Engineering a DNA World

California Institute of Technology
Center for Biological Circuit Design
Rock Auditorium, Broad Center for Biological Sciences
Pasadena, California

January 6 - 8, 2005


 

home program / speakers travel / accommodations registration contacts

 

program
posters
biosketches
abstracts

program

click on name for homepage
Thursday
8:00 am check-in and continental breakfast
8:55 am welcome
Session Chair: Stojanovic
9: 00 am

Ned Seeman
New York University
[bio] [abstract]

DNA: Not Merely the Secret of Life

Doug Turner
University of Rochester
[bio] [abstract]
Predicting RNA Secondary Structure

10:20 am coffee break
10:40 am

Gerald Joyce
Scripps Research Institute
[bio] [abstract]

Evolution of a DNA Enzyme from an RNA Enzyme

Erik Winfree
California Institute of Technology
[bio] [abstract]
Algorithmic Self-Assembly of DNA

Noon lunch and discussion (unstructured)
Session Chair: Winfree
2:30 Bernard Yurke
Lucent Bell Labs
[bio] [abstract]

Using DNA Nanomachines to Control the Bulk Mechanical Properties of Materials

Andy Ellington
University of Texas, Austin
[bio] [abs]
Requiem for a DNA World

Chengde Mao
Purdue University
[bio] [abstract]
Self-Assembly of DNA Nanostructures

4:30 poster session, cocktails
7:00 dinner in Old Pasadena (unstructured)
Friday
8:30 am continental breakfast
Session Chair: Seeman
9: 00 am

Ron Breaker
Yale University
[bio] [abstract]

Natural and Engineered Riboswitches

Paul Rothemund
California Institute of Technology
[bio] [abstract]
Design and Characterization of Programmable DNA Nanotubes

10:20 am coffee break
10:40 am

Niles Pierce
California Institute of Technology
[bio] [abstract]
Computational Nucleic Acid Engineering

Hao Yan
Arizona State University
[bio] [abstract]

DNA-Based Molecular Engineering: Pattern and Motion

Noon lunch and discussion (unstructured)
Session Chair: Joyce
2:30

Milan Stojanovic
Columbia University
[bio] [abstract]
Agile and Intelligent DNA Molecules

John Reif
Duke University
[bio] [abstract]
Self-Assembled DNA Nanostructures for Molecular Scale Patterning, Computation, and Motors

William Shih
Harvard University
[bio] [abstract]
Clonable DNA Nanotechnology

4:30 coffee break
4:45 panel discussion
Creating a DNA World: Science and Ethics
Leader: Joyce
Panel: Adleman, Ellington, Shih, Yurke
6:30 cocktails, Athenaeum
7:00 dinner, Athenaeum
Saturday
8:30 am continental breakfast
Session Chair: Schuster
9: 00 am

John SantaLucia
Wayne State University
[bio] [abstract]
Progress Toward Accurate 3D Structure Prediction of Nucleic Acids

Anne Condon
University of British Columbia
[bio] [abstract]
On Algorithms and Energy Models for Predicting Pseudoknotted DNA and RNA Secondary Structures

10:20 am coffee break
10:40 am

Hervé Isambert
Institut Curie - Section de Recherche
[bio] [abstract]
Designing DNA and RNA Switches under Out-of-Equilibrium Folding Control

Shi-Jie Chen
University of Missouri, Columbia
[bio] [abstract]
Exploring the Sequence-Dependent Complex Folding Kinetics of RNA Hairpins

Noon lunch and discussion (unstructured)
Session Chair: Pierce
2:30

Ashish Goel
Stanford University
[bio] [abstract]
Counting Using DNA Self-Assembly

Deborah Fygenson
University of California, Santa Barbara
[bio] [abstract]
Understanding and Controlling Self-Assembly of Nanotubes from DNA Tiles

Peter Schuster
University of Vienna
[bio] [abstract]
Designing Single- and Double-Stranded Nucleic Acids

4:30 coffee break
4:45 panel discussion
Engineering Tools for Nucleic Acid Nanoscience
Leader: Turner
Panel: Isambert, Mabuchi, SantaLucia, Schuster
6:00 thanks and goodbye!

posters

Poster dimensions: we can accommodate up to 3 feet by 3 feet.

biosketches and abstracts

Dr. Leonard Adleman received his Ph.D. in Computer Science from the University of California at Berkeley.   He was on the faculty of Mathematics at MIT before Joining the Computer Science Department at the University of Southern California in 1980.  He is best known for his co-discovery with Ron Rivest and Adi Shamir of the RSA public-key cryptosystem.  In 1994 he conducted the first DNA computation.  Professor Adleman is a member of the National Academy of Engineering and is the recipient of numerous awards including the ACM Turing Award.   He is the head of the Laboratory for Molecular Science at USC wherein he and his colleagues carry out experimental and theoretical research on self-assembly.

[top]

Dr. Ronald R. Breaker is a Professor in the Department of Molecular, Cellular and Developmental Biology at Yale University. His graduate studies at Purdue University were carried out with Dr. Peter Gilham and focused on the synthesis of RNA and the catalytic properties of nucleic acids. As a postdoctoral fellow at The Scripps Research Institute, Dr. Breaker worked with Dr. Gerald Joyce to develop a variety of in vitro evolution strategies for isolating novel nucleic acid enzymes. These efforts led to the discovery of catalytic DNAs or “deoxyribozymes”. Since establishing his laboratory at Yale in 1996, Dr. Breaker has continued to conduct research on the advanced functions of nucleic acids, including ribozyme reaction mechanisms, molecular switch technology, next-generation biosensors, and catalytic DNA engineering. Most recently, his laboratory has established the first proof that metabolites are directly bound by messenger RNA elements called riboswitches. In 2001, Dr. Breaker co-founded Archemix, a biotechnology company that is pursuing the commercial development of molecular switch technology for therapeutics and next-generation biosensor applications.

