A Novel, Ultra-Sensitive Multiple Input-Output System for Target Identification in Systems Metabolic Engineering of E. coli
Identification of relevant engineering targets is one of the key challenges in metabolic engineering. To avoid laborious try-and-error experiments, many studies have been focused on building genome-scale models to make predictions. One of the most popular systems-based tools is the genome-scale metabolic reconstruction and flux analysis. The recent studies in kinetic modeling show many difficulties and are limited to small scale networks. Metabolic and kinetic models lack the crucial information of regulation and interactions which are essential for prediction and reprograming of cellular functions. Although cells are composed of molecules and their viability relies on extracting and using energy to maintain them, they are not ‘just’ matter and energy. Information processing, also called “cellular computing”, is essential for cellular function. Previous studies have proved that such computational abilities could be used rationally. It’s interesting to ask the question “Can we rein the computation abilities of cells for systems-level prediction and optimization of microorganisms?” If so, we then avoid of building laborious mathematic models since the perfect “model” is the cell itself. The key issue is how to let the cells “to compute” the processes we are interested in and “to output” the right results corresponding to the different inputs?
Here, we aimed at reining the computation abilities of E. coli cells by designing multiple input and output (IO) systems. The input system is implemented by using various M13 phage derivatives which can carry out up or down regulations targeting various genes. Using a rationally designed synthetic circle, the signal changes within the cells after gene operations introduced by phage infection are linked to the phage reproduction process, which in turn is linked to the total phage population. Thus, the various signals are ‘recorded’ in form of the population of corresponding phage derivatives. Biological systems are complex but highly adaptive, meaning that the cells always try to reduce the perturbations introduced. For the traditional methods, we always do genetic modification and pick the expected colonies before detecting the signal changes. During this process, the cells have already made adaption and the final signal strength is lower than that while cells are making adaption. With the current IO system, the recording process happens immediately after the perturbations introduced. Thus, the signal changes during the whole adaption process could be captured, making current system more sensitive than the traditional methods principally.
For proof of concept, various gene operations related or not related to lysine biosynthesis in E. coli were used as inputs and the intracellular lysine concentration changes were used as output signals. Correct predictions of beneficial genetic manipulations for enhanced lysine production in E. coli were obtained. The IO system shows ultra-sensitivity in capturing the signal changes caused by the perturbations introduced. For example, while using operations that reduce TCA cycle activity as inputs, slight increase of intracellular lysine concentration could be detected although the strict allosteric regulations of lysine production pathways by lysine still work. This system developed in this work opens up new possibilities for target identification in metabolic engineering.