
I recently learned that the original genetic toggle switch, one of the first synthetic gene circuits ever made, used perhaps the worst possible design. For context: In 2000, two papers appeared back-to-back in Nature: the repressilator and the toggle switch. Both described a synthetic gene circuit and, together, they helped launch the field of synthetic biology. The toggle switch was the simpler design; it had only two genes, one encoding LacI and the other TetR. Each repressed the other, creating two stable states. If you add a small molecule, called an inducer, you could switch cells between those two states. (Really useful! It meant scientists could "program" cells and then switch those cells between different states or behaviors using small molecules.) The researchers built this toggle switch by placing the two promoters next to each other, with transcription pointing outward, like this: ← Gene A | Gene B → This was sensible, because RNA polymerase sometimes blows past the terminator at the end of a gene, thus transcribing into the next gene, too. So their concern was that putting the two genes facing the same direction could cause transcriptional crosstalk. This is why they pointed them away from each other. And the circuit worked OK. Five of six versions they built were bistable, but the switching was extremely slow. It would take 30 minutes to switch to, say, the LacI state, but then multiple hours to coax cells back into the TetR state! So in 2017, Enoch Yeung et al. rebuilt the circuit in three different orientations: divergent (← →, the original), convergent (→ ←), and tandem (→ →). And the original design was the worst one! So much so that increasing the inducer concentration 200-fold only increased gene expression by ~2.7-fold. With the convergent design, on the other hand, an 8x increase in inducer caused a 120-fold increase in expression, meaning it had much faster switching. So what was happening? Well, the original paper was correct about transcriptional bleed-through, but they ignored DNA supercoiling. As RNA polymerase moves along DNA, it overwinds DNA ahead of itself and underwinds DNA behind it. This basically inhibits transcription at nearby promoters, meaning that the original paper had used a design in which, when they added inducer, they were unknowingly repressing both genes via physical changes in the DNA. It's interesting to think about how this gene circuit might not have worked at all; they're lucky, in a way, that it did. Itt also shows how complicated gene circuit design is. You have to think about terminator read-through, promoter strength, DNA supercoiling, and all sorts of interactions; both chemical and physical. Textbooks tend to draw genes as simple arrows in regulatory diagrams, but they're physical objects on a piece of DNA, and require a lot of "know-how" to design well.
























