Introduction

The idea of using bacteria for cancer treatment is not novel. Their ability to regress tumors was decribed by William Coley back in the 19^th century. However, natural bacterial cytotoxicity alone, mediated by a general stimulation of the immune system and competition for nutrients, was proved to be of limited success. [1]

An ideal bacterial cancer therapeutic should therefore not only rely on the native bacterial cytotoxicity. To increase the effectiveness of the treatment, the bacteria should be genetically modified to express an anti-cancer agent that would be delivered specifically to the tumor site. [2]

requirements

One of the major challenges throughout the design period of our project was the selection of the anti-cancer payload. We came up with a list of properties our agent of choice should exhibit:

• the agent should be genetically encodable in order to achieve in vivo production of the drug within E. coli,
• limited post-translational machinery of bacteria should be taken in consideration, as proteins with crucial-for-activity post-translational modifications cannot be expressed by our bacterial cells,
• the agent should efficiently kill solid tumor cells with minor or ideally no side-effects on healthy cells,
• mode of action of the drug should be well-characterized in order to increase the safety of the treatment,
• the agent should possess a multivalent mode of action and thus the ability to target a variety of solid tumors and finally,
• incorporation of the agent into our project should be in agreement with the safety regulations of iGEM and our department.

Azurin - the chosen one

In the process of searching through literature about anti-cancer proteins expressed by bacteria, our interest was caught by a protein named azurin, a blue copper protein originating from the well-known human pathogen Pseudomonas aeruginosa. Azurin has already been successfully expressed in E. coli Nissle and used as an anti-cancer agent in vivo. [3]

Although azurin's role as an electron transfer agent has been recognized since its isolation in 1956 [4], it was only in the year 2000 that the its cytotoxic effect on macrophages and mast cells was identifed. Shortly after, the cytotoxicity was shown to affect cancer cells as well. [5] Further in vitro and in vivo studies focused on elucidating the exact mode of action against malignant tumors. The results showed a the multimodal effect of purified azurin against diverse cancer cell-lines (Table 1).

p28 - a subdomain of azurin

A sequence of 28 aminoacids from azurin, spanning from aminoacid 50 to 77, is frequently mentioned in the literature as a stand-alone cytotoxic agent p28 [5]. It has been established that aminoacids 50 to 67 represent the protein transport domain of azurin. As for the actual cancer-killin activity, p28 is proposed to exert its effects mainly through p53 stabilization, which promotes apoptotic cell death. Considering that approximately 50% of all solid tumors carry a mutated p53, rendering them incapable of apoptosis, this mechanism of action alone would not be attractive. However, recently it has been shown that p28 also exhibits anti-angiogenic effects through inhibition of vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF). [9]

These features of p28, taken together with the fact that it has already reached clinical trials with success [10] led us to the decision to incorporate both p28 and azurin into our experiments.

Azurin and CATE

In our design, azurin (or p28) is accumulated early, immediately after passing throughCheckpoint 1. However, it is not released until the MRI Contrast Agent, produced at the same time as azurin (Figure 1), alerts the physician to activate the Heat Sensor. Activation of the Heat Sensor leads to Cell Lysis, which releases azurin specifically to the tumor (Figure 2).