A synthetic genetic circuit that mimics the quorum-sensing systems used by bacterial populations to coordinate gene expression enables bacteria to deliver drugs to mouse tumours in repeated and synchronized cycles. See Letter p.81
Humans and bacteria have a long history of parasitic and symbiotic relationships. Now, Din et al.1 exploit a relationship between bacteria and diseased human tissue for a therapeutic purpose. On page 81, the authors outline a system in which engineered bacteria acting as drug-delivery vehicles simultaneously break down, releasing an antitumour drug in synchronized cycles to maximize delivery efficiency and minimize toxicity.
In the body, some niches for bacteria — such as the anaerobic lumen of the intestines — have low oxygen levels. Similar conditions are found in solid tumours because of increased oxygen demand owing to highly proliferative tumour cells and insufficient blood supply owing to a structurally and functionally abnormal tumour vasculature2. The hypoxic areas in a tumour are relatively protected from attacks by the body's immune system, further facilitating bacterial colonization and growth3.
The idea of using bacteria to fight cancer has been around for more than a century. In 1891, surgeon William B. Coley infected patients with Streptococcus bacteria in an attempt to activate the immune system to fight cancer4. The method was controversial because of inconsistent efficacy and the toxicity of streptococcal infection. But the idea resurfaced later, when more was known about the tumour microenvironment and genetic-engineering tools had emerged, raising the hope that more-potent and less-toxic (attenuated) bacterial strains could be generated. Several bacterial strains have now been developed as agents for cancer therapy and they are showing promising effects in experimental models5.
Bacteria can destroy diseased tissue by competing for nutrients, secreting toxins and eliciting host immune responses. They can also be genetically engineered to have extra antitumour activity. Compared with viruses, which have also been used in cancer treatments, bacteria have substantially greater capacity to carry non-native DNA. It is routine practice in molecular biology to introduce DNA stretches of several kilobases into a bacterial host, and bacterial artificial chromosomes greater than 300 kb can be transferred to, and maintained in, Escherichia coli6. In principle, therefore, bacteria can serve as efficient drug-delivery vehicles, carrying genetic circuits that encode and regulate therapeutic payloads.
An ideal drug-delivery system for cancer therapy should deliver the substance selectively to the tumour to minimize harm to healthy tissues, and should release the drug in a controlled manner. In their attempt to develop such a system, Din et al. focused on quorum-sensing circuits, which enable bacteria to communicate with one another, regulating gene expression in response to changes in population density.
In a previous study7, the authors of the current work used a synthetic-biology approach to construct a quorum-sensing genetic circuit in E. coli. Three components — LuxI, LuxR and acyl-homoserine lactone (AHL) — have crucial roles in this circuit. The enzyme LuxI catalyses synthesis of AHL molecules, and LuxR is an AHL receptor protein that activates a quorum-sensing transcriptional program. When bacterial population density is low, LuxI is expressed at a basal level. The AHL molecules synthesized as a result of LuxI expression do not accumulate in the cell, but instead rapidly diffuse out and become diluted in the extracellular environment. When population density rises, AHL accumulates in the cell owing to the lowered diffusion gradient across the cell membrane. On reaching a threshold concentration in the cell, AHL molecules bind to LuxR. In turn, LuxR activates a promoter DNA sequence called PluxI, which drives expression of target genes. Notably, because AHL can diffuse across the cell membrane, it reaches similar concentrations in all bacterial cells in the growing population, ensuring synchronized execution of the gene-expression program.
In the current study, Din et al. created a version of this genetic circuit that controlled synchronized and cyclical release of a bacterial toxin in attenuated Salmonella enterica serovar Typhimurium strains. In this system, PluxI promotes expression of genes encoding four components — LuxI; the drug; a fluorescent protein to enable monitoring of population dynamics and drug release; and protein E, a lysis protein from a bacterial virus called φX174 (Fig. 1). When the bacterial population reaches the critical density threshold, this PluxI-driven transcriptional program is turned on in almost all cells, leading to drug production and its subsequent release owing to breakdown of bacterial cells through lysis. A few outliers in the population survive and repopulate the niche. The result is periodic bacterial lysis and drug delivery. To demonstrate efficacy, Din and colleagues treated tumour-bearing mice with the engineered bacteria and showed that the bacteria exhibit synchronized cyclical population dynamics, and confer some therapeutic benefits either when administered alone or in combination with chemotherapy.
The authors did not directly compare the efficacy of their bacteria with that of microbes engineered in the conventional way to continuously secrete the therapeutic protein. Regardless, the new bacteria are notably different from the conventional ones. First, drug delivery is achieved through the simultaneous lysis of the entire population, rather than through continuous secretion by proliferating individuals. Second, the periodic lysis serves as a safety mechanism, because keeping the bacterial population to a defined size minimizes the risk of an adverse systemic inflammatory response that might harm the patient.
Despite these features, bacteria alone (whether engineered or not), are unlikely to eradicate tumours1,5. In the current study, treatment of mice with the engineered microbes in combination with chemotherapy did not destroy the tumour; instead tumours shrank for 18 days, after which regrowth occurred. A curative therapeutic approach would most likely involve further improvements to engineered bacteria or using bacteria in combination with immunotherapy or other, more-powerful anticancer agents.
Cyclical drug release could be more useful for treating people who have diseases that require periodic dosing, such as diabetes and high blood pressure. To treat non-cancerous diseases, perhaps natural niches could be targeted for cyclical bacterial colonization. Alternatively, an implantable, semipermeable cassette that can be traversed by proteins and small molecules but not by bacteria could be developed to host the engineered microbes. One challenge to using periodically lysing bacteria to treat non-cancerous diseases that require long-term treatment is that the bacterial-degradation by-products released in each lysis cycle might be absorbed into the blood and build up, causing toxic systemic effects. Choosing less-toxic bacterial strains or creating attenuated strains (for example, by deleting the msbB gene, which is involved in making endotoxin8) could overcome the problem. Footnote 1
Din, M. O. et al. Nature 536, 81–85 (2016).
Vaupel, P., Mayer, A. & Höckel, M. Methods Enzymol. 381, 335–354 (2004).
Yu, Y. A. et al. Nature Biotechnol. 22, 313–320 (2004).
McCarthy, E. F. Iowa Orthop. J. 26, 154–158 (2006).
Forbes, N. S. Nature Rev. Cancer. 10, 785–794 (2010).
Shizuya, H. et al. Proc. Natl Acad. Sci. USA 89, 8794–8797 (1992).
Danino, T., Mondragón-Palomino, O., Tsimring, L. & Hasty, J. Nature 463, 326–330 (2010).
Low, K. B. et al. Nature Biotechnol. 17, 37–41 (1999).
Under a licensing agreement between BioMed Valley Discoveries Inc. and the Johns Hopkins University, S.Z. is entitled to a share of royalties received by the University on sales of products related to therapy using C. novyi-NT bacteria. The terms of the arrangement are under ongoing management by the Johns Hopkins University in accordance with its conflict of interest policies.
About this article
IET Systems Biology (2020)
Biomaterials Science (2020)
National Science Review (2019)
Interactions between microbiome and lungs: Paving new paths for microbiome based bio-engineered drug delivery systems in chronic respiratory diseases
Chemico-Biological Interactions (2019)
Physical Review E (2019)