Separation of supercoiled from open circular forms of plasmid DNA, and biological activity detection

Abstract

To establish a cost-effective purification process for the large-scale production of plasmid DNA for gene therapy and DNA vaccination, a single anion-exchange chromatography (AEC) step was employed to purify supercoiled plasmid DNA (sc pDNA) from other isoforms and Escherichia coli impurities present in a clarified lysate. Two different size and conformation plasmids were used as model targets, and showed similar elution behavior in this chromatographic operation, in which sc pDNA was effectively separated from open circle plasmid DNA (oc pDNA) in a salt gradient. The process delivered high-purity pDNA of homogeneity of 95 ± 1.1% and almost undetectable levels of endotoxins, genomic DNA, RNA and protein, at a yield of 65 ± 8%. Furthermore, the transfection efficiency (29 ± 0.4%) was significantly higher than that (20 ± 0.1%) of a pDNA control. The present study confirms the possibility of using a single AEC step to purify sc pDNA from other isoforms and host contaminants present in a clarified E. coli lysate.

Introduction

The recent developments in nonviral gene therapy and DNA vaccination have fostered the development of efficient plasmid DNA (pDNA) purification process (Diogo et al. 2005; Bencina et al. 2004). The number of naked/plasmid DNA studies has dramatically increased accounting for over 17.7% of all gene therapy clinical trials (www.wiley.co.uk/genmed/clinical, 2009), just behind adenoviruses (23.9%) and retroviruses (20.8%). In comparison to viral vectors, non-viral ones like naked pDNA are considered to be safer, but they are less effective. Several lines of evidence indicate that high supercoiled levels are required for eliciting an effective immune response and ultimately, protection from infectious challenge (Cupillard et al. 2005; Pillai et al. 2008). So the separation of supercoiled (sc) and open circular (oc) isoforms is one of the key steps in the large-scale purification of pDNA vectors for gene therapy and DNA vaccination.

Plasmids are usually produced in an Escherichia coli host by fermentation and are purified by a sequence of operations (Prazeres et al. 1998). Although E. coli produces mainly the more compact sc pDNA isoform (Prazeres et al. 1998), oc, linear and denatured pDNA isoforms are usually present due to conformational changes that occur within the bacterial host or during processing of biomass (Schleef and Schmidt 2004). For instance, an intact and undamaged form of sc pDNA may give rise to oc or linear isoforms by random cleavage (e.g., enzymatic, chemical) of one or two opposing strands, respectively (Iuliano et al. 2002). Furthermore, irreversibly denatured covalently closed pDNA isoforms are typically generated during alkaline lysis if the pH is not kept below 12.5 (Prazeres et al. 1998). Whatever the case, the resulting isoforms are likely to be less efficient in transferring gene expression if cleavage or strand melting disrupts promoter or gene coding regions (Schleef and Schmidt 2004). For this reason, regulatory agencies specify that the homogeneity of a therapeutic pDNA product, expressed as percentage of the sc isoform, should be higher than 90% (Schleef and Schmidt 2004).

There are different chromatographic techniques implemented for pDNA purification, such as size-exclusion chromatography (SEC) (Horn et al. 1995; Ferreira et al. 1997), anion-exchange chromatography (AEC) (Prazeres et al. 1998; Hines et al. 1992), hydrophobic interaction chromatography (HIC) (Gurunathan et al. 2000; Diogo et al. 2000, 2001, 2003) or affinity chromatography (AC) (Wils et al. 1997; Schluep and Cooney 1998; Sandberg et al. 2004). These chromatographic methods work well in removing host impurities (protein, RNA, endotoxins, genomic DNA etc.), but they can not separate sc pDNA from oc pDNA just by a single chromatographic step.

The aim of our work was to investigate the possibility of removing host impurities and separating sc from oc pDNA conformation just by a single chromatographic step. To evaluate the novel method, two different size plasmids pCDNA3.1-GFP (7,200 bp) and pT-survivin 2B (3,500 bp) were used as model target.

Materials and methods

Materials

The E. coli host strain DH5α containing pCDNA3.1-GFP (7,200 bp), a plasmid encoding GFP, and the E. coli host strain DH5α containing pT-survivin 2B (3,500 bp), a plasmid encoding survivin 2B, were constructed by the Institute of Biopharmacology of Guangdong Pharmaceutical University.

DEAE Sepharose Fast Flow, Sephacryl S-500 High Resolution, a XK 16/40 column, AKTA Basic system and Frac-920 fraction collector were provided by Amersham Bioscience. Beckman Avanti J-25 centrifuge was purchased from Beckman Co. Tianli gel scanner was the product of Tianli Biotech.

Lipofectamine 2000 and RPMI 1640 were purchased from Invitrogen, fetal bovine serum from TBD, Luria broth (LB) from Sigma and agarose from FMC. All salts used were of analytical grade.

Methods

Bacterial culture

E. coil DH5α harboring pCDNA3.1-GFP and pT-survivin 2B were grown overnight, at 37 °C, in shake flasks containing 500 mL of LB medium supplemented with 100 μg/mL Ampicillin, at 200 rpm. Cells were harvested by centrifugation at late log phase.

