Large-Scale Manufacturing of Plasmid DNA for Gene Therapy and DNA Vaccination -- Part 1: The Suitability of Current Techniques to Purify Plasmid Without Adding RNase

BioPharm International, BioPharm International-08-01-2003, Volume 16, Issue 8

Although gene therapy and DNA vaccination suggest promising new approaches to disease treatment - and nonviral vectors (which are cheap and easy to manufacture) afford low immunogenicity, better safety profiles, and improved stability - commercial-scale purification of plasmid DNA remains difficult, particularly if bovine-derived ribonuclease A is left out of the process. This article series reviews the benefits and limitations of current plasmid DNA purification and suggests an RNase-free downstream process that is scalable, robust, and meets the requirements set by industry regulators.

Over the past 13 years, interest has increased in gene therapy and DNA vaccination, which offer new approaches to treating and preventing disease based on the transfer of foreign genes - transgenes - into human cells (1). These transgenes express proteins to replace defective ones in diseased tissues (gene therapy) or to elicit an immune response against a target antigen (DNA vaccination). DNA vaccines (2,3) have engendered both humoral and cellular immune responses in vivo, which suggest they might become therapeutics for poorly-treated diseases such as cancer, HIV, malaria, and tuberculosis. Gene transfer requires a vector, which can be either viral (adenovirus, retrovirus, adeno-associated virus, herpes simplex, or lentivirus) or nonviral (naked DNA, complexes of DNA–cationic lipids, or DNA-coated particles).

Plasmid Vector Purification

Although viral vectors result in better transfection efficiency and more durable expression, recent well-publicized events have raised concerns about their safety. Nonviral vectors are less efficient at gene delivery but have the advantages of low immunogenicity, better safety profiles, and improved stability. They are also easier and cheaper to manufacture. In these vectors, the therapeutic gene is inserted into a plasmid DNA molecule, which is then injected intramuscularly as a solution. Another approach is to coat the plasmid DNA molecule onto gold particles and deliver the vector into the epidermis with a gene gun (3). Therapeutic genes can be inserted into suitable plasmid vectors easily with recombinant DNA technology, followed by production of a high number of copies in bacterial cells such as

Escherichia coli

.

Table 1. Plasmid DNA purification techniques suitable for large-scale production.

Until the advent of gene therapy, plasmid purification was essentially confined to small bench-scale operations in molecular biology laboratories (4). Several manufacturers have developed plasmid purification kits that provide rapid, reliable isolation of plasmid DNA of adequate purity for use in molecular biology. Most currently available plasmid purification techniques, however, are unsuitable for the manufacture of plasmid DNA intended for human applications for two reasons. Plasmid purification uses either solvents (such as ethanol and isopropanol), toxic chemicals (cesium chloride, ethidium bromide, phenol, and chloroform, for instance), or animal-derived enzymes (such as ribonuclease A, lysozyme, or proteinase K) - all of which are either unapproved or not recommended by the agencies regulating the pharmaceutical industry. In addition, many of the processes used for plasmid purification in the laboratory are difficult to scale up for the production of high-quality plasmid DNA required for the treatment of large populations.

The challenge for the gene therapy industry is to design a downstream plasmid process that is scalable, robust, and meets purity, potency, identity, efficacy, and safety standards set by regulatory agencies (5,6). The plasmid product must be of high purity, essentially in its supercoiled form and free of host-cell proteins, chromosomal DNA, RNA, and endotoxins (7). It is generally believed that gene expression is conferred by plasmid in its supercoiled form, so other topologies - such as open circular, linear, or multimeric - should be kept to a minimum.

Bovine additives. Ribonuclease A (RNase A) from bovine pancreas is commonly added during plasmid processing to degrade RNA molecules that would otherwise be difficult to separate from the plasmid DNA because of their similarity in chemical composition and structure. This enzyme is very robust - withstanding extremes of pH - and is often added during alkaline lysis. However, the rise of variant Creutzfeldt–Jakob (vCJD) disease in the UK has meant bovine-derived products are no longer recommended for pharmaceutical production intended for humans (8).

Ideally, animal products should be avoided altogether, but without the addition of exogenous RNase, the levels of undegraded RNA in the clarified lysate are substantial - at least 20 to 25 times the amount of the plasmid, by weight. Long incubation in alkaline pH allows endogenous RNases to hydrolyze RNA (9,10), but it is time-consuming and can result in plasmid losses or plasmid conversion to undesirable open circular and linear forms. Techniques are therefore needed for removing RNA selectively in a plasmid downstream process without affecting plasmid recovery and quality.

