Is DNA an electrical conductor?

Viewpoint: Yes, DNA is an electrical conductor, despite inconsistent experimental results and disagreement about how it conducts.

Viewpoint: No, experiments have not conclusively proved that DNA is an electrical conductor; furthermore, there is no universally accepted definition of a wire conductor at the molecular level.

In 1941, long before the historic determination of the double helix structure of DNA by James Watson, Francis Crick, and Rosalind Franklin, Albert Szent-Györgyi, the Nobel laureate biochemist who had discovered Vitamin C, proposed that some biological molecules might exhibit a form of electrical conductivity. This proposal was based on the observation that x-ray damage—the knocking out of an electron by an x ray—at one part of a chromo-some could result in a mutation in a gene located some distance away through motion of the electrons in the molecule. As the structure and genetic function of DNA became understood, it seemed natural to expect that the proposed conductivity would be found in the DNA molecule itself.

DNA, the fundamental information storage molecule in all self-reproducing life-forms, is an interesting polyatomic assembly in its own right. DNA is not, however, a single compound, but rather a family of polymeric molecules in which strands of alternating deoxyribose and phosphate groups carrying purine and pyrimidine bases are (at low temperatures and in an aqueous medium at the right pH and ionic strength) bound to strands bearing complementary base sequences. With modern technology, DNA strands of any predetermined sequence can be synthesized and "cloned" in vitro at relatively low cost, and the availability of such carefully defined molecular material has led scientists to seek for novel applications of DNA outside the biological realm. A number of DNA-based computing schemes have been proposed and tested. There is also interest in using DNA as a building material on the nanometer scale.

Interest in DNA conductivity, apart from its biological function, has heightened in recent years, as computer-chip technology continues to develop ways of making chips from smaller and smaller components. The components still need to be wired together, and the size of the wires has to decrease if more powerful chips with larger numbers of components are to be made. Thus the prospect of being able to use individual DNA molecules as "wires" to connect the elements in future generations of integrated circuits (chips) is quite attractive.

To describe an individual molecule as a conductor requires carefully defining what being a conductor might mean on a molecular level. The most common form of electrical conductor is a macroscopic piece of a metallic element or alloy. Conduction in such materials can be viewed as the result of the component atoms participating in a special "metallic" form of bonding in which the electrons involved in bonding remain nonetheless free to travel long distances in response to an applied electric field. Some metallic conductors become superconductors below a characteristic temperature. In supercon-ductors, electrons travel in pairs, and their motion is correlated with the vibration of the component atoms in such a way that no energy is lost as heat. They exhibit no electrical resistance.

Semiconductors are another important class of electrically conducting materials. Semiconductors are elements or chemical compounds held together by localized electron-pair bonds from which electrons can escape if they gain a small amount of additional energy. Chemically pure semiconductors are devoid of conductivity at low temperatures but become conducting as they are warmed. Their conductivity is enhanced greatly if a small fraction of the atoms is replaced by an impurity, or "dopant" atoms, that have one more or fewer electrons. Other forms of conduction are also possible, for example, by a kind of electron "hopping" between impurity or defect sites.

Most of the proposed mechanisms for conduction in DNA would involve electron motion along the axis of the double helix, although whether the motion would resemble that in metallic conductors, superconductors, semiconductors, or some form of hopping, has not been definitively resolved. Conduction is assumed to take place at right angles to the planes of guanine-cytosine (G-C) and adenine-thymine (A-T) pairs that hold the two helical strands together.

The paired bases are themselves aromatic organic compounds, which means that some of their most loosely bound electrons occupy the so-called pi molecular orbitals that extend above and below the molecular planes. Depending on how the electrons in these pi orbitals interact, one could have any of the types of conduction just described, or none.

As the Yes essay indicates, it is clear that one can introduce ions between the stacked base pairs to render DNA conducting. Whether DNA is a conductor in the absence of such doping ions is less clear. Claims for normal conduction, hopping conduction, and super-and semi-conduction, have emerged from different laboratories. It is reasonable that DNA with different ratios of G-C to A-T pairs would have different conductivity characteristics, and it is not out of the question that DNA strands with certain base sequences will differ markedly from each other in conductivity. Measurement of the conductivity of a molecule will be influenced by the position and type of conducting contacts that have been made with it. The variation in results reported to date may in part be attributable to both differences in sequence and electrical contact.

