Elsevier

Journal of Chromatography A

Volume 1065, Issue 1, 11 February 2005, Pages 93-106
Journal of Chromatography A

Application of monoliths for plasmid DNA purification: Development and transfer to production

https://doi.org/10.1016/j.chroma.2004.12.007Get rights and content

Abstract

The demand of high-purity plasmid DNA (pDNA) for gene-therapy and genetic vaccination is still increasing. For the large scale production of pharmaceutical grade plasmids generic and economic purification processes are needed. Most of the current processes for pDNA production use at least one chromatography step, which always constitutes as the key-step in the purification sequence. Monolithic chromatographic supports are an alternative to conventional supports due to their excellent mass transfer properties and their high binding capacity for pDNA. Anion-exchange chromatography is the most popular chromatography method for plasmid separation, since polynucleotides are negatively charged independent of the buffer conditions. For the implementation of a monolith-based anion exchange step into a pDNA purification process detailed screening experiments were performed. These studies included supports, ligand-types and ligand-densities and optimization of resolution and productivity. For this purpose model plasmids with a size of 4.3 and 6.9 kilo base pairs (kbp) were used. It could be shown, that up-scaling to the production scale using 800 ml CIM Convective Interaction Media radial flow monoliths is possible under low pressure conditions. CIM DEAE was successfully implemented as intermediate step of the cGMP pDNA manufacturing process. Starting from 200 l fermentation aliquots pilot scale purification runs were performed in order to prove scale-up and to predict further up-scaling to 8 l tube monolithic columns. The analytical results obtained from these runs confirmed suitability for pharmaceutical applications.

Introduction

Although some technical and regulatory hurdles for DNA vaccines are reported to be still an issue [1] an increasing number of clinical trials for gene-therapy and genetic vaccination based on plasmid DNA (pDNA) reach the later clinical phases. The required amount of high-purity pDNA to feed these studies and finally the market have been underestimated in the past, since for clinical applications a trend from traditional vector systems, such as viruses, to safer but less efficient methods, such as naked pDNA and formulation as cationic complexes, can be observed [2], [3], [4], [5], [6], [7], [8]. Therefore industrial scale processes for the production of plasmids have to be suited for manufacturing grams or even several kilograms of purified pDNA per batch while meeting the appropriate quality standards requested by the national health agencies. Hence, the chromatographic supports used in such a process play a major role [9]. As productivity becomes a limiting factor chromatographic supports with a high dynamic binding capacity for pDNA are required.

pDNA applied as a DNA vaccine has to meet some typical quality specifications. In this context the product-quality is defined by the purity and the homogeneity as percentage of the supercoiled form compared to the total pDNA [10]. Supercoiled, also named covalently closed circular (ccc), pDNA [11] is the desired topological form since it induces the most efficient transfection and expression rate in eukaryotic cells. According to international regulations a content of ccc form higher than 90% is required [12], [13], [14]. Other undesired topological plasmid forms such as the open circular (oc) and the linear form, as well as dimers [15], [16], [17], reduce the homogeneity and should be removed by the separation process. Beside these, a variety of other host (E. coli)-related impurities such as genomic DNA (gDNA), RNA, proteins and endotoxins [23], [24], [25] have to be considered [8], [9], [18], [19], [20], [21], [22].

At laboratory scale, isolation of pDNA from crude cell lysates is well established [26], [27]. For scientific purposes, simple commercial small-scale kits of different suppliers are available. They are designed for purifying small quantities of pDNA in the range of μg to mg, yielding a final preparation of minor quality [27]. This is sufficient for the majority of laboratory applications, of molecular cloning, but not for therapeutic purposes. Such pDNA purification processes consist of the following steps: cell lysis using lysozyme, RNA removal by RNase, extraction and precipitation with organic solvents and ultracentrifugation in density gradients. Due to their initial design, they are very time consuming and not scaleable. Other problematic issues are the use of flammable liquids, materials that are not certified for application in humans, enzymes from avian or bovine origin and toxic reagents such as phenol, CsCl or CsBr. To meet the appropriate guidelines of the regulatory authorities [2], [28], [29], [30] such reagents have to be avoided in manufacturing of pharmaceuticals under current good manufacturing practice (cGMP) conditions.

An industrial manufacturing process for pDNA typically comprises fermentation, cell lysis, clarification, purification, polishing and final formulation and filling [2], [21], [30], [31], [32]. Liquid chromatography is considered as the downstream operation with highest resolution and is essential for pDNA production suited for therapeutic applications.

The requirements of a chromatographic support for pDNA separation are different from those for recombinant proteins, because these two classes of macromolecules differ significantly in their physico-chemical properties [33]. Plasmids are always negatively charged, are much larger in size and their shape resembles a long fiber. A typical plasmid is composed of 3–20 kilo base pairs (kbp), which corresponds to a relative molecular mass of 2 × 106–13 × 106 with a radius of gyration of 100 nm and higher [34]. The shape of the molecule was made responsible for the sensitivity against mechanical stress [9], [35], [36], [37], [38].

