Understanding Signal and Background in a Thermally Resolved, Single-Branched DNA Assay Using Square Wave Voltammetry

Abstract

Electrochemical bioanalytical sensors with oligonucleotide transducer molecules have been recently extended for quantifying a wide range of biomolecules, from small drugs to large proteins. Short DNA or RNA strands have gained attention recently due to the existence of circulating oligonucleotides in human blood, yet challenges remain for adequately sensing these targets at electrode surfaces. In this work, we have developed a quantitative electrochemical method which uses target-induced proximity of a single-branched DNA structure to drive hybridization at an electrode surface, with readout by square-wave voltammetry (SWV). Using custom instrumentation, we first show that precise control of temperature can provide both electrochemical signal amplification and background signal depreciation in SWV readout of small oligonucleotides. Next, we thoroughly compared 25 different combinations of binding energies by their signal-to-background ratios and differences. These data served as a guide to select the optimal parameters of binding energy, SWV frequency, and assay temperature. Finally, the influence of experimental workflow on the sensitivity and limit of detection (LOD) of the sensor is demonstrated. This study highlights the importance of precisely controlling temperature and SWV frequency in DNA-driven assays on electrode surfaces while also presenting a novel instrumental design for fine-tuning of such systems.

Figure 1

Single-branched cooperative DNA quantification structure. To elucidate the hybridization arrangement(s) giving optimal signal and minimized background, five different thiolated-DNAs were used (green region; n = 7, 9, 10, 12, and 14), each with five different target DNAs tested (red region; m = 6, 7, 8, 9, and 10), resulting in 25 combinations of binding energies and locations. The blue region was varied to maintain the target DNA at 21 base pairs (s + m = 21). Black regions show structural portions of the sequences not involved in binding.

Figure 2

Heat maps displaying peak height (color intensity) as a function both temperature (y-axis) and SWV frequency (x-axis). This display aids in visualizing the results of the 25 different complexes studied in this work and helps to determine the most optimal conditions for the assay format. For clarity, truncated complexes are shown above and to the right, although all studies included the full complexes as shown in Figure 1.

Figure 3

Electron-transfer kinetics as a function of temperature for selected complexes. Shifts in electrochemical critical times as a function of temperature can be visualized using normalized I_p/f_SWV displayed as a function of inverse SWV frequency (1/f_SWV), where the maximum represents the critical electron-transfer time (in seconds). Similar kinetic trends were observed as temperature was increased, independent of complex stability on the surface (stable hybridization at n = 14; weak hybridization at n = 7). The weakest complex (n = 7, no target) prevented kinetic measurements simply because the complex dissociated/melted from the surface.

Figure 4

Heat maps displaying signal peak height minus background peak height (signal-to-background difference; depicted as color intensity) as a function both temperature (y-axis) and f_SWV (x-axis). Excluding the leftmost maps which showed overwhelming background, signals showed a complex relationship on m, n, temperature, and f_SWV, suggesting that careful evaluation of these parameters may be necessary in surface-confined, DNA-driven electrochemical assays. Conditions identified for further study are labeled as regions (1) through (4).

Figure 5

Heat maps displaying signal peak height divided by background peak height (signal-to-background ratio; depicted as color intensity) as a function both temperature (y-axis) and f_SWV (x-axis). Again, signals showed a complex relationship on m, n, temperature, and f_SWV, but the ratio was clearly optimal at low background stabilities (n = 7). The highest ratio was observed at n = 7, m = 10, f_SWV = 100 Hz, and T = 30 °C; this region was labeled as (3) in the maps.

Figure 6

Single- and two-step quantification of target DNA with the selected single-branched complexes and condition sets from the heat maps above, (1) through (4). The left side with blue background is for a single-step calibration workflow, whereas the right side with red background is for a two-step calibration workflow. The red curves fitted to all data are four-point logistic curve fits. The standard deviation is presented for all cases (n = 6 electrodes).

Figure 7

Limit of detection comparison for the three sets of conditions with responsive assays, (2) through (4). Results show the important effects of binding energy, assay temperature, and workflow/procedure on LOD.