J. Phys. Chem. B 2006, 110, 18661-18664
18661
Structures of Nucleobases Trapped within Au Triangles and Its Effects on Hydrogen Bonding in Base Pairs of DNA P. J. Mohan, Ayan Datta, Sairam S. Mallajosyula, and Swapan K. Pati* Theoretical Sciences Unit, Chemistry and Physics of Materials Unit, and the DST Unit on Nanoscience, Jawaharlal Nehru Center for AdVcanced Scientific Research, Jakkur Campus, Bangalore, India ReceiVed: June 22, 2006
Nucleobases (adenine (A), thymine (T), cytosine (C), and guanine (G)) trapped within two metal clusters such as Au3 undergo expansion. Our investigation reveals that this primarily arises due to the concomitant increase in all the bond lengths in molecules. Such expansion of the molecules can be qualitatively understood on the basis of classical harmonic potentials in the bonds and loss of aromaticity in the rings. Specifically, the highly electronegative O and N elements in the base pairs anchor to Au atoms and form X-Au bonds, which leads to charge redistribution within the molecules. As a very important consequence of this, the nature of the hydrogen bonds (in Au3-A‚‚‚T-Au3 and in Au3-G‚‚‚C-Au3) change substantially within these electrodes in comparison to gas-phase structures. These hydrogen bonds have a single-well potential energy profile (of the type N‚‚‚H‚‚‚O and N‚‚‚H‚‚‚N) instead of double-well potentials (like N-H‚‚‚O or N-H‚‚‚N/ N‚‚‚H-N types). A detailed energy calculation along the proton movement pathway supports our conclusions.
Microscopic understanding of the basis of interaction between organic molecules and metal electrodes is an important aspect in the emerging area of mesoscopic materials.1,2 The subtle changes along the molecule-metal interfaces critically effect exotic phenomena in these materials such as molecular conductance and spintronics.3 Thus, it is fundamentally important to understand the chemical nature of interactions between the molecules and the electrodes. In a similar context, recently there has been a huge interest in studying the conducting and electronic states of biomaterials such as DNA.4,5 The exact nature of the quasi-particles involved in long-range non-Marcus kinetics of electron transfer (ET) in DNA is poorly understood, with different research groups suggesting importance of different bound states such as polarons, bipolarons, excitons, and kondolike bound states.6 However, the actual performance of DNA as a molecular electronics material will depend crucially on the interaction of DNA with the metal contacts.7 To understand the exact nature of DNA-metal interactions, we have performed quantum-chemical calculations on Au3base-Au3 (base ) adenine (A), guanine (G), cytosine (C), and thymine (T)) systems. The structures of these bases held within these metal contacts show an expansion in the volume of the molecule, increase in bond lengths, and red shift in their stretching frequencies. More interestingly, for the cases of the base pairs Au3-A‚‚‚T-Au3 and Au3-G‚‚‚C-Au3 the hydrogen bonds become bridged type of the form X‚‚‚H‚‚‚Y rather than of the conventional form X-H‚‚‚Y or X‚‚‚H-Y thereby having a single-well potential rather than an asymmetric double-well potential energy profile that is associated with conventional hydrogen bonds. The geometries of the bases and base pairs were retrieved from the Protein Data Bank (PDB) for B-DNA (PDB code: 1BDNA) .8 After removal of the sugar and phosphate backbones, the molecular structures for A, T, G, and C were optimized at * Author to whom correspondence should be addressed. Phone: +9180-22082839. Fax: +91-80-22082766. E-mail:
[email protected].
the B3LYP/6-31+(d) level using the Gaussian 03 suite of programs.9 Frequency analyses were performed to verify the absence of vibrational instabilities. As a general trend, we observe that the bases and base pairs remain almost unchanged from those in the crystal structure other than of course the position of the hydrogens that are not unambiguously determined from crystallography. We model the electrodes by considering the first layer of the 111 face-centered cubic (FCC) layer of bulk gold crystal as an equilateral triangle with an Au-Au distance of 2.9 Å.10 Modeling the electrodes through a small cluster is a good assumption because the base pairs effectively interact with only this first layer of the gold surface. It is however important to note that this equilateral triangle is a model for the crystal of bulk Au. Next, we optimize the geometries of the bases and the base pairs within these two equilateral Au triangles using the optimized gas-phase geometries as the initial starting point. This is done at the B3LYP level through split basis functions with the LANL2MB basis set for the Au atoms and the 6-31+G(d) level for all the atoms of the bases and base pairs followed by vibrational analysis for removal of any unstable vibrational modes. So as to check the reliability of these split basis functions, we have redone the geometry optimizations for all the systems (bases, base pairs, and the Au3-base pair-Au3) and frequency analysis at the B3LYP/LANL2MB level, which has been shown to be quite relieable.11 We observe that the structures remain unaffected. In Figure 1, the optimized structures of the four bases within the gold clusters are shown. As clearly seen, the bases interact asymmetrically with the gold atoms in the clusters on either side. As a general trend, the Au atoms interact with the more electronegative atoms such as N and O that possess a lone pair of electrons suggesting a Lewis acid (Au)-base (N or O) type of interaction between the metal electrodes and the DNA bases. For example, in adenine, the Au clusters are bonded with the free N atoms instead of the N atom in the -NH or -NH2 group in both the five- and six-membered rings. In thymine, the
10.1021/jp0639041 CCC: $33.50 © 2006 American Chemical Society Published on Web 08/23/2006
18662 J. Phys. Chem. B, Vol. 110, No. 37, 2006
Mohan et al.
