Computational Biology and Chemistry 27 (2003) 461–467

Commentary

The DNA double helix fifty years on Robert B. Macgregor Jr.∗ , Gregory M.K. Poon Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 19 Russell Street, Toronto, Ontario, Canada M5S 2S2 Received 16 June 2003; received in revised form 8 August 2003; accepted 12 August 2003

Abstract This year marks the 50th anniversary of the proposal of a double helical structure for DNA by James Watson and Francis Crick. The place of this proposal in the history and development of molecular biology is discussed. Several other discoveries that occurred in the middle of the twentieth century were perhaps equally important to our understanding of cellular processes; however, none of these captured the attention and imagination of the public to the same extent as the double helix. The existence of multiple forms of DNA and the uses of DNA in biological technologies is presented. DNA is also finding increasing use as a material due to its rather unusual structural and physical characteristics as well as its ready availability. © 2003 Published by Elsevier Ltd. Keywords: DNA; Double helix; Molecular biology; Nanotechnology

1. Introduction 1953 witnessed the birth of a science icon with the publication of James Watson and Francis Crick’s proposal of a double helix structure of DNA in the journal Nature (Watson and Crick, 1953a). In their original publication they proposed that DNA consists of two molecules that are wound around each other to form a right-handed helix that is stabilized by hydrogen bonding interactions between complementary base pairs between the two molecules. They pointed out that this proposed structure provided a hint as to how DNA could be self-replicating (Watson and Crick, 1953b). The proposal rationalized and accommodated a great deal of current experimental information and pointed the way to other experiments that could verify it. The process of verification provided additional revelations about the molecular mechanisms of cellular processes, and it provided a model on which many other ideas could be based. 2. The path to molecular biology In the 19th century, the Moravian monk Gregor Mendel had given a solid quantitative grounding to the idea of ge∗ Corresponding author. Tel.: +1-416-978-7332; fax: +1-416-978-8511. E-mail address: [email protected] (R.B. Macgregor Jr.).

1476-9271/$ – see front matter © 2003 Published by Elsevier Ltd. doi:10.1016/j.compbiolchem.2003.08.001

netic transmission of characteristics. At about the same time as Mendel was carrying out his studies, DNA was discovered by the Swiss physiologist Friedrich Miescher. For the first several decades after its discovery the role of DNA in cellular processes remained unexplained and for the most part uninvestigated. During the early part of the 20th century, various biochemical pathways and principles were brought to light; however, the question remained, which cellular molecule provides the basis for Mendellian genetics? The different disciplines interested in cellular processes, biochemistry, genetics, and microbiology took very different approaches in their investigations of this question; however, none focused on DNA. Proteins and enzymes were known to be the molecular phenotype of cells; however, the link between the gene-carrying molecule and these proteins was anything but clear. DNA was generally considered to be chemically, and thus structurally, too simple to contain the presumably complex information necessary for heredity. This view changed over the course of about 20 years starting in 1928 when Fred Griffith found that a benign strain of pneumococcal bacteria could be transformed into a pathogenic strain by exposure to a cell-free extract of the pathogenic strain. In the mid 1930s Oswald Avery and coworkers set out to purify this transforming factor. Chemical analysis of the extract showed that DNA had transformed the pneumococcal bacteria and in 1944 he published the first results demonstrating that DNA was the genetic material (Avery et al., 1944). A few years later, the results of Hershey

462

R.B. Macgregor Jr., G.M.K. Poon / Computational Biology and Chemistry 27 (2003) 461–467

