Premium
Studies of DNA dumbbells. III. Theoretical analysis of optical melting curves of dumbbells with a 16 base‐pair duplex stem and T n end loops ( n = 2, 3, 4, 6, 8, 10, 14)
Author(s) -
Paner Teodoro M.,
Amaratunga Mohan,
Benight Albert S.
Publication year - 1992
Publication title -
biopolymers
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.556
H-Index - 125
eISSN - 1097-0282
pISSN - 0006-3525
DOI - 10.1002/bip.360320714
Subject(s) - dumbbell , chemistry , duplex (building) , enthalpy , base pair , thermodynamics , crystallography , melting point , ionic strength , dna , physics , aqueous solution , medicine , biochemistry , physical therapy , organic chemistry
Optical melting curves of seven DNA dumbbells with the 16 base‐pair duplex sequence 5 ′G‐C‐A‐T‐A‐G‐A‐T‐G‐A‐G‐A‐A‐T‐G‐C 3 ′ linked on both ends by T n ( n = 2, 3, 4, 6, 8, 10, and 14) loops measured in 30, 70, and 120 m M Na + are analyzed in terms of the numerically exact statistical thermodynamic model of DNA melting. The construction and characterization of these molecules were described in the previous paper (Amaratunga et al., 1992). As was recently reported for hairpins (T. M. Paner, M. Amaratunga, M. J. Doktycz, and A. S. Benight, 1990, Biopolymers , Vol. 29, pp. 1715‐1734) theoretically calculated melting curves were fitted to experimental curves by simultaneously adjusting the parameters representing loop and circle formation to optimize the fits. The systematically determined empirical parameters provide evaluations of the free energies of hairpin loop formation Δ G loop ( n ) and single‐strand circles Δ G circle ( N ), as a function of end loop size, n = 2–14, and circle size, N = 32 + 2 n . The dependence of these quantities on solvent ionic strength over the range from 30 to 120 m M Na + was evaluated. An approximately analytical expression for the partition function Q ( T ) of the dumbbells was formulated that allowed a means for determining the transition enthalpy Δ H ° and entropy Δ S ° for every dumbbell, revealing the dependence of these quantities on loop size. In this multistate approach a manifold of partially melted intermediate microstates are considered and therefore no assumptions regarding the nature of the melting transitions (that they are two‐state) are required. The transition thermodynamic parameters were also determined from a van't Hoff analysis of the melting curves. Comparisons between the results of the multistate analysis and the two‐state van't Hoff analysis revealed significant differences for the dumbbells with larger end loops, indicating that the melting transitions of the larger looped dumbbells deviate considerably from two‐state behavior. Results are then compared with published melting studies of much larger DNA dumbbells (D. B. Naritsin and Y. L. Lyubchenko, 1990, Journal of Biomolecular Structure and Dynamics , Vol. 8, pp. 1–13), of small hairpins (Paner et al., 1990; M. J. Doktycz, T. M. Paner, M. Amaratunga and A. S. Benight, 1990, Biopolymers , Vol. 30, pp. 829–845) and another dumbbell (A. S. Benight, J. M. Schurr, P. F. Flynn, B. R. Reid, and D. E. Wemmer, 1988) Journal of Molecular Biology , Vol. 200, (pp. 377–399). Δ G loop ( n = 4) and ΔG circle (40) were also evaluated from analysis of the melting curves of the 15 DNA dumbbells with 16 base‐pair stems presented in the first paper (Doktycz et al., 1992). With the exception of the dumbbell with the central sequence 5 ′‐T‐T‐A‐A‐ 3 ′, these quantities were found to be virtually independent of the sequence identity of the central 4 base pairs of the stem.