Natural and Engineered Riboswitches
Abstract: Biological systems must accurately sense metabolites to maintain a delicate balance of complex biochemical processes. Although most known genetic factors that bind metabolites are made of protein, we have recently identified numerous RNA domains that selectively bind target compounds and control gene expression. These “riboswitches” serve as precision genetic switches that permit genes to be regulated in response to changing conditions with performance characteristics that are similar to those of protein factors. The number of known riboswitches also continues to expand, and includes novel forms of gene control such as metabolite-induced ribozyme activation. Furthermore, many organisms use riboswitches to control key metabolic pathways, suggesting that this mechanism of genetic control is ancient and might represent a biochemical “lost world” of RNA.

Prior to the discovery of natural riboswitches, we had embarked on a series of molecular engineering studies to explore the potential for RNA and DNA to function as molecular switches. Both modular rational design and directed evolution strategies have been used to create numerous RNA switches that combine the functions of aptamer and ribozyme domains. Some of these engineered ribozymes have properties that are strikingly similar to some of the more complex natural riboswitches identified recently. For example, one riboswitch class makes use of a metabolite-responsive self-cleaving ribozyme, whereas many engineered RNA switches harness the activity of the hammerhead self-cleaving ribozyme. Also, a natural glycine-sensing riboswitch uses cooperative ligand binding similar to that of an RNA switch we engineered previously. The utility and mechanisms of both natural and engineered RNA switches will be discussed.

[top]

The long-term goal of the research in Dr. Shi-Jie Chen's lab is to uncover the fundamental mechanism of RNA folding/misfolding, and, eventually, to predict RNA structures and conformational changes from sequences using theoretical and computational methods. Specifically, the research has addressed (1) the development of computational models that are detailed enough to capture the microscopic details of RNA tertiary structures, but is simple enough to permit efficient calculations for the folding thermodynamics, (2) the accurate prediction of the kinetic pathways, rates, intermediates and transition states of RNA folding based on the complete conformational ensemble, and (3) the electrostatic interactions in RNA folding, including the effect of electrostatic fluctuations and correlations in multivalent ion salt solutions.

Exploring the Sequence-Dependent Complex Folding Kinetics of RNA Hairpins
Abstract:
Previous understanding of hairpin folding kinetics is limited to a few isolated experimental and computational studies. We go beyond the previous studies by exploring much broader and complex RNA hairpin folding landscapes. The complexity arises from the interplay between the formation of the loops, the disruption of the misfolded states, and the formation of the rate-limiting base stacks. The model is validated through direct tests against several experimental measurements. The study reveals the general physical mechanism for RNA hairpin folding kinetics as well as the dependence of the kinetics on the sequence/length of the hairpin loop and the helix stem. For example, (1) the folding would slow down when a stable GC base pair moves to the middle of the stem, (2) an hairpin with GC base pair close to the loop would fold/unfold faster than the one with GC close to the tail of the stem, (3) within certain range of the stem length, longer stem can cause faster folding, and (4)certain misfolded states can assist folding through the formation of scaffold structures to lower the entropic barrier for the folding. The results from the present study suggest that (1) previous experimental findings based on the individual hairpins revealed only a small fraction of much broader and more complex RNA hairpin folding landscapes, (2) even for structures as simple as hairpins, universal folding time scale and pathways do not exist, and (3) to treat the loop size as the sole factor to determine the hairpin folding rate is an oversimplification. In addition, our results can be useful for molecular design to achieve desirable fast/slow-folding hairpins, hairpins with/without specific misfolded intermediates, and hairpins that fold along designed pathways.

[top]

Dr. Anne Condon is a Professor in the Department of Computer Science at U. British Columbia. Her research interests are in computational complexity theory, algorithms, and their applications, including new computational methods for prediction and design of nucleic acid structures. Anne received her B.Sc. degree (1982) in Mathematics and Computer Science at University College Cork, Ireland and her Ph.D. (1987) from the University of Washington, Seattle. Prior to her position at UBC, she was a faculty member of the Computer Sciences Department at U. Wisconsin at Madison from 1987-1999.  She won an ACM Distinguished Dissertation award (1988), an NSF National Young Investigator Award (1992) and the University College Cork Distinguished Alumna award (2001) for her work.

On Algorithms and Energy Models for Predicting Pseudoknotted DNA and RNA Secondary Structures
Abstract: Algorithms based on free energy minimization or partition function calculation are widely used to predict the secondary structure of a DNA or RNA molecule from its base sequence.  There is a strong interplay between the quality and efficiency of an algorithm and its underlying free energy model.  Currently, there is quite a bit of variation in the energy models used for pseudoknotted secondary structure prediction. We will describe the energy models of dynamic programming algorithms for pseudoknotted secondary structure prediction as sum-of-loops models, show how the energy of a structure can be calculated in linear time, and comment on other properties of these models.  We will also discuss the generality of dynamic programming algorithms, in terms of the types of pseudoknots that can be handled by the algorithms.

