DNA nanotechnology: a future perspective
© Zahid et al.; licensee Springer. 2013
Received: 15 January 2013
Accepted: 21 February 2013
Published: 4 March 2013
Skip to main content
© Zahid et al.; licensee Springer. 2013
Received: 15 January 2013
Accepted: 21 February 2013
Published: 4 March 2013
In addition to its genetic function, DNA is one of the most distinct and smart self-assembling nanomaterials. DNA nanotechnology exploits the predictable self-assembly of DNA oligonucleotides to design and assemble innovative and highly discrete nanostructures. Highly ordered DNA motifs are capable of providing an ultra-fine framework for the next generation of nanofabrications. The majority of these applications are based upon the complementarity of DNA base pairing: adenine with thymine, and guanine with cytosine. DNA provides an intelligent route for the creation of nanoarchitectures with programmable and predictable patterns. DNA strands twist along one helix for a number of bases before switching to the other helix by passing through a crossover junction. The association of two crossovers keeps the helices parallel and holds them tightly together, allowing the assembly of bigger structures. Because of the DNA molecule's unique and novel characteristics, it can easily be applied in a vast variety of multidisciplinary research areas like biomedicine, computer science, nano/optoelectronics, and bionanotechnology.
Nucleic acids (e.g., deoxyribonucleic acid (DNA) and ribonucleic acid (RNA)) encode the genomes of all living things on earth. Of these, DNA has become a key biological molecule in the study of genetics, medicine, and biotechnology. It possesses the natural ability to self-assemble and interacts with a wide range of molecules. Besides its importance in genetic studies and its application in various biological fields like biomedicine, cancer research, and genetic engineering, DNA has also become a preferred material for nanotechnologists because of its unique properties of structural stability, programmability of sequences, and predictable self-assembly. Nanobiotechnology is made up of two words: ‘nano’ pertains to the study or development of structures in the 1 to 100-nm size range in at least one dimension, while ‘biotechnology’ refers to technological tools associated with the development of living things or biological molecules. Thus, components of natural biological systems are scrutinized by nanobiotechnologists to engineer innovative nanodevices .
This has led to the development of several macroscopic structures with nanometer-size features [4–7]. DNA nanotechnology has also been used to produce various kinds of reprogrammable functionalized devices and sensors, some of which will be discussed in this review.
The history of nanoarchitecture is fairly short. In the early 1990s, Seeman and colleagues first described a process by which DNA could be hybridized in more than one way to create self-assembling nanostructures. They created tiles made up of DNA with sticky ends which were allowed to hybridize to form a cube-like structure [8, 9]. Yurke et al. experimented with the interesting idea that a single DNA strand can undergo multiple hybridizations through strand displacement cycles using a toehold or hinge made up of the DNA itself. Instead of using proteins and other bio-supportive molecules to build their structures, they demonstrated that DNA strand displacement and hybridization was enough to coax molecular-level changes in the structure of DNA. They achieved this by exploiting two double helical arms of DNA connected by another short DNA sequence acting as a ‘hinge’. This ‘hinge’ repeatedly cycled the two strands into an opened and closed state by consecutive addition of two single-stranded DNA molecules . This method made it possible to form a variety of nanostructures based on differences in sequence, rather than being dependent on the influence of changes in the environment surrounding the DNA (pH, salt, and temperature) [11, 12].
DNA-modifying enzymes can also be used to generate and manipulate DNA nanostructures. Although studies in this area have so far been limited, many design tools have been developed for the application of these enzymes to alter DNA in a sequence-specific manner. Most of these enzymes work like small nanofactories and are, hence, highly specific in their actions, based on various biological processes .