If DNA, at least with some base sequences, is shown to be conducting, there is still the question of whether the conductivity has anything to do with the suitability of DNA as the genetic material. Does the conductivity in some way reduce the incidence of harmful mutations or the frequency of mistakes in reproduction? The true extent and full significance of DNA conductivity may not be fully understood for some time.

—DONALD R. FRANCESCHETTI

Viewpoint: Yes, DNA is an electrical conductor, despite inconsistent experimental results and disagreement about how it conducts.

Does DNA (deoxyribonucleic acid) conduct electricity? Yes. Researchers are not entirely in agreement as to how it conducts, but there is convincing evidence that it does. DNA research has attracted attention around the globe. Swiss researchers have demonstrated that DNA conducts electricity in the same way as a wire, while Dutch researchers have found it is a semiconductor. The Swiss team also found a clue that might explain why there have been inconsistent results in the research to determine the conductivity of DNA. Scientists at labs in France and Russia working together have found DNA conducts electricity, and they believe contradictory results may be related to how it is connected. Canadian scientists are patenting a novel form of DNA that they developed by chance, which has definite commercial potential.

The Canadian Connection

In 2001 a Saskatchewan provincial government agency provided $271,000 (Canadian) to manufacture and test new light-based electronic transistors that will employ a novel conductive form of DNA, dubbed M-DNA by the team of researchers who discovered it at the University of Saskatchewan under the direction of Jeremy Lee, professor of biochemistry at the university's College of Medicine. A Toronto-based company is adding another $277,000 (Canadian) to develop and commercialize a new biosensor tool based on the M-DNA molecule.

The novel form of DNA was an unexpected discovery, according to Dr. Lee. In what he describes as curiosity driven research, the group found that DNA readily incorporates metal ions at a high pH (a very basic solution). Conducting metal ions such as zinc, cobalt, or nickel were incorporated into the center of the DNA helix between the base pairs. Researchers then found the new DNA not only conducts electricity, but it does so without losing its ability to bind to other molecules.

DNA normally exists as two intertwined strands of many connected nucleotides. Each nucleotide consist of a 5-carbon sugar, a nitrogen-containing base attached to the sugar, and a phosphate group. The bases are adenine, cytosine, guanine and thymine; but are usually identified as A, C, G, and T. The DNA double helix looks much like a twisted ladder with the rungs formed by connecting G-C and A-T base pairs. The metal ions in M-DNA are in the center of the rungs.

In March 2000, Lee said of their discovery in an interview for SPARK (Students Promoting Awareness of Research Knowledge), "M-DNA is the smallest wire that you can imagine because it's only one molecule thick. And the beauty of DNA is that it self-assembles. You don't need a machine to put it together. It can make itself. You throw the sequences together and the base pairs automatically match up." SPARK is part of NSERC, the Natural Sciences and Engineering Research Council of Canada, a national body for making strategic investments in Canada's capability in science and technology. Among its functions, NSERC provides research grants.

Canadian, U.S., European, and Japanese patent applications have been filed on M-DNA. The University of Saskatchewan Technologies, Inc. (UST), the technology commercialization arm of the University of Saskatchewan, listed electronic applications along with biosensing and microarray applications, as among available technologies for partners to develop or license in 2001. An M-DNA molecule acts as a semiconductor. It has distinct advantages in the miniaturization of electronics in its size, and its ability to self-assemble and create highly organized and predictable structures.

Biosensors have applications in medicine, environmental monitoring, biological research, process control, security, and national defense. When groups of biosensors that can test for multiple substances are assembled, they are called microarrays. M-DNA has advantages over current DNA biosensors in that it could potentially test samples from more sources and is more sensitive and versatile. The electronic signal can be quantitatively as well as qualitatively measured, which would increase the data provided by an M-DNA biosensor, according to the researchers.

Specific applications of M-DNA in biosensing could include screening for genetic abnormalities. It could also be used to identify environmental toxins, drugs, or proteins, and also to search for new antitumor drugs that work by binding to DNA. As promising as the applications of M-DNA to biosensors are, the potential to use M-DNA as "wires" in integrated circuits, may be even more lucrative.