[top]

Research in the laboratory of Dr. Andrew Ellington, Fraser Professor of Chemistry and Biochemistry at UT Austin, uses in vitro evolution to engineer biopolymers and cells. The Ellington lab has developed diagnostic and therapeutic reagents known as aptamers (nucleic acid binding species) and aptazymes (allosteric ribozymes). The lab focuses on real-world applications of these reagents, including inhibiting HIV replication and detecting biothreat agents such as the toxin ricin. The wholescale manipulation of organismal genomes is also being undertaken in order to engineer cellular signal transduction pathways and generate organismal ‘sentinels’ that can sense environmental signals, and develop organisms with unnatural genetic codes, proteomes, and biochemistries. These various manipulations are culminating in the development of synthetic biology tools and approaches, including the engineering of new virus-like replicators. The lab maintains core interests in evolution, robotics, analytical biochemistry, bioinformatics, and functional genomics.

Requiem for a DNA World
Abstract: The use of nucleic acids as tools, in circuits, and as materials is recommended by several engineering features, including the simplicity of Watson-Crick base-pairing as a means of design, the possibility that conformational change can be engineered in a quantitative way, the possibility of ordered or concerted self-assembly, and the ability of nucleic acid shapes to recognize other shapes. All of these features can be robustly implemented in both RNA and DNA; several examples of tools made of DNA will be cited. However, the notion that nucleic acids can be formulated as devices the rival those made of silicon or, indeed, most other materials is ill-founded. The fact that nucleic acids are first and foremost biological storage devices severely limits their applications in fields other than biology and biotechnology. Therefore, it would seem that the only viable role for DNA in devices would be as an interface with other biological materials (for example, as sensor elements) and that the chief task in developing a DNA world is how to best meld the properties of this component molecule with the properties of those electronic devices that are already supreme at computation and communication; examples of device interfaces will again be touted. Alternatively, it may prove possible to engineer DNA devices for in vivo use, but even in this instance DNA can better serve as a template for the production of more serviceable molecules (RNA, proteins) than as a device in and of itself.

[top]

Dr. Deborah Fygenson is an Assistant Professor in Physics at the University of California, Santa Barbara. Her research concerns the interplay between structure, mechanics and dynamics of self-assembling macromolecules, both biological (e.g., microtubules) and designed (via DNA tilings). These studies are aimed at enabling the rational design of macromolecular devices that emulate or expand upon the natural nanotechnology of cellular biology. Prior to joining the faculty at UCSB in 1998, Dr. Fygenson was a Jane Coffin Childs Postdoctoral Fellow in Molecular Biology at the University of Southern California, a postdoc at the Center for Studies in Physics and Biology at the Rockefeller University in New York City, and a visiting researcher at NEC Labs in Princeton, New Jersey. She received a B.S. in physics from M.I.T. in 1989, and a Ph.D. in physics from Princeton in 1998.

Understanding and Controlling Self-Assembly of Nanotubes from DNA Tiles
Abstract: Nanotubes are an especially versatile structure in which to study self-assembly. The enhanced stiffness of their long dimension makes it possible to monitor individual assemblies floating freely in solution using light microscopy. We take advantage of a tile design that generates tubular polymers that are ~10 nm in diameter and correspondingly stiff (persistence length >5 µm) to characterize the self-assembly of DNA tiles. Our observations indicate that both energetics and kinetics of the self-assembly are available for rational design and enzymatic manipulation in this elegant equilibrium polymer system.

[top]

Dr. Ashish Goel is an Assistant Professor of Management Science and Engineering and (by courtesy) Computer Science at Stanford University. He received his PhD in Computer Science from Stanford in 1999, and was an Assistant Professor of Computer Science at the University of Southern California from 1999 to 2002. His research interests lie in the design, analysis, and applications of algorithms. Professor Goel is a recipient of an Alfred P. Sloan faculty fellowship (2004-06), a Terman faculty fellowship from Stanford, and an NSF Career Award (2002-07).

Counting Using DNA Self-Assembly
Abstract: Simulating counters is a basic tool for the self-assembly of many interesting structures. We will use counting to illustrate recent advances in the algorithmic theory of self-assembly. We will start with the basic binary counter construction of Rothemund and Winfree. We will explain how to modify it to obtain optimum assembly time and design complexity. We will present general techniques for analyzing assmebly time. Finally, we will discuss the issues of robustness and error-correction which are critical to precise simulation of counters.

[top]

Dr. Hervé Isambert is a CNRS scientist at Institut Curie, Paris. Trained as a soft condensed matter physicist, his interests for nucleic acids started with theoretical and simulation studies on RNA folding and unfolding pathways including pseudoknots (HDV ribozymes, tetrahymena group I intron, E coli 16S rRNA). The Kinefold webserver (http://kinefold.curie.fr) now offers online RNA folding simulations including pseudoknots and entangled helices. Our more recent researchs primarily focus on experimental designs of DNA and RNA switches operating under out-of-equilibrium conditions.

