The effects of 3' single-strand dangling-ends of different lengths, sequence identity of hairpin loop, and hairpin loop biotinylation at different loop residues on DNA hairpin thermodynamic stability were investigated. Hairpins contained 16 bp stem regions and five base loops formed from the sequence, 5'-TAGTCGACGTGGTCC-N5-GGACCACGTCGACTAG-E(n)-3'. The length of the 3' dangling-ends (E(n)) was n = 13 or 22 bases. The identities of loop bases at positions 2 and 4 were varied. Biotinylation was varied at loop base positions 2, 3 or 4. Melting buffers contained 25 or 115 mM Na+. Average t(m) values for all molecules were 73.5 and 84.0 degrees C in 25 and 115 mM Na+, respectively. Average two-state parameters evaluated from van't Hoff analysis of the melting curve shapes in 25 mM Na+ were DeltaH(vH) = 84.8 +/- 15.5 kcal/mol, DeltaS(vH) = 244.8 +/- 45.0 cal/K.mol and DeltaG(vH) = 11.9 +/- 2.1 kcal/mol. In 115 mM Na+, two-state parameters were not very different at DeltaH(vH) = 80.42 +/- 12.74 kcal/mol, DeltaS(vH) = 225.24 +/- 35.88 cal/K.mol and DeltaG(vH) = 13.3 +/- 2.0 kcal/mol. Differential scanning calorimetry (DSC) was performed to test the validity of the two-state assumption and evaluated van't Hoff parameters. Thermodynamic parameters from DSC measurements (within experimental error) agreed with van't Hoff parameters, consistent with a two-state process. Overall, dangling-end DNA hairpin stabilities are not affected by dangling-end length, loop biotinylation or sequence and vary uniformly with [Na+]. Consider able freedom is afforded when designing DNA hairpins as probes in nucleic acid based detection assays, such as microarrays.
