The structure of poly(dA).poly(dT) currently arouses great interest, mainly because dAn.dTn stretches are associated with considerable DNA bending. Until recently the heteronomous DNA described by Arnott et al., with the poly(dA) and poly(dT) chains in A and B conformations respectively, was the only detailed model of this structure. Following our earlier studies of the interaction of DNA and monovalent ions, we examined the X-ray diffraction of the bivalent Ca2+ salt of poly(dA).poly(dT) (Ca-poly(dA).poly(dT)) and found no sign of a heteronomous structure: Ca-poly(dA).poly(dT) in fibres shows fully equivalent B-type conformations of the opposite sugar-phosphate chains. A revision of the structure of the sodium salt, Na-poly(dA).poly(dT), based on this result, yields only a slightly heteronomous structure with each chain in a B-type conformation, which is in much better agreement with the experimental data underlying the original heteronomous model. Both structures, Ca- and Na-poly(dA).poly(dT), have a minor groove narrower than that of the B form: this peculiarity seems to be very important for the interaction of poly(dA).poly(dT) and biologically significant molecules (including proteins and antibiotics). The specific base-pair positions in poly(dA).poly(dT) may account for the DNA bending adjacent to dAn.dTn tracts.

No abstract

DNA films with psi +/- CD spectra have been investigated. X-ray analysis has shown the sign of the psi spectra to be independent of the secondary structure of DNA. The appearance of the psi spectra is attended by the formation of a characteristic polygonal texture of the cholesteric type in the DNA film.

An X‐ray diffraction study of CsDNA has been carried out with the use of synchrotron radiation of wavelength λ = 1.2 Å. The geometry corresponds to a cylindrical specimen brought into a cylindrical beam of the same diameter and the absorption factor and the primary‐beam attenuation factor have been calculated as functions of specimen μR for this geometry. The optimum size of a specimen is here about 1.5 times greater than that in the case of a plane‐parallel beam used in International Tables for X‐ray Crystallography [Vol. II, (1967), 2nd ed., p. 295. Birmingham: Kynoch Press]. Experimental and calculated primary‐beam attenuation factors have been compared and the experimental value of the linear absorption coefficient μ of CsDNA has been estimated for various humidities. The agreement between μc = 78.1 and μo = 78.2 ± 3.6 cm−1 (relative humidity 66–75%) indicates that the geometrical model used is suitable to describe polymer fibres in a cylindrical beam of nearly the same size. It is seen from the observed μ values that there is one caesium ion per nucleotide (monomer unit of DNA molecule) in the specimens used.

No abstract