Gene Circuits That Correct Crosstalk with Crosstalk

A major challenge to understanding natural regulatory networks and engineering synthetic biological circuits is crosstalk between pathways. A common approach in synthetic gene network design is to try to minimize crosstalk at the input stage and then process information with orthogonal Boolean logic. However, completely insulating biological signal-processing networks is challenging, especially as gene networks increase in size and complexity. Here, we show that analog gene circuits can accurately measure inputs despite unwanted crosstalk with other gene networks by incorporating circuits that compensate for the crosstalk. We demonstrate this principle by engineering gene circuits in Escherichia coli to sense reactive oxygen species (ROS). The initial ROS-sensing circuits exhibited unwanted crosstalk between two different inputs. We reduced this crosstalk via gene circuits that intentionally introduced counter-crosstalk, resulting in networks that were capable of discriminating between the analog concentrations of the distinct inputs. Thus, we show that by integrating signals instead of isolating signals from each other, gene networks can be engineered to buffer against crosstalk. We anticipate that this approach of correcting undesirable interactions with additional compensatory circuits can be generalized to design artificial gene networks with optimized behaviors, even in the context of complicated natural signaling networks. In addition, this strategy raises interesting hypotheses and avenues of investigation into how endogenous cellular networks may integrate signals and account for crosstalk to accurately assess their environments.