Dynamically Amorphous Character of Electronic States in Poly(dA)-Poly(dT) DNA James P. Lewis,* ,² Thomas E. Cheatham, III, ‡ Eugene B. Starikov, § Hao Wang, | and Otto F. Sankey | Department of Physics and Astronomy, Brigham Young UniVersity, N233 ESC P.O. Box 24658, ProVo, Utah 84602-4658, Departments of Medicinal Chemistry and of Pharmaceutics and Pharmaceutical Chemistry, UniVersity of Utah, 30 South 2000 East, Room 201, Salt Lake City, Utah 84112-5820, Karolinska Institutet Department of Biosciences at NOVUM Center for Structural Biochemistry, S-141 57 Huddinge, Sweden and Institute for Crystallography, Free UniVersity of Berlin, Takustr. 6 D-14195 Berlin, Germany, and Department of Physics and Astronomy, Arizona State UniVersity, Tempe, Arizona 85217-1504 ReceiVed: August 15, 2002; In Final Form: January 2, 2003 We present theoretical work on the electronic states in a model DNA double helix of poly(dA)-poly(dT) (10 base pairs) as the molecule undergoes thermal fluctuations at room temperature. We couple state-of-the art empirical force field molecular dynamics (MD) simulations with an ab initio tight-binding formalism based on density-functional theory [Lewis et al. Phys. ReV.B 2001, 64, 195103-1]. The dynamical features of the charge density distributions and the electronic structure are presented. The periodic structure exhibits extended HOMO-LUMO electronic states; however, equivalent states are quite localized in the aperiodic structures generated as snapshots from the MD simulation. Our results show strong Anderson localization in DNA as a result of the disorder due to structural changes promoted by the thermal fluctuations. 1. Introduction The mechanisms of electron or hole transport in deoxyribo- nucleic acid (DNA) have been examined intensely within the past few years. Fundamental attempts to answer questions have generated debate as to whether DNA is an insulator, a semiconductor, a metal, or a superconductor. Pioneering experi- ments have produced different interpretations and lively discussion. 1-8 The latter interest is driven by DNA’s supreme importance in biology and life as carriers of genetical informa- tion. Also, charge migration is critical in oxidation and reduction processes related to DNA radiational biology, and there is recent interest in DNA as the active element in potential molecular electronics devices. This last proposal is especially attractive, since advanced synthetic methods exist that produce, on-demand, a wide variety of complex DNA sequences and structures. There are highly contradictory issues concerning long-range charge transfer along nucleic acid polymers. Resolution of these issues bears a profound technological significance for nucleic acid nanotechnology, as well as for creating new hybrid DNA- polymer materials (see, for example, refs 9 and 10). Of course, charge transfer in DNA has possible biological and physiological importance, for example, in connection with DNA repair of oxidative damage, see for example, refs 11-14). There are proposals concerning the therapeutic significance of DNA charge transfer, 15 including the repair of well-known 16,17 mutagenic photolesions in DNA (cis-syn-thymine [2+2] photodimers), although such proposals are vigorously debated. 18 DNA charge-transfer experiments are devoted to two main physicochemical aspects: (i) intrinsic conductivity and photo- conductivity and (ii) induced conductivity of nucleic acids. Measuring the intrinsic conductivity and photoconductivity of nucleic acids is a mature 19 but still challenging proposal since nucleic acids appear to conduct electrical current comparably to conventional conjugated polymers such as polyacetylene. 4,20 Similarly, measuring the induced conductivity of nucleic acids can be performed in a variety of ways; 21,22 however, despite Herculean efforts, a complete understanding of the mechanisms of charge transfer through DNA remains a challenge. 23,24 A debatable issue is whether DNA double helices can be consid- ered as molecular wires or not. The problem is whether DNA charge transfer has a long-range or short-range nature, as several experiments and simulations observe both mechanisms. 25-28 The notion of a molecular wire seems to apply to the DNA double helix because of its unique π-electron system of bases stacked upon each other. At first glance this is reminiscent of certain charge-transfer molecular metals such as TTF-TCNQ. More- over, this speculation is vivid, since, as mentioned earlier, the nature of base interactions within the stacks is not yet completely understood. To analyze the experimental data on DNA charge transfer, one conventionally uses Markus-Levich-Jortner theory of charge transfer, 29 which has proven to be successful for various charge transfer processes including those in proteins. According to Markus theory, an electronic mixing is required between the initial (donor) and final (acceptor) states. In most intra- and intermolecular charge transfer reactions of (bio)organic chem- istry, the charge is localized only on the initial and final states, so that coherent (superexchange or tunneling) transfer applies. Intermediate (bridge) states are off resonant (higher or lower in energy than the donor and acceptor states). This produces a transfer rate that is exponentially dependent on the distance between the donor and acceptor. If the bridge states are close in energy to each other and to the donor and acceptor states, * Corresponding author. ² Brigham Young University. ‡ University of Utah. § NOVUM Center for Structural Biochemistry and Free University of Berlin. | Arizona State University. 2581 J. Phys. Chem. B 2003, 107, 2581-2587 10.1021/jp026772u CCC: $25.00 © 2003 American Chemical Society Published on Web 02/21/2003