Figure 1. Minimum energy structures for adenine (A), thymine (T), guanine (G), and cytosine (C). Atom color code: blue, N; black, C; white, H; red, O; yellow, Au.
bonding is through the oxygen atom of the CdO group. Similar modes of binding are observed for T and G as well. However, for C, the binding is through the N atom in NH2 group as there exists no other electronegative atom in proximity to the Au cluster. A major structural change observed in the molecules that are anchored to Au3 is that there is an overall increase in all the bond lengths in the bases, leading to an expansion of the volumes within the electrodes. For example, the CdC bond lengths increase from 1.411 Å in the free bases to 1.463 Å in the Au3 coupled bases. Similarly, there is an increase in C-N and CdO bond lengths from 1.386 and 1.221 Å to 1.459 and 1.291 Å in the free and complexed forms, respectively. The C-H and N-H bond lengths also increase from 1.011 and 1.01 Å to 1.047 and 1.045 Å, respectively. Thus, overall there is an increase of ∼0.07 Å for the bonds between heavy atoms and ∼0.03 Å for C-H and N-H bonds. In agreement with the increase in the bond lengths in the bases and base pairs, the molecular volumes also increase from the gas phase to the anchored geometries (deduced by removing the gold clusters from the ground-state geometries in the Au3base/base pair-Au3 configurations). Such large expansion arises due to some form of an electrostatic cavity driven by gold triangles. Our configuration mimics the cases of base/base pairs of DNA trapped within real gold electrodes. We also verify that there are similar expansions in the bases when trapped within two equilateral Ag3 and Cu3 triangles (Ag-Ag distance ) 2.7 Å and Cu-Cu distance ) 2.6 Å as derived from their first layer in FCC lattice). The fact that we observe similar expansion for all the bases and base pairs suggests that our results are candidates for experimental verification even for other electrodes (Supporting Information). Another possibility for error may arise due to the finite-size nature of our electrodes. We have eliminated this possibility to an extent by performing calculations on an Au6-A-Au6 system (see Supporting Information for structure) where again we observe a similar expansion in the molecule within the electrodes. The red shifts in the C-C bonds and C-N bonds are on the order of ∼10 and 8 cm-1, respectively, in all the molecules. We point out that small molecules such as H2O and HF trapped within spherical potentials such as fullerenes are known to exhibit contraction in their volumes with blue shifts in stretching frequencies and associated shortening of bond length.12,13 For a proper understanding of the origin of such molecular expansion, we next critically examine the energetics and bonding
Figure 2. Highest occupied molecular orbitals for adenine, thymine, guanine, and cytosine in gas-phase and the Au-complexed states.
in these systems. As a first assumption, one can model the molecular expansion within a harmonic approximation for which the potential energy takes a simple form of ∆E ) 1/2Σall bondsK(∆x)2 where ∆x is the increase in the bond lengths in the Aucomplexed systems and K is the force constant for the corresponding bond. From infrared spectroscopy, we find K for the CdN, CdO, CdC, C-H, and N-H bonds14 in the bases. The elastic energy associated with such expansion is derived as +6.5, +7.25, +9.00, and +10.86 kcal/mol for adenine, thymine, guanine, and cytosine, respectively. Whereas the expansion energy derived from the quantum-chemical calculations for adenine, thymine, guanine, and cytosine are +10.7, +6.94, +13.06, and +8.30 kcal/mol, respectively. Note that, although we use only the harmonic potential and approximate force constants, the model compares fairly well with quantumchemical calculations. To understand the contribution from other factors, we quantify the aromatic nature of the rings (at their ring centers) of the bases before and after complexation to the Au clusters. Purine bases (adenine and guanine) and pyrimidine bases (thymine and cytosine) show distinct aromatic behavior. On complexation with gold clusters purine bases becomes less aromatic whereas pyrimidine bases become more aromatic. For example, in adenine, the nucleus independent chemical shift (NICS)15 in the five- and six-membered rings are calculated as -7.11 (-4.68) ppm and -12.46 (-11.06) ppm and for thymine -1.33 (-3.67) ppm before (after) complexation to Au3, respectively. Similar features are also observed for the other bases (Supporting Information). This accounts for the electronic effects other than elastic stretching.
