October 1, 2013 (Vol. 33, No. 17)

Engineering E. Coli Host Strains for Optimized Fermentation and Production of DNA

Plasmid DNA vaccines (pDNA) are receiving renewed interest as initial problems of low antigenicity are being solved. These vaccines are particularly attractive for their fast development cycle, temperature stability and inexpensive manufacture by E. coli fermentation.

However, traditional strains of E. coli contain mobile DNA elements including up to 65 Insertion Sequences (IS) and up to 11 species of partially defective bacteriophage in their genomes (Table). Both phage and IS elements can be activated during production causing inconsistent fermentation results.

Moreover, IS elements can transpose into the plasmid product, changing the sequence and function of the therapeutic agent, leading to the possibility that bacterial IS elements could make their way into the mammalian genome post-administration.

Fortunately, these problems can all be eliminated by the use of IS-free E. coli in all steps of pDNA production.

Table. IS and Prophage in popular E. coli strains: Each box shows the number of copies of the element in the genome. Note: these counts represent a snapshot in time. Strains that have been sub-cultured multiple times may differ in their IS count or contain different complements of IS elements.

Fermentation Induces Instability

The bacteria’s reaction to stress during fermentation causes mobilization of IS elements and prophages, sometimes leading to cell lysis, spilling cell contents including pDNA into the medium, thus making the plasmid product difficult or impossible to recover.

Mobilization of IS elements (plasmid or chromosomal) to new locations disrupts genes in the chromosome and the resident plasmids. Any disruption that relieves the stress confers a growth advantage, enabling the resulting mutant to take over the culture.

Higher stress levels increase IS transposition and error-prone DNA replication rates. For example, plasmids with secondary structure such as viral Long Terminal Repeats (LTR) or short-hairpin RNA (shRNA) grow faster when the secondary structure (and its function) is disrupted.

Genetic instability can be triggered simply by growing a culture of commonly used bacterial strains from a cell bank. IS elements also serve as homologous targets of recombination when present in multiple copies. Deletions, insertions and large-scale rearrangements generated by IS10 have been reported.

Even the synthetic genome of Mycoplasma mycoides was found to be contaminated with IS1 from a DNA molecule grown in DH10B in an intermediate step. The extent of the IS and prophage problem in common E. coli strains is illustrated in the Table.

Transposition of some IS elements involves covalently closed active intermediates, “mini-circles”, which co-purify with plasmid DNA made by standard isolation techniques. Normally below gel detection level, they can be demonstrated by inside-out PCR, where the primers are designed to extend outwards past the end of the element.

Since the template is circular, amplimers are obtained. Unless specific methods are applied, IS transpositions generally remain undetected and their ability to cause problems has been under-appreciated. However, investigators are beginning to recognize the significance of the problem.

For example van der Heijden et al. reported that 12.85% of an HPV pDNA vaccine plasmid contained an IS2 although none had been detected in their master cell bank. They calculated that “it would have to be at a frequency below 1 in 10,000” to be undetectable. This study shows that massive expansion of IS elements from a very low starting level can be catastrophic. The authors conclude that the only way to avoid the problem is to use an IS-free host.

Stability Is Key in DNA Vaccines

If an IS-contaminated vaccine is given to a human patient, is there a chance that an IS element can jump into the human genome? At present we do not know. There are reports of bacterial IS elements transposing across Phyla that are genuine insertions rather than sequencing artifacts. In 2004, Merck scientists reported that whole plasmids could integrate into mouse genomes after intramuscular injection for gene therapy. The insertions sites were not targeted and the insertions appeared to be random.

If bacterial IS elements can do this too, such an event would be difficult to detect unless it were tumorigenic. The possibility of transposition of bacterial elements into mammalian genomes remains an under-appreciated and uninvestigated possibility. Indeed, a fatal tumor was caused by a genomic insertion of a viral gene therapy vector into a patient’s genome.

Though safer vectors have now solved the problem, it was a salient lesson. The fact that pDNA vaccines are delivered in multiple high doses adds to the potential for unintended transpositions.

IS-Free E. coli Is the Stable Solution

To completely eliminate this risk, a unique platform consisting of the Clean Genome® E.coli strains has been developed by Scarab Genomics by systematically deleting segments from the genome. The strains are called MDS (multiple deletion strains). More than 70 deletions have been made, removing all mobile elements (prophages and IS elements) and repeat sequences.

Many other genes have also been deleted including those for surface antigen genes and endotoxin. These strains are not only IS-free and prophage-free but are engineered for robust growth in both rich and minimal media resulting in reproducible, high yield fermentations.

In addition, advanced strains with reduced background mutation rates have been constructed which further enhance fermentation stability.

Figure 1. Detection of IS in plasmids grown in different host strains: The White Glove PCR test for IS1, IS2, and IS5 was performed on 1 ng pUC19 DNA prepared from different host strains by alkaline lysis and cesium chloride density gradient purification. PCR products were analyzed by agarose gel electrophoresis.

IS-Free Stability Is Necessary at Every Step

Routine plasmid preparations, including those purchased and used as cloning vectors, contain IS molecules detectable by PCR. The White Glove Kit consists of a set of specific primers for amplification of all known E. coli IS elements.

Figure 1 shows distinct amplimers of the predicted size, demonstrating multiple IS contaminants in all plasmids grown on non-MDS strains. Figure 2 illustrates the steps necessary to avoid IS contamination in pDNA products compared with current practice.

Using a minimum amount of DNA for transformation, IS-free transformants can be identified by testing individual colonies with the kit. A second or third round of transformation and testing may be necessary to obtain a plasmid preparation that is completely free from IS elements including mini-circles.

DNA to be cloned into the plasmid vector should also be subjected to this cleaning routine. This effort is crucial to prepare a clean master cell bank to ensure stability, because a “dirty” plasmid will “infect” a clean host.

Thus, the entire plasmid construction process should be carried out in a clean host to guarantee that every plasmid molecule is an authentic replicate of the original molecule.

Figure 2. Workflow for pDNA vaccine production: Extra care in the first steps will ensure trouble-free production.

David Frisch, Ph.D. ([email protected]), is a senior scientist, Robert Novy ([email protected]) is senior manager of product development, Charles Landry, Ph.D. ([email protected]), is a research and development scientist, Heather Kirkpatrick ([email protected]) is director of marketing, and Frederick R. Blattner, Ph.D. ([email protected]), is CEO and director of research at Scarab Genomics. György Pósfai, Ph.D. ([email protected]), is director of the Institute of Biochemistry, Biological Research Centre of the Hungarian Academy of Sciences.

Previous articleFirms Ripe for Takeover
Next articleMerck & Co. Axes 8,500 in R&D Restructuring