Despite these advances in DNA nanotechnology, it remains in the development phase. Generally, only about 30% of the assembled DNA molecules are similar to the original design . This presents a great challenge for the development of techniques to fabricate modern DNA nanostructures, especially in the DNA computational area. Researchers compare this process with the complicated and eventually successful development of electronics, computers, and automobiles. Besides errors in the ‘designed’ genetic sequences, another shortcoming is that prolonged thermal cycling for up to 24 h is required to produce a useful nanodevice. In case of automobiles, it took over a decade to produce the first functional prototype. Hopefully, the development of potent nanomaterials will not take as long. Here, we review some of the functional challenges and exciting future prospects of developing nanobiotechnology with a special focus on DNA nanotechnology.
In addition to the use of liposome-based nanoparticles to carry miniscule amounts of chemotherapeutic agents to affected cancer sites, albumin-bound nanostructures may be used to enhance permeability of the endoplasmic reticulum for breast cancer therapy . Most nanostructures, however, are considered insufficient for effective treatment of cancer cells. This has led to the development of potent ‘nano-systems’, generally possessing four basic qualities: firstly, they can themselves be therapeutic or diagnostic and thus in theory can be designed to carry a hefty therapeutic cargo deliverable to the tumor site. Secondly, more than one targeting ligand can be attached to these nanosystems, providing high affinity and specificity for target cells. Thirdly, these nanosystems have the advantage of being able to house more than one type of therapeutic drug, thereby providing multivalent drug therapy. Finally, most nanosystems that are designed from biological materials such as DNA and RNA are ‘programmed’ to be able to evade most, if not all, drug-resistance mechanisms. Based on these properties, most nanosystems are able to deliver high concentrations of drugs to cancer cells while curtailing damage to surrounding healthy cells .
Porous inorganic particles can now be loaded with an assortment of drugs contained in organic nanomicelles that can target very specific cells and tissues in the body. Some of these carbon nanotubules are very potent drug delivery vehicles for cancer treatment . The tubular structure of nanotubules allows for both carrying and protection of drugs from external influences. Therapeutic applications which involve nanomaterials combined with cytotoxic materials such as antineoplastic or chemotherapy agents are a key area of development for science and technology .
Research is also being conducted on the use of highly organized DNA lattices to detect biological activity of various molecules. Amin and colleagues have developed a biotinylated DNA thin film-coated fiber optic reflectance biosensor for the detection of streptavidin aerosols. DNA thin films were prepared by dropping DNA samples into a polymer optical fiber which responded quickly to the specific biomolecules in the atmosphere. This approach of coating optical fibers with DNA nanostructures could be very useful in the future for detecting atmospheric bio-aerosols with high sensitivity and specificity .
Nucleic acid nanotechnology has many other applications besides medical diagnosis and drug therapy. Synthetic polymers such as dendriworms are made up of dendrimer units of magnetic nanoworms and are being used for intercellular delivery of small interfering RNA (siRNA). These siRNA carriers are assembled from magnetic as well as fluorescent nanoparticles.
siRNAs have been widely acknowledged as a potent new class of therapeutics, which regulate gene expression through sequence-specific inhibition of mRNA translation. siRNA delivery vehicles such as lipid and polymer nanoparticle-based dendrimers have proven effective in improving the stability, bioavailability, and target specificity of siRNAs following systemic administration in vivo . Other important applications have included the activation of enzyme cascades on topologically active scaffolds. This process makes use of DNA self-assembly and uses DNA as a scaffold. Enzymes or cofactor enzymes are attached to this scaffold and then plays an active role in improving the biological efficiency of the system . Bionanotechnology has also been applied in the field of contraception. Where traditional methods have employed over-the-counter drugs and an assortment of widely available contraceptives, bionanotechnology aims to develop drugs that may be effective in targeting the fallopian tubes while anti-implantation drugs can be employed in the uterus to foil pregnancy without influencing other organs. Current studies are centered on manipulating follicle stimulating hormone (FSH) and its inhibitor known as FSH binding inhibitor in mice  and monkeys .
