Howard M. Salis

Automated design of synthetic ribosome binding sites to control protein expression

Microbial engineering often requires fine control over protein expression—for example, to connect genetic circuits or control flux through a metabolic pathway. To circumvent the need for trial and error optimization, we developed a predictive method for designing synthetic ribosome binding sites, enabling a rational control over the protein expression level. Experimental validation of >100 predictions in Escherichia coli showed that the method is accurate to within a factor of 2.3 over a range of 100,000-fold. The design method also correctly predicted that reusing identical ribosome binding site sequences in different genetic contexts can result in different protein expression levels. We demonstrate the methods utility by rationally optimizing protein expression to connect a genetic sensor to a synthetic circuit. The proposed forward engineering approach should accelerate the construction and systematic optimization of large genetic systems.

A Synthetic Genetic Edge Detection Program

Edge detection is a signal processing algorithm common in artificial intelligence and image recognition programs. We have constructed a genetically encoded edge detection algorithm that programs an isogenic community of E. coli to sense an image of light, communicate to identify the light-dark edges, and visually present the result of the computation. The algorithm is implemented using multiple genetic circuits. An engineered light sensor enables cells to distinguish between light and dark regions. In the dark, cells produce a diffusible chemical signal that diffuses into light regions. Genetic logic gates are used so that only cells that sense light and the diffusible signal produce a positive output. A mathematical model constructed from first principles and parameterized with experimental measurements of the component circuits predicts the performance of the complete program. Quantitatively accurate models will facilitate the engineering of more complex biological behaviors and inform bottom-up studies of natural genetic regulatory networks.

Translation rate is controlled by coupled trade-offs between site accessibility, selective RNA unfolding and sliding at upstream standby sites

The ribosome’s interactions with mRNA govern its translation rate and the effects of post-transcriptional regulation. Long, structured 5′ untranslated regions (5′ UTRs) are commonly found in bacterial mRNAs, though the physical mechanisms that determine how the ribosome binds these upstream regions remain poorly defined. Here, we systematically investigate the ribosome’s interactions with structured standby sites, upstream of Shine–Dalgarno sequences, and show that these interactions can modulate translation initiation rates by over 100-fold. We find that an mRNA’s translation initiation rate is controlled by the amount of single-stranded surface area, the partial unfolding of RNA structures to minimize the ribosome’s binding free energy penalty, the absence of cooperative binding and the potential for ribosomal sliding. We develop a biophysical model employing thermodynamic first principles and a four-parameter free energy model to accurately predict the ribosome’s translation initiation rates for 136 synthetic 5′ UTRs with large structures, diverse shapes and multiple standby site modules. The model predicts and experiments confirm that the ribosome can readily bind distant standby site modules that support high translation rates, providing a physical mechanism for observed context effects and long-range post-transcriptional regulation.

The ribosome binding site calculator.

The Ribosome Binding Site (RBS) Calculator is a design method for predicting and controlling translation initiation and protein expression in bacteria. The method can predict the rate of translation initiation for every start codon in an mRNA transcript. The method may also optimize a synthetic RBS sequence to achieve a targeted translation initiation rate. Using the RBS Calculator, a protein coding sequences translation rate may be rationally controlled across a 100,000+ fold range. We begin by providing an overview of the potential biotechnology applications of the RBS Calculator, including the optimization of synthetic metabolic pathways and genetic circuits. We then detail the definitions, methodologies, and algorithms behind the RBS Calculators thermodynamic model and optimization method. Finally, we outline a protocol for precisely measuring steady-state fluorescent protein expression levels. These methods and protocols provide a clear explanation of the RBS Calculator and its uses.