RNA Removal from Plasmid DNA

Part 1 of this article series reviews plasmid purification techniques, such as selective precipitation, chromatography, filtration, and aqueous two-phase extraction, with specific emphasis on RNA removal. In Part 2, we will show how these different techniques can fit into a purification process that maximizes RNA removal. Finally, we will propose a process stream suitable for the scale-up and production of gram quantities of pure plasmid. Table 1 summarizes the techniques suitable for the large-scale purification of plasmid DNA. Various plasmid purification techniques and their suitability for RNA removal are summarized in Table 2.

Table 2. Plasmid DNA purification techniques' suitability for RNA removal.

Bacterial Lysis

The first step in a plasmid purification process is to release the molecule of interest from its bacterial host. Mechanical disruption of

E. coli

cells is usually avoided because of the shear sensitivity of plasmid DNA. However, bead milling has been shown to provide high recovery of intact plasmid under optimized operating conditions (11). Although extraction by boiling has also been reported (12), alkaline lysis based on the method developed by Birnboim and Doly (13,14) is, by far, the most common.

In the alkaline lysis method, resuspended cells are lysed by adding a solution containing an alkali such as sodium hydroxide and a detergent, usually sodium dodecyl sulfate (SDS). By targeting a certain pH range (12.0 to 12.5), chromosomal DNA is denatured irreversibly, while plasmid DNA denaturation remains reversible. As the lysed solution is neutralized (by adding acidic sodium or potassium acetate), the denatured chromosomal DNA aggregates and precipitates, whereas the compact structure of the plasmid allows its complete renaturation.

There is evidence some RNA and protein molecules precipitate with chromosomal DNA, but RNA precipitation is limited. The addition of RNase for complete degradation of RNA is a common adaptation of the alkaline lysis protocol. Alkaline lysis is a useful purification step: Most of the chromosomal DNA is removed at this stage, which would be difficult otherwise because of its similarity to the plasmid DNA in chemical composition and structure. The protein and RNA burden is also reduced during alkaline lysis.

Using alkaline lysis in an industrial environment, however, raises several regulatory and manufacturing concerns. Regulators discourage the addition of animal-derived enzymes (such as lysozyme, proteinase K, and RNase A). Scaling up the alkaline lysis method is a manufacturing challenge: For instance, mixing must be carefully controlled to prevent shearing of the DNA, which would result in plasmid loss and chromosomal DNA contamination (15). Impellers have been designed, however, to provide gentle but thorough mixing of very viscous lysate solutions in batch operation.

In another approach, static mixers have been developed that rapidly but gently mix two continuous flows of solutions, providing efficient lysis with minimal shearing (16,17). Cell suspension and the lysis solution are pumped simultaneously into a static mixer in which lysis occurs very rapidly. Then, the resulting lysate is pumped into a second device, along with the neutralizing solution, before collecting the neutralized lysate in a stirred tank.

The precipitate that results from alkaline lysis also contains cell debris and aggregates of chromosomal DNA, RNA, and protein-SDS, which can then be removed by centrifugation or filtration. Continuous centrifugation is not recommended, however, because of plasmid DNA's shear sensitivity, so time-consuming and labor-intensive batch centrifugation or dead-end filtration must be employed.

Precipitation

After clarification of the lysate, many protocols precipitate the plasmid with ethanol or isopropanol and remove soluble contaminants. But solvents are not recommended for large-scale manufacturing because large volumes of solvent require spark-proof facilities and increased safety measures in handling. However, precipitation techniques that separate plasmid DNA from RNA can be useful at this stage. The various techniques to accomplish this can be divided into those that selectively precipitate plasmid DNA and those that selectively precipitate RNA.

Plasmid can be precipitated selectively by the addition of isopropanol (18,19), polyethylene glycol (7,20), spermine (21,22), spermidine (21), or cetyltrimethylammonium bromide (23,24). The disadvantage of this strategy is that the plasmid precipitate needs to be recovered by centrifugation or filtration and resuspended, which makes scale-up more difficult.

Selective precipitation of RNA removes the impurity while the molecule of interest (the plasmid) remains in solution. An antichaotropic salt at high concentration - such as lithium chloride (18,25), sodium acetate (26), ammonium acetate (27), ammonium sulfate (28), magnesium chloride (29–31), or calcium chloride (32, 33) - is the most common RNA precipitant. Lithium chloride is the most widespread RNA precipitant used in molecular biology laboratories but is not recommended for the manufacture of pharmaceutical-grade plasmid DNA because of toxicity issues. A literature survey of RNA precipitants reveals that not only is the list of precipitating agents extensive, but precipitant concentration and length of incubation vary widely.