and Chase (1952) reinforced Avery’s finding by demonstrating that DNA mediated the production of progeny virus in bacteriophage-infected Escherichia coli bacteria. During this time, George Beadle and Edward Tatum had shown in the early 1940s that enzyme synthesis in cells was controlled by genes and that there was one gene for each enzyme (Beadle and Tatum, 1941). Taken together these were extremely important discoveries putting an end to nearly a century of speculation about the chemical nature of genes. It was interesting to know that genetic information somehow resided in DNA; however, there was no way to incorporate it into the current understanding of cellular mechanisms. It was not clear how the genes on DNA could be transformed into proteins; several pieces of the puzzle were still missing. This was the state of knowledge in biology during which the work on the structure of DNA began. In the early 1950s several prominent scientists had turned their efforts toward elucidating the structure of DNA, the perceived importance and prestige of these studies was enhanced by the fact that Linus Pauling, the chemist who received the Nobel Prize for his descriptions of chemical bonding, was working on the structure of biological molecules, including DNA. Studies of the structure of molecules had been made possible by the advent of X-ray diffraction, a technique that had been recently developed by physicists. And although X-ray diffraction worked well for small molecules, work on large biological molecules like proteins or DNA was slow for a number of reasons. In the case of analyzing the structure of DNA the difficulties were exacerbated by the fact that there were no crystals of DNA available meaning that the diffraction images were acquired using oriented fibers of DNA. The diffraction patterns obtained using fibers is poor and only gross details of the structure can be determined. For example, it was known that DNA was helical and that the bases were oriented perpendicular to the axis of the helix. However, neither the relative orientation of the molecular groups nor the handedness of the helix could be ascertained. When Watson and Crick combined the known data about DNA, the results of the X-ray diffraction analysis, and a great deal of imagination they struck upon a structure that satisfied all of the requirements. Because of the earlier work of Avery, Hershey and Chase, and Beadle and Tatum, when Watson and Crick published their proposed structure the scientific world was ready. Their two publications in 1953 offered an entirely new way of understanding the molecular mechanisms underlying many cellular processes, such as cell division, genetic inheritance, and protein synthesis. However, despite the fact that the advances between 1927 and 1953 had revolutionized the understanding of the cell a number of very large questions had also been raised. In the mid- to late-1950s, after the publication of the Watson and Crick structure the biggest difficulty posed by the results of the previous twenty-five years was how the genes present in the double helical structure of the DNA

give rise to the proteins for which they code. At that time there was no known molecule or mechanism that would link these two structures. A proposal by Mahlon Hoagland suggesting the formation of a complementary structure between DNA and RNA appeared in the magazine Scientific American in 1959. The next year experimental data from the laboratory of Sol Spiegelman (Nomura et al., 1960) showed the involvement of a DNA–RNA hybrid molecule that was crucial to protein synthesis; this molecule was messenger RNA (mRNA). Oddly, although RNA was known to form helical structures similar to DNA, the necessity or likelihood of formation of a transient DNA–RNA hybrid had not been considered. With the discovery of mRNA the modern central dogma of molecular biology, DNA makes RNA makes protein, was established.

3. The success of the Watson–Crick model Synthesis, verifiability, and extrapolation are hallmarks of great scientific ideas. Many other experiments in biology that occurred after the proposal of the double helix are considered to be direct consequences of the Watson and Crick double helix. This may be overstating the case somewhat; many seminal discoveries about the molecular biology of the cell had already been made and would have been made with or without a working model for the structure of DNA. However, it is clear that their proposal is a major landmark in the progression of understanding of biological systems that began in the 19th century. There have been many other scientific milestones but none have so consistently captured the imagination of the public and scientific community. In purely structural terms, the DNA double helix was the first model for a major macromolecular cellular component and it remains the greatest success of structural biology to date. Although there are currently thousands of protein structures known and the three dimensional structures of the other cellular components have also been elucidated, this was not the case in the 1950s. A working model for any large cellular molecule was a great novelty at the time; the first protein structure was still about ten years away. DNA is much simpler than any protein and Watson and Crick’s model was consistent with all the known information about the composition and properties of DNA. Although detailed information concerning protein structure is often useful in guiding further experimentation about a particular protein, such knowledge does not have the generality, and thus, does not have the impact of the originally proposed double helical structure for DNA. This is true because it is very difficult to extract general principles from most protein data, the details are important. In contrast to this the details of the structure of the double helix are not as important as the concepts that arose from it and the challenges it presented. The Watson–Crick model was a simple and elegant model that facilitated the subsequent understanding of many biological processes such as DNA replication, gene