DNA computing was first proposed as a means of solving complex problems by Adleman in 1994. He recognized that the incredible storage capacity of DNA could be used to solve complex computational problems. For this, he picked a common mathematical problem normally referred to as the ‘traveling sales man problem’ and was able to solve it using strands of DNA . In 1996, a new technology called the ‘sticker DNA’ model was introduced by Roweis and colleagues. This model applies to random access memory and requires no enzymes or strand extension. This method, thus, has the capability of becoming the universal method for DNA computation. A controlled robotic work station helped not only in implementing the sticker model but also in reducing error rates . Since then, many technologies which make use of DNA to resolve basic mathematical equations and pure computational problems have been developed.
Inspired by Adelman's experiment, researchers have been able to solve a diverse group of mathematical problems using DNA molecules. In 2011, Qian and Winfree were able to calculate square roots using ‘seesaw’ logic gates. The idea behind these gates is that a single stretch of DNA can pair up with various molecules, thus allowing competition for binding sites. Once a molecule is attached, it can be replaced instantly to allow other molecules to fasten themselves to the resident sequence, which itself can be displaced again. This system allows ‘gates’ to be loaded with several input molecules and generates logical output molecules as a result. The various DNA strands can come to represent numbers, of which output can yield the square root result as answers .
In another attempt to mimic smart biological computations, the Qian group has developed an artificial neural network. This model employs the use of four neurons. A neuron in its natural environment is susceptible to many incoming inputs, and it ‘reacts’ or ‘fires’ when it reaches a certain threshold. Based on their previous development of logic gates, Qian and his colleagues were able to construct Boolean logical circuits and other circuits which could store memories. The DNA logic circuits were not only able to recall memory using incomplete information but also to determine when conflicting answers were obtained . In other instances, scientists have also used sticker-based DNA to solve the independent set problem . Unlike the earlier sticker DNA system, this model had a random access memory and, thus, required no extension of its strands and enzymes .
Inspired by Roweis and Adelman's methods, Taghipour and colleagues  set out to unravel the independent set problem through the use of DNA computing. In the beginning, a solution space was created using memory complexes made up of DNA. Then, by the application of a sticker-based parallel algorithm, the independent set problem was solved in polynomial time. Other biological molecules besides DNA have also been used for computation. Faulhammer and colleagues used RNA to solve an assortment of chess problems through DNA computing . Bandyopadhyay and colleagues were able to apply the same reasoning and used 2,3-dichloro-5,6-dicyano-p-benzoquinone which is capable of transforming between four different states to mimic natural phenomenon such as diffusion of heat and detection of cancer growth .
DNA has also been applied for the development of pure computational methods. While many techniques are available to use DNA for computation, the most widely used technique involves the manipulation of mixtures of DNA on a support. A DNA molecule which encodes all possible solutions to a designed problem is synthesized and attached to this supportive surface. Repeated hybridization cycles and action of exonuclease enzymes are used to digest, identify, and eliminate non-solution strands of DNA. Upon completion of this step, several polymerase chain reaction (PCR) reactions are used to amplify remaining molecules, most of which are then hybridized to an array of molecules . Recent progress in DNA computation has been remarkable. Although these advances may be far off to be equivalent of the today's computational capacities of computers, the long-term goal of this research would be DNA computing, overriding everyday computing with great perfection.
The term nanoelectronics refers to the use of nanotechnology for the use and development of electrical components and circuits. Nanoscale electronics have been developed at the molecular level. Such devices are referred to as molecular electronics . Nanoelectronics had been highly dependent on the complementary-symmetry metal-oxide semiconductor (CMOS) technology. CMOS has been vital in analogue circuits such as image sensors, data convertors, and logic-based devices such as digital logic circuits, microcontrollers, and microprocessors . However, CMOS is being replaced as the demand for further miniaturization and processing speeds increase. CMOS circuitry has limitations that can greatly influence the size and shape of computers and other electronics.
