New gene circuit design strategy to advance synthetic biology

Over the last 17 years, scientists and engineers have developed synthetic gene circuits that can program the functionality, performance and behaviour of living cells. Now, US researchers have developed an approach to gene circuit design that is far more efficient than current techniques.

Analogous to the integrated circuits that underlie electronics, engineered gene circuits can be used to generate defined dynamics, rewire endogenous networks, sense environmental stimuli and produce valuable biomolecules. They are said to hold great promise in medical and biotechnological applications, such as combating superbugs, producing advanced biofuels and manufacturing functional materials.

However, most circuits to date have been constructed through a trial-and-error manner, which relies heavily on a designer’s intuition and is often inefficient. As noted by Associate Professor Ting Lu, from the University of Illinois at Urbana-Champaign, “With the increase of circuit complexity, the lack of predictive design guidelines has become a major challenge in realising the potential of synthetic biology.”

Lu explained that typical models regard gene circuits as isolated entities that do not interact with their hosts and focus only on the biochemical processes within the circuits. He said, “Although highly valuable, the current modelling paradigm is often incapable of quantitatively, or even qualitatively sometimes, describing circuit behaviours. Increasing experimental evidences have suggested that circuits and their biological host are intimately connected and their coupling can impact circuit behaviours significantly.”

Lu and his colleagues addressed this challenge by constructing an integrated modelling framework for quantitatively describing and predicting gene circuit behaviours, with the results published in the journal Nature Microbiology. Using E. coli as a model host, the framework consists of a coarse-grained but mechanistic description of host physiology that involves dynamic resource partitioning, multilayered circuit-host coupling and a detailed kinetic module of exogenous circuits.

The team demonstrated that, following training, the framework was able to capture and predict a large set of experimental data concerning the host and simple gene overexpression. For instance, they discovered that ppGpp-mediated effects are the key to understanding constitutive gene expression under environmental changes, including both nutrient and antibiotic changes. The team also demonstrated the utility of the platform by applying it to examine a growth-modulating feedback circuit whose dynamics is qualitatively altered by circuit-host couplings and revealing the behaviours of a toggle switch across scales from single-cell dynamics to population structure and to spatial ecology.

Although Lu’s framework was established using E. coli as the model host, it has the potential to be generalised for describing multiple host organisms. “For example, we found that, by varying only a single parameter, the framework successfully predicted several key host metrics, including RNA-to-protein ratio, RNA contents per cell and mean peptide elongation rate, for Salmonella typhimurium and Streptomyces coelicolor,” said Lu.

According to Lu, this work advances the quantitative understanding of gene circuit behaviours and facilitates the transformation of gene network design from trial-and-error construction to rational forward engineering. By systematically illustrating key cellular processes and multilayered circuit-host interactions, it further sheds light on quantitative biology towards a better understanding of complex bacterial physiology.