In a comparative study, we tested five antichaotropic salts of high molal surface tension - ammonium sulfate, sodium sulfate, tripotassium citrate, ammonium acetate, and calcium chloride - for their potential to precipitate RNA (34). We found calcium chloride precipitated RNA rapidly and efficiently (85% removal) when compared with the other salts. Calcium chloride also substantially reduced chromosomal DNA and endotoxin levels. However, RNA precipitation with a high salt concentration was incomplete: Small RNA molecules remained soluble.

The selective precipitation of RNA in the presence of antichaotropic salt may be explained by the flexible single-stranded structure of the molecule, which means more hydrophobic bases are exposed, resulting in aggregation and "salting out." Supercoiled plasmid DNA, by contrast, is less hydrophobic because the bases are sheltered within the compact double helix structure. This may also explain why denatured chromosomal DNA coprecipitates with RNA.

Chromatography

Purification of plasmid DNA can also be accomplished using size-exclusion, reversed-phase, hydrophobic-interaction, ion-exchange, hydroxyapatite, silica, and triple-helix affinity chromatography.

Size-exclusion chromatography (SEC), also called gel-permeation chromatography, separates plasmid DNA from RNA by size (35). Complete resolution has been achieved on a Sephacryl S-1000 column, resulting in the recovery of pure plasmid, free of contaminating RNA (33,36). Removal of endotoxins, proteins, and chromosomal DNA has also been reported. However, several RNA reduction steps were required before the chromatography, including isopropanol, ethanol, ammonium acetate, PEG (36), and calcium chloride (33) precipitation. SEC is, therefore, essentially a polishing step, necessitating product prepurification and concentration. Scale-up of an SEC plasmid purification process is limited by the slow linear flow rate necessary for optimal resolution and is further complicated by the small loading volume, which requires prior concentration and a large-sized column.

Reversed-phase chromatography (RPC) exploits the difference in hydrophobicity between plasmid DNA and RNA. Elution is achieved with a gradient of increased solvent. Separation of plasmid from RNA has been achieved with silica-based C18 resins (28) and polymeric resins (27,37) such as polystyrene divinylbenzene or PolyFlo (Puresyn, Inc.). Silica-based resins are usually restricted to nonalkaline pH conditions, whereas polymeric resins have the advantage of a wide operational pH range, allowing sanitization with sodium hydroxide.

Resolution of plasmid DNA from RNA on polymeric resins requires an ion-pairing agent, usually tetrabutylammonium phosphate (TBAP) or triethylammonium acetate (TEAA), giving rise to the term "ion-pair reversed-phase chromatography" (27). Endotoxins have a high affinity for reversed-phase media, and high endotoxin clearance has been reported (27).

The disadvantages associated with RPC include the need to elute with solvents (usually acetonitrile, although ethanol can also be used), the RNA reduction steps required before chromatography to ensure sufficient resolution, and a low capacity for plasmid DNA.

Hydrophobic-interaction chromatography (HIC) is similar to RPC, in that molecules are separated on the basis of surface hydrophobicity. The difference is HIC resins are less hydrophobic (with C2 to C6 alkyl or with phenyl ligands), and elution is achieved with a decreasing gradient of antichaotropic salt (usually ammonium sulfate). Separation of plasmid from RNA has been achieved on a butyl-Sepharose column by loading in 2.5 M of ammonium sulfate (38,39). The plasmid was recovered in the column breakthrough, whereas bound impurities - RNA, chromosomal DNA, and denatured plasmid DNA - were eluted with a decreasing gradient of ammonium sulfate. The presence of exposed bases on single-stranded nucleic acids, such as RNA and chromosomal or plasmid DNA denatured during alkaline lysis, results in higher hydrophobicity when compared with double-stranded, supercoiled plasmid DNA that has its bases shielded within the double helix.

The main drawback to HIC is that, because only impurities bind to the column, contaminants must be significantly reduced before the chromatographic separation. Otherwise, column capacity will be exceeded, and the breakthrough of pure plasmid will become contaminated. Significant amounts of RNA and chromosomal DNA precipitate in the presence of high concentrations of ammonium sulfate required for loading. This, however, may result in insufficient reduction of impurity levels, which would then affect column capacity. In addition, the ammonium sulfate salt in the breakthrough needs to be removed from the final product by diafiltration or SEC.