DNA provides the molecular blueprint for all living cells. It is also now recognized as an ideal tool for making nanoscale devices. The term nano comes from the Greek word for dwarf. It is also a prefix meaning one billionth, as in the word nanometer (nm), one billionth of a meter (m). To put that into perspective, a nanometer is about the width of 3 to 5 "average" atoms, or 10 hydrogen atoms, hydrogen being the smallest of all atoms. DNA molecules are about 2.5 nm wide. To the world of technology, nano is key to a scientific revolution based on very small things.

DNA as a Conductor Suggested by a Nobel Laureate

The biochemist Albert Szent-György (1893-1986) was awarded the 1937 Nobel Prize for Physiology and Medicine for his discoveries about the roles played by organic compounds, especially vitamin C, in the oxidation of nutrients by the cell. Born in Hungary, he emigrated to the United States in 1947 for political reasons.

The idea that DNA might work like a molecular wire can be traced back to 1941, when Szent-György suggested that biological molecules could conduct electricity, although he added, "It cannot be expected that any single observation will definitively solve this problem." As evidence Szent-György offered instances of genetic mutation that occur in one place when something such as irradiation with x rays is inflicted on chromosomes some distance away. He noted it was as if the radiation had sent an electric signal along the DNA to cause disruption at a distance. Szent-György saw the question of whether DNA conducts electricity as important to understanding the mechanics of genetic mutation. In more recent work, optical experiments with fluorescence quenching using DNA molecules has encouraged research into DNA as an electrical conductor.

Another Nobel Laureate Suggests the Nano Connection

The nano part of the DNA story can be traced back to 1959 and another Nobel laureate, Richard P. Feynman, who presented a classic lecture at the California Institute of Technology that year titled "There Is Plenty of Room at the Bottom." Feynman used biological systems as an example of working on a very small scale. He said, "[Biological systems] manufacture substances; they walk around; they wiggle; and they do all kinds of marvelous things—all on a very small scale." Feynman predicted there would be a day when we could arrange atoms the way we want, once we got the tools. These tools started to be available in 1981, when an IBM team invented a scanning tunneling microscope.

Part of the momentum for the rise in the "nanoage" has come from the semiconductor industry's concerns that Moore's law is reaching its limits. Moore's law is not a law of nature or government, but an observation made in 1965 by Gordon Moore, the cofounder of Intel, an American manufacturer of semiconductor computer circuits, when he plotted the growth of memory chip performance versus time. Moore observed that the number of transistors that can be fabricated on a single integrated circuit was doubling every 18 to 24 months. However, smaller has limits. As components get down to the 100 nm (0.00001 cm) range, they approach the quantum mechanical world of atoms and molecules, and the laws of physics for larger structures no longer apply. Research on materials that can be used in the quantum range, such as DNA, is going on in key labs around the globe. Jeremy Lee's Saskatchewan team sees DNA as a self-replicating semiconductor with a very attractive potential for future molecular computers.

The Swiss Story

In 1999, physicists Hans-Werner Fink and Christian Schonenberger of the University of Basel in Switzerland reported that they were able to measure conductivity in bundles of DNA that were 600 nm (0.00002 in, 0.00005 cm) long. They ground one end of the bundle on a carbon grid and applied a voltage to the other end through a tungsten tip, then measured the conductivity as the voltage was varied. The team reported electrical measurements that suggested DNA is a good linear conductor, and as efficient as a good semiconductor. Their interests in DNA as a conductor are particularly focused on its use in wiring ultra-small electronic devices.

Also at the University of Basel, Bernd Giese and colleagues suggested why there are differing results being reported on DNA as a conductor. They suggest the difference may be in the sequencing of the nucleotide base pairs, A-T, and G-C. Their research indicates that charge is best carried by G bases, and that G-C pairs work best where they are not separated by many A-T pairs. They concluded the sequencing made a difference, although they agreed that more research is needed.

The Dutch Connection

Research by a team of scientists at the Delft University of Technology in the Netherlands, including Cees Dekker, a professor of physics and the recipient of awards for his work in nanotechnology, and researchers from the Dutch Foundation for Fundamental Research into Matter devised an experiment to study the conductivity of DNA. They prepared an artificial DNA fragment 10.4 nm long to bridge an 8 nm gap between two electrodes and demonstrated it consistently acted like a semiconductor over a range of conditions. The conductivity was observed at ambient conditions, in vacuum, and at cryogenic temperatures. Their work was reported in Letters to Nature in 2000.