The isolation and purification of large polynucleotides, such as pDNA, is hampered by the low performance of commercially available chromatographic supports, which are mainly based on highly porous particles. Most of the chromatographic supports were tailor-made for the high adsorption capacity of proteins with a particle pore diameter of typically 30–400 nm, since proteins have diameters typically lower than 5 nm [34]. In columns packed with such supports, large molecules such as pDNA with a size of 100 nm to over 300 nm in diameter adsorb only at the beads outer surface [2], [21], [39], [40], [41], [42]. Consequently capacities are on the order of tenths of milligrams plasmid per milliliter of chromatographic support compared to 200 mg/ml reported for proteins [2]. Thus, pDNA purification columns need to be large. Relatively low flow rates together with low capacity result in low productivities. Since most of the supports are reused a total pDNA recovery of about 100% after each cycle is mandatory to avoid carry-over from one purification batch to the next batch. Harsh cleaning conditions with up to 1 M NaOH are also preferred.

Traditional liquid chromatography is a rather slow, diffusion-controlled process. It often causes significant product loss due to oxidative degradation and enzymatic attack [43], [44], [45]. On the other hand, the efficient isolation of labile, valuable biomolecules requires a fast, reliable and affordable separation process under mild conditions.

For the purification of pDNA several chromatographic methods based on particulate supports have been reported [2], [9], [30], [46]. Beside conventional techniques such as anion exchange [29], hydrophobic interaction [47] and size exclusion chromatography (SEC) [48] other methods were tested with more or less success. As examples triple-helix affinity [49], thiophilic interaction [50], reversed phase silica [51] or polymeric [52] and hydroxyapatite chromatography [53] have to be mentioned.

Alternatives to porous particles are the use of membrane- and monolith-technology, which reflect technological advances in fixed-bed liquid chromatography [54], [55], [56] based on favorable hydrodynamic properties compared to conventional supports. Membranes are very thin beds and can be considered as monolithic columns with an extreme aspect ratio. They provide a reduced pressure drop along the chromatographic unit, allowing increased flow rates and consequently higher productivity [54], [55], [56]. The problems with membranes are uniform flow distribution, a relatively large dead volume and scalability. To increase capacity membranes have been stacked into a column, which introduces additional void spaces.

A typical monolith is a continuous bed consisting of a single piece of a highly porous solid material [57], [58]. Similar to membranes the most important feature of this support is that all the mobile phase is forced to flow through the large pores of the monolith [33]. As a consequence, mass transport is enhanced by convection, dramatically reducing the long diffusion time required by conventional particle-based chromatographic supports. Therefore, the chromatographic separation process on monoliths is practically not diffusion-limited [40], [61], [62], [63], [64]. The “large” channels (pores) of about 700–1000 nm of these monoliths allow binding of large molecules such as pDNA [43], [65], [66]. The high porosity of more than 50% leads to a low pressure drop [33].

Three types of monolithic separation-supports [59], [60] are currently commercially available: Silica gel based monolithic beds [67] (Merck: “Chromolith”), polyacrylamide based monolithic beds [68] (Bio-Rad: “Uno”) and rigid organic gel based monolithic beds [69]. Polymethacrylate based short monolithic columns as stationary phases for biochromatography [43] have been developed in the early 1990s. They are currently distributed under the trade name “Convective Interaction Media (CIM)” as disk monolithic columns and tube monolithic columns [70] (BIA Separations). It has been previously shown, that this material can be used for pDNA purification [40], [43], [66], [71], [72].

At laboratory scale a plasmid purification process using CIM columns was developed and implemented in a pilot scale. All critical elements of existing pDNA purification processes such as enzymes, detergents and organic solvents could be avoided. As a result a modern generic pDNA purification process which fulfills all regulatory requirements, delivering pDNA of high quality could be developed [31], [73].

Section snippets

Materials

The plasmid pRZ–hMCP1 [4.9 kbp; host: E. coli K12 JM108, ATCC No. 47107] and another model plasmid (6.9 kbp; host: E. coli K12 DH5-alpha, Invitrogen) were produced in the laboratory according to the procedure described in Section 2.2.1.