Designing DNA and RNA Switches under Out-of-Equilibrium Folding Control
Abstract:
Beyond its central role in biology, DNA has long been recognized as a versatile molecule that can self-assemble into nanoscale objects or form well-controlled supramolecular scaffolds. More recently, DNA was also shown to provide switchable nanomechanical devices whose equilibrium states can be controlled by changing, for example, the medium ionic strength or by using additional "fuel" and "waste" DNA molecules. In this talk, I will present an alternative design, based on folding kinetics control, to drive and maintain a DNA nanomechanical devices out-of-equilibrium under fixed chemical conditions. As a proof-of-concept, we designed a unimolecular DNA switch that folds, following heat denaturation, into its lowest energy conformation under moderate cooling rate while an alternative, fast folding conformation is kinetically formed upon rapid cooling (>100°C/ms). Heating and cooling cycles rapidly and reversibly interchange the two conformations, inducing atomic displacements of 20-110A, while their spontaneous equilibration can take several weeks at room temperature through nucleation and branch migration of a Holliday junction.

We have also designed bistable RNA switches folding out-of-equilibrium into either one of their stable structures under co-transcriptional folding control. RNA co-transcriptional folding is known to play an active role in helping proper native folding of ribozymes and regulatory structural motifs in mRNA UTRs. Yet, the underlying mechanisms and coding requirements for efficient co-transcriptional folding remain unclear. Using bistable RNA switches with symmetrical helices conserved under sequence reversal, we could demonstrate that native and transiently formed helices encoded on the sequence can efficiently guide co-transcriptional folding into either long-lived structure of these RNA switches. In addition, co-transcriptional folding paths may also be redirected through transient antisense interactions. Hence, transient intra- and intermolecular base pair interactions can effectively regulate the folding of nascent RNA molecules into different native structures. This constitutive coupling between RNA synthesis and RNA folding regulation may have enabled the early emergence of autonomous RNA-based regulation networks.

[top]

The research of Dr. Gerald Joyce concerns the biochemistry of nucleic acids and the development of novel nucleic acid enzymes through in vitro evolution. Like their protein counterparts, nucleic acid enzymes assume a well-defined structure that is responsible for their catalytic activity. Unlike proteins, nucleic acids are genetic molecules that can be amplified and mutated in the test tube. We have learned to exploit the dual role of nucleic acids as both catalyst and genetic molecule to develop RNA- and DNA-based evolving systems that operate entirely in vitro. At best, we can carry out 100 “generations” of test-tube evolution in a day, employing a population of trillions of nucleic acid molecules. This allows us to evolve nucleic acid enzymes far more rapidly than whole organisms evolve in nature.

Evolution of a DNA Enzyme from an RNA Enzyme
Abstract: The ability of RNA to catalyze the replication of RNA molecules is an activity that would have been essential for RNA-based evolution during the early history of life on Earth (the so-called “RNA world”). We previously developed an RNA enzyme, termed the R3C ligase, which catalyzes the template-directed joining of two RNA molecules. This ribozyme was converted to a format that allows it to produce additional copies of itself through the joining of two component subunits. More recently, the self-replicating ribozyme was converted to a cross-catalytic format whereby two ribozymes direct each other’s synthesis from a total of four component substrates. In contemplating the possibility of a “DNA world”, one faces similar functional requirements, but without the constraint of historical plausibility. Despite lacking the 2´-hydroxyl group, DNA is comparable to RNA in its catalytic efficiency and versatility. There are now many examples of DNA enzymes (deoxyribozymes), all of which have been obtained through in vitro evolution. No ribozyme, when prepared as DNA, retains its catalytic activity; and no deoxyribozyme, when prepared as RNA, retains its catalytic activity. Employing in vitro evolution, however, we were able to convert the R3C ribozyme to a corresponding deoxyribozyme. The evolved DNA contains 11 mutations relative to the starting RNA and exhibits a similar catalytic rate. From the perspective of engineering, rather than natural history, one should feel free to choose either RNA or DNA, whichever is more appropriate for a particular application.

[top]

Dr. Hideo Mabuchi is Associate Professor Physics and Control & Dynamical Systems at Caltech. His research group adopts a combined theoretical and experimental approach to the characterization, control, and fabrication of atomic, molecular, and quantum-optical systems. In particular, they build upon methods from estimation, control, and dynamical systems theory to better understand topics ranging from quantum measurement to cell-scale biochemical networks. Their experiments utilize techniques from cold atom physics, precision metrology, cavity QED, and single-molecule spectroscopy to explore topics in quantum and classical stochastics.

[top]

Dr. Chengde Mao is currently Assistant Professor of Analytical Chemistry at Purdue University.

Self-assembly of DNA Nanostructures
Abstract: Structural control at the nanometer scale is key to the development of nanotechnology. Supramolecular self-assembly is one promising approach to achieve this goal. Among many self-assembly molecular systems, DNA stands out as one of the best choices. Because DNA is the universal genetic materials, its structure and physical/chemical properties have been extensively studied, and a rich array of manipulation tools have been developed. DNA has excellent molecular recognition capability. Its structure can be precisely predicted. And branched DNA motifs have also been constructed. Combining all these factors together, DNA-based nanostructures have been rapidly developed. My talk will focuses on the recent development of DNA nanostructures in my group: (1) static structures, (2) dynamic structures, and (3) transferring DNA structures into metallic structures.

[top]

Dr. Niles Pierce received a BSE in Mechanical & Aerospace Engineering from Princeton University in 1993 and a DPhil in Applied Mathematics from Oxford University in 1997. He subsequently performed postdoctoral research in numerical analysis at Oxford University and computational protein design at Caltech before joining the Caltech faculty in 2000. The Pierce Lab combines computational and experimental approaches to design and construct nucleic acid devices. New algorithms for analyzing and engineering nucleic acid strands contribute to the long-term goal of creating a compiler for molecular function.  These software tools guide experimental studies of new mechanisms for implementing autonomous sensing and transport at the molecular level.