R.B. Macgregor Jr., G.M.K. Poon / Computational Biology and Chemistry 27 (2003) 461–467

recombination, transcription, mutation, and in turn important developments such as chemo- and antisense therapies. Nonetheless, it is possible to imagine a world in which molecular biology exists with little or no knowledge of the structural nature of DNA. The validity of this assertion is seen in the fact that the first x-ray crystal structure of a DNA molecule was not published until more than 25 years after the first appearance of Watson and Crick’s double helix (Wing et al., 1980). The intervening years were hardly fallow for our understanding of the cell. A great deal of biology had been done and would have been possible without a model of the structure. As an example, after the discovery that DNA carried the genetic message it was clear that the message had to be encoded somehow. A logical exercise, with or without the double helical model DNA, would lead to the same conclusion: a minimum of three bases are necessary to encode the twenty amino acids that constitute proteins. Chargaff had shown that adenosine and thymine bases always occurred in the same proportions as do guanines and cytosines. With this knowledge, one can conjecture that it would have been possible to propose a working model of DNA, consistent with all of the data, without the double helix model. Two other technological breakthroughs played equally important roles in the development of molecular biology. The first of these was the ability to rapidly sequence DNA. The methods published by Sverdlow et al. (1973), Maxam and Gilbert (1977), and Sanger et al. (1977) led to the discovery of the complexity of eukaryotic genomes and eventually to the sequencing of the genomes of several organisms, including humans. The second significant technological feat was the emergence of a rapid, inexpensive, method of synthesis of short strands of DNA (Gait and Sheppard, 1977; Hutchinson et al., 1978). Many other scientific and practical uses of DNA have arisen as a consequence of these two innovations. Automated, rapid sequencing, the polymerase chain reaction, forensic and archeological uses of DNA, are some of the examples of the uses of DNA that were unimaginable a short time ago but were made possible by the convergence of the knowledge of several different scientific disciplines in Watson and Crick’s proposal of a structure of DNA.

4. The rise of information science and the double helix It is also interesting to consider the era in which Watson and Crick lived. The middle of the twentieth century saw the rise of two other factors that provided an interesting backdrop to their work, namely, the concomitant rise in information and computer science. Physicists, mathematicians, and engineers were interested in the information content of different types of signals. Their interest was driven in part by certain consequences of quantum mechanics, discovered earlier in the century and in part because of the increasing amount and value of information. This led to the rise of information science, Claude Shannon and Alan Turing, early pi-

463

oneers in this new science made important discoveries about the nature of information and noise. For the first time, information and its quality could be expressed in quantitative terms. The increasing use and power of digital computers accompanied these theoretical insights. We are not aware of any contact between information scientists and Watson or Crick, nor did computers play any direct role in the genesis of their idea. However, Watson and Crick’s description of DNA as a double helix and the subsequent deciphering of the genetic code by other researchers began to move biology, or molecular biology, to a level with these more quantitative sciences. The cell was, and in many ways remains, a black box, but for the first time there was at least the possibility to treat biology as a branch of knowledge that at least someday could be understood in mathematical terms. It provided an arrow, a directionality to the processing of information in the cell: DNA makes RNA makes protein. And although it is not digital it is quite clearly quantized.

5. Alternatives to the Watson–Crick proposal Other alternative structures have expanded the conformational repertoire of DNA. Indeed, shortly after the publication of the double-helix structure, it was apparent that unwinding the two strands in order to replicate the DNA prior to cell division would pose a very significant topological problem. The so-called side-by-side model essentially eliminates the topological linking of the two strands of DNA by allowing the base pair interactions to occur with the backbones of the two molecules alternating between leftand right-handed helices (Millane and Rodley, 1981). Another topologically unlinked model of DNA has also been proposed by Biegeleisen (2002). Although these alternative models rectify certain difficulties inherent in the double helical structure of DNA, they are inconsistent with great deal of experimental evidence that is described by Watson and Crick’s model. It should be pointed out that the structure of DNA inside of a cell remains a mystery and it is entirely possible that DNA adopts conformations in the cell that have not or cannot be observed in a test tube. A few years after the proposal of the double helix, an alternative type of interaction between the complementary bases was put forward by Hoogsteen. The alternative base pairing interactions apparently is found only in unusual DNA structures. Several of these unusual, or non-canonical, structures were discovered in the 1950s as increasing numbers of biochemists and physical chemists turned their attention to DNA. The most easily interpretable experiments investigating the structure of DNA relied on the use of simple synthetic polymers consisting of repeats of one base for the entire length of the polymer, for example polyadenosine (poly(A)) or polyuracil (poly(U)). As chance would have it, these simple polymers, exhibit some very interesting behavior; titrations of complementary strands with each other, for example poly(A) and poly(U) showed that the stoichiom-

464

R.B. Macgregor Jr., G.M.K. Poon / Computational Biology and Chemistry 27 (2003) 461–467

etry of the interacting strands could be either 1:1, the expected proportion, or 1:2. Another interesting example is the self-association of polyguanine which was found to associate into four-stranded structures. Although the stoichiometry of these molecules differed from that of the double helix, i.e. three or four individual DNA strands instead of two, the Watson–Crick double helix was used as a starting point for interpreting the results. Thus, the unusual stoichiometry between poly(A) and poly(U) was assumed to arise from the formation of a three-stranded helix. In this particular case, model building and then crystallographic results showed that the bases interacted via two different types of hydrogen bonding patterns: which are now known as Watson–Crick binding and Hoogsteen hydrogen binding.