Hydroxyapatite chromatography (HAC). Hydroxyapatite is a crystalline mineral of calcium phosphate [Ca5(PO4)3OH] in which the presence of highly-ordered positively-charged calcium ions and negatively-charged phosphate groups gives unique selectivity. The exact mechanism of interaction with the hydroxyapatite resin is not fully understood, but anion exchange, cation exchange, and calcium coordination appear to play a part (40).

Ceramic hydroxyapatite (cHA) is a form of the mineral sintered at high temperature, modifying its structure from a crystalline to a ceramic form and making a rigid bead. Elution is achieved by an increasing phosphate gradient.

That HAC resolves plasmid DNA from RNA has been known for several years (41-43). Macro-Prep cHA type II (Bio-Rad Laboratories) is the standard media for industrial HAC applications. We have found cHA to provide excellent resolution of plasmid from RNA, even when RNA was not previously reduced from the samples before chromatography. Loading crude lysate directly on a column is not recommended, however, because chelating agents can degrade the crystal structure. A diafiltration or precipitation step (to exchange the buffer of the plasmid solution) is, therefore, preferable.

In our experience, pH has an important effect on the separation of plasmid DNA from RNA - a difference of only 0.2 in pH resulted in significant loss of resolution. The phosphate counter ion in the elution buffer is also important: Sodium phosphate provided better resolution than potassium phosphate. Potassium phosphate is more soluble than sodium phosphate, however, and may be preferred for the final plasmid elution.

Despite impressive resolution, HAC media have one main disadvantage: low capacity. Because RNA has a lower affinity for hydroxyapatite than plasmid DNA, dynamic capacity can be increased by loading in a phosphate buffer at a concentration that prevents RNA binding, thereby maximizing plasmid capture. Even under these conditions, in our experiments, dynamic capacity never exceeded 400 µg/mL.

Anion-exchange chromatography (AEC) is the most common chromatographic technique in plasmid purification. AEC is easy to use and offers a wide choice of media from different manufacturers. AEC columns are easily sanitized, and no solvents are required. Methods have been developed for both packed (44-46) and expanded-bed (47,48) operations. Strong anion-exchangers - essentially those with quaternary (Q) functional groups (that is, where the positive charge is maintained over a wide pH range) - are usually preferred. Weak anion-exchangers, primarily diethylaminoethyl (DEAE) charged groups (where the degree of ionization is dependent on pH), are also used. Many matrix compositions are available including agarose, polystyrene-divinylbenzene (PS-DVB), methacrylate, cellulose, and acrylamide.

Resolution of RNA from plasmid DNA by AEC is less than from SEC or HAC. Binding is achieved through electrostatic interaction of the negatively-charged phosphate groups on the nucleic acid backbone with the positively charged ligands on the chromatography support. When the salt gradient is applied, the more weakly anionic molecules elute first. With nucleic acids, overall charge is a function of size so RNA molecules are fractionated according to their molecular weight. Plasmids are larger than the RNA molecules remaining after alkaline lysis and would be expected to have a higher affinity for anion-exchangers. However, the supercoiled structure of plasmids affects their overall charge density, and as a result, plasmids coelute with high molecular weight RNA molecules. The difference in charge densities between open circular and supercoiled isoforms also allows their separation on anion-exchangers. Pretreatment of the plasmid solution to reduce the RNA content is, therefore, necessary, often involving the addition of RNase (44-47).

Anion-exchangers, like other chromatography media, suffer from a low capacity for plasmids. Chromatography media were originally developed for protein purification. Porous beads, with a pore size of up to 1,000 Å were designed to increase the surface area available for protein binding. Plasmids - which are several orders of magnitude larger than globular proteins - can have molecular weights of more than 1,000 kDa and are too bulky to penetrate the pores of most commercial chromatography media. The only ligands available for plasmid binding are on the surface of the bead, resulting in far less capacity when compared with proteins.

Several manufacturers have developed chromatography media more appropriate for the purification of large molecules. POROS media (Applied Biosystems) are made of beads with large through-pores up to 8,000 Å diameter. Fractogel (Merck) and Sepharose XL media (Amersham Biosciences) have tentacular structures grafted onto beads, making charged ligands more accessible to large molecules. Ceramic Hyper D media (BioSepra) use a polymeric cross-linked hydrogel homogeneously distributed within a porous ceramic material. With this new generation of chromatography media, capacity for plasmid DNA has increased from a few hundred micrograms up to 5 mg/mL.