Computational Nucleic Acid Engineering
Abstract: RNA and single-stranded DNA are versatile construction materials that can be programmed to self-assemble into nanoscale devices driven by the free energy of base pair formation.  This talk will describe computational algorithms for analyzing and designing the underlying free energy landscapes that govern device function. These methods have been used to develop metastable hybridization chain reaction (HCR) systems that perform triggered amplification for biosensing applications.

[top]

Dr. John Reif is Hollis Edens Distinguished Professor in Trinity College of Arts and Sciences at Duke University since 2003 and Professor of Computer Science at Duke University since 1986. Previously he was Associate Professor, Harvard University. He received a Ph.D. Applied Mathematics in 1977 from Harvard University, an M.S. Applied Mathematics in1975 Univ. and was awarded a magna cum laude B.S. in Applied Math & CS in 1973 from Tufts University. He is Fellow, Association for the Advancement of Science (AAAS) since 2003, Fellow of IEEE since 1993, Fellow of ACM since 1996. Fellow of Inst. of Combinatorics since 1991. Although originally primarily a theoretical computer scientist, he also has made a number of contributions to practical areas of computer science including parallel architectures, data compression, robotics, and optical computing. He has also worked for many years on the development and analysis of parallel algorithms for various fundamental problems including the solution of large sparse systems, sorting, graph problems, data compression, etc. He has developed algorithms and lower bounds for a large variety of robotic motion planning problems, and provided the first known computational complexity results for a robotic motion planning problem. He has also developed a wide range of efficient parallel algorithms, particularly randomized parallel algorithms. Recently Reif has worked on DNA computing and DNA nanostructures. He is the author of over 200 papers and has edited three books on synthesis of parallel and randomized algorithms. Recently Reif has worked on DNA computing and DNA nanostructures.

Self-Assembled DNA Nanostructures for Molecular Scale Patterning, Computation and Motors
Abstract: Self-assembly is the spontaneous self-ordering of substructures into superstructures driven by the selective affinity of the substructures. DNA provides a molecular scale material for programmable self-assembly, using the selective affinity of pairs of DNA strands to form DNA nanostructures. DNA self-assembly is the most advanced and versatile system that has been experimentally demonstrated for programmable construction of patterned systems on the molecular scale. The methodology of DNA self-assembly begins with the synthesis of single-strand DNA molecules that self-assemble into macromolecular building blocks called DNA tiles. These tiles have sticky ends that match the sticky ends of other DNA tiles, facilitating further assembly into larger structures known as DNA tiling lattices. In principle, DNA tiling assemblies can form any computable two or three-dimensional pattern, however complex, with the appropriate choice of the tiles' component DNA. Two-dimensional DNA tiling lattices composed of hundreds of thousands of tiles have been demonstrated experimentally. These assemblies can be used as scaffolding on which to position molecular electronics and robotics components with precision and specificity. This programmability renders the scaffolding have the patterning required for fabricating complex devices made of these components. We overview the evolution of DNA self-assembly techniques from pure theory, through simulation and design, and then to experimental practice. We will begin with an overview of theoretical models and algorithms for DNA lattice self-assembly. Then we describe our software for the simulation and design of DNA tiling assemblies and DNA nanomechanical devices. As an example, we discuss models and algorithms for the key problem of error control in DNA lattice self-assembly, as well as the computer simulation of these methods for error control. We will then briefly discuss our experimental laboratory demonstrations, including those using the designs derived by our software. These experimental demonstrations of DNA self-assemblies include the assembly of patterned objects at the molecular scale, the execution of molecular computations, and freely running autonomous DNA motors.

[top]

Dr. Paul Rothemund received his B.S. in Biology and Computer Science at Caltech, and his Ph.D. in Computer Science under Len Adleman at the University of Southern California. He has been working on DNA computation, in one form or another, since 1992 when an undergraduate professor challenged his computer science class to construct a molecular computer along the lines suggested by Charles Bennett. This stimulated Dr. Rothemund to propose a DNA and restriction enzyme Turing machine, for which he was awarded a U.S. patent. In 2001 he was awarded a Beckman Foundation postdoctoral fellowship. He is currently a post-doctoral fellow with Caltech's Center for the Physics of Information.

Design and Characterization of Programmable DNA Nanotubes
Abstract: DNA self-assembly provides a programmable bottom-up approach for the synthesis of complex structures from nanoscale components. Although nanotubes are a fundamental form encountered in tile-based DNA self-assembly, the factors governing tube structure remain poorly understood. We report and characterize a new type of nanotube made from DNA double-crossover molecules (DAE-E tiles). Unmodified tubes range from 7 to 20 nm in diameter (4 to 10 tiles in circumference), grow as long as 50 microns with a persistence length of approximately 4 microns, and can be programmed to display a variety of patterns. A survey of modifications (1) confirms the importance of sticky-end stacking, (2) confirms the identity of the inside and outside faces of the tubes, and (3) identifies features of the tiles that profoundly affect the size and morphology of the tubes. Supported by these results, nanotube structure is explained by a simple model based on the geometry and energetics of B-form DNA.

