Effect of the DNA End of Tethering to Electrodes on Electron Transfer in Methylene Blue-Labeled DNA Duplexes Elaheh Farjami, † Rui Campos, †,‡ and Elena E. Ferapontova* ,†,‡ † Interdisciplinary Nanoscience Center (iNANO) and ‡ Center for DNA Nanotechnology (CDNA), Science and Technology, Aarhus University, Gustav Wieds Vej 1590-14, DK-8000 Aarhus C, Denmark * S Supporting Information ABSTRACT: Electron transfer (ET) in redox-labeled double- stranded (ds) DNA tethered to electrodes through the alkanethiol linker at either the 3′ or 5′ DNA end and bearing methylene blue (MB) conjugated to the opposite end of DNA is shown to depend on the DNA end of tethering to electrodes. For 3′ tethering, a nanoscale diffusion of the positively charged MB redox probe (and thus of the individual DNA molecules) to the negatively charged electrode surface provided the highest apparent diffusion and ET rates as a result of the tilting of 3′-tethered DNA (as compared to 5′-tethered DNA) versus the normal to the surface. Dynamic values of the tilting angle varied between 57 and 45° for 16-mer and 22-mer 3′-tethered DNA, and 5′-tethering was correlated with an upright orientation of DNA at the electrode surface. The values of the diffusion coefficient D MB corrected for tilting angles were similar for 5′- and 3′-tethered DNA and ranged between 5.4 × 10 -12 and 2.5 × 10 -12 cm 2 s -1 , whereas the ET rate constant k ET dif fit the 4.7 × 10 -6 -10.3 × 10 -6 cm s -1 range for 22-mer and 16-mer dsDNA, respectively. Those values, when related to the nanometer (10 -7 cm) diffusion distances (the length of the studied DNA), allow relatively fast diffusion-limited ET at an apparent rate that may exceed the rate of the corresponding surface- confined ET process. This phenomenon is of particular importance for molecular electronics and electrochemical genosensor development. ■ INTRODUCTION Interfacial and electron transfer (ET) properties of individual DNA molecules tethered to electrodes crucially affect both molecular electronics 1 and electrochemical biosensor applica- tions of DNA. 2,3 Mediated by the DNA double helix, directional ET was successfully used for the sensitive analysis of DNA, 4 DNA mutations, 5-9 and interactions of DNA with anticancer drugs and xenobiotics. 10,11 In such genosensors, 12 redox-active indicators intercalate into double stranded (ds) DNA, thus providing an electronic coupling between the indicator and DNA base-pair π stack. 13 That results in DNA- mediated ET, with an efficiency (in terms of the ET rate constants) that ranges between 1.5 and 40 s -16,14 and depends on the presence of mismatches in the DNA duplex. The situation is quite different when the redox probe is conjugated to DNA through a linker and intercalation of the redox probe into DNA does not take place. In this case, the interfacial state and ET reactions of redox-labeled dsDNA essentially depend on the potential window in which experiments are performed 15,16 and such experimental conditions as the potential scan rate 17-20 and DNA surface coverage. 21,22 At the positively charged electrode surface (under conditions corresponding to the ferrocene (Fc) redox label), negatively charged dsDNA was shown either to lie flat 15 or to be orientated upright on the condition that the ionic strength of the working solutions was high enough to provide sufficient screening of electrostatic interactions between the electrode and the negatively charged phosphate residues of DNA. 23 Then, depending on the experimental conditions, ET in Fc- labeled DNA correlated with either a surface-confined 24,25 or a diffusion limited ET reaction. 18,19,26 DNA elastic bending (C 2 - electrode tethering linker) 18 and the free rotational motion of dsDNA (C 6 -electrode tethering linker) 26 were considered in the mechanistic models elaborated to describe diffusional ET in such essentially complicated system as Fc-labeled dsDNA. Both models fit the experimental data in 1 M electrolyte solutions well and implied a mechanism of ET independent of the type of redox probe or Fc linker used, 26 because within this modeling approach the motion of dsDNA was considered to be randomly thermally driven as a result of the intrinsic flexibility of the rodlike DNA. At the negatively charged electrode surface (at potentials corresponding to the methylene blue (MB) redox trans- formation) the results vary, and both high 20,27 and low ET efficiencies, 28,29 commonly discussed in terms of the varying Received: August 9, 2012 Revised: October 21, 2012 Published: October 29, 2012 Article pubs.acs.org/Langmuir © 2012 American Chemical Society 16218 dx.doi.org/10.1021/la3032336 | Langmuir 2012, 28, 16218-16226