Session U26: Physics in Synthetic Biology

Spatiotemporal dynamics in synthetic microbial consortia

Synthetic microbial consortia have an advantage over isogenic synthetic microbes because they can apportion biochemical and regulatory tasks among the strains. However, it is difficult to coordinate gene expression in spatially extended consortia because the range of signaling molecules is limited by diffusion. In this talk I will discuss how spatiotemporal coordination of gene expression can be achieved even when the spatial extent of the consortium is much greater than the diffusion distance of the signaling molecules. To do this, we examined the dynamics of a two-strain synthetic microbial consortium that generates coherent oscillations in small colonies. In large colonies, we found that temporally coordinated oscillations across the population depend on the presence of an intrinsic positive feedback loop that amplifies and propagates intercellular signals. These results demonstrate that synthetic multi-cellular systems can be engineered to exhibit coordinated gene expression using only transient, short-range coupling among constituent cells.
  

Synthetic genetic circuits using plant protoplasts

The complexity of plant biology has slowed the development of synthetic gene circuits in plants. However plant synthetic biology can form the basis for sustainable green technologies. The rational design of synthetic networks requires quantitative characterization of components for predictive modeling. This poses special difficulties for plants, since stably transforming plants is time consuming. We developed an experimental system for rapid quantitative measurements of synthetically designed repressors using plant protoplasts but found that protoplast assays show significant experimental batch effects that lead to incorrect quantitative results. With the help of a mathematical model coupled with stochastic simulations, we were successful in explaining and normalizing the batch effects to make quantitative comparisons between different inducible repressors, and approximately predict quantitative properties of synthetic circuits in stably transformed plants. We tested hundreds of repressible promoters, and carried out a statistical analysis of the quantitative data to uncover design principles for building synthetic inducible repressors in plants (Nature Methods, v13, pp94–100, (2016)). Our mathematical model may be broadly applicable to a wider class of cells and assays.   

Construction and Characterization of a Tunable Plasmid Copy Number System

Plasmids have evolved elegant mechanisms to ensure stable replication which balance the probability of plasmid loss with their metabolic burden on host cells. The pUC19 plasmid controls its copy number through a replication priming RNA (RNAII) and an inhibitory antisense RNA (RNAI). Through randomized mutagenesis of the promoters controlling these RNAs, we have created a library of over 1000 plasmids with copy numbers spanning a range greater than 2 orders of magnitude, allowing plasmid copy number to be used as a tunable parameter in several systems. We demonstrate the broad applicability of this system by using it to tune the behavior of simple genetic circuit elements and determine the fitness cost of green fluorescent protein and several CRISPR proteins. We additionally infer transcription rates for over 1000 different promoters of both RNAI and RNAII by combining our copy number measurements with the results of stochastic molecular simulations.   

Engineering Transcriptional Interference for Genetic Logic Gates

Transcriptional Interference (TI) is widespread in the genomes in all kingdoms of life and serves to regulate important cellular decisions. TI can occur through the collisions of RNA Polymerases (RNAPs) in tandem or in convergent orientation, and through the collision of RNAPs with protein roadblocks. Mathematical modeling and experiments have characterized several naturally occurring TI systems, but the design rules for constructing TI-based genetic devices are not yet well-defined. Here, we show that rationally controlling RNAP traffic on DNA through TI can lead to diverse gene expression profiles and facilitate the construction of TI-based logic gates. We demonstrate that tuning the dissociation constant of a protein roadblock enables optimized AND and OR logic gates, and that gate behavior can be predicted and validated through mathematical modeling. We then show that protein roadblocks coupled with RNAP collisions can produce NAND and NOR gates, and that the magnitude of gene repression from RNAP collisions can be tuned through inhibition of Rho-dependent transcription termination. These results expand the potential for TI as a novel tool for synthetic biology and offer insights into an important but relatively unexplored gene regulatory mechanism.   

Multicellular Drug Resistance from Synthetic Mitosis Control in Budding Yeast

Experimental modeling plays a vital role in understanding microbial drug resistance. Multicellular, biofilm- or clump-forming microbes encounter smaller local antibiotic concentrations and can survive harsher drug treatment than unicellular populations. However, how drug resistance depends on specific characteristics of multicellular populations, such as gene expression, and clump size distribution, is poorly understood. By deleting the AMN1 (antagonist of mitotic exit network) gene from multicellular S. cerevisiae, we obtained a drug-sensitive unicellular yeast strain. Assuming that AMN1 levels control clumping, we implemented a negative feedback-based gene circuit to control a synthetic copy of AMN1, mitotic progression, and consequently, clump sizes and drug resistance. This technology can be applied to extract parameters for computational models of multicellular drug resistance in yeast and has potential for testing the effect of novel drugs on a custom range of discrete multicellular yeast phenotypes.
  

FAST, Smart Therapeutics

The rapid rise of multidrug-resistant (MDR) superbugs and the declining antibiotic pipeline are serious challenges to global health. Rational design of therapeutics can accelerate development of effective therapies against MDR bacteria. In this talk, I will describe multi-pronged systems, synthetic biology, and nano-biotechnology based approaches being devised in our lab to rationally engineer therapeutics that can overcome antimicrobial resistance. We have developed Synthetic biology-based approach dubbed “Controlled Hindrance of Adaptation of OrganismS” or “CHAOS” to slow the evolution of antibiotic resistance by interfering with the processes involved in adaptive resistance. To translate our findings into the clinical setting, we have engineered antisense therapeutics that can block translation or increase transcription of any desired gene in a species-specific manner for targeted inhibition. Using this approach we have built a Facile Accelerated Specific Therapeutic (FAST) platform for the accelerated therapeutics in less than a week. Finally, I will also present a nano-biotechnology based approach involving development of a unique semiconductor material based quantum dot-antibiotic (QD ABx) which, when activated by stimuli, release reactive oxygen species to eliminate a broad range of MDR bacterial clinical isolates. The CHAOS, FAST and QD Abx platforms and inter-disciplinary approaches presented in this talk offer novel methods for rationally engineering new therapeutics to combat disease challenges.
  