Nucleobases within Au Triangles
J. Phys. Chem. B, Vol. 110, No. 37, 2006 18663
Figure 3. Potential energy profiles for the adenine‚‚‚thymine and guanine‚‚‚cytosine base pairs in gas-phase and Au-complexed states. The energies (in kcal/mol) are scaled so as to make the minimal energy configurations as zero in energy (corresponding to N-H‚‚‚N for the gas phases and N‚‚‚H‚‚‚N in complexed states). The proton position (X-axis) for the minimum energy state corresponds to ∆R ) 0, and movement to either side involves positive and negative ∆R.
However, the bases and base pairs form stable coordinated structures with the Au clusters. The net stabilization due to coordination ∆Ecoordination ) EAu-cluster-(base/base-pair)-Au-cluster [Ebase/base-pair(complexed) - Ebase/base-pair(gas-phase)] EAu-cluster(right) - EAu-cluster(left) for adenine, thymine, cytosine, guanine, A‚‚‚T and G‚‚‚C are -84, -89, -85, -84, -90, and -87 kcal/mol, respectively. The large stabilization energies in these clusters suggest strong coupling between the Au3 triangles and the molecules. In Figure 2, we plot the highest occupied molecular orbitals (HOMOs) for adenine and thymine in the gas phase and the Au-complexed states. As can be seen, there exists strong intermixing between the electrons of the molecule and the Au atoms. In fact, it increases due to complexation. The base pairs, A‚‚‚T and G‚‚‚C, also exhibit similar molecular expansion with the increase in the bond lengths. However, apart from covalent bonds, A and T are connected through two hydrogen bonds (N‚‚‚H-N + N-H‚‚‚O), and G and C are connected through three hydrogen bonds (N‚‚‚H-N + 2N-H‚‚‚O). Thus, it is interesting to ask what happens to the hydrogen-bonding patterns when the base pairs are anchored to the two Au clusters. We find that as expected due to expansion the D and A distance increases in D-H‚‚‚A for the complexed geometries, but the hydrogen atom becomes equally positioned between the D and A atoms, and it behaves as a bridged hydrogen with interaction of the type D‚‚‚H‚‚‚A. We have optimized the positions of the hydrogen atoms that are involved in the hydrogen-bonding interactions at the cc-pVQZ/B3LYP level (see Supporting Information for geometries and bridged H positions in D‚‚‚H‚ ‚‚A). Calculations at this level of theory also show that there is an overall weakening of the hydrogen bonds within the clusters
and the hydrogen atoms takes the form of bridged site in the D‚‚‚H‚‚‚A. The existence of a stable bridged H between two D and A centers suggests a single-well potential profile for the hydrogen position.16 This is different from the conventional hydrogen-bonding energy profile, which shows a double-well potential with the bridged H form as a transitional state between two minimal energy states (D-H‚‚‚A and D‚‚‚H-A).15 For quantitative estimation of the hydrogen-bonding profiles in the base pairs, we plot the energy profiles for the central N-H‚‚‚N bond for A‚‚‚T and G‚‚‚C in Figure 3. In the gas-phase geometry, the profile follows a double-well potential with the N‚‚‚H‚‚‚N structure as the high-energy transition state within the N-H‚‚‚N and N‚‚‚H-N states (verified through frequency calculations, see Supporting Information). Indeed, N-H‚‚‚N is the ground-state structure with H on the N of thymine interacting with the free N of adenine. However, for the case of Au3-A‚‚‚T-Au3, the hydrogen-bonding profile is symmetric with a single minima at the N‚‚‚H‚‚‚N geometry and both the conventional N-H‚‚‚N and N‚‚‚H-N structures being the highenergy structures. Similarly, the G‚‚‚C pairs also show similar stabilization of the N‚‚‚H‚‚‚N structures within the Au clusters. Also for the N-H‚‚‚O hydrogen bonds molecular expansion leads to stabilization of the N‚‚‚H‚‚‚O structures. We have additionally calculated the infrared spectra for the A‚‚‚T and G‚‚‚C base pairs with and without the Au clusters. These calculations confirm the red shifts in the hydrogen bonds. For the case of the A‚‚‚T pair, the N-H stretching frequency in the N-H‚‚‚N hydrogen bond is red-shifted by 22 cm-1 and the N-H‚‚‚O hydrogen bond by 7 cm-1, respectively. For the G‚‚‚C pair, the N-H stretching frequency in the N-H‚‚‚N
18664 J. Phys. Chem. B, Vol. 110, No. 37, 2006
Mohan et al. Supporting Information Available: Structures, Cartesian coordinates energies, HOMO plots, and NMR values (NICS) for all the systems. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes
Figure 4. Infrared spectrum for the (A) A‚‚‚T pair in the gas-phase (black points) and Au3-G‚‚‚C-Au3 (red points) geometries and (B) G‚‚‚C pair in the gas-phase (black points) and Au3-G‚‚‚C-Au3 (red points) geometries.
