DNA Flexibility does not show Appreciable Temperature Dependence

Abstract

DNA fragments shorter than the persistence length display about five orders of magnitude higher flexibility than predictions of the semi-flexible polymer models based on a persistence length of 50 nm (∼150 bp). One experimental method to quantitatively compare the flexibility of different constructs at various conditions is via single molecule cyclization assays. Briefly, short double stranded DNA oligos of interest with single stranded overhangs are sparsely immobilized on a slide surface. Linear and circular constructs are distinguished by the FRET values of two fluorophores covalently attached at the two termini. This observation is repeated at regular time intervals after introduction of a high salt buffer to promote looping. Rate constants can be deduced from the relaxation towards the new equilibrium state.We performed cyclization experiments at different temperatures to investigate the affect of temperature on the looping kinetics. Our results suggest a significant temperature dependence as both looping and unlooping rates increase dramatically as temperature is increased. We then further investigated whether this effect was due to the change of the flexibility of the DNA or other temperature dependent factors inherent in our setup such as viscosity and overhang annealing, which we addressed by measuring the bimolecular linear dimerization rate under similar conditions. We also observed highly temperature dependent annealing and melting rates of the same overhangs. We then quantified the j-factors (aka Jacobson-Stockmayer factors), which measures the effective concentration of one end of a polymer at the location of the opposite end. We observed that j-factors poorly correlated with temperature, within the accuracy of our technique. This is in contradiction to our expectations if extreme bendability is mediated by the formation of transient kinks or bubbles, whose incidence should be highly correlated with temperature.