Various spectroscopic methods were also used to investigate DNA conductivity. The movement of electrons was detected at the level of single molecules by fluorescence decay. Varying fluorescence levels indicated how electrons may have been transferred along the DNA chains [68, 69]. Contact methods can be used to measure conductivity directly. Molecules are laid directly on top of gold electrodes, and current flowing across these circuits is plotted on a graph to ascertain levels of conductivity. However, with this method, it is often difficult to determine whether DNA molecules are in direct physical contact with the electrodes. It is thought that weak physical contact between the DNA and electrode produces an insulating effect and, thus, accounts for varying resistance across the circuit. An expansion in experimental methodology to measure conductivity by a contactless approach will improve understanding of this process .
This model is essentially a grain connected by two tunnel junctions to a voltage source. The DNA molecule is not very conductive; however, it does possess a large energy gap which makes single electron transfer possible. In order for this circuit to operate as a transistor, the voltage supplied to the circuit is varied around threshold levels. This voltage can be varied if the tunneling rates of electrons between the two junctions are different or if there is a gap in the density of the energy states of the grain. The natural energy gap of the DNA can be enhanced using a longer strand of DNA having more than one grain. Longer chains of DNA tend to have more non-linear effects. As a result, more charges are formed. A large uncoated DNA molecule is, thus, used as compared to one that is entirely coated with a metal sheath. The tunneling rates of electrons, however, are about the same as the two phosphate bonds are identical. To counter this effect, a chemical group may be attached to one of the phosphate bonds, thus altering its properties and making electron transport and transistor behavior possible .
Some studies have reported the formation of three-dimensional structures such as switches  and motors ; devices such as DNA-based capacitors are also being contemplated. Biological polymer-based DNA hybrids have intriguing electrical characteristics such as a high dielectric constant, dielectric breakdown behavior, and good resistivity. These are encouraging signs for the development of DNA-based capacitors . In another DNA-bioploymer-based study, Nakamura and colleagues developed a light-emitting diode based on a DNA/polyaniline/Ru(bpy)32+ and tris(8-hydroxyquinolinato)aluminum complex. The voltage across the hybrid circuit was increased from 5 to 14, 16, and finally 18 V. The light emitted varied in color, ranging from green, yellow, orange, and finally to red. This was the result of electron transfer in the DNA hybrid molecule with increasing voltage . Other important DNA-based nanoscale devices that have recently been developed include highly conductive nanowires , quantum dots with carbon nanotubules , and even radically advanced devices which detect single-nucleotide polymorphism and conduct nucleotide sequence mutation analysis . With added progress in this field, it could be possible to use DNA-based electronics for both DNA-based diagnostics and sophisticated nanoscale electrical devices.
In another study, Kim and colleagues attempted to construct a biosensor based on graphene and polydimethylsiloxane (PDMS) . An evanescent field shift occurred in the presence of chemical or biological structures which were very sensitive in the refractive index. They were able to monitor the target analyte by attaching the selective receptor molecules to the surface of the PDMS optical waveguide resulting in a shift of the optical intensity distribution. Hence, they monitored the electrical characteristics of graphene in the dark and under PDMS wave-guided illumination. Changes in the resulting photocurrent through the graphene film showed that the fabricated graphene-coupled PDMS optical waveguide sensor was sensitive to visible light for biomolecular detection . This finding can be used for the development of optical biosensor for the detection of various biological molecules in future biological assays.
The rise of DNA-based nanobiotechnology has led to an increase in demand for synthetic DNA. DNA can be synthesized from nucleotides into small molecules such as ssDNA up to entire viral genomes. In spite of these accomplishments, the time and cost of synthesizing such molecules have somewhat limited the use of DNA as a current research tool. Another significant drawback in this technology has been the significant error rate of synthetic DNA sequences . The reduction and correction of errors are, thus, essential for the synthesis of long DNA molecules. The correction of these errors is, however, very time-consuming and expensive. There are several approaches to develop error-free sequences in synthesized populations of DNA.