Dekker said the team used a prepared double-stranded DNA molecule of 30 G-C pairs because they learned from the earlier research that this configuration was most likely to conduct. Normal DNA contains both A-T and G-C pairs with a maximum of 15 repeats of one pair. As a semiconductor, the DNA strand acts much like the silicon used in computer chips. According to Dekker, it may be possible some day to make smaller chips using DNA. However, he adds, there is still research to be done to determine exactly how DNA transports electricity.

More on Global Research

In 2000, researchers under the direction of physics professor Tomoji Kawai at Osaka University in Japan prepared networks of DNA strands linked together in a single layer on mica. The researchers have been able to change the thickness of the DNA network, making DNA networks with 10-to 100-nm mesh up to 1.7874 in (4.5 cm) square. Kawai says their work could possibly lead to a method to produce high-density electronic devices and ultimately to integrated circuits created out of DNA. In fact, he is optimistic that a DNA memory device could be made using their techniques to deposit DNA networks early in the twenty-first century.

Kawai suggests DNA conducts electricity in complex ways, depending on the oxidation-reduction potential of the bases and the distance between them. The group tested specific complementary pairs of bases within their DNA networks. They attached gold particles to serve as one electrode, while the tip of an atomic force microscope made the second contact.

In early 2001, Alik Kasumov and colleagues at the Solid Physics Laboratory in Orsay, France, and the Moscow Academy of Science, Russia, demonstrated DNA conducts as a metal conducts at temperatures above-457.87 ∞ F (1K). Below that temperature, DNA molecules connected to superconducting electrodes 0.5 micrometers (1 micrometer is approximately 0.000039 in) apart become what they describe as proximity induced superconductors.

In 2001 and 2002, Jacqueline K. Barton and her colleagues at the California Institute of Technology published reports that offer considerable evidence to indicate that DNA conducts electricity like a metal wire. They report on the charge transport—the conductivity—of DNA in a variety of multiple-stranded DNA assemblies. The Barton Group is focusing on understanding how DNA conducts. They too have suggested there is a relation between conductivity and how the base pairs are stacked. The group is also exploring the design and application of DNA-based electrochemical sensors. They are using DNA films to develop a completely new family of DNA-based sensors.

—M. C. NAGEL

Viewpoint: No, experiments have not conclusively proved that DNA is an electrical conductor; furthermore, there is no universally accepted definition of a wire conductor at the molecular level.

Physicists, chemists, and radiation biologists have long been fascinated with the question "Is DNA a conductor or insulator?" In 1941, Nobel laureate Albert Szent-György (1893—1986) proposed that biological molecules such as chromosomes (which are composed of DNA) could conduct electricity along a chainlike form for a certain distance, after observing that x-ray radiation focused on one part of a chromosome could lead to damage on another section of the chromosome. How did the radiation get from one section of the chromosome to another? A rationalized answer—the radiation kicked off an electron that traveled along the DNA chain. In 1962, after the 1953 James Watson, Francis Crick, and Rosalind Franklin x-ray crystallography discovery of the double-helix structure of DNA, physicists Daniel Eley and D. I. Spivey measured by the DC conductivity of dried DNA samples. Their results suggested that DNA, with its unique stacking of bases, could serve as an electrical conductor.

In the 1990s scientists began to vigorously reinvestigate the electron transfer conductivity and insulating properties of DNA with a variety of research techniques. These physicists, chemists, and radiation biologists particularly focused on experiments that study how the DNA nucleotide sequence, structure, length, buffer, conducting material, and environment (e.g., liquid, air, vacuum) affect the electrical properties of DNA. The results of the many experiments have been inconclusive. Since DNA has both insulating and conducting properties, today many scientists agree that DNA is a semiconductor. According to Gary Schuster at the School of Chemistry and Biochemistry at the Georgia Institute of Technology in Atlanta, "The problem is that there is no universally accepted definition of a wire on a molecular scale. Certainly, DNA is not a classical wire in the sense that copper is."

