Metabolism is precisely coordinated, with the goal of balancing fluxes to maintain robust growth. However, coordinating fluxes requires information about rates, which can only be inferred through concentrations. While flux-sensitive metabolites have been reported, the design principles underlying such sensing have not been clearly elucidated. Here we use kinetic modeling to show that substrate concentrations of thermodynamically constrained reactions reflect upstream flux and therefore carry information about rates. Then we use untargeted multi-omic data from Escherichia coli and Saccharomyces cerevisiae to show that the concentrations of some metabolites in central carbon metabolism reflect fluxes as a result of thermodynamic constraints. We then establish, using 37 real concentration-flux relationships across both organisms, that in vivo Delta G degrees >= 4 kJ/mol is the threshold above which substrates are likely to be sensitive to upstream flux(es). SIGNIFICANCE Some metabolites naturally have concentrations that are dependent on local metabolic fluxes. The relationship between fructose-1,6-bisphosphate (FBP) concentration and glycolytic flux is a well-characterized example of this phenomenon. These so-called flux sensors provide a mechanism for cells to measure fluxes and perform control action in response to perturbations. However, the underlying mechanism(s) that lead to such concentration-flux relationships is/are not understood, so engineers have no principles to guide their search for existing unknown flux sensors and/or to build flux-sensing controllers. This study outlines a thermodynamic constraint hypothesis explaining the emergence of flux sensors. Multi-omics data from both Escherichia coli and Saccharomyces cerevisiae are analyzed to show that natural concentration-flux relationships are associated with thermodynamically constrained reactions.

Small-molecule regulation modulates enzyme activity and is widespread in metabolic networks. However, the organization of small-molecule regulatory networks and its generalized role is not well understood. We analyze the structure of the genome-wide Escherichia coli small-molecule regulatory network (SMRN) to reveal that it optimizes controllability in the metabolic network. This is achieved by conserved, highly overabundant incoherent feedforward loops. Using multi-omics data, we characterize loop examples in central carbon metabolism. These use signals from hypothesized flux-sensing metabolites phosphoenolpyruvate, α -ketoglutarate, citrate, and malate to distinguish between glycolysis, gluconeogenesis, and glyoxylate shunt activity to differentially couple fluxes across these major modes of metabolism. Our results suggest that coupling of fluxes by direct modulation of enzyme activity is an emergent property of the SMRN that depends heavily on both regulatory structure and metabolic context via the metabolome, and further that flux sensing and coupling may be a global property of the metabolic network.

Metabolism is a precisely coordinated phenomenon, the apparent goal of which is to balance fluxes to maintain robust growth. However, coordinating fluxes requires information about rates, which is not obviously reconcilable with known regulatory mechanisms in which concentrations are sensed through metabolite binding. While flux sensor examples have been characterized, the fundamental principles underlying the phenomenon in general are not well understood. Specifically, the questions of which fluxes can be sensed, and the mechanism by which they are remain open. We address this by showing that the concentrations of substrates of thermodynamically constrained reactions reflect upstream flux and therefore carry information about rates which can be propagated through regulatory interactions to control other fluxes in the network. Using fluxomic, metabolomic, and thermodynamic data in E coli, we show that the concentrations of a few metabolites in central carbon metabolism reflect their producing fluxes and demonstrate that they can transmit information about these rates because of their positions in the network and their roles as effectors.

Itaconate (ITA) is an emerging powerhouse of innate immunity with therapeutic potential that is limited in its ability to be administered in a soluble form. We developed a library of polyester materials that incorporate ITA into polymer backbones resulting in materials with inherent immunoregulatory behavior. Harnessing hydrolytic degradation release from polyester backbones, ITA polymers resulted in the mechanism specific immunoregulatory properties on macrophage polarization in vitro. In a functional assay, the polymer-released ITA inhibited bacterial growth on acetate. Translation to an in vivo model of biomaterial associated inflammation, intraperitoneal injection of ITA polymers demonstrated a rapid resolution of inflammation in comparison to a control polymer silicone, demonstrating the value of sustained biomimetic presentation of ITA.

Dynamic control is a common approach to solve the tradeoff between productivity and yield that exists in engineered microbial metabolisms. Here we explore the possibility of implementing a dynamic control strategy based on direct activity modulation of a hypothetical optogenetic enzyme. With this system we sought to understand whether such a strategy is practical for controlling flux partitioning between biomass and production pathways, and whether it could be used to explore the effect of switching time on the performance of dynamic control strategies. We find that, while a protein-level control system is likely feasible in a model metabolism, several barriers to implementation and performance exist. Based on these limitations we suggest that careful balancing of protein expression at the biomass-production split node is required to implement such a system.

Abstract –As the bio-based economy expands, Chemical Engineering graduates will find themselves in new contexts for which they must be prepared. The broad shift toward including biology in departmental research and teaching activities reflects this, but relatively little formal thought has been given to the pedagogy of biology within Chemical Engineering curricula. The case study presented here is centered on the use of a biological control system in a lab setting as the means by which advanced control concepts can be taught to upper-year and graduate students within a constructivist framework. This approach was successfully applied to achieve all of the learning outcomes for the lab, but student feedback indicated that structured collaboration and metacognitive activities should have been given higher priority to improve student experiences. A re-iteration of this framework for upper-year lab curriculum design based on student feedback is presented.

Significant evidence suggests protein-level or metabolic control is widespread and important in metabolic networks. However, the biophysical interactions responsible for flux control at the metabolic level are not nearly as well-characterized as those which are responsible for control at other biological levels, such as transcriptional regulation. This knowledge gap is a limiting factor in the application of engineered protein-level regulation in Metabolic Engineering for the rational and sensitive control of pathway flux. Here we apply an in silico dynamic numerical optimization approach to a representative branched pathway to understand how engineered allosteric regulation could be used to control flux. We consider inhibition sensitivity as a hypothetical tunable parameter to demonstrate that integration of allosteric and transcriptional regulation is necessary to stably achieve arbitrary targets for both downstream metabolite concentrations. We further show that the steady-state ratio of these metabolites can be controlled by tuning the sensitivity of allostery at the branch point. Finally, we demonstrate that system dynamics dictate which type of engineered control is optimal. This work has implications for the co-optimization of transcriptional and allosteric regulatory systems in metabolic networks and provides a framework for the design of allosteric regulation in engineered metabolisms.

Accurate predictions of protein stability have great potential to accelerate progress in computational protein design, yet the correlation of predicted and experimentally determined stabilities remains a significant challenge. To address this problem, we have developed a computational framework based on negative multistate design in which sequence energy is evaluated in the context of both native and non-native backbone ensembles. This framework was validated experimentally with the design of ten variants of streptococcal protein G domain β1 that retained the wild-type fold, and showed a very strong correlation between predicted and experimental stabilities (R2 = 0.86). When applied to four different proteins spanning a range of fold types, similarly strong correlations were also obtained. Overall, the enhanced prediction accuracies afforded by this method pave the way for new strategies to facilitate the generation of proteins with novel functions by computational protein design.