At first these non-standard conformations were considered biophysical chemical oddities with little or no biological relevance. However, it would now appear that most, if not all of the structures listed in Table 1 can and do exist in cells. An interesting footnote in the progress towards the structure of DNA was provided by the publication of the first high-resolution x-ray structure by the group of Alex Rich in 1979. Much to the surprise of everyone, the DNA in their study, the hexanucleotide, d(CGCGCG), crystallized as a left-handed helix (Wang et al., 1979). It has since been shown that this particular sequence motif, alternating CG residues, will adopt left-handed helicity in the appropriate solvent conditions and that, in general DNA exists as a right-handed helix for other sequences.

6. One molecule, many different helices

7. DNA as a material

These early unusual structures provided only a hint of the conformational plasticity of nucleic acids. In the majority of cases, the novel structures are based upon the double helix as originally proposed; a partial list of the types of structures is given in Table 1. In many cases the different types of structures result from relatively minor alterations in the helix structures, e.g. the difference between the A and B conformation (Fig. 1). Other changes are much more dramatic and differ radically from the double helix structure proposed by Watson and Crick. Perhaps the two best examples of this class of structures are Z-DNA and parallel-stranded DNA. In Z-DNA the helical sense of the molecule is reversed, resulting in a left-handed helix (Fig. 1). In parallel DNA, the relative orientation of the two complementary strands of DNA is parallel as opposed to the more standard antiparallel arrangement observed in most other structures.

Although it may be argued that much of molecular biology might have existed in something resembling its present form without Watson and Crick’s proposed structure, the fact that DNA is indeed a double stranded structure has had implications in areas of science that probably were not expected 50 years ago. The structural linearity of DNA, its ability to form predicable complementary interactions between two molecules, and the advent of solid-phase DNA synthesis have made DNA one of the most attractive building blocks for the synthesis of what is now called nanostructures. Nanostructures are molecular species with dimensions on the nanometer scale that are synthesized from the “bottom up.” That is, these structures are made starting with smaller molecules that are allowed to associate in a controlled manner in order to construct the final, nanometer-scale structure. Several laboratories have exploited the base

Table 1 A sample of the different types of DNA structures. Most of the entries in the Table have sub-classes and not all of the structures necessarily have any known biological relevance. All of the known structures involve some form of a helix although it may differ radically from Watson and Crick’s proposal. Structure

Description

A-DNA

Duplex isoform found in dehydrated environments. First fibre defraction by Roselind Franklin was of A-DNA. Only isoform accessible to RNA and RNA/DNA duplexes. Duplex isoform found in most biological environments. Original structure proposed by Watson and Crick. Not adopted by structures containing RNA. Left-handed duplex favored by GC-rich sequence under high salt. Discovered in first single-molecule crystallographic structure of DNA (Wang et al., 1979) Putative duplex isoform in which WC base pairs coordinate divalent ions (e.g., Zn2+ , Ni2+ and Co2+ ) at alkaline pH. (Aich et al., 1999). Noted for electrical conductive properties. Once postulated structure of DNA by Linus Pauling, with an interwound phosphate backbone and exposed bases (Pauling and Corey, 1953). Observed in positively supercoiled B-DNA duplex under a stretching force (Allemand et al., 1998). Three-stranded isoform formed by binding of a third strand to the major groove of a B-DNA duplex. Formed by two homopyramidine and one homopurine strands and favored by low pH or high ionic strength. Intramolecular triplex formed composed of a hairpin duplex by one strand and Hoogsteen base paring by another. Sequences exhibit mirror symmetry along the strands. Four-stranded isoform supported by Hoogsteen base pairing among four guanine residues, coordinating a monovalent ion (e.g., K+ , Na+ , NH4 + ). Contacts may be intra-strand (as in telomeric ends of chromosomal DNA) or propagate linearly (G-wire). Multistranded superstructures formed by oligonucleotides with long consecutive guanine runs (Protozanova and Macgregor, 1998). Non-guanine nucleotides branch out and may be functionalized.