Inconclusive Experimental Results

A series of DNA electron transfer donor-acceptor experiments (conceptually similar to having a cathode and anode attached by a wire, where the piece of DNA is the wire) in the 1990s support the hypothesis that DNA is not a conducting wire. The electron transfer rate is used to calculate beta (ß), a constant that can be used to characterize the electrical conductivity of materials. A lower beta value indicates higher electrical conductivity. In 1994, chemists Anne M. Brun and Anthony Harriman followed up on Jacqueline Barton's 1993 initial electron transfer donor-acceptor experiments at the California Institute of Technology and investigated how the length of a double-stranded DNA helix affected the rate of electron transfer along the DNA chain. They attached electron donor and acceptor groups to the DNA and measured the electron transfer rate on varying lengths of pieces of DNA. For a 17-A° piece of DNA, the beta constant was 0.9^-1. The Brun and Harriman beta value of 0.9 A °-1beta value. In 1997, Frederick D. Lewis and colleagues at Northwestern University in Illinois used photo-oxidation techniques and investigated electron transfer between a donor molecule, in this case a guanine base in the DNA strand, and an attached acceptor. The photo-oxidation experiments yielded a beta value of 0.7 A °-1. The 1998 oxidation electron transfer experiments of Keijiro Fukui and Kazuyoshi Tanaka at Kyoto University, Japan, yielded a beta value of 1.42 A°-1 between an intrinsic guanine base (donor) and an introduced dye molecule (acceptor). At the University of Basel, Switzerland, Bernd Giese's electron-transfer-hopping work in 1998 with guanine radicals as donors and guanine bases as acceptors in different lengths of DNA sequences yielded a beta value of 1.0 A°-1. In 1998, Satyam Pryadarshy and Steven M. Risser at the University of Pittsburgh performed theoretical quantum chemical calculations using the same experimental electron transfer donor-acceptor DNA systems and found beta values in the range of 0.6 A °-1 differs from Barton's 0.2 A°-1—1.4A °-1. The experimental and theoretical beta values for the DNA electron donor-acceptor systems are in contrast to the beta values of conducting carbon nanowires, which are 0.0 A °-1—0.2 A °-1.

Mobility, a measure of an electron's travel velocity in a medium, can also be calculated from electron transfer experiment data. For conducting materials (metals), the mobility is on the order of 103. For semiconductor materials, the mobility is 10^-6—10². For insulating materials, the mobility is less than 10-14. The calculated mobilities in the DNA systems are in the range of 10^-5 to 10^-7 cm²/V/s which characterizes the DNA electron-transfer systems as semiconductors, not conductors.

The Work of Fink and Schonenberger

In 1999, physicists Hans-Werner Fink and Christian Schonenberger at the Institute of Physics in Basel, Switzerland, continued the investigation of the importance of the length of DNA and conductivity. Fink and Schonenberger were able to set-up a 600-nanometer (nm) (0.00002 in, 0.00005 cm) piece of DNA and attempt to measure a current through the DNA with a fine needle-like tip of tungsten. The experimental results showed that the resistivity values were comparable to those of conducting polymers. Therefore, DNA is transporting electrical current similarly to a good semiconductor.

The many scientists have also been interested in deciphering the role that the sequence of nucleotides in a strand of DNA plays in modulating the electrical insulating or conductivity properties. Bernd Giese's electron transfer work with guanine bases shows that the electron "hopping" transfer between guanosine-cytosine (GC) base pairs is less efficient the more adenosine-thymidine (AT) pairs that are in between. Cees Dekker and his colleagues at the Delft University of Technology in the Netherlands have used varying lengths of strands of DNA that are made only of guanosine-cytosine base-pairs in electrostatic trapping and scanning tunneling experiments that measure the resistance as greater than 10¹³ ohms for pieces of DNA that are 40 nm and longer. Erez Braun and coworkers at the Technion in Israel have also observed DNA that is 16 um long acting as a conductor. Researchers in Tomoji Kawai's group at Osaka University in Japan also showed that sequence and length of DNA affected the resistance properties and measured values that range from 10⁹ ohm to 10¹² ohm using scanning tunnelling microscopy.

