Minicells on Target
by Elio
The gold ring everyone in cancer chemotherapy is reaching for is the ability to selectively kill cancer cells without damaging normal ones. Easier said than done. So far, clever attempts at delivering potent drugs straight to the cancer cells using techniques such as conjugating them to antibodies specific to those cells, have been inconclusive at best. Now, a group of Australian investigators report promising results using bacterial minicells as the drug delivery system. (This research is highlighted in Nature Review: Drug Discovery.)
Thin-section electron micrograph showing a minicell producing E. coli cell, magnification ×66,000. Source
Several items here are of microbiological interest, the most surprising of which is that minicells can be loaded with anticancer drugs with the greatest of ease. Just place the minicells in a solution containing the drugs and presto! The minicells imbibe with such gusto that, depending on the drug, between 200,000 and 10 million molecules are stowed per minicell. No efflux takes place, so once loaded, the minicells become a stable delivery system. It seems to make no difference whether the drugs are hydrophilic (e.g., irinotecan), hydrophobic (e.g., paclitaxel and cisplatin), or amphipatic (e.g., doxorubicin, vinblastin, and 5‑FU). Magic!
Of course, stopping at this point would not result in selective targeting of the drugs. Therefore, the authors coat their minicells with a bispecific antibody. One antibody arm binds to the O‑antigen of the lipopolysaccharide on the minicell surface, the other to a specific surface receptor on the cancer cell. The minicells are now ready to go.
Minicells are rapidly taken up by cancer cells and broken up internally, thus delivering the drugs where intended. For a variety of human tumors, treatment with drug laden minicells destroys the tumors in vitro and also in vivo in mice carrying such tumors. The effect was much greater than with the drugs alone. Minicells also handily beat out drug-loaded liposomes. Mice, dogs, and pigs exhibited essentially no ill effects from the administration of the minicells. Impressive, indeed, and cheering. Bacterial minicells hold out significant promise for the development of tolerable anticancer therapies. And there promise to be yet other uses. Already minicells are being used as a safe delivery system for vaccines against lymphocytic choriomeningitis virus in mice.
One wonders if minicells might possibly be prepared totally free of bacteria so that one could avoid injecting whole live bacteria into patients. The authors use a combination of centrifugation and filtration steps to separate the minicells from their mother cells. They achieve a cleaner separation of the two cell populations by making the parent bacteria filamentous, hence larger and easier to retain on filters. The final minicell preparation contains no bacteria capable of forming colonies. Interestingly, they don't use the usual techniques for inducing filaments (adding murein-synthesis inhibitors such as β‑lactams or using min mutants that allow continuous production of minicells from filaments). Instead, they use high salt (5% NaCl) for 4 hours, which also works. As a personal aside, I'd like to quote from a prehistoric review I wrote in 1975 with Martin Salter: "It seems possible that any chemical at some concentration, whether attainable in the laboratory or not, could cause filament formation."
These tiny structures have held the attention of microbiologists since they were first described in detail by Howard Adler in 1966. In the early days, specific pieces of DNA were introduced into minicells via phages or plasmids, thus converting them into factories for manufacturing specific proteins. Later on, minicells led to the 1989 discovery of the Min system by Piet de Boer, Larry Rothfield, and colleagues, a discovery which materially increased our understanding of bacterial cell division. So, who says a cell has to have DNA to be exciting?
I am sure that vectors have to be safe (experience in the virus field underlines this) and one should be a bit cautious. I am sure I would be unhappy following an incautious dose of LPS along with minicells of Gram negative bacteria. I note that there is at least one paper that had this in mind and used protoplasted, achromosomal bacterial minicells for DNA transfer purposes.( M.J. Giacolone et al., Cell. Microbiol. 8:1624, 2006) They recognised this as one of a number of necessary steps before there is utility. On the other hand maybe a tiny dose of some such stimulus primes useful processes.