Centrifugation Application Notes

Gradients used to separate components that have large differences in density, such as a mixture of DNA, RNA, and protein in CsCl, will have large sample capacities, as much as 0.5 mg per mL of gradient. A narrow-tipped pipette or syringe equipped with a 20-gauge needle plus a steady hand are usually sufficient to layer the sample. It is helpful to place the delivery tip at the interface between the tube wall and the liquid surface, and rest the pipette on the edge of the tube. Once the sample has been applied, the run should be started as soon as possi- ble in order to minimize the effects of diffusion. Sample molecules will diffuse causing zone broad- ening. More importantly, however, is the danger of a density inversion caused by the more rapid diffu- sion of the smaller gradient molecules into the sam- ple zone where their concentration is much lower. also allows one to maximize sample load, is to introduce the sample as an inverted gradient (4). The published technique involved mixing sample and gradient solution in a gradient-forming appara- tus. Decreasing amounts of sample are mixed with a sucrose solution which is less dense than the top of the gradient but dense enough to support the sample load above it. For today’s users, this is probably overly complicated. A nice simplification of this is described by Williamson (5). The sample, in buffer, is taken up in a pipette tip. Then, with the pipette held at an angle, an equal volume of sucrose (for example, 4% sucrose for a 5-20% gradient) is taken into the tip. without additional mixing, the contents are layered on the gradient, forming a rough invert- ed sample gradient. A somewhat elaborate technique which not only avoids diffusion-induced density inversions, but

change with concentration. In the case of DNA, where increasing sample concentrations are associ- ated with decreasing sedimentation coefficients and zone-broadening, concentrations appropriate for other samples may give unsatisfactory results. Fritsch(6) recommends that the concentration of DNA in the sample zone be no greater than 10 µ g/mL for DNA with molecular weights less than 2 x 10 7 daltons (about 30 kbp) and even less for high- er molecular weight DNA. The lack of a method for reliably predicting opti- mum sample concentration for a given sample on a given gradient can pose a significant problem when separation protocols are being developed. The two best and simplest approaches to resolving this with a minimum of trial and error are (1) to use the low- est possible sample concentration that is compatible with preparative needs and/or sample detection and (2) if necessary, change to a steeper gradient to accommodate an increased sample load. References 1. Berman, A. National Cancer Institute Monograph 21, 41-76 (1966) 2. Vinograd, J., Bruner, R. Biopolymers 4 , 157-170 (1966) 3. Flamm, W.G. Subcellular Components: Preparation and Fractionation , 2nd ed. London, Butterworth, 1972 4. Britten, R.J., Roberts, R.B. Science 131 , 32 (1960) 5. Williamson, R. Separations With Zonal Rotors , pp. M-5.1—M-5.8. Surrey, Eng., Wolfson Bioanalytical Centre, University of Surrey, 1970 6. Fritsch, A. Preparative Density Gradient Centrifugation . Geneva, Beckman Instruments International, 1975.

The preceding discussion has treated the sample as one with sedimentation properties which do not

© 2005 Beckman Coulter,Inc .

B2005-6989