William A. Bernhard and his biophysicist colleagues at the University of Rochester School of Medicine in New York have approached the question of whether DNA is a conductor or an insulator by measuring the electrical conductivity (resistance) of DNA at different temperatures. The biophysicists irradiated crystalline DNA with x rays at 4K (-452.47°F,-269°C) and measured electrons that were trapped in the DNA with electron paramagnetic resonance. They found that the crystalline DNA trapped as much as 60% of the electrons. The measurements of trapped electrons are characteristic of an insulator or poor conductor rather than a conductor. The scientists argue that the total number of radicals trapped in the DNA appears to be relatively independent of factors such as DNA conformation, sequence, and water content, and primarily a function of the density of packing. Mobility measurements from the x-ray irradiated crystalline DNA indicate an increase in charge mobility for DNA that is warmed to room temperature, which is characteristic of a semiconductor.

With all the experimental and theoretical data about DNA conductivity that does not classify DNA as a classical conductor or insulator, trying to understand the mechanism by which electrons are traveling through the DNA can help better define DNA as a molecular wire or conductor. The electron transfer experiments by Jacqueline Barton and coworkers led the scientists to propose that electrons delocalize and transfer through what she terms the "pi-way" orbital structure of the duplex DNA. The work of Giese and other researchers led scientists to suggest a parallel super-exchange sequential charge-hopping mechanism where the electrons hop as discrete electronic entities where there is not significant electronic overlap between the piorbitals of the adjacent base pairs. The photo-chemistry experiments of Henderson, Schuster, and colleagues has led to a suggested mechanism of phonon-assisted polaron-like hopping. In the polaron-like hopping mechanism, the researchers propose that small domains (perhaps a string of five bases) form a delocalized polaron, and the polaron hops from domain to domain on the DNA duplex strand. However, DNA does not have enough degree of delocalization of the polaron or a sufficient rate of hopping to be classified as wirelike conductance similar to conducting polymers such as doped polyacetylene and polythiophere, or copper wire.

Conclusion

What will scientists do with the information about the electrical conductivity and resistance properties of DNA? Understanding these properties leads to developments in the field of nanotechnology. Nanoelectronics consists of wires, transistors, and other components that have dimensions measured in billionths of a meter. Today, DNA is already being used in biosensor technology. The role the DNA plays in each device is different. Scientists hope to build more specific and precise nanocircuits that may be DNA-like molecules by using knowledge about DNA's electrical conductivity and resistance.

—LAURA RUTH

KEY TERMS

ANGSTROM:

1 x 10^-10 (one ten-billionth) of a meter.

BASE:

Molecule or ion that can combine with a hydrogen ion. The four bases in DNA are all nitrogen-containing organic compounds; "organic" means they contain combined carbon.

BASE-PAIR:

The pairing of two nucleotide bases. The base guanine pairs only with cytodine, creating the G-C or C-G base-pair. The base adenine pairs only with thymidine, creating the A-T or T-A configuration. The chemical bonds between a string of nucleotide base pairs hold the two strands of DNA material together in a shape described as a double helix.

CONDUCTOR:

Material or substance that transfers electricity, heat, or sound.

CRYSTALLINE:

In a solid state (not a liquid, solution, or gas).

DNA:

Deoxyribonucleic acid.

INSULATOR:

Substance that prevents or reduces the transfer of electricity, heat, or sound.

KELVIN:

Scale of temperature (abbreviated as K, with no degree symbol) in which the size of the degree is the same as in the Celsius system, but where zero is absolute zero, not the freezing point of water. Absolute zero is the theoretical point where a substance has absolutely no heat energy (i.e., there is no molecular motion). 32°F (0°C) is 273K.

SEMICONDUCTOR:

Substance that can conduct electricity under some, but not all, conditions. The properties of a semiconductor can depend on the impurities (called dopants) added to it. Silicon is the best-known semiconductor and forms the basis for most integrated circuits used in computers.

NANOMETER:

1 x 10^-9 (one one-billionth) of a meter.

NUCLEOTIDE:

Nitrogen-containing molecules, also called bases, which link together to form strands of DNA. There are four nucleotides, named for the base each contains: adenine (A), thymidine (T), cytodine (C), and guanine (G).

OHM:

Unit of electrical resistance.

OXIDATION-REDUCTION:

Any process that makes an element, molecule, or ion lose one or more electrons; where reduction is the gain of one or more electrons. Whereas electrons have a negative charge, gaining electrons reduces (i.e., makes less positive) the charge on the element, ion, or molecule. Oxidation, being an opposite process, raises the charge (i.e., makes more positive).