Silica chromatography. The property of silica oxide that causes it to bind DNA in the presence of chaotropic agents has been exploited for a number of years in DNA purification and is the basis of popular commercial kits. Until the advent of commercial kits, chromatography columns were prepared from diatomaceous earth (49,50), glass powder (51), or purified silica oxide (52,53). Binding DNA to silica initially required the use of toxic chemicals such as sodium iodide, sodium perchlorate (51), guanidinium thiocyanate, or guanidinium hydrochloride (49,52), but sodium chloride was found to be just as effective (54). The high salt content after alkaline lysis is likely to be sufficient for binding plasmid DNA to silica oxide without requiring additional binding agents. Plasmid is easily eluted with an aqueous, low-salt buffer.

Silica does not differentiate between double-stranded DNA, single-stranded DNA, or RNA. As a result, RNase is usually added to the cell-resuspension buffer, which is the case for QIAGEN plasmid purification kits. Lakshmi et al. (54) reported that by replacing SDS with Triton X-100 as the detergent for alkaline lysis and adding sodium chloride to 2 M final concentration for binding to silica, only traces of RNA were detected in the column eluate.

The main issue with silica chromatography is the absence of published information on the purification of pharmaceutical-grade plasmid DNA. Silica columns mentioned in the literature are usually prepared in-house rather than purchased from a reliable supplier. The application of this technique seems to be limited to high-throughput, laboratory-scale preparations.

Triple-helix affinity chromatography. Purification of plasmid DNA by affinity chromatography relies on the formation of a triple helix between an oligonucleotide covalently linked to a chromatographic matrix and a duplex sequence present on the plasmid DNA (55,56). A triplex is formed when a homopyrimidine on the oligonucleotide binds to a homopurine in the major groove of duplex DNA, forming Hoogsteen hydrogen bonds: thymine (T) on the oligonucleotide recognizes adenine–thymine (A•T) base pairs forming T•A•T triplets, and protonated cytosine (C+) recognizes guanine–cytosine (G•C) base pairs forming C•G•C+ triplets. The plasmid to be purified needs to be engineered to contain a target sequence of polypurine such as (GAA)17(55).

The advantage of affinity chromatography is the specificity of the interaction: only plasmids carrying the engineered target sequence are purified, while impurities including chromosomal DNA, RNA, proteins, and endotoxins are cleared in the column breakthrough. Plasmid binding requires a low pH and high salt environment, allowing direct loading of clarified lysate.

Initial assessments of triple-helix affinity chromatography have raised issues regarding the industrial application of this technique. Either batch adsorption or load recycling is required because of the long time (1–2 h) solute must be in contact with chromatography resin. Plasmid yields of less than 50% are also a concern, especially because the presence of plasmid in the column breakthrough and wash has not been explained. Capacity for plasmid DNA of only 28 µg/mL has been reported (56) - a major limitation for scale-up.

A promising recent adaptation of the triple-helix interaction involves precipitation rather than chromatography for recovery of plasmid product (57). In this technique, the oligonucleotide is covalently linked to a stimulus-responsive polymer, therefore a small change in temperature results in selective precipitation of the plasmid DNA affinity macroligand, while impurities remain in solution. This adaptation uses the highly selective triple-helix interaction without the low-capacity limitations of a chromatography resin.

Filtration

Filtration in most plasmid purification processes is for clarification or sterilization; it is rarely regarded as a separation technique. Recent advances in charged-membrane technology (58–60) offer an alternative to AEC. Membranes with large convective through-pores overcome the limitations found in porous bead structures and allow rapid purification at high capacity. RNA has been separated from plasmid in clarified lysate using Mustang Q (Pall) charged membranes and sequential elution with buffers of increasing ionic strength (just like in AEC). However, contaminant RNA could not be completely removed without the use of RNase (59,60).

Tangential flow filtration (TFF), also called cross flow filtration (CFF), is another technique commonly found for volume reduction and buffer exchange in downstream processing, which has potential in separation. TFF appears well suited for separating plasmid DNA from much smaller RNA molecules that remain after alkaline lysis. Kahn et al. (10) purified plasmid DNA from clarified lysate using a single TFF step but had to extend the lysis step (to 24 hours incubation at alkaline pH) to completely degrade RNA.

We have found significant amounts of RNA can be cleared using TFF directly after standard alkaline lysis (61). For optimal RNA removal, careful control of membrane pore size, transmembrane pressure, conductivity of the diafiltration buffer, diafiltration volumes, and plasmid load are required. TFF has the added benefits of clearing small molecules (including proteins), concentrating the product, and providing buffer exchange. However, we could not completely clear RNA directly after alkaline lysis, especially high molecular weight molecules, in one step.

Aqueous Two-Phase Extraction

In aqueous two-phase systems, molecules are partitioned between a top phase, rich in polyethylene glycol (PEG), and a dextran or salt-rich bottom phase. Separations of this type are highly dependent on system conditions, including PEG molecular mass, PEG concentration, salt type, salt concentration, pH, and the addition of detergents or chaotropic agents (62).

