(436f) Size-Exclusion Chromatography of Plasmid DNA Isoforms

Size exclusion chromatography (SEC) is attractive for both analytical separations and preparative purification of plasmid DNA. This includes the use of SEC for separation of the different plasmid isoforms, or topological morphologies of DNA, that occur naturally during plasmid production and manufacture. However, most studies of SEC have been performed with only a single isoform from clarified bacterial lysates, with the results providing only qualitative information on the effects of plasmid size and morphology on peak shape and resolution. The objective of this study was to obtain quantitative data for the behavior of the supercoiled, open-circular, and linear isoforms of several different size plasmids during SEC.

Plasmids from 3.0 to 17.0 kbp were obtained in the supercoiled form. Linear and open-circular isoforms were prepared by enzymatic digestion of the supercoiled isoform. The effectiveness of the digestions was examined by agarose gel electrophoresis. SEC experiments were performed with a 10 x 310 mm glass column flow-packed with Sephacryl S-1000 SF resin and an Agilent HPLC system. The mobile phase was a Tris-EDTA buffer with varying amounts of NaCl or MgCl₂. Elution data were obtained over a range of flow rates and buffer compositions.

Resolution between the different isoforms was relatively poor under all conditions, with significant peak overlap. The chromatograms for the larger plasmids displayed significant positive skewness due to flow non-uniformity and mass transfer limitations. The elution volume was thus evaluated using the first statistical moment of the chromatogram, with the first moment being relatively independent of the elution flow rate. The elution volumes were used to evaluate the partition coefficient, K_P, the ratio of average solute concentration in the stationary phase to that in the mobile phase. The partition coefficient decreased as plasmid size increased. For example, K_P decreased from 0.299 to 0.045 as the size of the supercoiled plasmid increased from 3.0 to 17.0 kbp. For any given plasmid size, the supercoiled isoform had the highest K_P followed by the open-circular and then the linear isoform. For example, K_P for the supercoiled 3.0 kbp isoform was 60% larger than that for the linear isoform, and this difference increased to a factor of three for the 9.8 kbp plasmid. Additional insights into the SEC behavior of the different plasmid isoforms was obtained by analyzing the experimental data using available theoretical models for partitioning of flexible linear and cyclic polymers. There was good qualitative agreement between the models and data, although there were clear discrepancies between the predicted and experimental values of K_P. These results provide important fundamental insights into the potential use of size exclusion chromatography for the analysis and purification of plasmid DNA.