machines It is also desirable for nanomachines to beautonomous This section describes our initial efforts to address this chal-lenge The key of our strategy is the introduction of a DNA enzyme domain,which extracts chemical energy and powers a DNA machine We have testedthis notion first with a construct that performs a simple opening and closingmotion This DNA machine undergoes continuous, autonomous motion in thepresence of a fuel strand The motion is controlled by the addition of a brakestrand [7, 5] Fig 3 illustrates this process This motor has a triangular shape

It contains a V-shaped dual arm spaced by a single DNA strand at the top.The single strand has a special sequence, corresponding to the core part of aDNA enzyme (E) and its flanking recognition arms on both sides When thisstrand is base-paired with its substrate (S), which is a DNA strand with twoRNA bases in the middle, the V-shaped arms will be opened owing to theincrease in the rigidity of the single-stranded linkage after forming a duplexwith its substrate The DNAzyme then cleaves its substrate, and the cleavedproducts are short and dissociate from the DNAzyme, which virtually closesthe two arms of the motor If there is substrate in the solution, the aboveopening and closing process will continue until all substrate molecules (fuel)have been consumed This process can be regulated by addition of a brakestrand (B) The brake molecule is a DNA analog of the substrate, but hasbase pairs extending into the catalytic core of the enzyme The brake strandcan form a slightly longer duplex with the DNAzyme than the substrate does.The DNA machine will preferentially bind brake strands and further incor-poration of fuel strands is then disabled Therefore, the motor will be frozen

in its open state Note that the brake strand has an unpaired tail, which isdesigned for removal of the brake Upon addition of a removal strand (R),the removal strand completely base-pairs with the entire brake strand As aresult, the brake will be removed from the motor system, and the motion ofthe machine will be resumed

A DNAzyme-Containing DNA Walker

Recent work has shown that relatively complex motions can be realized withDNA nanocontructions, including walkers and gears In the following, weshow further that we can introduce a DNAzyme into a DNA walker Such

a DNAzyme-containing walker can move autonomously in either directionalong a linear track in a controllable fashion [20] This design takes advantage

of the RNA-cleaving function of a DNAzyme Details of this walker and itsmechanism of movement are presented in Fig 4 The enzyme strand (the redparts are base-recognizing arms, and the orange part is the catalytic core)

A Fresh Look at DNA Nanotechnology 27

Fig 3 Schematic of an autonomous DNA nanomotor based on a DNA enzyme The

DNA motor consists of two single strands, E and F The strand E contains a 10–23DNA enzyme domain, which is colored purple The strand F has a fluorophore atthe 5 end (labeled as a solid green circle) and a quencher at the 3end (labeled by

a solid black circle) (Reproduced from [5] with permission)

base-pairs with one of its substrates on the linear track The blue dots on thegreen substrates depict the cleavage points, where the RNA bases are located.After the enzyme cleaves the substrate, the shorter product will be releasedinto the solution owing to its relatively weaker bonding This gives an oppor-tunity for the exposed part of the enzyme strand to seek another substratewithin its vicinity Gradually, through branch migration, the whole enzymestrand will shift to the next neighboring substrate on the track, and the aboveprocess will be repeated until the DNA enzyme moves to the other end of thetrack This process can be purposely chosen to start from either end of thetrack and will continue to the opposite end, but the walker cannot move back-wards once it has been determined which end of the track it will start from,because the enzyme will destroy all the substrates that it has passed

2.3 Can DNA Motors Work with Inexpensive Fuel Molecules? — pH-Switched DNA Motor Based on a Duplex–Triplex Transition

In the DNA machines described so far, DNA and RNA are used as fuel,but they quite expensive It would be desirable to use inexpensive commonchemicals as fuel This motivation has guided us to design a pH-triggeredDNA machine Under certain conditions, a DNA triplex rather than a duplex

28 Z Deng, Y Chen, Y Tian, C Mao

Fig 4 Schematic of a walking DNAzyme and its track (a) Principle of walking.