Lysis and primary isolation

Cells were lysed using the alkaline method. Cells were centrifuged at 5,000g for 15 min at 4 °C by Beckman Avanti J-25 centrifuge. Supernatants were discarded and pellet resuspended in 25 mL of 50 mM glucose, 10 mM ethylene-diamine tetraacetic acid (EDTA), 25 mM Tris/HCl, pH 8.0. Lysis was performed on ice by adding 25 mL of a prechilled 200 mM NaOH, 1% (w/v) sodium dodecylsulfate solution (SDS). Cellular debris, genomic DNA, and proteins were precipitated by gently adding 20 mL of prechilled (on ice) 3 M potassium acetate, pH 5.0. The precipitate was removed with centrifugation at 12,000g (20 min, 4 °C).

The neutralized lysate was precipitated by slowly adding solid CaCl₂ to a final concentration of 0.4 M with constant mixing. Suspension was left at 4 °C for 15 min, then centrifuged at 12,000g for 20 min at 4 °C. Supernatant was filtered through 0.45 μm filter and decreased ionic strength (around 34 mS/cm) by adding dH₂O, then loaded directly on the AEC columns.

Anion-exchange chromatography (AEC)

An XK 16/40 (40 cm × 1.6 cm) column was packed according to the manufacturer’s instructions with DEAE Sepharose Fast Flow, a weak anion exchanger. Chromatography was performed with AKTA Purifier system. The column was equilibrated with 4 column volumes (CV) of loading buffer (0.3 M sodium chloride in 10 mM Tris/HCl, pH 8.0) at a flow rate of 7 mL min⁻¹. The plasmid sample was loaded on the column at a flow rate of 7 mL min⁻¹. The column was washed with 2 CV of loading buffer. A linear gradient from 0.3 to 0.5 M was carried out with sodium chloride solution in 10 M Tris/HCl, pH 8.0 in 10 CV at a flow rate of 7 mL min⁻¹. Absorbance was monitored at 254 nm. Plasmid-containing fractions were collected and kept for further analysis. The column was cleaned with 2 CV of 1 M NaOH.

Desalting

Plasmid samples were desalted prior to protein assay and electrophoresis to prevent the interference of the salt with the analysis. Sephacryl S-500 High Resolution column was used with 10 mM Tris/HCl, pH 8.0, as running buffer.

Agarose gel electrophoresis

Samples from AEC were analyzed by horizontal electrophoresis using 0.8% agarose gels run at 100 V, with TAE buffer (40 mM Tris base, 20 mM acetic acid and 1 mM EDTA, pH 8.0), and stained with ethidium bromide (0.5 μg mL⁻¹).

Homogeneity analysis

The homogeneity of pDNA was analyzed by gel scanning with Tianli gel scanner according to the manufacturer’s instructions. The homogeneity was expressed as the percentage of the sc pDNA in total pDNA.

Protein analysis

Protein concentration was determined using the BCA (bicinichoninic acid) assay from Bioteke (China). Twenty microliter samples were pipetted into the wells of a microplate and 200 μl of the working reagent was added to each well. After homogenization, the microplate was incubated at 37 °C for 30 min. Absorbance was measured at 570 nm in a microplate reader. A calibration curve was made using bovine serum albumin as the protein standard.

Endotoxin analysis

Endotoxins were quantitatively assayed by the kinetic-QCL Limulus amebocyte lysate assay kit from Biowhittaker (Walkersville, MD), according to the manufacture’s instructions. The detection level for the method used was 0.010 EU/mL.

Purity analysis

Purity considered as the relative DNA/protein content was estimated by spectrophotometry, and expressed as the ratio of absorbance at 260 nm and at 280 nm respectively (A₂₆₀/A₂₈₀).

Biological activity detection

The plasmid’s biological activity was detected by liposomes transfection experiment with 7,402 cell. 4 × 10⁵ cells were plated in 2 mL of growth medium without antibiotics and cultured overnight to a confluence of 90–95%. 4 μg plasmid DNA and appropriate amount Lipofectamine 2000 were, respectively diluted in 250 μl RPMI 1640 medium without serum, incubated for 5 min at room temperature, combined by gently mixing, then incubated for 20 min at room temperature. All of the 500 μl complexes were added to each well containing cells and medium, mixed gently by rocking the plate back and forth, then incubated at 37 °C in a CO₂ incubator for 24 h. Medium may be changed after 4–6 h. Each experiment was performed six times. The transfected cells were detected by inverted fluorescence microscope (Olympus IX71), then digested by trypsin and adjusted to a cell concentration of 5 × 10⁶/mL. The transfection efficiency was detected using flow cytometry (Beckman).

Results

Separation of sc pDNA from oc pDNA

Two different size plasmids pCDNA3.1-GFP (7,200 bp) and pT-survivin 2B (3,500 bp) were used as model plasmids. Cells were lysed and clarified by conventional “alkaline lysis procedures”. In order to further remove impurities from host cells in the clarified lysates, a CaCl₂ precipitation step was taken before polishing.

