According to Jim Collins, a synthetic biologist at Boston University who is exploring the use of transcription factors as key components in synthetic gene circuits, bioengineers come in two flavors.
“In their youth, those that take things apart become systems biologists, while those that put things together and tinker go into synthetic biology.” Collins finds himself in the latter class.
A physicist by training, Collins was led somewhat inadvertently into the world of synthetic biology by his BU colleagues. “It was suggested that I take my physics background and apply those skills to reverse engineer natural genetic networks,” recalls Collins. This was in 1996 and while microarray technology had been developed, there were still very few large-scale genetic datasets that could be used as a starting point in any reverse engineering effort. “In the end, we ran away from the problem.”
But not entirely. While Collins did not have the tools to take apart and reverse engineer naturally occurring networks, he and his graduate student at the time Tim Gardner came to realize that they did have the capabilities to assemble some basic molecular components, proteins, into genetic circuits that could then function in cells — akin in some ways to an electrical engineer wiring a light switch in a house.
In 2000, Collins and Gardner's forward engineering efforts paid off, resulting in a publication in Nature detailing the construction of a genetic toggle switch in Escherichia coli, the first synthetic, bistable gene-regulatory network to be described. That same issue of Nature also featured a report by Michael Elowitz and Stanislas Leibler describing a synthetic oscillating gene network in E. coli that periodically induced synthesis of green fluorescent protein as a readout of cell state. These two studies were the first demonstrations of the potential of engineering basic gene circuits, and with their publication, a new era of genetic circuit design in synthetic biology started.Lessons in complexity
Collins’ bistable genetic toggle was simple, a switch composed of two repressors and two promoters with each promoter inhibited by the repressor that is transcribed by the opposing promoter. But this simple toggle also turned out to be robust (i.e. exhibiting bistability over a range of parameters), a key property in circuit design. In fact, in the concluding paragraph of the article, Collins even makes mention of the idea that such toggles could find use in gene therapy and biotechnology applications in the future. As it turns out, in hindsight though, moving from this E. coli toggle toward more complex circuitry introduces new problems and challenges — as well as possibilities.
A genetic circuit needs a couple basic components to function properly. First, there needs to be a sensor module capable of identifying a specific input. From there, the sensor needs to be connected to a computational module. These modules, sometimesreferred to as logic gates, take the input and “calculate” the appropriate output. Logic gates can make basic decisions based on the input provided. For example, a NOT logic gate will result in one output over another while the AND logic gate will result in the same output. Finally, an output component is needed in the circuit to register the activation or repression of the circuit (for example, GFP was the output used in that early oscillator network study from Elowitz and Leibler). And that's not all, these circuit components should be both robust (exhibiting responses over a range of parameters like Collins’ bistable toggle) and sensitive, not to mention tunable or adaptable to new functions.
If it is not obvious by this point, creating a synthetic genetic circuit is no small task. The good news here is that parts or building blocks, basic circuit components, are available — the bad news is that components generated for other studies cannot simply be added to new circuit designs — this is often not a direct plug-and-play deal.
“If I purchased 1000 transistors, I would not test every transistor prior to use,” says Collins. “However, when it comes to synthetic biology, you do need to test every genetic component prior to use in a particular circuit.” Why? Synthetic gene circuits, unlike the electrical circuits in your house, operate in the context of a whole cell; they interact with other proteins and metabolites as well as the unique cellular environment. And this is the reason that increasing circuit complexity, well, increases circuit complexity.