[top]

Dr. John SantaLucia's group studies nucleic acid structural biology, biophysical chemistry, biological function, as well as biotechnology applications. His group has determined a database of thermodynamic parameters for Watson-Crick base pairs, all possible mismatches, and a variety of loop motifs including hairpins, bulges and internal loops. These parameters are required for accurate prediction of DNA hybridization and secondary structure. Knowledge of DNA structure is important for designing oligonucleotide microarrays for mRNA expression profiling, primers for PCR reactions, and probes for various genotyping diagnostics. This knowledge has been integrated into software that allows accurate prediction of DNA behavior and automated design of DNA-based applications. Most recently, he has developed new methods for RNA and DNA 3D structure prediction and design of biomaterials. His other research includes NMR structure determination of RNA with bound drugs and proteins. Dr. SantaLucia is the Founder and Chief Scientist of DNA Software, Inc.

Progress Toward Accurate 3D Structure Prediction of Nucleic Acids
Abstract: Methods for macromolecular structure determination cannot keep pace with the exponential growth of biological sequence databases. Thus, there is a need to develop tools for structure prediction from sequence and constraint information. Our goal is to fill this need by developing homology modeling and de novo structure prediction methods for RNA and DNA from natural or artificial sources. Homology modeling uses the known coordinates one sequence (i.e. the "template") and automatically and optimally accommodates the substitutions, deletions, and insertions a different but similar sequence (i.e. the "query"). We have developed a new alignment method that incorporates secondary structure constraints. We have also developed new methods for optimizing highly distorted initial geometries that include long bond lengths due to deletions and overlaps due to substitutions and insertions. The homology modeling project is important for converting biological sequence databases into 3D structure databases. For RNA and DNA molecules with new folds, de novo structure prediction is required. Our method of de novo structure prediction combines the strengths of dynamic programming algorithms for secondary structure prediction, a novel BUILDER algorithm, and modified classical molecular dynamics simulations. The BUILDER algorithm uses a motif database derived from fragments of known X-ray and NMR structures. The classical dynamics calculations allow for structural distortions to be resolved using a customized forcefield and optimization algorithms. These methods are being used to design new DNA/RNA architectures and potentially new catalysts.

[top]

Dr. Peter Schuster was born in Vienna in the year 1941. He studied chemistry and physics at Vienna University. In the year 1967 he received his PhD in chemistry and then worked as PostDoc researcher at the Max-Planck-Institute for Biophysical Chemistry in Göttingen (Germany). Afterwards, in 1973, he returned to Vienna where he accepted an offer as full professor of Theoretical Chemistry at Vienna University. In the years 1992 – 1995 he took a leave of absence from Vienna and was founding director of the Institute for Molecular Biotechnology in Jena (Germany). Since 1990 he is a member of the external faculty of the Santa Fe Institute in Santa Fe (U.S.A.). His main research interests are the theory of hydrogen bonds and proton transfer reactions, the exploration of ribonucleic acid structures, the study of mechanisms of biological evolution by means of computer assisted molecular models as well as the development of algorithms for the design of RNA and DNA molecules for predefined purposes.

Designing Single- and Double-Stranded Nucleic Acids
Abstact: The relation between sequences and minimum free energy structures of nucleic acids is understood as a mapping from sequence space onto a space of structures. On the level of secondary structures this mapping is always many to one and thus not (uniquely) invertible. The preimage of a structure in sequence space is called its neutral network. Based on this notion of structure and sequence space an algorithm for inverse folding has been designed: Given a structure, sequences are derived that form this structure under the minimum free energy criterion. The design of nucleic acid structures is extended to suboptimal conformations and then allows for the design of RNA and DNA molecules that form two or more long-lived conformations. Another extension leads to cofolding of two or more nucleic acid molecules. It provides tools for the design of double stranded DNA and RNA complexes.

[top]

Nadrian C. Seeman was born in Chicago in 1945. Following a BS in biochemistry from the University of Chicago, he received his Ph.D. in biological crystallography from the University of Pittsburgh in 1970. His postdoctoral training, at Columbia and MIT, emphasized nucleic acid crystallography. He obtained his first independent position at SUNY/Albany, where his frustrations with the macromolecular crystallization experiment and his awareness of the fatal series—no crystals, no crystallography, no crystallographer—led him to the campus pub one day in the fall of 1980. There, he realized that the similarity between 6-arm DNA branched junctions and the flying fish in the periodic array of Escher's 'Depth' might lead to a rational approach to the organization of matter on the nanometer scale, particularly crystallization. Ever since, he has been trying to implement this approach and its spin-offs, such as nanorobotics and the organization of nanoelectronics; for the last sixteen years he has worked at New York University. When told in the mid-1980s that he was doing nanotechnology, his response was similar to that of M. Jourdain, the title character of Moliere's Bourgeois Gentilehomme, who was delighted to discover that he had been speaking prose all his life.

DNA: Not Merely the Secret of Life
Abstract: Structural DNA nanotechnology uses the concept of reciprocal exchange between DNA double helices or hairpins to produce branched DNA motifs, like Holliday junctions, or related structures, such as double crossover (DX), triple crossover (TX), paranemic crossover (PX) and DNA parallelogram motifs. We have been working since the early 1980's to combine DNA motifs, using sticky-ended cohesion, to produce specific structures; more recently, we have also used PX cohesion. The key strength of sticky-ended cohesion is that it produces predictable adhesion combined with known structure. From branched junctions, we have constructed DNA stick-polyhedra, whose edges are double helices, and whose vertices are the branch points of DNA branched junctions. These include a cube, a truncated octahedron, and an irregular graph. This approach has also rendered accessible several topological targets, such as deliberately designed knots and Borromean rings. Recently, we have begun to template the topology of industrial polymers, such as nylon with DNA-like scaffolds.

