Dependence upon temperature of corrected sedimentation coefficients measured in a Beckman analytical ultracentrifuge

Abstract

The corrected sedimentation coefficient ( s 20, w) has been examined as a function of temperature for bovine liver catalase and for a small linear duplex DNA in a Beckman Model E analytical ultracentrifuge. These measurements were carried out using both the standard Beckman rotor temperature indicator and control system and a modified version of this system which was designed to reduced thermal gradients in the rotor [Hearst, J. E., and Gray, H. B., Jr. (1968) Anal. Biochem. 24, 70–79)]. The temperature dependence of s 20, w for both the protein and the DNA were identical between the two systems. The slight increase of s 20, w with increasing temperature observed for the DNA is in agreement with that predicted on the basis of the temperature dependence of the Kuhn statistical length for DNA. Intrinsic viscosity measurements as a function of temperature were also carried out for the DNA in a low-shear viscometer. These data indicate a slight decrease of intrinsic viscosity with increasing temperature. Previous investigations, which have suggested that the actual temperature of the ultracentrifuge cell is substantially higher [Incardona, N. L., Notarius, H., and Flanegan, J. B. (1971) Anal. Biochem. 40, 267–280] or lower [Rowe, A. J., and Khan, G. M. (1972) Anal. Biochem. 45, 488–497] than that indicated by the temperature monitoring system of the Model E centrifuge at temperatures below ambient, are not supported by the present studies. Discrepancies between the actual temperature of the sedimenting sample and the indicated temperature are thus shown in this work not to be an inherent problem of the design of the Beckman instrument.