(b) A construction where the walking DNAzyme is at one end of its track Black

lines, template (T); green lines, substrate (S); red–gold lines: a 10–23 DNAzyme;gold lines, the catalytic core Blue dots indicate the bonds to be cleaved by theDNAzyme (Reproduced from [20] with permission)

is a more stable conformation for certain DNA molecule assemblies On thebasis of this phenomenon, a DNA motor with a structure as shown in Fig

5 can be generated [4] This figure shows a DNA assembly that contains along strand L (red) and two short strands (black) in an open and closed state.Strand S forms a duplex with one segment of strand L At pH 8.0, this duplex

is the dominant conformation When the pH is switched to 5.0, one originallydangling single-strand segment within the strand L base-pairs back with theduplex part formed between S and L to give a triplex structure The formation

of this triplex contracts the whole assembly into a closed state In the closedstate, a prelabeled fluorescent dye (green) and a quencher (black) are broughttogether, resulting in efficient fluorescence quenching Therefore, by measuringthe fluorescent emission of the sample, the real-time operation of this motorcan be easily monitored

A Fresh Look at DNA Nanotechnology 29

Fig 5 A DNA nanomotor based on a DNA duplex–triplex transition The DNA

machine consists of three strands: a strand with a fluorescent label (strand F), a longstrand (strand L), and a short strand (strand S) The open and solid circles represent

a fluorophore and a quencher, respectively Note the formation and dissociation of

a DNA triplex involving the S and L strands upon change of the pH of the solution.(Reproduced from [4] with permission.)

2.4 Can DNA Motors Perform any Useful Work? — Programming Chemical Reactions by a Two-State DNA Switch

Various DNA machines have been demonstrated, but there are very few ports of their applications Very fundamentally, we would like to ask: are theyuseful? This section describes one of the few reported examples of an attempt

re-to address this question DNA-templated organic reactions has been pursuedover the years [13]; the following example will demonstrate that a DNA mo-tor can be employed to control the path of a chemical reaction [6] As shown

in Fig 6a, the DNA structure used consists of three strands: a long strand

c (red), which can be divided into three domains, c1, c2, and c3 There areanother two shorter strands, N1 (green) and N2 (blue) that base-pair withdomains c1 and c2, respectively These three strands are modified with eitheramine or carboxylic groups, as indicated in Fig 3 A pH change between 5.0and 8.0 determines whether a triplex structure will be formed or dissociatedbetween domain c3 and the duplex formed between N2 and domain c2 At pH8.0 or 5.0 (Fig 6b), the carboxylic group on strand c will be brought closer

to the amine group on strand N1 or on strand N2 respectively, and thus allow

a corresponding amide bond to be formed at the designated position uponaddition of a condensing reagent

30 Z Deng, Y Chen, Y Tian, C Mao

Fig 6 Schematic illustration of the switching of a chemical reaction based on a

DNA duplex–triplex transition (a) DNA sequences, and the positions of the amino

groups and carboxylate group of interest Note that there is a string of unpairedT6 at the 5 end of strand N2 Addition of the extra six bases to strand N2 causesstrands N1 and N2 to have different molecular weights and electrophoretic mobilities,which allows identification of strands N1 and N2 by polyacrylamide gel electrophore-

sis (PAGE) (b) Switching of a chemical reaction by switching the location of the

carboxylate group This behavior is triggered by a change of the pH value of thesolution Note the formation and dissociation of a DNA triplex (Reproduced from[6] with permission)