For the purification of plasmid DNA directly from lysate, PEG–salt systems are preferred, using either ammonium sulfate (62) or potassium phosphate (63) as the salt. A PEG molecular mass of 600 to 1,000 at a concentration of 15% to 20% (w/w) partitions plasmid DNA to the salt-rich bottom phase. No proteins and significantly-reduced RNA and chromosomal DNA content were observed at a PEG molecular mass of 600 (63), but plasmid recovery was low (,50%). Increasing the PEG molecular mass to 1,000 resulted in higher plasmid recovery but lower product purity.

A New Approach to RNA Removal

A radical new approach to removing RNA bypasses traditional purification methods altogether. Cooke et al. (8) have designed an E. coli strain that encodes a chromosomal RNase A expression cassette. The enzyme accumulates in the periplasm during growth and is then released during alkaline lysis, hydrolyzing the cytoplasmic RNA. Despite this significant development, RNA is incompletely degraded.

Designing a large-scale RNase-free plasmid purification process that completely removes RNA remains a challenge. Part 2 of this article will discuss a robust, two-step purification process that produces gram quantities of pure plasmid without adding RNase.

References

(1) Mountain, A., "Gene Therapy: The First Decade,"

Trends Biotechnol.

18, 119-128 (2000).

(2) Gurunathan, S. et al., "DNA Vaccines: A Key for Inducing Long-Term Cellular Immunity," Curr. Opin. Immunol. 12, 442-447 (2000).

(3) Robinson, H., "DNA Vaccines: Basic Mechanism and Immune Responses (Review)," Int. J. Mol. Med. 4, 549-555 (1999).

(4) Marquet, M., Horn, N.A., and Meek, J.A., "Process Development for the Manufacture of Plasmid DNA Vectors for Use in Gene Therapy," BioPharm 8(7), 26-37 (September 1995).

(5) CBER, Points to Consider on Plasmid DNA Vaccines for Preventive Infectious Disease Indications (FDA, Rockville, MD, 1996).

(6) CBER, Guidance for Human Somatic Cell Therapy and Gene Therapy (FDA, Rockville, MD, 1998).

(7) Prazeres, D.M.F. et al., "Large-Scale Production of Pharmaceutical-Grade Plasmid DNA for Gene Therapy: Problems and Bottlenecks," Trends Biotechnol. 17, 169-174 (1999).

(8) Cooke, G.D. et al., "Purification of Essentially RNA-Free Plasmid DNA Using a Modified Escherichia Coli Host Strain Expressing Ribonuclease A," J. Biotechnol. 85, 297-304 (2001).

(9) Monteiro, G.A. et al., "Analysis and Use of Endonuclease Activities in Escherichia Coli Lysates During the Primary Isolation of Plasmids for Gene Therapy," Biotechnol. Bioeng. 66, 189-194 (1999).

(10) Kahn, D.W. et al., "Purification of Plasmid DNA by Tangential Flow Filtration," Biotechnol. Bioeng. 69, 101-106 (2000).

(11) Carlson, A. et al., "Mechanical Disruption of Escherichia Coli for Plasmid Recovery," Biotechnol. Bioeng. 48, 303-315 (1995).

(12) Holmes, D.S. and Quigley, M., "A Rapid Boiling Method for the Preparation of Bacterial Plasmids," Anal. Biochem. 114, 193-197 (1981).

(13) Birnboim, H.C. and Doly, J., "A Rapid Alkaline Extraction Procedure for Screening Recombinant Plasmid DNA," Nucleic Acids Res. 7, 1513-1523 (1979).

(14) Birnboim, H.C. "A Rapid Alkaline Extraction Method for the Isolation of Plasmid DNA," Meth. Enzymol. 100, 243-255 (1983).

(15) Levy, M.S. et al., "Biochemical Engineering Approaches to the Challenges of Producing Pure Plasmid DNA," Trends Biotechnol. 18, 296-305 (2000).

(16) Wan, N.C. et al., "Methods for Lysing Cells," International Patent Application No. WO 97/23601 (1997).

(17) Nochumson, S. et al., "Process and Equipment for Plasmid Purification," United States Patent Application No. US2001/0034435 A1 (2001).

(18) Wallace, D.M., "Precipitation of Nucleic Acids," Methods Enzymol. 152, 41-48 (1987).

(19) Chakrabarti, A., Sitaric, S., and Ohi, S., "A Procedure for Large–Scale Plasmid Isolation Without Using Ultracentrifugation," Biotechnol. Appl. Biochem. 16, 211-215 (1992).

