Thermodynamique des associations de poly A et poly U en milieu neutre et alcalin.

The complexes of poly A and poly U are interesting models for studying the structure of nucleic acids. In the first part of this paper, the thermal stability of these complexes has been studied as a function of both pH and ionic strength in alkaline solution. The thermal stability is measured by the half transition temperatures, or Tm's. There are four different kinds of transitions: Tm2-1 refers to the dissociation of the two-stranded complex, poly (A + U); Tm2-1 to the dissociation of the three-stranded complex, poly (A + 2U) [4]. Tm8-2 is the temperature of half dissociation of poly (A + 2U) into poly (A + U) [1], and Tm2-3 is the temperature of rearrangement of 2 poly (A + U) into poly (A + 2U) [3] (Fig. 1). It is found that Tm2-1, and Tm8-1 decrease as pH increases above neutrality (Fig. 2 and 3), but that Tm2-8 does not vary (Fig. 3). This effect is readily explained if it is assumed that the ionisation of uracil brings a change in the state of the free poly U, but that the complexes themselves are not modified. In the second part of the paper a simple thermodynamic formulation is given which represents satisfactorily the relationship between the four equilibria, as observed at neutral pH (Fig. 1). A more complex treatment is then presented. It includes quantitation of electrostatic interactions, according to the theory of Kotin [18] and of the free enthalpy term corresponding to the organisation (stacking) of poly A. This leads to an excellent representation of the experimental data (Fig. 5). In alkaline solution, one can calculate the free enthalpy change associated with the ionisation of poly U at the Tm. According to our observations, it is likely that this is the only factor responsible for the decrease of Tm. Therefore, we can write that the sum of the free enthalpy of complex formation in the same conditions, but at pH 7, and of the free enthalpy of ionisation is zero. This leads to a determination of ΔG both as a function of temperature and ionic strength: the variation of ΔG with these factors is shown to be in good agreement with our detailed model. However, the absolute values are different, probably because of uncertainties concerning the data for poly A stacking enthalpy and entropy. Apart from providing evaluation of the thermodynamic parameters for double and triple helix formation, our results lead us to emphasize the importance of the thermodynamic state of the unassociated polynucleotides. This may be particularly relevant when one considers the partially complementary secondary structures of RNA's. [ABSTRACT FROM AUTHOR]