New Synthetic Regulatory Circuit Can Identify and Kill Cancer Cells

Collaborating scientists at the Department of Biological Engineering at MIT and the Center for Systems Biology at Harvard University have created an engineered biological system that senses and integrates multiple inputs and can precisely regulate the biology of a living cell. This type of approach could be useful to engineer anti-cancer therapies that are able to distinguish a cancer from a non-cancer cell, inducing apoptosis in the cancerous cells.

Geniom RT Analyzer for microRNA quantification analysis for biomarkers of disease progression

Zhen Xie and Professor Ron Weiss of MIT, Yaakov Benenson of Harvard and the ETH Zurich Institute and their colleagues have demonstrated how a “transcriptional synthetic regulatory circuit” can sense the expression levels of endogenous microRNAs within a cell and trigger an engineered cellular response. The research is published in Science (Xie et al. Science Sept. 2, 2011 333: 1307-1311). The applications of this system include cancer and other disease treatment as well as further elucidating important biological processes.

To induce cancer cell-killing, the researchers designed a logic circuit made up of genes that detect molecules that are specific to the HeLa cervical cancer cell line. The gene circuit is delivered to cells and if a specific set of endogenous molecules are present at a particular level, the apoptotic protein, Bcl-2–associated X protein (hBax) is expressed from the circuit and apoptosis is induced. If the molecules are not detected, nothing happens.

The circuit senses microRNAs, and essentially functions as a “classifier” of a specific cell type, executing the pre-specified output of apoptosis. CancerNetwork spoke with Professor Ron Weiss of MIT about the development of this system. “He explained that the circuit is made up of three essential components: the sensing component, the computational component, and the actuation component.  The sensing component is comprised of several sensors that can detect levels of microRNA of interest in the cell; microRNA that has been identified as being indicative of the presence of cancer. The sensors then feed into the computational portion or the logic processing portion of the circuit. That portion of the circuit integrates multiple pieces of information from the sensors and makes the decision about whether the profile is indicative of the cancer that we hope to eliminate; it does this by looking at the sensor levels and computing “and” or “not” operations. The sensors are fed into this computational core, which then determines whether the levels of the sensors, looked upon as a group, are actually indicative of the cancer. Weiss went on to explain that "If the answer is yes,  then the computational core causes a production of a killer protein – the hBax protein – which ends up killing the cell.  If the answer is no then that protein is not expressed and eventually the genetic circuit just goes away."

The published research brings together two important concepts to create a system that has high utility. The first is tissue-specific signaling, which has previously been used to restrict, or at least partially restrict, a therapeutic agent’s action specifically to cancer cells. The second is the demonstration that multi-input information processing can work in living cells.

The researchers chose a set of low and high expression HeLa microRNAs whose expression is drastically different from a healthy, wild-type cell as markers. To decide which markers to use to differentiate cancerous from normal cells, the researchers created a mathematical model and experimentally derived responses of the individual sensors to their microRNA inputs. Multiple combinations of microRNAs were tested until an efficient output in HeLa cancerous cells, but not in normal cells was achieved.

The selective output of hBax expression followed by apoptosis in HeLa cells but not in the control normal cells was demonstrated by quantitating the level of apoptosis. This was done by using a fluorescent reporter assay; the protein was expressed when apoptosis is induced and this allowed a quantitative measure of apoptosis. Importantly, the system was demonstrated as being both specific and selective in a heterogenous cell population consisting of HeLa and non-cancerous, reference cells. A difference in the level of a single microRNA profile was enough to prevent the cell death output. However, further evolution and optimization of this circuit system is still necessary, to decrease false-positive cell detection and false-negative cell death in normal cells.

There are clearly therapeutic applications for this type of synthetic modulation of biological cell function. The new technology offers the possibility of designing a system to selectively induce cell death of any cancer type, without adverse effects on the surrounding healthy tissue. Because the circuit can be easily manipulated by adding any gene of interest, it may yield new treatments or diagnostics nobeyond cancer.

"This is a general technology for disease-state detection," says Professor Weiss. “We can basically target any biomarker. Certain existing therapies may utilize a small molecule that has to bind to a specific enzyme that's involved in the progression of cancer--it has to interfere with the progression of the cancer itself, whereas our system is actually a very different approach: We’re not interfering with the pathways that are responsible for the cancer, we're creating a circuit that reacts to the changes in the cell, and then we kill the cell,” professor Weiss explained.

The approach is not limited to targeting microRNA levels, which lends further flexibility to the potential uses of this type of system. However, the authors note that microRNAs make great markers and also because the data on micro RNA expression in cancer cells is readily available.

There are also potential in vitro applications of this type of system, including drug screening and the monitoring of developmental processes, as the authors suggest in the paper. The in vitro applications will likely be developed in a much shorter time frame compared to therapeutic applications.

The researchers are currently creating new circuits to identify other cell types, with the idea of moving this approach into a pre-clinical animal system. The original choice of using HeLa cells was based on the availability and ease of culture of these cells, and because this cell line is a “good testing ground”. They are also modulating the output. “In this case hBAX worked for us so we decided to use, it but we’re certainly exploring others, either separately or in conjunction with hBAX” Weiss said.

“We’re asking ourselves, how much we need to optimize this current circuit before we’re actually able to test this in a mouse model?” said Weiss. One of the current limitations is the efficient delivery of the DNA circuit to cells and the body. The groups is currently looking into different methods to package and deliver the DNA that makes up the circuit, using vesicles or a virus.

The ultimate goal is to advance this technology to therapeutic in-human trials. “We are planning animal testing in the near term," says Dr. Yaakov Benenson of Harvard. "We also need to solve the issue pertaining to efficient in vivo DNA delivery, as well perform additional circuit optimization that will be required for successful in vivo function. We have to wait for animal testing results to say anything about clinical potential."