Nanorobotics are key to the success of nanotechnology. We have used two DX molecules to construct a DNA nanomechanical device by linking them with a segment that can be switched between left-handed Z-DNA with right-handed B-DNA. PX DNA has been used to produce a robust sequence-dependent device that changes states by varied hybridization topology. The sequence-dependent nature of this device means that a variety of them attached to a motif can all be addressed individually. Recently, we have used this device to make a translational device. A protein-activated device that can be used to measure the ability of the protein to do work, and a bipedal walker have both been built.

A central goal of DNA nanotechnology is the self-assembly of periodic matter. We have constructed micron-sized 2-dimensional DNA arrays from DX, TX and two kinds of parallelogram motifs. We can produce specific designed patterns visible in the AFM from DX and TX molecules. We can change the patterns by changing the components, and by modification after assembly. In addition, we have generated 2D arrays from DNA parallelograms. These arrays contain cavities whose sizes can be tuned by design. Recently, we have used new motifs to produce honeycomb-shaped arrays.

The key structural challenge in the area is the extension of the 2D results obtained so far to 3D systems with a high degree of ordering. Several motifs have been produced that can produce 2D arrays in each of the three directions normal to the vectors that span the 3D space. Crystals with dimensions as large as a millimeter, ordered to 10 Å resolution (as determined by X-ray diffraction) have been produced. Ultimately, we expect to be able to produce high resolution crystals of DNA host lattices with heterologous guests, leading to well-ordered bio-macromolecular systems amenable to diffraction analysis. Other challenges are to incorporate DNA nanomechanical devices in periodic and aperiodic lattices and to use the lattices to organize nanoelectronic components, such as metallic nanoparticles or carbon nanotubes.

[top]

As an undergraduate at Harvard, Dr. William Shih worked with Dr. Tom Kirchhausen (Department of Cell Biology) on characterizing clathrin-coated vesicles. As a graduate student at Stanford, he worked with Dr. James Spudich (Department of Biochemistry) on measuring conformational changes in the molecular motor myosin. As a postdoctoral fellow at The Scripps Research Institute, Shih worked with Dr. Gerald Joyce (Departments of Chemistry and Molecular Biology) on creating a virus-sized wireframe octahedron that folds from a long single chain of DNA. He is currently an Assistant Professor at Harvard Medical School in the Department of Biological Chemistry and Molecular Pharmacology. Shih's lab, which works on DNA nanotechnology, resides in Dana-Farber Cancer Institute in the Department of Cancer Biology.

Clonable DNA Nanotechnology
Abstract: A key property of DNA — its ability to be amplified exponentially by polymerases — facilitates the large-scale clonal production of individual sequences. This property also makes possible the directed evolution of sequence lineages toward optimized behaviors. Previous examples of three-dimensional geometric DNA objects, however, were built using architectures that are not amenable to copying by polymerases. We have developed a strategy for encoding DNA cages as single strands that are amplifiable by polymerases and that can be folded into a target structure by a simple denaturation-renaturation procedure. Our demonstration of a clonable DNA octahedron represents a large step toward making the use of DNA scaffolds more practical and more versatile.

[top]

Dr. Milan Stojanovic holds a B.A. in Chemistry froim Beogradski Univerzitet and a Ph.D. in Chemistry from Harvard University. He held a Post-Doctoral position in Biochemistry at Columbia University. Currently he is Assistant Professor of Medical Sciences, Department of Medicine, Columbia University.

Agile and Intelligent DNA Molecules
Abstract: Networks of allosterically regulated nucleic acid catalysts can perform reasonably complex Boolean calculations. But what else can they do?

[top]

Doug Turner is a Professor of Chemistry at the University of Rochester and has a secondary appointment as Professor of Pediatrics in the Center for Human Genetics and Molecular Pediatric Disease. He received his A.B. from Harvard, his Ph.D. from Columbia, and was a postdoc with I. Tinoco at Berkeley before moving to Rochester. He has held both Sloan and Guggenheim fellowships. Dr. Turner’s research program focuses on the interactions that drive RNA folding and the use of this information to predict structure from primary sequence. The group has also worked recently on RNA-based therapeutics. Including self-citations, publications from his lab since 1983 have been cited over 6000 times.

Predicting RNA Secondary Structure
Abstract: RNA structure is important for many biological functions and for designing therapeutics. The structures of RNAs are encoded in the genome sequence of an organism. Recent results will be presented that can facilitate interpretation of this code by predicting the secondary structure of an RNA by free energy minimization. Thermodynamic and NMR experiments provide insight into approximations for free energy parameters for internal loops of 3X3 nucleotides. These data also provide bench marks for testing programs for predicting stabilities and local three dimensional structures of regions that are not Watson-Crick base paired.

The lack of knowledge of the sequence dependence of loop stabilities limits the accuracy of predictions for RNA secondary structure by free energy minimization. Predictions can be improved, however, by constraining folding with results from chemical modification experiments.

[top]

Erik Winfree is an Assistant Professor in Computer Science and Computation and Neural Systems at Caltech. His research concerns the theory and engineering of autonomous biochemical algorithms using in vitro systems of DNA and enzymes, including programmable DNA self-assembly, DNA and RNA conformational switches and devices, and RNA transcriptional circuits. Such systems are envisioned as embedded information-processing and control for bottom-up nanofabrication, nanorobotics, biochemical diagnostics, and other biochemical processes. Prior to joining the faculty at Caltech in 1999, Winfree was a Lewis Thomas Postdoctoral Fellow in Molecular Biology at Princeton, and a Visiting Scientist at the MIT AI Lab. Winfree received a B.S. in mathematics and computer science from the University of Chicago in 1991, and a Ph.D. in computation and neural systems from Caltech in 1998.

