Feedback regulation mode of gene circuits directly affects the detection range and sensitivity of lead and mercury microbial biosensors.

Whole cell biosensors offer high potential for the detection of heavy metals in a manner that is simple, rapid and low-cost. However, previous researchers have paid little attention to the impacts of construction models on the performance of these biosensors, thereby limiting the achievement of rational design and the optimization of detection characteristics. Herein, for the first time, three basic models of lead and mercury detection circuits, namely feedback coupled, uncoupled and semi-coupled models, have been constructed and compared to explore the effects of uncoupling the topology of sensing circuits on the reporter signals. The results demonstrated that the uncoupled model had better sensitivity for both lead (50 nM) and mercury (1 nM), while the feedback coupled circuits had a wider detection range for mercury (10 nM - 7.5 μM). Introducing the semi-coupled model into the comparison revealed that both the type and location of promoters for regulatory protein genes were key factors for sensitivity. Moreover, the detection characteristics of the uncoupled biosensors were robust, as conditions such as induction time, the concentration of microbial cells, and the concentration of antibiotics had little interference on the performance of the microbial biosensors. This study also established a novel and simple pre-treatment method for sample detection by biosensors. When the uncoupled microbial biosensor was put into practice, the concentration levels of mercury in milk and lead in sewage were determined quickly and accurately. Our study, therefore, provides a strategy for the rational design of whole cell heavy metal biosensors and has developed the potential of their application.