Function B: MRI Contrast Agent

This is a detailed description of an individual function of our circuit. To access other functions and get an overview of the whole circuit, visit the Circuit page.

Introduction

Since the beginnings of modern medicine, doctors have been eager to see what's happening on the inside of the body rather than to just treat blindly.

Therefore, it comes as no surprise that one of the characteristics that the ideal bacterial cancer therapeutic should posses is external detectability. The ability to see from the outside what's happening on the inside would allow the doctor to get critical information about the state of the tumor, the success of localization and the efficacy of treatment. [1]

Figure 1. MRI scan of the brain, a non-invasive way to take a peak at what's going on inside of the head. From pixabay.com

The Imaging Modality

To achieve this in CATE, the first step was to choose the imaging modality. Several methods have already been used to visualize contrast-producing bacteria, such as Positron Emission Tomography (PET), Bioluminescence Imaging (BLI), Fluorescence Imaging, Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). [2][3][4] To pick one, we carefully evaluated advantages and disadvantages of these different techniques (Table 1).

Our final decision: MRI. We chose MRI because it is a well-established imaging method in humans, readily available in the clinics. It offers good spatial resolution and does not use ionizing radiation, which means it is in no way harmful. Finally, focused ultrasound, an integral part of our Heat Sensor, uses exclusively MRI image guidance. Therefore, for both theoretical and practical reasons, we decided to incorporate an MRI reporter gene into CATE.

Table 1. Advantages and disadvantages of different imaging modalities. BLI = bioluminescence imaging, PET = positron emission tomography, MRI = magnetic resonance imaging, CT = computed tomography
	BLI	FLUORESCENCE	PET	MRI	CT
ADVANTAGES	sensitive non-toxic	sensitive non-toxic	suitable for humans very sensitive	suitable for humans spatial resolution non-toxic	suitable for humans spatial resolution temporal resolution
DISADVANTAGES	only small animals	only small animals external light source limited depth	radioactive source expensive spatial resolution	slow (ca. 30 min/scan) expensive less sensitive than PET	ionizing radiation

The MRI Reporter

FIXME — Figure 2. Effects of overexpression of different ferritin-like proteins in E. coli Nissle on T2 relaxation rate. R2 relaxation = reciprocal of T2 relaxation, GFP = green fluorescent protein, bfr = bacterioferritin, fri = Dps-type ferritin, ftn = archetypal ferritin, bfr M52L and M52H = heme-deleted bacterioferritins
From Hill, Philip J., et al. *PLoS One* 6.10 (2011): e25409 under the terms of the Creative Commons Attribution License.

We have looked into several different MRI reporters and consulted professor Markus Rudin, an expert in molecular imaging from the Institute of Biomedical Engineering at ETH Zurich. We considered reporters based on chemical exchange saturation transfer (CEST) [5], iron-based and enzyme-based reporters. [3] After consultation with professor Rudin and careful consideration, we chose to work with a ferritin-like bacterial protein, bacterioferritin. Features of bacterioferritin that influenced our choice include: [6]

superior sensitivity,
the lack of a need to supply an exogenous substrate for contrast generation since iron is endogenously available and
the fact that it can be easily overexpressed in E. coli. [7]

In general, ferritins are iron storage proteins found in different species. Through ferroxidase, they convert iron from its toxic, ferrous state (Fe²⁺) into a non-toxic, ferric state (Fe³⁺). When present in its ferric state, iron acts as a paramagnetic agent and causes a shortening of the transversal relaxation (T2) in MRI. Depending on the concentration of iron, this can lead to a visible change in contrast. [8]

Bacterioferritin is one of the three forms of ferritin-like proteins found in bacteria. It has been shown that overexpression of bacterioferritin in E. coli Nissle 1917 can lead to a visible contrast change in MRI, which allows for visualization of the bacteria (Figure 2). [7]

Bacterioferritin and CATE

In our design, we make bacterioferritin a reporter of passing through Checkpoint 1. Once enough bacteria have colonized the tumor and overproduction of lactate has been sensed, the AND-gate of the Tumor Sensor is activated. This leads to expression of the MRI contrast agent and the Anti-Cancer Toxin (Figure 3).

While the Anti-Cancer Toxin will accumulate inside of CATE until it is ready to be released during Cell Lysis, the MRI Contrast Agent will take up iron and create a change in the MRI signal. The change in the signal will alert the doctor of the following:

the bacteria have colonized the correct location,
there are enough bacteria to produce a full dose of the Anti-Cancer Toxin, so the risk of sub-dosing and eventually producing a drug-resistant tumor is minimized, and
the Anti-Cancer Toxin, located on the same plasmid and under the control of the same promoter as the MRI Contrast Agent, has started to accumulate.

References