3 DNA Encoded One-Dimensional Array of Nanogold

A very challenging aspect of nanotechnology is the development of an cient and potentially universal way to organize nanosized building blocks intodesigned architectures Among the various possible materials, DNA is a su-perior molecule for this purpose for the following reasons: (1) DNA can bemade to form well-defined nanostructures by rational design; (2) DNA can

effi-be chemically modified and operated on by enzymes; (3) DNA itself is an vironmentally benign biochemical reagent By choosing gold nanoparticles asmodel materials for the assembly, we have demonstrated the successful prepa-ration of 1D gold nanoparticle arrays with lengths up to 4 m [10] It has beenpreviously shown that gold nanoparticles can be assembled into small, discretestructures [1, 16, 12, 24] through hybridizing mono-DNA-modified gold parti-cles with a DNA template However, creating a gold nanoparticle array con-taining hundreds of nanoparticles does not seem to be an easy matter because

en-of difficulties with the availability en-of long, single-stranded DNA templates.Fortunately, a rolling-circle DNA polymerization technique [11, 15] developed

A Fresh Look at DNA Nanotechnology 31

ten years ago can help With the help of this rolling-circle polymerization,

we can obtain a single-stranded DNA template with a tandemly repetitive quence defined by the circular DNA template (Fig 7) Also, gold nanoparticlescan be modified with a thiolated single-stranded oligonucleotide, and mono-DNA-modified particles can be isolated simply by agarose gel electrohporesisusing a protocol developed by Alivisatos et al [1, 16, 12, 24] After combiningthe mono-DNA-modified gold particles with the rolling-circle-synthesized longDNA template, 1D gold nanoparticle linear arrays with lengths up to severalmicrometers can be obtained

se-Fig 7 Synthesis of an extended gold nanoparticle array by combining

DNA-encoded self-assembly and rolling-circle polymerization of DNA (Reproduced from[10] with permission)

4 DNA as Templates for Nanofabrication

4.1 Oriented Metallic Nanowire Networks Templated by DNA

Besides the use of self-assembly to form various structures, DNA moleculescan also be used as scaffolds for nanofabrication The first example that wehave demonstrated is related to DNA metallization (Fig 8) It is known thatlambda-phage DNA, a linear DNA with a natural length of 16μm, can bealigned on a surface [2] It is also known that DNA strands can be metallizedthrough electroless metal reduction in solution or on a surface [3, 18] Thisprovides a fundamental possibility of fabricating 1D or 2D oriented metalwire networks The method described here integrates a molecular combingtechnique and DNA metallization [8] In the first step, DNA is aligned on amica surface by a fluid flow in the presence of magnesium ions, which enhancethe binding between the DNA and the mica surface and thus minimize DNAdetachment during the metallization process Note that both the DNA and the

32 Z Deng, Y Chen, Y Tian, C Mao

mica surface are negatively charged After alignment, the DNA sample is thenused for metal deposition Palladium was chosen as the metal for this purpose

To avoid the formation of nanowires with abundant branches, removal of thepalladium solution from the surface before adding the reduction bath solution

is helpful The incubation time for the reduction process must be controlledwithin a range of several minutes to as short as tens of seconds, otherwiseDNA will begin to detach from the surface, and the originally created DNAnetwork structures will be partially destroyed

Fig 8 AFM images of 2D aligned Pd nanowires (a) and the corresponding precursor

DNA molecules (b) The insets in (a) and (b) give closer views of a 2D square of metal nanowires and of DNA molecules, respectively Height scale: (a) 30 nm and (b) 3.0 nm (Reproduced from [8] with permission).

4.2 Molecular Lithography with DNA Nanostructures

Another example of DNA-templated nanofabrication is the molding of DNApatterns with a metal film, resulting in a negative replica of the DNA structure(Fig 9) [9] The first step in realizing the replication of DNA nanostructures

is to deposit DNA samples onto a mica surface Since mica has an atomicallyflat surface, DNA structures on the mica surface show significant topographicpatterns even though they are only about 1 nm high Immediately after thesample has been deposited and the surface has been dried, a layer of goldmetal is thermally evaporated onto the mica surface until a continuous filmwith a thickness of about 20 nm is formed The weak bonding between thegold film and the mica surface offers the possibility to easily peel off thegold film to release the replica By this strategy, DNA structures, both one-dimensional and two-dimensional can be successfully transferred to a metalsubstrate Since good control over the DNA structure could be achieved byrational design, it is not a dream that in the future we might use the replicated