Purification of plasmid DNA using membrane-based processes

Open Access
Espah Borujeni, Ehsan Allah
Graduate Program:
Chemical Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
November 07, 2014
Committee Members:
  • Andrew Zydney, Dissertation Advisor/Co-Advisor
  • Andrew Zydney, Committee Chair/Co-Chair
  • Themis Matsoukas, Committee Member
  • Michael John Janik, Committee Member
  • William O Hancock, Committee Member
  • Phillip E Savage, Committee Member
  • Plasmid ultrafiltration
  • Plasmid isoforms separation
  • Plasmid purification
  • Membrane fouling
  • Backpulsing in ultrafiltration
There is significant interest in therapeutic applications of plasmid DNA, requiring new separations technologies for the purification of plasmids. Recent studies have demonstrated the potential of using ultrafiltration for separating plasmid topological isoforms; however these data were obtained over short times with very dilute solutions. The overall objective of this dissertation was to address these issues thereby providing a more extensive evaluation of the potential of ultrafiltration in the large-scale production of plasmid DNA. Data obtained at higher plasmid concentrations showed a significant reduction in plasmid transmission during ultrafiltration even under conditions where the filtrate flux remained relatively constant. These results were explained using a simple model in which the membrane pores become partially blocked by plasmids that are trapped at the pore entrance. These “blocked” pores remain permeable to fluid due to the very open structure of the plasmid. The extent of plasmid fouling increased with increasing plasmid size, consistent with the greater probability of having a “knot” in the long DNA molecule. Model calculations were in excellent agreement with the experimental data over a wide range of conditions in both constant pressure and constant flux ultrafiltration. The effects of plasmid fouling could be dramatically reduced by backpulsing. Experiments were performed over a range of pulse duration, frequency, and amplitude to identify the optimal set of operating parameters. High performance could be achieved using a pulse with zero back-pressure, i.e. by simply clamping the permeate exit line while maintaining crossflow, suggesting that the trapped plasmids were removed from the pore primarily by the crossflow. A diafiltration process with pulsing provided more than 98% recovery of the plasmid in the permeate compared to only a 35% yield without pulsing. The performance of the ultrafiltration process could be further improved by altering the membrane pore structure / morphology. Data obtained with asymmetric ultrafiltration membranes in the reverse orientation showed much greater plasmid transmission at the same filtrate flux. This increase in transmission was due to the pre-extension of the plasmid which facilitates elongation and transmission of the plasmid through the small pores in the membrane skin. Membrane fouling was significantly reduced in this reverse filtration, possibly due to the “un-knotting” of the plasmid in the large pores within the membrane support. A laboratory-scale centrifugal ultrafiltration process was developed as a high-throughput method for quickly screening ultrafiltration for plasmid purification. A simple model was developed to evaluate the filtrate flux as a function of centrifugal conditions. Plasmid transmission in the centrifugal device was similar to that seen in pressure-driven ultrafiltration under comparable conditions. The results obtained in this thesis provide important insights into plasmid ultrafiltration, while furthering the development of ultrafiltration processes for the purification of plasmid DNA isoforms.