Most biopharmaceuticals are made in recombinant bacteria that are engineered with antibiotic resistance genes which are closely allied to the gene coding for the plasmid in turn that creates the target protein.
That creates a simple method of separating bacterial cells that have been successfully altered to carry the desired gene, as only those that have the related resistance gene survive when exposed to antibiotics. But widespread use of resistance genes raises the risk of gene transfer to other organisms, potentially those that could cause infections.
This approach is also unsuitable for production of vaccines and gene therapy products, and is costly for industrial scale production, according to the scientists. Non-antibiotic-based systems are available, but involve the use of mutant host strains, defined media or expensive reagents.
The researchers, from the Karolinska Institute in Sweden and Royal Veterinary College in the UK, used an alternative approach.
In their study the bacteria were engineered to over-express a cell membrane synthesis gene called fab1, which is required for survival when the cells are grow in a medium containing triclosan, a compound with biocidal properties that is widely-used in bacteria-resistant consumer goods.
While working on gene targeting in bacteria, RVC researchers found that the interaction between fab1 and triclosan could be exploited for strain selection. Surprisingly, triclosan selection performed better than conventional antibiotic selection in a study involving Escherichia coli, which is commonly used in biopharmaceutical production.
“We think this simple technology is well suited for industrial scale fermentations that produce a range of valuable products, including bio-fuels and bio-pharmaceuticals,” said Dr Liam Good, the lead researcher on the project who works at the Royal Veterinary College.
In addition to avoiding the use of resistance genes and antibiotics, the triclosan approach had higher growth, yields and gene containment.
“More importantly, the new system is relatively safe and inexpensive, because the gene is native in all bacteria and triclosan is approved for use in many household applications.”
There are still concerns that need to be answered before the technique can be widely adopted, according to the researchers.
For instance, while triclosan is not used as a systemic antibiotic, there is a concern that its widespread use could generate resistance, and contribute to antibiotic resistance. That said, regulatory agencies continue to back the use of the product in many applications, and while resistant strains have been generated in the lab they have not been seen in non-lab conditions.
Another question concerns the issue of plasmid stability. In conventional large-scale production of proteins, plasmids must be stable in the absence of antibiotics, as it is important that residual antibiotics do not carry onward in the processing chain. In the current study triclosan was required to keep the plasmid stable, albeit at low levels.
Good and research partner Shan Goh of the Karolinska Institute speculate that a low level of residual triclosan may be acceptable, given the widespread use of the compound.
“Nevertheless, pFab stability is a potential problem and vector construct or process adjustments may be needed during scale-up,” they suggest.