B-DNA Z-DNA M-DNA P-DNA Triplex H-DNA G-quartet Frayed wires

R.B. Macgregor Jr., G.M.K. Poon / Computational Biology and Chemistry 27 (2003) 461–467

465

Fig. 1. A world of helices. From left to right, the crystal structures of three families of DNA structure: A-DNA, B-DNA, and Z-DNA. The A and B forms are right handed double helices that differ in the details of the parameters such as the orientation of the base pairs relative to the helix axis and the number of base pairs per turn of the helix. The Z form is a left handed helix. Watson and Crick had arrived at the structure that today is called the B form from model building based their data. Although their data could not discriminate between the two different helix senses, they chose the right-handed orientation because of they found it impossible to build a left handed double helix that obeyed the known structural constraints. (These structures are taken from the Protein Data Base, ID Numbers 440D, 5DNB, and 1DJ6.)

pairing capability of DNA for this purpose. Many creative structures have been synthesized using only DNA by the laboratory of Nadrian Seeman. Among the accomplishments of his group is the synthesis of a DNA molecule with the connectivity of a cube (Fig. 2), and “nanomechanical device” based upon the B-Z conformational change (Fig. 3). Other DNA-based molecular machines have also been created (Fig. 4) (Yurke et al., 2000). In 1994, Leonard Adleman took advantage of the parallel nature and specificity of DNA basepairing to obtain a DNA-based solution to the Travelling Salesman Problem (Adleman, 1994). Although this problem can be solved much more quickly and efficiently by electronic digital computers, the specificity of DNA basepairing and the very large number of parallel interactions that occur in a double helix make DNA (or RNA) an interesting tool for solving certain very large combinatorial problems.

Fig. 2. DNA as a material: a schematic of the backbone of a DNA molecule whose helix axes have the connectivity of a cube. These nanostructures were synthesized in the laboratory of N. Seeman (from Seeman, 1998).

Watson and Crick’s proposal of a double helical structure for DNA double helix facilitated and greatly accelerated the subsequent scientific discoveries by providing a simple accurate model. The discoveries during the 30 years between roughly 1930 and 1960 set the ground work for the scientific and technological advances of the past 40 years. It was one of the most important steps in the development of molecular biology in the latter half of the twentieth century. It would be difficult to say that the advent of the double helix was the single most important of the discoveries; the discovery must be considered in the context in which it was made. On the other hand, it is abundantly clear that Watson and Crick’s proposal remains the most widely popularized of the giant steps in biology that occurred in the middle of the last century. Much of modern molecular biology would have been possible perhaps without it but many key experiments central to our knowledge would have been much slower in coming. The structure originally proposed has now been verified and it remains one of the greatest successes of the elucidation of function by consideration of the structure. The standard Watson–Crick structure also inspired a number of developments that authors probably had not foreseen; perhaps the most intriguing of these is the use of DNA as a material for the synthesis of molecule-sized devices and machines. Note: There are several excellent books describing the history of molecular biology from different perspectives. These include: A History of Molecular Biology, by M. Morange, Harvard University Press, Cambridge,1998; Life Science in the Twentieth Century, by G. E. Allen, Wiley, New York, 1975; The Path to the Double Helix, by R. Olby, Macmillan, London, 1974.

466

R.B. Macgregor Jr., G.M.K. Poon / Computational Biology and Chemistry 27 (2003) 461–467

Fig. 3. A DNA-based nanomechanical device. Top: A molecular model of a molecule constructed entirely from right-handed B-DNA. At the center of the connecting helix is a 20-nucleotide region (yellow) that can adopt either the left-handed Z conformation or the right-handed B conformation. When the B-Z transition takes place, this same yellow portion becomes left-handed Z-DNA (bottom). The large green and red circles indicate fluorescent dyes covalently attached to the DNA that change their relative positions, when the B-Z transition is induced. The change in the position of the dyes is reversible and may be cycled (from Mao et al., 1999).

Fig. 4. DNA as a device: using it to create molecular tweezers. The operation of the tweezers, closing and opening, proceeds by alternative annealing of DNA strands which are complementary to different parts of the structure (from Yurke et al., 2000).

