In this study, we innovatively developed a high-efficiency, cost-effective, and easy-to-operate system for the lab-scale separation and purification of the pcDNA-CCOL2A1 vaccine. Furthermore, we confirmed that the residual E. coli protein, genome, RNA, and endotoxins in the final supercoiled pcDNA-CCOL2A1 vaccine product completely conformed to the international criteria for DNA vaccines [13,14,15]. Our results will not only provide sufficient high-quality and high-yield pcDNA-CCOL2A1 vaccine for preclinical research but also promote further pilot-scale and even industrial-scale production of pcDNA-CCOL2A1 vaccine. Notably, some important advances worthy of in-depth discussion emerged from this study.

The production of genetic engineering products requires a complex biological engineering system that includes upstream and downstream process technologies. After successfully optimising the upstream process technologies, the downstream process technologies are more critical to the industrialisation of genetic engineering products. Downstream process technologies in genetic engineering typically include large-scale cultures of engineering bacteria (cells) and the separation, purification, and identification of expression products that meet clinical use standards [12]. In recovering genetically engineered products, we should not only pay attention to the use of highly selective separation and purification methods but also consider the reasons that affect the biological activity of the final product and the low utilisation rate of the culture medium during fermentation. Thus, several separation and purification steps are necessary, making downstream process technologies more complex than upstream process technologies. Because a large amount of the pcDNA-CCOL2A vaccine is required for efficacy studies, safety analyses, pharmacokinetic research, and vaccine stability investigations in preclinical trials, we have successfully established a three-tier cell bank and demonstrated the genetic stability of the engineered E. coli DH5α carrying the pcDNA-CCOL2A1 plasmid to produce this DNA vaccine with high potential [26]. Furthermore, we have systematically optimised the fermentation process for the engineered E. coli strain and greatly increased the yield of plasmid DNA by 51.9%, with the plasmid DNA yield per unit of bacterial liquid reaching 16.97 mg/L [18]. Statistically, the protein content of cell lysates obtained after fermentation is the largest in the unit dry weight, accounting for approximately 55% of the total weight, followed by RNA, which accounts for approximately 21%, and other impurities, such as endotoxins and genomic DNA, which account for approximately 21%; the plasmid DNA only accounts for approximately 3% [27]. Therefore, maximising the yield of plasmid DNA is the first goal of the purification process. In addition, plasmids larger than 10 kb will increase the difficulty of the purification process because larger plasmids are easily affected by shear force. Moreover, obtaining a relatively high proportion of supercoiled plasmid DNA is difficult. Similarly, the yield and purity of the plasmid may also be reduced. Therefore, to facilitate subsequent downstream purification, the pcDNA 3.1( +) expression plasmid, which has a length of only 5.428 kb, was used in our therapeutic pcDNA-CCOL2A1 vaccine [5]. These advances provide a solid foundation for further large-scale separation and purification of the pcDNA-CCOL2A1 vaccine with high quality and yield.

Downstream separation and purification technologies of genetic engineering should meet some requirements. First, the technical conditions should be mild to maintain the biological activity of the target product. Second, the approach should exhibit good selectivity and effectively separate the target product from the complex mixture to achieve a high purification ratio. Additionally, the yield should be high, and the two technologies should connect directly without the need to process or adjust the materials, which can reduce the number of process steps. Finally, the entire separation and purification process should be fast, meeting the requirements of high productivity. Thus, different separation and purification strategies and technical routes are usually formulated for the target plasmid DNA depending on the application, such as lab-scale, pilot-scale, and industrial-scale stages. For example, at the lab-scale stage, purifying plasmid DNA usually involves using a commercial purification kit or the hexadecane trimethylammonium precipitation method. However, these two methods have numerous shortcomings. For example, the quality of the final product is not controllable. Plasmid DNA obtained by different technicians in different batches showed instability and poor reproducibility regarding yield and residual impurities. To purify different plasmid DNA, it is necessary to repeatedly explore the best purification conditions [28]. Additionally, the purification cost is high. The endotoxin removal solution is a patented component of the kit product and cannot be recycled and reused, significantly increasing the purification cost. The chromatogram column used in purification is also non-renewable; thus, the chromatography step greatly increases the purification cost [29, 30]. The approach is also time-consuming, and increasing the reaction to a larger scale is difficult. The use and residual presence of some solvents may also cause safety hazards. Therefore, chromatography techniques are undoubtedly one of the best methods for large-scale plasmid DNA purification, with the advantages of high resolution and high separation efficiency. Commonly used methods include affinity chromatography (AC), size-exclusion chromatography (SEC), and AEC. AC is more sensitive in terms of specificity and selectivity and has therefore become an essential step in the separation of plasmid DNA isomers. SEC is more suitable for the purification of plasmid DNA as a part of downstream purification in combination with other purification methods because the existing media cannot effectively separate the isomers of plasmid DNA. Owing to it versatile functions, AEC can remove a wide range of impurities. However, it is limited by sample volume and quality; therefore, it is suitable for use in the last purification step to achieve the final purification of residual impurities that have not been completely removed in previous purification steps [23, 24]. Overall, the purification of plasmid DNA with different quality requirements can be achieved via chromatography alone or in combination with other methods. Hence, it is imperative to select different chromatographic processes to obtain plasmid DNA that meets the international quality standard. However, two or more chromatography steps will increase costs while decreasing the recovery of plasmid DNA [31].