These methods may include, but are not limited to, physical separation which may apply the use of metals to chelate partially denatured purine bases and allow elimination of errors  or PCR-based approaches such as hairpin PCR, which completely separates genuine mutations from polymerase mis-incorporations. Hairpin PCR operates by converting a DNA sequence to a hairpin following ligation of oligonucleotide caps to DNA ends. Conditions are such to allow a DNA hairpin to be efficiently PCR‐amplified so that during DNA synthesis, the polymerase copies both DNA strands in a single pass. Consequently, when a mis-incorporation occurs, it forms a mismatch following DNA amplification and is distinguished from genuine mutations that remain fully matched .
New approaches in the production of error-free DNA exploit the use of self-assembly and natural error correction proteins. Among these proteins, celery I nuclease enzyme (CEL I; Surveyor, Transgenomic, Inc., Omaha, USA) endonuclease has been very useful . Hughes and colleagues  found CEL I to be a reasonably effective at reducing synthetic DNA errors up to six times. The enzyme is added to previously amplified PCR product, and this mixture is subjected to a second round of thermal cycling at the end of which it is put through gel electrophoresis, quantified, and cloned. CEL I is a naturally occurring enzyme that cleaves mismatched DNA sequences [93–95]. It is, thus, most effective at removing common insertions and deletions that may occur during DNA synthesis .
Another tactic in dealing with error-prone DNA synthesis is changing the way we synthesize premeditated DNA. Usually, the formation of synthetic DNA requires the use of PCR-based technologies, but microarrays are now also used to synthesize DNA . In this case, DNA synthesis typically relies on spatial confinement of reactions to certain regions on a silica chip since this technology employs the addition of picoliters of reagents to the silica chip. Error rates can be reduced by controlling the locations on the chip where the reagents eventually end up. Another possibility could be directing reacting reagents through the use of photochemistry. In this way, light can be used to block or restrict reactions at potential error sites. Directing redox reactions only at desirable sites in the forming DNA is another approach. All these strategies can help reduce error rates from 1 in 200 bases to 1 in 600 bases .
DNA is one for the most useful engineering materials available in nanotechnology. It has the potential for self-assembly and formation of programmable nanostructures, and it can also provide a platform for mechanical, chemical, and physical devices. While the formation of many complex nanoscale mechanisms has been perfected by nature over the course of millennia, scientists and engineers need to aggressively pursue the development of future technologies that can help expand the use of DNA in medicine, computation, material sciences, and physics. It is imperative that nanotechnology is improved to meet the need for better detectors in the fields of biological and chemical detection and for higher sensitivity. In terms of DNA-based nanostructures, there is an urgent need to develop sophisticated architectures for diverse applications. Currently, much progress is being made in modelling DNA into various shapes through DNA origami, but the next step is to develop intelligent and refined structures that have viable physical, chemical, and biological applications. Despite the fact that DNA computation may be in its infancy with limited forays into electronics and mathematics, future development of novel ways in which DNA would be utilized to have a much more comprehensive role in biological computation and data storage is envisaged. We are hopeful that the use of DNA molecules will eventually exceed expectations far beyond the scope of this review.
SHP is working as an assistant professor in the Department of Physics and SKKU Advanced Institute of Nanotechnology (SAINT) at the Sungkyunkwan University, Suwon, Korea. His research interests span experimental nanobio sciences including but not limited to physical and biological circuit design and device fabrication using nanoscale materials; design, fabrication, and testing of micro/nanomechanical devices; electrical and mechanical characterization of circuits, sensors and devices; and biophysics, especially in DNA bottom-up self-assembly and its applications. RA is working as an assistant professor in the Interdisciplinary Research Center in Biomedical Materials (IRCBM) at COMSATS Institute of Information Technology, Lahore, Pakistan. His research interests are in the field of artificially designed DNA nanostructures and their applications in different fields, especially in biosensor applications, nanodevices designing and fabrication, and tissue engineering, especially in assisting burn patients.
Celery I nuclease enzyme
Complementary metal-oxide semiconductor
Polymerase chain reaction
Small interfering ribonucleic acid
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MEST) (2012-005985).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.