The effect of time-dependent drive and delayed feedback loop in two-dimensional gene regulatory network

Recent advances in reprogramming a differentiated cell back to a pluripotent cell fate has rekindled interest in a quantitative understanding of the epigenetic landscape describing cellular differentiation. Reprogramming is a multistep process, involving multiple feedback loops. While the importance of feedback loops is well appreciated, most models assume instantaneous feedback, while biological feedback often involves a time delay between the signal and the response. In the present work, we propose a theoretical model based on a two-gene regulatory motif to investigate the role of time delay in the regulation of gene expression level. In particular, we focus on the interplay between time delayed feedback loops and time-dependent external chemical drive and their effect on dynamics. We observed that the concentration of the two transcription factors can undergo sustained oscillations and we speculate that this oscillatory state may provide an explanation of certain puzzling experiments on the reprogramming process. We also observe transdifferentiation-like behavior, where one differentiated state transitions to another without passing through an intermediate stem cell state.   

Synthetic gene circuits enable quantitative comparison of proteosynthesis rates in mammalian cells

Transcription and translation may vary significantly among mammalian cell types. Such differences may be important for comprehensively understanding cellular homeostasis and responses in various conditions. Here, we establish a synthetic biological framework to enable such comparisons. To minimize the random effects of genomic integration, we performed site-specific recombination （SSR）by applying FLPe recombinase-mediated cassette exchange (RMCE) to integrate single genes or gene circuits into the same “genomic safe harbor” locus AAVS1 in multiple mammalian cell lines. Upon establishing stable cell lines, we charted and compared the constitutive and inducer-dependent proteosynthesis levels using the eGFP reporter, and characterized the fold change, coefficient of variation (CV), expression kinetics based on the proliferation rate of each cell type. Overall, our study provided a systematic approach to quantify the variation of proteosynthesis due to the activity of the transcription and translation machinery in various mammalian cell types, allowing a better understanding of cell state diversity, which may be useful in mammalian biosystem modeling.
  

Synthetic Gene Circuits Reveal how KRAS(G12V) Affects Cell Proliferation & Migration Patterns

Small chemicals and light can serve as stimuli for controlling synthetic gene circuits to probe biological networks relevant to systems biology and biological physics. We previously developed Light-Inducible Tuner (LITer) gene circuits that have a wide dynamic range of gene expression, low gene expression noise, and an ability to respond with gene expression changes to both chemicals and light. As a foundation for exploring gene regulatory networks, we have created LITer variants controlling half-a-dozen functional genes. Here we present results from perturbing the levels of proto-oncogene KRAS(G12V). We explore a range of cellular features including cellular distance travelled, velocity, acceleration, proliferation, and invasion versus KRAS(G12V) levels. This study reveals that functional genes can have nontrivial effects at intermediate levels of activity, illustrating the utility of analog synthetic gene circuits to complement existing knock-out and overexpression genetic approaches. Furthermore, these tools may reveal quantitatively biological thresholds for functional genes to produce cellular phenotypes.   

Improved CRISPRi gene circuit function via antisense RNA sequestration

By using the binding of the catalytically-dead CRISPR protein dCas12a, we can create programmable gene circuit elements in E. coli. These ‘CRISPRgates’ are simple NOT gate elements which can target genes or other CRISPRgate elements and in principle can be combined to create complex genetic circuits, a fundamental goal of synthetic biology. While natural transcription factors have built-in advantages for transcriptional regulation (e.g. cooperativity), they are not programmable or individually orthogonal and a limited number of them exist. Repression with CRISPR is advantageous because we can in principle repress many different targets simultaneously without crosstalk. However, such gene circuit elements behave poorly when placed in series due to signal loss that occurs due to leaky repression (e.g. NOT NOT NOT != NOT) and retroactivity effects due to a shared pool of Cas proteins. By utilizing antisense RNAs to sequester guide RNA transcripts, we demonstrate a mechanism to suppress leaky CRISPRi repression and restore logical gene circuit function when elements are used in series. Further, we explore utilizing this system to enhance cooperativity in CRISPR-based gene circuits to create better dynamic gene circuit elements such as oscillators and toggle switches.   

Uncovering thermodynamic determinants of CRISPR-Cas gene circuit design.

The versatility of CRISPR-Cas endonucleases as a tool for synthetic biology has lead to diverse applications in gene editing, programmable transcriptional control, and nucleic acid detection. Most CRISPR-Cas systems, however, suffer from off-target effects and unpredictable non-specific binding that negatively impact their reliability and broader applicability. To better evaluate the impact of mismatches on DNA target recognition and binding, we develop a massively parallel CRISPR interference assay to measure the binding energy between tens of thousands of CRISPR RNA and target DNA sequences. By developing a general thermodynamic model of CRISPR-Cas binding dynamics, our results unravel a comprehensive map of the energetic landscape of CRISPR-Cas as it searches for its DNA target. Our generalizable approach provides a mechanistic understanding of target recognition and DNA binding by CRISPR-Cas variants, which should contribute to the advancement of recent synthetic biology efforts to repurpose dCas as gene circuit elements that behave orthogonally and operate independently without crosstalk.