Efficient search, mapping, and optimization of multi‐protein genetic systems in diverse bacteria

Developing predictive models of multi‐protein genetic systems to understand and optimize their behavior remains a combinatorial challenge, particularly when measurement throughput is limited. We developed a computational approach to build predictive models and identify optimal sequences and expression levels, while circumventing combinatorial explosion. Maximally informative genetic system variants were first designed by the RBS Library Calculator, an algorithm to design sequences for efficiently searching a multi‐protein expression space across a > 10,000‐fold range with tailored search parameters and well‐predicted translation rates. We validated the algorithms predictions by characterizing 646 genetic system variants, encoded in plasmids and genomes, expressed in six gram‐positive and gram‐negative bacterial hosts. We then combined the search algorithm with system‐level kinetic modeling, requiring the construction and characterization of 73 variants to build a sequence‐expression‐activity map (SEAMAP) for a biosynthesis pathway. Using model predictions, we designed and characterized 47 additional pathway variants to navigate its activity space, find optimal expression regions with desired activity response curves, and relieve rate‐limiting steps in metabolism. Creating sequence‐expression‐activity maps accelerates the optimization of many protein systems and allows previous measurements to quantitatively inform future designs.

Automated physics-based design of synthetic riboswitches from diverse RNA aptamers

Riboswitches are shape-changing regulatory RNAs that bind chemicals and regulate gene expression, directly coupling sensing to cellular actuation. However, it remains unclear how their sequence controls the physics of riboswitch switching and activation, particularly when changing the ligand-binding aptamer domain. We report the development of a statistical thermodynamic model that predicts the sequence-structure-function relationship for translation-regulating riboswitches that activate gene expression, characterized inside cells and within cell-free transcription–translation assays. Using the model, we carried out automated computational design of 62 synthetic riboswitches that used six different RNA aptamers to sense diverse chemicals (theophylline, tetramethylrosamine, fluoride, dopamine, thyroxine, 2,4-dinitrotoluene) and activated gene expression by up to 383-fold. The model explains how aptamer structure, ligand affinity, switching free energy and macromolecular crowding collectively control riboswitch activation. Our model-based approach for engineering riboswitches quantitatively confirms several physical mechanisms governing ligand-induced RNA shape-change and enables the development of cell-free and bacterial sensors for diverse applications.

Kinetic Buffering of Cross Talk between Bacterial Two-Component Sensors

Two-component systems are a class of sensors that enable bacteria to respond to environmental and cell-state signals. The canonical system consists of a membrane-bound sensor histidine kinase that autophosphorylates in response to a signal and transfers the phosphate to an intracellular response regulator. Bacteria typically have dozens of two-component systems. The key questions are whether these systems are linear and, if they are, how cross talk between systems is buffered. In this work, we studied the EnvZ/OmpR and CpxA/CpxR systems from Escherichia coli, which have been shown previously to exhibit slow cross talk in vitro. Using in vitro radiolabeling and a rapid quenched-flow apparatus, we experimentally measured 10 biochemical parameters capturing the cognate and non-cognate phosphotransfer reactions between the systems. These data were used to parameterize a mathematical model that was used to predict how cross talk is affected as different genes are knocked out. It was predicted that significant cross talk between EnvZ and CpxR only occurs for the triple mutant DeltaompR DeltacpxA DeltaactA-pta. All seven combinations of these knockouts were made to test this prediction and only the triple mutant demonstrated significant cross talk, where the cpxP promoter was induced 280-fold upon the activation of EnvZ. Furthermore, the behavior of the other knockouts agrees with the model predictions. These results support a kinetic model of buffering where both the cognate bifunctional phosphatase activity and the competition between regulator proteins for phosphate prevent cross talk in vivo.

Rational design of a synthetic Entner-Doudoroff pathway for improved and controllable NADPH regeneration