References Adleman, L.M., 1994. Molecular computation of solutions to combinatorial problems. Science 266, 1021–1024. Aich, P., Labiuk, S.L., Tari, L.W., Delbaere, L.J., Roesler, W.J., Falk, K.J., Steer, R.P., Lee, J.S., 1999. M-DNA: A complex between divalent

metal ions and DNA which behaves as a molecular wire. J. Mol. Biol. 294, 477–485. Allemand, J.F., Bensimon, D., Lavery, R., Croquette, V., 1998. Stretched and overwound DNA forms a Pauling-like structure with exposed bases. Proc. Natl. Acad. Sci. USA 95, 14152–14157. Avery, O.T., MacLeod, C., McCarty, M., 1944. Studies on the chemical nature of the substance inducing transformation of pneumococcal types. J. Exp. Med. 79, 137–158. Beadle, G.W., Tatum, E.L., 1941. Genetic control of biochemical reactions in Neurospora. Proc. Natl. Acad. Sci. USA 27, 499–506. Biegeleisen, K., 2002. Topologically non-linked circular duplex DNA. Bull. Mathem. Biol. 64, 589–609. Gait, M.J., Sheppard, R.C., 1977. Rapid synthesis of oligodeoxyribonucleotides: a new solid-phase method. Nucleic Acid Res. 4, 1135–1158. Hershey, A.D., Chase, M., 1952. Independent functions of viral proteins and of nucleic acids in the growth of the bacteriophage. J. Gen. Physiol. 36, 39–56. Hutchinson III, C.A., Phillips, S., Edgell, M.H., Gillam, S., Jahnke, P., Smith, M., 1978. Mutagenesis at a specific position in a DNA sequence. J. Biol. Chem. 253, 6551–6560. Mao, C., Sun, W., Shen, Z., Seeman, N.C., 1999. A nanomechanical device based on the B-Z transition of DNA. Nature 397, 144–146. Maxam, A.M., Gilbert, W., 1977. A new method for sequencing DNA. Proc. Natl. Acad. Sci. USA 74, 560–564. Millane, R.P., Rodley, G.A., 1981. Stereochemical details of the side-by-side model for DNA. Nucleic Acid Res. 9, 1765–1773. Nomura, M., Hall, B.D., Speigelman, S., 1960. Characterization of RNA, synthesized in Escherichia coli after bacteriophage T2 infection. J. Mol. Biol. 2, 306–326. Pauling, L., Corey, R.B., 1953. A proposed structure for the nucleic acids. Proc. Natl. Acad. Sci. USA 39, 84–97. Protozanova, E., Macgregor Jr., R.B., 1998. Frayed wires: a thermally stable form of DNA with two distinct structural domains. Biochemistry 35, 16638–16645. Sanger, F., Nicklen, S., Cuolson, A.R., 1977. DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467.

R.B. Macgregor Jr., G.M.K. Poon / Computational Biology and Chemistry 27 (2003) 461–467 Seeman, N.C., 1998. DNA nanotechnology: Novel DNA constructions. Annu. Rev. Biophys. Biomol. Struct. 27, 225–248. Sverdlow, E.D., Monastyrskaya, W., Chestukhin, A.V., Budowsky, E.I., 1973. The primary structure of oligonucleotides, partial apurination as a method to determine the position of the purine and pyrimidine residues. FEBS Lett. 33, 15–17. Wang, A.H., Quigley, G.J., Kolpak, F.J., Crawford, J.L., van Boom, J.H., van der Marel, G., Rich, A., 1979. Molecular structure of a left-handed double helical DNA fragment at atomic resolution. Nature 282, 680– 686.

467

Watson, J.D., Crick, F.H.C., 1953a. A structure for deoxyribosenucleic acid. Nature 171, 737–738. Watson, J.D., Crick, F.H.C., 1953b. Genetical implications of the structure of deoxyribonucleic acid. Nature 171, 964–967. Wing, R., Drew, H., Takano, T., Broka, C., Takana, S., Itakura, K., Dickerson, R.E., 1980. Crystal structure analysis of a complete turn of B-DNA. Nature 287, 755–758. Yurke, B., Turberfield, A.J., Mills Jr., A.P., Simmel, F.C., Neumann, J.L., 2000. A DNA-fuelled molecular machine made of DNA. Nature 406, 605–608.