Penn 2012 iGEM Wiki

Image Map

Design of our System

To construct platform for a bacterial therapy, we broke the system into two main components: light-induced production of a therapeutic and highly-specific targeting. We chose Escherichia coli BL21(DE3) cells as our chassis due to their optimization of protein expression as well as availability.

Light-Induced Production of a Therapeutic

In the midst of our brainstorming, Yi Yang and his colleagues published a paper in Nature Methods describing their construction of a light-switchable transgene system in mammalian cells. This system, named GAVPO, was a fusion protein consisting of three parts: a Vivid LOV domain that dimerizes under blue-light activation , the GAL4 DNA binding domain that only binds when dimerized, and the p65 transactivation domain. Upon induction by blue light, the Vivid domain of one GAVPO protein would dimerize with the Vivid domain of another protein, the GAL4 DNA binding domains would bind to the GAL4 specific upstream activation sequence (UAG) located upstream of the reporter gene, and the p65 transactivation domain would recruit the machinery necessary for transcription. The paper reported significant light-controlled gene expression using the GAVPO system.

Inspired by their results, we set out to create a similar system in bacteria using red light, since the wavelength is able to penetrate deeper within the body. Countless literature searches later, we realized that transactivation of transcription in bacteria was not well characterized. However, we identified a few systems that utilized an inverter to turn repression into activation. Among these was a red-light sensitive system using the phytochrome Cph1 and the EnvZ/OmpR two-component signaling pathway (Tabor et al, 2011), and a blue light sensitive system using the YtvA LOV protein and the FixL/FixJ two-component signaling pathway (Moglich paper). Further research showed that the YtvA-FixL fusion protein (named YF1) and the FixJ response regulator had been optimized for a greater on-off ratio, so we chose to implement that system (named pDawn) for proof of concept.

As we set out to build the pDawn system, we identified that all of the components existed in the registry. A lacIQ constitutive promoter drives expression of YF1 and FixJ. In the absence of blue light, YF1 phosphorylates FixJ, which activates transcription under the FixK2 promoter. To invert this process, the lambda phage repressor cI is placed under the FixK2 promoter and the gene of interest is placed under the lambda promoter pR, which is repressed by the lambda repressor. Therefore, in the presence of blue light, the FixK2 promoter is off, there is no expression of the lambda repressor, the pR promoter is activated, and the gene of interest is being expressed. Luckily, one of our advisors had recently received the pDawn plasmid through an MTA, and we were able to use this system in our experiments. This allowed us to focus our cloning on the second component of our system: targeting.


The second component of our system was targeting our bacterial therapeutic to a specific disease. In order to accomplish this, we had to design a surface display system that would allow us to display a targeting protein, peptide, or antibody fragment on the cell membrane of our bacteria. Depending on what is displayed on the surface, the bacteria would be able to bind to cells expressing the desired target molecule.

We identified various surface display proteins through literature searches, and finally settled on using an ice-nucleation protein (INPNC). These proteins have been well characterized as surface-display molecules in e.coli and have been used to display large proteins used for biocatalyst reactions. They had not been used as a display protein for cell targeting, however. We could fuse our targeting molecule of choice to the INPNC surface display protein using a GS linker on the C terminus, since that is the terminus that is outside of the cell. This system allowed us to surface display any protein we desired.

Proof of Concept: Light-Induced Cytolysis of Her2-overexpressing Cancer Cells

With our light-induction and surface display platforms designed, we chose to target Breast Cancer as a proof of principle to show that our system works. For our targeting mechanism, we cloned a Her2-specific DARPin protein into our surface display plasmid. The Her2 DARPin protein is very well characterized, has been expressed in E.coli, and binds with high specificity to Her2, a growth factor that is over expressed in many breast cancer cell lines. A neighboring lab gave us access Her2-overexpression cells (SK-BR-3), which would make allow us to assay the targeting mechanism at a lower cost. In addition, breast cancer tumors are generally closer to the surface of the human body. Since our induction system uses blue light as a proof of concept, and since blue light cannot penetrate too deeply into the body, targeting a disease that occurs closer to the surface of the body was also beneficial.

For our therapeutic of choice, we chose to use the cytoxin ClyA, a hemolytic, pore forming toxin that kills target cells by inserting channels into their membranes. ClyA is a simple yet elegant solution as a bacterial therapeutic. It is native to E.coli and has been expressed in large quantities. It doesn’t require invasion or internalization by the target cell and is therefore easy to assay by measuring cell lysis. In addition, there is much research being done on the use of ClyA as a cancer therapeutic, with some preclinical animal models currently in progress.

Next Steps: One-Plasmid System

We have successfully built and classified the two components of our system: the light-induced production of E. coli and the surface display of DARPin. The next steps are to verify that the DARPin binds to Her2 when surface displayed, and to finalize our system into a single, comprehensive plasmid. The construction of this plasmid is currently under progress with hopes of being completed and tested in time for the Jamboree.