NADPH is an essential cofactor for the biosynthesis of several high-value chemicals, including isoprenoids, fatty acid-based fuels, and biopolymers. Tunable control over all potentially rate-limiting steps, including the NADPH regeneration rate, is crucial to maximizing production titers. We have rationally engineered a synthetic version of the Entner-Doudoroff pathway from Zymomonas mobilis that increased the NADPH regeneration rate in Escherichia coli MG1655 by 25-fold. To do this, we combined systematic design rules, biophysical models, and computational optimization to design synthetic bacterial operons expressing the 5-enzyme pathway, while eliminating undesired genetic elements for maximum expression control. NADPH regeneration rates from genome-integrated pathways were estimated using a NADPH-binding fluorescent reporter and by the productivity of a NADPH-dependent terpenoid biosynthesis pathway. We designed and constructed improved pathway variants by employing the RBS Library Calculator to efficiently search the 5-dimensional enzyme expression space and by performing 40 cycles of MAGE for site-directed genome mutagenesis. 624 pathway variants were screened using a NADPH-dependent blue fluorescent protein, and 22 were further characterized to determine the relationship between enzyme expression levels and NADPH regeneration rates. The best variant exhibited 25-fold higher normalized mBFP levels when compared to wild-type strain. Combining the synthetic Entner-Doudoroff pathway with an optimized terpenoid pathway further increased the terpenoid titer by 97%.

Induction and Relaxation Dynamics of the Regulatory Network Controlling the Type III Secretion System Encoded within Salmonella Pathogenicity Island 1

Bacterial pathogenesis requires the precise spatial and temporal control of gene expression, the dynamics of which are controlled by regulatory networks. A network encoded within Salmonella Pathogenicity Island 1 controls the expression of a type III protein secretion system involved in the invasion of host cells. The dynamics of this network are measured in single cells using promoter-green fluorescent protein (gfp) reporters and flow cytometry. During induction, there is a temporal order of gene expression, with transcriptional inputs turning on first, followed by structural and effector genes. The promoters show varying stochastic properties, where graded inputs are converted into all-or-none and hybrid responses. The relaxation dynamics are measured by shifting cells from inducing to noninducing conditions and by measuring fluorescence decay. The gfp expressed from promoters controlling the transcriptional inputs (hilC and hilD) and structural genes (prgH) decay exponentially, with a characteristic time of 50-55 min. In contrast, the gfp expressed from a promoter controlling the expression of effectors (sicA) persists for 110+/-9 min. This promoter is controlled by a genetic circuit, formed by a transcription factor (InvF), a chaperone (SicA), and a secreted protein (SipC), that regulates effector expression in response to the secretion capacity of the cell. A mathematical model of this circuit demonstrates that the delay is due to a split positive feedback loop. This model is tested in a DeltasicA knockout strain, where sicA is complemented with and without the feedback loop. The delay is eliminated when the feedback loop is deleted. Furthermore, a robustness analysis of the model predicts that the delay time can be tuned by changing the affinity of SicA:InvF multimers for an operator in the sicA promoter. This prediction is used to construct a targeted library, which contains mutants with both longer and shorter delays. This combination of theory and experiments provides a platform for predicting how genetic perturbations lead to changes in the global dynamics of a regulatory network.

A predictive biophysical model of translational coupling to coordinate and control protein expression in bacterial operons

Natural and engineered genetic systems require the coordinated expression of proteins. In bacteria, translational coupling provides a genetically encoded mechanism to control expression level ratios within multi-cistronic operons. We have developed a sequence-to-function biophysical model of translational coupling to predict expression level ratios in natural operons and to design synthetic operons with desired expression level ratios. To quantitatively measure ribosome re-initiation rates, we designed and characterized 22 bi-cistronic operon variants with systematically modified intergenic distances and upstream translation rates. We then derived a thermodynamic free energy model to calculate de novo initiation rates as a result of ribosome-assisted unfolding of intergenic RNA structures. The complete biophysical model has only five free parameters, but was able to accurately predict downstream translation rates for 120 synthetic bi-cistronic and tri-cistronic operons with rationally designed intergenic regions and systematically increased upstream translation rates. The biophysical model also accurately predicted the translation rates of the nine protein atp operon, compared to ribosome profiling measurements. Altogether, the biophysical model quantitatively predicts how translational coupling controls protein expression levels in synthetic and natural bacterial operons, providing a deeper understanding of an important post-transcriptional regulatory mechanism and offering the ability to rationally engineer operons with desired behaviors.