Algorithmic Self-Assembly of DNA
Abstract: Crystals can compute as they grow.  Thus, information encoded in a small seed crystal can dictate the growth of arbitrarily complex forms, in principle -- a simple physical analog of organismal development.  In theory, this provides a Turing-universal mechanism for controlling the morphology of molecular structures.  In practice, algorithmic crystals of DNA double-crossover tiles have been created for simple algorithmic growth processes. Major experimental challenges are improving control over nucleation and growth error rates.

[top]

Dr. Hao Yan. Our research interests center on self-assembly of nanostructures, particularly using DNA as an assembly element. We would like to use this new technology to develop molecular motors, sensors, and templates for more complex nanostructural systems.

DNA-based Molecular Engineering: Pattern and Motion
Abstract: In recent years, a number of research groups have begun developing nanofabrication methods based on DNA self-assembly. DNA is an extraordinarily versatile material for designing nano-architectural motifs, due in large part to its programmable G-C and A-T base pairing into well-defined secondary structures. These encoded structures are complemented by a sophisticated array of tools developed for DNA biotechnology: DNA can be manipulated using commercially available enzymes for site-selective DNA cleavage (restriction), ligation, labeling, transcription, replication, kination, and methylation. DNA nanotechnology is further empowered by well-established methods for purification and structural characterization and by solid-phase synthesis, so that any designer DNA strand can be constructed.

Here we present our recent experimental progress to utilize novel DNA nanostructures for self-assembly as well as for templates in the fabrication of functional nano-patterned materials. We have prototyped a new nanostructured DNA motif known as a cross structure. This nanostructure has a 4-fold symmetry which promotes its self-assembly into tetragonal 2D lattices. Each unit cell can be considered as an individual pixel; if unique DNA labels can be assigned to each cross structure, they can be used to construct 2D arrays with individually addressable binding sites. We have also demonstrated a DNA barcode lattice composed of DNA tiles assembled on a long scaffold strand; the system translates information encoded on the scaffold strand into a specific and reprogrammable barcode pattern which is visible by atomic force microscopy. We have achieved gold nanoparticle linear arrays templated on DNA arrays comprised of triple crossover (TX) molecules. We have designed and demonstrated a 2-state DNA lattice which display expand/contract motion switched by DNA nanoactuators. We have also developed an autonomous DNA motor executing unidirectional motion along a linear DNA track. The above experimental demonstrations provide evidence that precise nanoscale control of pattern and motion can be achieved through molecular engineering using DNA nanostructured building blocks.

[top]

Dr. Bernard Yurke received his Ph.D. from Cornell University in 1982 for work in low-temperature physics. He has since been employed as a research physicist at Bell Laboratories. He initially served as the theorist for the Bell Labs squeezed light effort and conducted his own experimental research on generating squeezed states at microwave frequencies using Josephson parametric amplifiers. In addition to quantum optics, he has worked in a variety of fields including liquid crystals, biophysics, and MEMS. Currently, he is building DNA-based molecular motors and exploring nanoscale assembly strategies. In 1997 he became a Distinguished Member of Technical Staff. He is a fellow of the Optical Society of America, the American Physical Society, and the American Association for the Advancement of Science. In 2001, for work in quantum optics, he received the Max Born Award from the Optical Society of America.

Using DNA Nanomachines to Control the Bulk Mechanical Properties of Materials
Abstract: Movement is a conspicuous manifestation of life, exhibited by animals, which results from the coordinated activity of trillions of molecular motors in muscle tissues. There is considerable interest in devising active materials that could serve as synthetic analogues of muscle tissue. The availability of DNA, functionalized in a manner that allows ready incorporation into gels, provides a means by which DNA-based nanodevices can be incorporated into gels. This makes possible the investigation of active material driven by DNA-based molecular motors. Here we describe work on DNA-crosslinked polyacrylamide gels in which a simple hybridization motor is used to control the stiffness of the gel. The gel is prepared as follows: First acrydite functionalized oligomers with a nucleotide sequence A are copolymerized with acrylamide monomers to produce strands of polyacrylamide with DNA side branches. Because of the absence of crosslinking, the result is a viscous fluid. In a second batch, acrydite functionalized oligomers with a nucleotide sequence B are copolymerized with acrylamide monomers which also produce a viscous fluid. When polymers A and B are mixed together the result is again a viscous fluid. However, when C oligomers, having a region that is complementary to A and a region complementary to B, are added to the mixture, crosslinks are formed between the A and B polymer strands and gelation occurs. Between the domains complementary to A and B, the C oligomers possess a domain M that remains single-stranded during the formation of the gel. This domain functions as a hybridization motor. DNA strands F complementary to M can bind with M to stiffen and increase the effective length of the crosslink. This causes the acrylamide strands to be stressed and results in the stiffening of the gel. The gel can be restored to its initial state through the removal of the F strands from the crosslink by toehold-mediated strand exchange. We have been able to construct DNA-crosslinked gels that manifest a threefold change in stiffness upon application of the F strands. Electrophoresis was used to introduce F or F’s complement into the gel.

[top]