One of the major difficulties in calculating transition curves for the “melting” of helical polynucleotides is to take account of the fact that all of the base pairs do not have the same stability, a point not considered in detail in the early theoretical treatments of the transition. In this paper it is shown that the resulting intra-molecular heterogeneity has a profound effect on calculated melting curves, and that transition curves in satisfactory agreement with experiment can be calculated only if the theory recognizes the difference between base pairs. Because of the mathematical complications which accompany this extension of the theory, a simplified model is used: it is assumed that the molecules are short enough for all of the bonded base pairs in a molecule to be clustered together, with unbonded bases only at the ends. This model is a good approximation when the molecular weight is below a certain upper limit, found in practice to be several hundred thousand. Two kinds of base pairs of differing stability, corresponding to adenine—thymine and guanine—cytosine, are recognized by the model used. It is found that the calculated transition curves depend on the sequence in which they are arranged; for the purpose of calculation a random sequence is assumed. Transition curves were obtained with computer random-sampling techniques. Experimental melting curves for a series of shear-degraded bacteriophage T2 DNA samples have been compared with theoretical curves, with good agreement at low molecular weights. Separation of the strands is taken into account, both in the theory and in the experiments.
Read full abstract