A DEAE-Sepharose AEC was carried out for resolving proteins and nuclease contaminants and separating sc pDNA from oc pDNA. Two different size plasmids showed similar chromatogram in which there were a big flow-through peak and two elution peaks, a small but clear peak (first elution peak) followed by a big sharp peak (second elution peak), during the salt gradient elution process (Fig. 1). The results from agarose gel electrophoresis showed that the flow-through peak, first elution peak and second elution peak corresponded to non-nuclease impurities, oc pDNA and sc pDNA, respectively, and the yield of sc pDNA was 65 ± 8% (Fig. 2).

Fig. 1.

Chromatographic separation of pDNA isoforms with AEC. Linear gradient was performed at 7 mL/min by increasing the NaCl concentration in the eluent from 0.3 to 0.5 M in 10 mM Tris/HCl, pH 8.0. a pCDNA3.1-GFP (7,200 bp), b pT-survivin 2B (3,500 bp)

Fig. 2.

Agarose gel electrophoresis analysis. a pCDNA3.1-GFP (7,200 bp). Lane 1 DNA ladder; Lane 2 lysate; Lane 3 flow-through; Lane4–5 first peak; Lane 6–7 second peak. b pT-survivin 2B (3,500 bp). Lane 8 DNA ladder; Lane 9 lysate; Lane 10 flow-through; Lane 11 first peak; Lane 12 second peak

The quality of the purified pDNA was estimated by a series of analytics. The final preparations of pDNA had an A₂₆₀/A₂₈₀ ratio of 1.93 ± 0.05, which was regarded as high purity. The contaminants of genomic DNA and RNA were not detected in the agarose gel electrophoresis. The homogeneity, expressed as percentage of the sc isoform, of the obtained pDNA was estimated to be 95 ± 1.1% by gel scanning analysis. The protein contaminants were not detected by the BCA method. Endotoxins analysis revealed that the level of endotoxins in the preparations was lower than 0.003 EU μg⁻¹ pDNA, which well complied with the specification of <0.1 EU ug⁻¹ pDNA for gene therapy.

Biological activity detection

The results from in vitro transfection tests with 7,402 cells showed that the sc pDNA purified in the present work had higher transfection efficiency than the control pDNA, prepared with the OMEGA kit which is widely used for the small-scale preparation of pDNA for gene therapy in the world. The inverted fluorescence microscope analysis revealed that more cells were transfected by sc pDNA than by the control pDNA (Fig. 3), and the transfection efficiencies, estimated by flow cytometry analysis, of sc pDNA and control pDNA were 29 ± 0.4% and 20 ± 0.1%, respectively (Fig. 4).

Fig. 3.

Result of inverted fluorescence microscope (×100) a Plasmid purified using the OMEGA kit b Plasmid purified with the protocol of the present work

Fig. 4.

Transfection efficiency detected by flow cytometry a negative b Plasmid purified with the protocol of the present work c Plasmid purified using the OMEGA kit

Discussions

For the production of pharmaceutical-grade pDNA at an industrial scale three main requirements have to be met. First, the product must be of high purity, free from bacterial genomic DNA, RNA, Protein, endotoxins. Second, the purified pDNA must be of high homogeneity, meaning most plasmid in supercoiled form. Last, the process must be cost-effective, scalable and easy to control quality.

Alkaline lysis procedures are routinely used in harvesting and partially purifying of plasmid. By this method, E.coli cells can be effectively broken up, and the majority of impurities from host cells such as genomic DNA, host protein, and RNA was precipitated and removed, but further purification and separation of sc pDNA from oc pDNA is essential to meet the requirements of pharmaceutical-grade pDNA. Chromatography is easy to scale up and considered as the method with highest resolution, therefore being essential for producing pDNA for therapeutic applications. To remove trace contaminants and separate sc pDNA from oc pDNA, multiple chromatographic steps are frequently employed. However, just one single AEC step was employed to achieve above goal in our novel process. It is reasonable to conclude that DEAE Sepharose Fast Flow used in the process is the ideal AEC medium for separation sc pDNA from oc pDNA, and a CaCl₂ precipitation step maybe have significant benefits for the subsequent purification steps.

The novel purification process for pDNA developed in present work met all above requirements for producing pharmaceutical-grad plasmid. Results from a series of analytics showed that the final purified pDNA contained extremely low amounts of genomic DNA, RNA, Protein and endotoxins, and had homogeneity higher than the specification of 90% (Schleef and Schmidt 2004). The biological activity of the purified pDNA was also testified to be significantly higher than that of the control pDNA by In vitro transfection.

The process developed in this work is cost-effective and scalable since just one single chromatographic step is employed, allowing trouble-free transfer from lab-scale preparation to large-scale production. Furthermore, it has significant advantages over traditional laboratory methods because the use of enzymes or animal-derived products is avoided, and there are no known or suspected toxic, mutagenic, carcinogenic, teratogenic, or otherwise harmful compounds used in the process.