Overview of UV melting

The temperature-induced transition between native and random coil states of a nucleic acid can be conveniently monitored by ultraviolet (UV) absorbance (see (4, 9, 10) for reviews). The reason for this is that stacked bases have a smaller absorption per base than unstacked bases this is called hypochromicity (11,12) which is defined as  

The simplest way to derive thermodynamic parameters from UV melting data is to apply a van t Hoff analysis of the data by assuming a two-state model (i.e. native and denatured states) and that the difference in heat capacities of the native and denatured states, ACp°, is zero (13-16) (more complex models are described in Section 5). At each temperature the absorbance can be used to calculate the fraction of strands in the native and denatured states, thereby allowing the calculation of an equilibrium constant (10). Thus, the absorbance versus temperature profile is used to determine the temperature dependence of 

In practice, however, to obtain accurate thermodynamics from the shape of a melting ciurve it is necessaiy to subtract upper and lower temperature baselines because the extinction coefficients of the denatured and folded states are temperature dependent (9,14,16). The details of one commonly used program for fitting UV melting ciurves were published recently (17). 

For sequences that form selfcomplementaiy duplexes, the melting temperature, Tm, is calculated using equation 3. 

Observing the transition with a different technique (e.g. calorimetry, NMR, or circular dichroism) can test the two-state assumption. The observation of isosbestic or isodichroic points in the absorbance or circular dichroism spectra, respectively, is diagnostic of a two-state transition (6). Alternatively, the melting profile can be monitored at a different wavelength, so that contributions from different nucleotides are emphasized. Deriving thermodynamics from the concentration dependence of the Tm (see below) can also test the two-state assumption. If all methods give the same results, the two-state approximation is validated since the methods have different sensitivities to each species. 

---