(20) Lis, J.T. and Schleif, R., "Size Fractionation of Double Stranded DNA by Precipitation with Polyethylene Glycol," Nucleic Acids Res. 2, 383-389 (1975).

(21) Murphy, J.C. et al., "Purification of Plasmid DNA Using Selective Precipitation by Compaction Agents," Nature Biotechnol. 17, 822-823 (1999).

(22) Mourich, D.V. et al., "Spermine Compaction Is an Efficient and Economical Method of Producing Vaccination-Grade DNA," J. Immunol. Methods 274, 257-264 (2003).

(23) Ishaq, M., Wolf, B., and Ritter, C., "Large-Scale Isolation of Plasmid DNA Using Cetyltrimethylammonium Bromide," BioTechniques 9, 19-24 (1990).

(24) Lander, R.J. et al., "Fractional Precipitation of Plasmid DNA from Lysate by CTAB," Biotechnol. Bioeng. 79, 776-784 (2002).

(25) Cathala, G. et al., "A Method for Isolation of Intact, Translationally Active Ribonucleic Acid," DNA 2(4), 329-335 (1983).

(26) Evans, J.K. et al., "Simultaneous Purification of RNA and DNA from Liver Using Sodium Acetate Precipitation," BioTechniques 24, 416-418 (1998).

(27) Green, A.P. et al., "Preparative Purification of Supercoiled Plasmid DNA for Therapeutic Applications," BioPharm 10(5), 52-62 (1997).

(28) Sparks, R.B. and Elder, J.H., "A Simple and Rapid Procedure for the Purification of Plasmid DNA Using Reverse–Phase C18 Silica Beads," Anal. Biochem. 135, 345-348 (1983).

(29) Palmiter, R.D., "Magnesium Precipitation of Ribonucleoprotein Complexes: Expedient Techniques for the Isolation of Undegraded Polysomes and Messenger Ribonucleic Acid," Biochemistry 13, 3606-3615 (1974).

(30) Skovgaard, O., "Selective Precipitation of RNA with Mg2+ Improves the Purification of Plasmid DNA," Trends Genet. 6, 140 (1990).

(31) Ugarov, V.I. et al., "Plasmid Purification Using Hot Mg2+ Treatment and No RNase," BioTechniques 26, 194-198 (1999).

(32) Mukhopadhyay, M., and Mandal, N.C., "A Simple Procedure for Large-Scale Preparation of Pure Plasmid DNA Free from Chromosomal DNA from Bacteria," Anal. Biochem. 133, 265-270 (1983).

(33) Raymond, G.J. et al., "Large-Scale Isolation of Covalently Closed Circular DNA Using Gel Filtration Chromatography," Anal. Biochem. 173, 125-133 (1988).

(34) Eon-Duval, A. et al., "Precipitation of RNA Impurities with High Salt in a Plasmid DNA Purification Process: Use of Experimental Design to Determine Reaction Conditions," Biotechnol. Bioeng. 83, 544-553 (2003).

(35) McClung, J.K. and Gonzales, R.A., "Purification of Plasmid DNA by Fast Protein Liquid Chromatography on Superose 6 Preparative Grade," Anal. Biochem. 177, 378-382 (1989).

(36) Horn, N.A. et al., "Cancer Gene Therapy Using Plasmid DNA: Purification of DNA for Human Clinical Trials," Hum. Gene Ther. 6, 565-573 (1995).

(37) Green, A.P., DNA Vaccines: Methods and Protocols, D.B. Lowrie and R.G. Whalen, Eds. (Humana Press Inc., Totowa, NJ, 2000), pp. 1-9.

(38) Diogo, M.M. et al., "Purification of a Cystic Fibrosis Plasmid Vector for Gene Therapy Using Hydrophobic Interaction Chromatography," Biotechnol. Bioeng. 68, 576-583 (2000).

(39) Diogo, M.M. et al., "Scale-Up of Hydrophobic Interaction Chromatography for the Purification of a DNA Vaccine Against Rabies," Biotechnol. Lett. 22(17), 1397-1400 (2000).

(40) Gagnon, P., "An Enigma Unmasked: How Hydroxyapatite Works and How to Make It Work for You," Validated Biosystems Quarterly Web Newsletter (1998).

(41) Colman, A. et al., "Rapid Purification of Plasmid DNAs by Hydroxyapatite Chromatography," Eur. J. Biochem. 91, 303-310 (1978).