hydrogen bond is red-shifted by 35 cm-1 and the same in the two N-H‚‚‚O hydrogen bonds by 5 and 3 cm-1. In conclusion, we have shown in this work that nucleobases and base pairs trapped within two Au clusters that mimic the Au electrodes exhibit overall molecular expansion and red shifts in the stretching frequencies of the bonds. Also, such a trapping stabilizes the high-energy bridged hydrogen-bond structure that otherwise is a transition state for conventional hydrogen bonds. This we believe is a novel result since the trapping of bases and base pairs serves as a strategy to stabilize bridged hydrogen bonds in DNA and may have interesting consequences in molecular electronics through biomaterials. Acknowledgment. S.K.P. acknowledges support from the CSIR and DST, India. S.S.M. and A.D. thank the CSIR for junior and senior fellowships.
(1) Ratner, M.; Ratner, D. Nanotechnology: A Gentle Introduction to the Next Big Idea; Prentice Hall: Upper Sadde River, NJ, 2002. (2) Tour, J. M. Molecular Electronics: Commercial Insights, Chemistry, DeVices, Architecture and Programming; World Scientific: River Edge, NJ, 2003. (3) (a) Dutta, S. Electronic Transport in Mesoscopic Systems; Cambridge Universiry Press: Cambridge, U.K., 1997. (b) Clusters and Nanomaterials: Theory and Experiment; Kawazoe, Y., Kondow, T., Ohno, K., Eds.; Springer: Berlin, 2002. (4) (a) Schuster, G. B. Acc. Chem. Res. 2000, 33, 253. (b) Pascaly, M.; Yoo, J.; Barton, K. J. Am. Chem. Soc. 2002, 124, 9083. (5) Storhoff, J. J.; Lazarides, A. A.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R. L.; Schatz, G. C. J. Am. Chem. Soc. 2000, 122, 4640. (6) (a) Barnett, R. N.; Cleveland, C. L.; Joy, A.; Landman, U.; Schuster, G. B. Science 2001, 294, 567. (b) Endres, R. G.; Cox, D. L.; Singh, R. R. P.; Pati, S. K. Phys. ReV. Lett. 2002, 88, 166601. (b) Lakshmi, S.; Datta, A.; Pati, S. K. Phys. ReV. B. 2005, 72, 045131. (c) Beljonne, D.; Pourtois, G.; Ratner, M. A.; Bredas, J. L. J. Am. Chem. Soc. 2003, 125, 14510. (7) (a) Kryachko, E. S.; Remacle, F. Nano Lett. 2005, 5, 735. (b) Kryachko, E. S.; Remacle, F. J. Phys. Chem. B. 2005, 109, 22746. (8) Drew, H. R.; Wing, R. M.; Takano, T.; Broka, C.; Tanaka, S.; Itakura, K.; Dickerson, R. E. Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 2179. (9) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (c) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (10) Wells, A. F. Structural Inorganic Chemistry, 5th ed. Oxford University Press: New York, 1984. (11) Datta, A.; John, N. S.; Kulkarni, G. U.; Pati, S. K. J. Phys. Chem. A. 2005, 109, 11647. (12) (a) Shameena, O.; Ramachandran, C. N.; Satyamurthy, N. J. Phys. Chem. A. 2006, 110, 2. (b) Sen, K. D. J. Chem. Phys. 2005, 122, 194324. (13) Waz, D. D.; Diercksen, G. H. F.; Klobukowski, M. Chem. Phys. Lett. 2001, 349, 215. (14) (a) Silverstein, R. M.; Webster, F. X.; Kiemle, D. Spectrometric Identification of Organic Compounds, 7th ed.; Wiley-Interscience: Hoboken, NJ, 2006. (b) Schleyer, P. V. R.; Maerker, C.; Dransfeld, A.; Jiao, H.; Eikema-Hommas, N. J. R. V. J. Am. Chem. Soc. 118, 6317, 1996. (15) Lowdin, P.-O. ReV. Mod. Phys. 1963, 35, 724. (16) Scheiner, S. Hydrogen Bonding: A Theoretical PerprectiVe; Oxford University Press: New York, 1997.
