Scientists reported this week that they have developed a novel way to drive cancer cells to suicide: by delivering what they termed a "gene circuit" that sensed the levels of half a dozen miRNAs and, if they matched the expected profile for cancer cells, induced expression of a protein that drove the cells to apoptosis.
For now, the approach is nowhere near ready for clinical testing – it is not, in fact, even in animal models to date. The experiments were performed on HeLa and other cancer cell lines.
But the work is an example of how the field of synthetic biology is changing the very concept of what cells are, and what can be done with – or to – them.
Rather than thinking of cells as beings whose behavior can be influenced in only the crudest of ways, "we start to see a cell as a programmable entity," co-corresponding author Ron Weiss told BioWorld Today.
And co-corresponding author Yaakov Benenson elaborated that "we are trying to solve a very basic problem of detecting a cell's state, and doing something with this detection."
In general, Benenson told BioWorld Today, their "gene circuit" approach is a noninvasive way to see what's up – and down – inside cells.
When cells do change their state, this is most often not advertised by enormous changes in one or two biomarkers. Instead, "there are many subtle changes happening."
The approach, which Benenson, Weiss and their colleagues described in the Sept. 2, 2011, edition of Science, basically consists of delivering a set of sensors for the levels of several different biomarkers.
If the cell matches the template, the second part of the gene circuit is activated to induce apoptosis in the cell.
In their Science paper, the authors specifically described using a set of five markers to identify HeLa cells. Two of those markers occur at high levels in HeLa cells and three others at low levels. The circuits delivered to those cells were able to sense whether their template matched the pattern of RNAi levels they encountered. If the levels match the template, the gene circuit starts to produce the proapoptotic protein Bax, which ultimately induces the cell to kill itself.
Delivery takes the form of DNA plasmids that encode messenger RNAs as well as some proteins. Those sensors can tell the level of different mRNAs inside a cell – Benenson said the process builds heavily on the principles of RNA interference – and match those levels to a template. Benenson described the process "not unlike gene expression arrays, except that we are not destroying the sample in the process." For the time being, it also measures the levels of many fewer markers. The six-part circuit described in the Science paper is currently a fairly big one as synthetic circuits go.
But Weiss said that the capacity for making such circuits is increasing rapidly, in something of a Moore's law of synthetic biology: "There's a technology curve in synthetic biology that allows us to create more and more complex circuits."
One of the challenges for picking the biomarkers, Benenson said, was to find markers whose expression levels differ in normal and cancer cells not just at the cell population level, but in each individual member of the population. A miRNA that is three times higher in cancer cells because every third cell has nine times the normal amount (with two out of three having normal levels) would not make a useful sensor. Such data, Benenson said, are still hard to come by: There's "lots of ensemble data . . . [but] not much data available at the single cell resolution."
Benenson said that long-term, gene circuits could be used therapeutically; although the delivery challenges are considerable, gene delivery in cancer – and other indications is "not unheard of."
In the near term, though, in vitro applications such as drug screening may be the first out of the gate.
Benenson, Weiss and their team plan to work on those delivery challenges, and are working to take the approach in vivo. Benenson said that there is also "a lot of fine-tuning to be done from an engineering perspective" to make the sensors more precise.