(42) Johnson, T.R. and Ilan, J., "Large-Scale Isolation of Plasmid DNA and Purification of l Phage DNA Using Hydroxyapatite Chromatography," Anal. Biochem. 132, 20-25 (1983).

(43) Giovannini, R. and Freitag, R., "Continuous Isolation of Plasmid DNA by Annular Chromatography," Biotechnol. Bioeng. 77, 445-454 (2002).

(44) Hines, R.N. et al., "Large-Scale Purification of Plasmid DNA by Anion-Exchange High-Performance Liquid Chromatography," BioTechniques 12, 430-434 (1992).

(45) Chandra, G. et al., "Large-Scale Purification of Plasmid DNA by Fast Protein Liquid Chromatography Using a Hi-Load Q Sepharose Column," Anal. Biochem. 203(1), 169-172 (1992).

(46) Prazeres, D.M.F., Schluep, T., and Cooney C., "Preparative Purification of Supercoiled Plasmid DNA Using Anion-Exchange Chromatography," J. Chromatogr. A 806(1), 31-45 (1998).

(47) Theodossiou, I., Sýgaard, M., Thomas, O.R.T., "Design of Expanded Bed Supports for the Recovery of Plasmid DNA by Anion Exchange Adsorption," Bioseparation 10, 31-44 (2001).

(48) Thwaites, E. et al., "Impact of the Physical and Topographical Characteristics of Adsorbent Solid-Phases upon the Fluidised Bed Recovery of Plasmid DNA from Escherichia Coli Lysates," J. Chromatogr. A 943, 77–-0 (2002).

(49) Carter, M.J. and Milton, I.D., "An Inexpensive and Simple Method for DNA Purifications on Silica Particles," Nucleic Acids Res. 21, 1044 (1993).

(50) Hansen, N.J.V. et al., "A Fast, Economical and Efficient Method for DNA Purification by Use of a Homemade Bead Column," Biochem. Mol. Biol. Int. 35, 461-465 (1995).

(51) Marko, M.A., Chipperfield, R., and Birnboim, H.C., "A Procedure for the Large-Scale Isolation of Highly Purified Plasmid DNA Using Alkaline Extraction and Binding to Glass Powder," Anal. Biochem. 121, 382-387 (1982).

(52) Boom, R. et al., "Rapid and Simple Method for Purification of Nucleic Acids," J. Clin. Microbiol. 28, 495-503 (1990).

(53) Neudecker, F. and Grimm, S., "High-Throughput Method for Isolating Plasmid DNA with Reduced Lipopolysaccharide Content," BioTechniques 28, 106-108 (2000).

(54) Lakshmi, R., Baskar, V., and Ranga, U., "Extraction of Superior-Quality Plasmid DNA by a Combination of Modified Alkaline Lysis and Silica Matrix," Anal. Biochem. 271, 109-112 (1999).

(55) Wils, P. et al., "Efficient Purification of Plasmid DNA for Gene Transfer Using Triple-Helix Affinity Chromatography," Gene Ther. 4, 323-330 (1997).

(56) Schluep, T. and Cooney, C.L., "Purification of Plasmids by Triplex Affinity Interaction," Nucleic Acids Res. 26, 4524-4528 (1998).

(57) Costioli, M.D. et al., "DNA Purification by Triple-Helix Affinity Precipitation," Biotechnol. Bioeng. 81, 535-545 (2003).

(58) Van Huynh, N. et al., "Sequential Elution of Denatured Proteins, Hydrolyzed RNA, and Plasmid DNA of Bacterial Lysates Adsorbed onto Stacked DEAE-Cellulose Membranes," Anal. Biochem. 211, 61-65 (1993).

(59) Grunwald, A.G. and Shields, M.S. "Plasmid Purification Using Membrane-Based Anion-Exchange Chromatography," Anal. Biochem. 296, 138-141 (2001).

(60) Teeters, M.A. et al., "Adsorptive Membrane Chromatography for Purification of Plasmid DNA," J. Chromatogr. A 989(1), 165-173 (2003).

(61) Eon-Duval, A. et al., "Removal of RNA Impurities by Tangential Flow Filtration in an Free-Free Plasmid DNA Purification Process," Anal. Biochem. 316, 66-73 (2003).

(62) Cole, K.D., "Purification of Plasmid and High Molecular Mass DNA Using PEG-Salt Two-Phase Extraction," BioTechniques 11, 18-24 (1991).

(63) Ribeiro, S.C. et al., "Isolation of Plasmid DNA from Cell Lysates by Aqueous Two-Phase Systems," Biotechnol. Bioeng. 78, 376-384 (2002). BPI