Purified pDNA for the determination of the dynamic binding capacity was also produced in the laboratory according to a modified laboratory-scale protocol [27] using a conventional chromatographic support. The purity of the pDNA solution was estimated to be around

Results and discussion

In order to get a first impression about the suitability of different support types for an economic pDNA production process their dynamic binding capacity was evaluated. The characteristics of the tested supports are summarized in Table 1. All materials listed here do meet the regulatory guidelines for production of biopharmaceuticals. They can be sanitized by NaOH. The particle diameter ranged from 20 to 90 μm except for the monoliths with an apparent particle diameter of 1.5 μm. The pore size

Conclusions

The high capacity of the CIM supports for the chromatographic purification of pDNA at high flow rates was the intention to set up a down-stream process based on CIM monoliths. They exert a high productivity, due to excellent mass transfer properties and a large number of accessible binding sites for pDNA, which both result from the monolith specific structure. Furthermore the selected AIEC monolith exerts a good resolution, which is maintained also at increased linear velocities. CIM supports

Acknowledgments

This work was supported by a grant from the Austrian Forschungsförderungsfonds, project no. 804349. The authors acknowledge the support of the Process Science team (Wolfgang Buchinger) of Boehringer Ingelheim Austria GmbH (Hans Huber for fermentation, Robert Schlegl for down stream processing, Harald Paril for down stream pilot and Franz Kollmann for process analysis) and at BI Austria Quality Control business center (Sandra Zsifkovits). Special thanks go to Christine Ascher, Helga Wöhrer and

References (83)

  • J. Donnelly et al.

    Int. J. Parasitol.

    (2003)
  • A. Mountain

    Tibtech

    (2000)
  • D.L. Montgomery et al.

    Pharmacol. Ther.

    (1997)
  • U.A. Hasan et al.

    J. Immunol. Methods

    (1999)
  • M.S. Levy et al.

    Tibtech

    (2000)
  • K.N. Kreuzer et al.

    Cell

    (1980)
  • D.M.F. Prazeres et al.

    Tibtech

    (1999)
  • A. Eon-Duval et al.

    J. Chromatogr. B

    (2004)
  • G.N.M. Ferreira et al.

    Tibtech

    (2000)
  • M.S. Levy et al.

    Nucl. Acids Res.

    (2000)
  • M.S. Levy et al.

    Chem. Eng. Sci.

    (1999)
  • Q.M. Mao et al.

    J. Chromatogr.

    (1991)
  • Q.M. Mao et al.

    J. Chromatogr.

    (1993)
  • D. Josic et al.

    J. Chromatogr. B

    (1997)
  • A. Lyddiatt et al.

    Curr. Opin. Biotechnol.

    (1998)
  • L.M. Sandberg et al.

    J. Biotechnol.

    (2004)
  • R.B. Sparks et al.

    Anal. Biochem.

    (1983)
  • T.R. Johnson et al.

    Anal. Biochem.

    (1983)
  • T.B. Tennikova et al.

    J. Chromatogr.

    (1993)
  • F. Svec et al.

    J. Chromatogr. A

    (1995)
  • D. Josic et al.

    J. Chromatogr. B

    (2001)
  • K. Branovic et al.

    J. Chromatogr. B

    (2004)
  • J.-L. Liao et al.

    J. Chromatogr.

    (1991)
  • A. Podgornik et al.

    J. Biochem. Biophys. Methods

    (2004)
  • P.A. Shamlou

    Biotechnol. Appl. Biochem.

    (2003)
  • F.D. Ledley

    Hum. Gene Ther.

    (1995)
  • M. Blaese et al.

    Cancer Gene Ther.

    (1995)
  • N. Weir

    Biotechology

    (1999)
  • United States Food Drug Administration

    Hum. Gene Ther.

    (1996)
  • R.R. Sinden et al.

    DNA Structure and Function

    (1994)
  • United States Food and Drug Administration, Addendum to The Points to Consider in Human Somatic Cell and Gene Therapy,...
  • United States Food and Drug Administration, Points to Consider on Plasmid DNA Vaccines for Preventive Invectious...
  • M. Marquet et al.

    BioPharm

    (1997)
  • D.K. Summers

    The Biology of Plasmids

    (1996)
  • R. Martin

    Gel Electrophoresis: Nucleic Acids

    (1996)
  • M. Schleef
  • T.R. Strick et al.

    Biophys. J.

    (1998)
  • C.R. Middaugh et al.

    J. Pharm. Sci.

    (1998)
  • United States, Food and Drug Administration, Guidance for Industry: Guidance for Human Somatic Cell Therapy and Gene...
  • G. Sofer et al.
  • J.-C. Jansen et al.
  • Cited by (156)

    • Application of an electro elution system for direct purification of linear covalently closed DNA fragments

      2023, Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences
    • Ion-exchange membrane chromatography

      2023, Ion-Exchange Chromatography and Related Techniques
    • Affinity chromatography for vaccines manufacturing: Finally ready for prime time?

      2019, Vaccine
      Citation Excerpt :

      The study from Forcic et al. demonstrated that purification of 145 mL of rubella virus suspension can be completed at around 1 h using a disk-shaped anion-exchange monolithic column compared to >10 h with only 10 mL of viral suspension loading using anion exchange resin [142,157]. Ion exchange monolith columns have been reported to be implemented in clinical-scale production for influenza vaccine and plasmid DNA [158,159]. Affinity monolith columns have been available for analytical use for various biomolecules and lab-scale pDNA, virus and VLP purification, but there is little information on their application in industrial processes [47,92,160–164].

    View all citing articles on Scopus
    View full text