Theoretical and Experimental Investigation of the Impact of Surfaces on DNA Melting Temperature.

Abstract

The design of microarrays rely on studies geared towards sequence-specific recognition between complementary probe and target molecules in bulk solution. However, this proves to be insufficient to understand the duplex formation reaction on solid-phase. In this dissertation, influence of the surface on DNA duplex stability and melting temperature were theoretically and experimentally investigated. The theoretical approach represents electrostatic and entropic repulsions experienced by hybridizing targets. Electrostatic blocking stemming from surface charge was modeled through Electric Double Layer Theory and Surface Partition Model. Entropic blocking due to steric effects was modeled using polymer physics. Investigated experimental parameters were target concentration, spacer length and probe density. All the experiments gave reproducible melting temperatures with values lower on-surface than in-solution. In a representative set, a target concentration increase from 0.5nM to 15nM with 0.82pmoles of probe at 5*10^12 molecules/cm2 density on 15 dT spacer resulted in approximately 8°C decrease in melting temperature, compared to 5°C increase in solution. This decreasing trend was supported by theory with increasing steric and electrostatic effects at increasing target concentrations leading to higher hybridization efficiencies. Additionally, at low target concentrations (0.0165nM), we observed a multiple melting process in low temperature domains of melting curves due to low stability truncated probes; an indirect indication of synthesis quality. It was observed that as spacer length increases from 2 dT to 25 dT with 0.82pmoles of probe at 5*10^12 molecules/cm2 density with target concentrations ranging from 0.36nM to 1nM, melting temperature increases; an observation theoretically explained by possible entropic blocking dominance. Probe density effect was tested at 5*10^12 molecules/cm2 and 5*10^13 molecules/cm2, on 15 dT spacer and target-to-probe concentration ratios of 0.61:1 to 1.7:1. It was observed that high probe density resulted in lower melting temperature. This trend was theoretically supported by increasing electrostatic and crowding effects. Previously observed dependence of melting temperature on target concentration was also confirmed in all experiments. Melting temperature dependence on probe density seems to be stronger than the dependence on spacer length. The results of this work would lead to better experimental design and correct use of microarrays.