In our study, following long-term screening and comparison experiments, we optimised and combined the best separation and purification methods, PEG/MgCl2 precipitation and Triton X-114. We initially chose a commercial purification kit alone or a chromatography purification method alone, both of which yielded unsatisfactory outcomes. Single AEC not only fails to effectively purify the plasmid DNA but also requires a combination of multi-step chromatography and ultrafiltration in addition to needing a series of high-end instruments. Besides the high costs and yield of plasmid DNA obtained, a single commercial kit also failed to meet the requirements of some preclinical experiments. The cost of producing 1 mg supercoiled plasmid DNA pcDNA-CCOL2A1 using the Qiagen Endo-Free Plasmid Mega kit was US$ 31.67; however, using a combination of PEG/MgCl2 precipitation and Triton X-114, the cost was only US$ 1.13. Moreover, by combining PEG/MgCl2 precipitation with Triton X-114, an average yield of 11.81 ± 1.03 mg supercoiled plasmid DNA can be purified from 1 L fermentation broth, with a concentration of 670.6 ± 57.42 mg/L, which was significantly higher than that obtained using AEC or the commercial purification kit alone. The plasmid DNA isolated and purified from 1 L fermentation broth could meet the demand of 100 CIA rat models for in vivo efficacy studies, pharmacology assays, and toxicity experiments [8]. In particular, the supercoiled plasmid DNA separated and purified by combining PEG/MgCl2 precipitation with Triton X-114 had a high purity and the same biological activity as the plasmid obtained from an internationally used commercial kit. These results indicated that our method was highly efficient, cost-effective, and easy to operate for lab-scale separation and purification of the pcDNA-CCOL2A1 vaccine.

The clinical applications of therapeutic DNA vaccines have broad prospects; however, their safety must be guaranteed before they can be applied to humans. First, quality control mainly considers the purity and consistency of the plasmid DNA vaccine as well as the presence of E. coli residual proteins, genome, and endotoxins. Second, the final product must reach a certain purity. The establishment and verification of the quality standard of the purified supercoiled plasmid DNA are essential steps in the overall purification process evaluation and are also the most critical issues for ensuring the safety and effectiveness of subsequent use [11, 12, 16]. Different countries and regions have formulated their quality standards for the quality control of gene products used for therapeutic purposes. Both FDA and EAEM have issued a series of strict evaluation principles and quality standards as follows: host RNA cannot be detected by 0.8% agarose gel electrophoresis, protein ≤ 1 ng/µg plasmid, genomic DNA ≤ 0.002 µg/µg plasmid, endotoxin ≤ 10 EU/mg plasmid, and supercoiled plasmid DNA ≥ 90%, with a total purity of A260/A280 ≥ 1.75 [11,12,13,14,15]. The identification test was performed in accordance with the restriction map after restriction enzyme electrophoresis (Fig. 4).

One of the strengths of this study is that the combined purification method of PEG/MgCl2 and Triton X-114 effectively eliminated residual impurities from E. coli in the final supercoiled plasmid DNA product, thereby conforming to the international guidance for DNA vaccines [11,12,13,14,15]. The final purity of the supercoiled plasmid DNA obtained by combining PEG/MgCl2 precipitation with Triton X-114 was 94.98%, which was far higher than the standards of FDA and EAEM (90%). Notably, the efficiency of combining PEG/MgCl2 precipitation and Triton X-114 to remove endotoxins was higher than that of AEC and the commercial purification kit. Removal of endotoxin is always one of the bottlenecks in the purification process of plasmid DNA. This is because an endotoxin has a saclike structure, and its molecular weight, charge, and hydrophobicity are similar to those of plasmid DNA. Moreover, the molecular weight of an endotoxin ranges from hundreds of thousands to tens of millions, with a large number of negative charges. Therefore, reducing the content of endotoxin to a safe level using the molecular sieve and AC technologies currently used for the large-scale preparation of pharmaceutical plasmid DNA is challenging. In particular, E. coli, an engineered bacterium used to prepare DNA vaccines, contains a large amount of endotoxin. Usually, a concentration of 10% wet bacteria can produce tens of thousands of EU/mL of lipopolysaccharide (LPS), which is beyond the tolerance range of the human body. Because LPS has a strong heat source, a small amount can cause fever, blood circulation disorders, and even death from septic shock. Moreover, the presence of endotoxins can significantly affect the transfection efficiency of cells [32, 33]. Therefore, it is essential to remove endotoxin from the supercoiled plasmid DNA and ensure conformity to the relevant safety standards for the clinical application of plasmid DNA vaccines.