LBNC | Laboratory of Biological Network Characterization

Program

Page banner

Abstract Booklet

Click here to download the abstract booklet (PDF).

Participants List

Click here to download the participants list (XLSX).

Invited Talks


Esther Amstad, EPFL

On-chip mixing and encapsulation of lysates

The recent progress achieved in the cell free synthetic biology opens up exciting possibilities to design artificial cells. To enable performing such reactions in volumes similar to that of a cell these reagents must be loaded into pl-sized reaction vessels. Micrometer-sized drops are attractive containers that mimic the cell size. Such drops of a well-defined size can be conveniently assembled using microfluidics. However, the loss of fluids during the start-up of these devices makes the use of this technique for the encapsulation of lysates expensive. To overcome this shortcoming and to enable rapid screening of synthesis conditions, we developed new microfluidic drop makers that contain pressure valves. These devices allow encapsulation of volumes as low as 10 µl without any reagent loss. We employ these devices to produce emulsion drops loaded with different amounts of lysates and analyze the rate and quantities of proteins synthesized in 40 µm diameter drops under different conditions.


Roy Bar-Ziv, Weizmann Institute of Science

Programmable On-Chip DNA Compartments as ‘Artificial Cells’

The assembly of artificial cells capable of executing DNA programs has been an important goal for basic research and technology. We assemble 2D DNA compartments fabricated in silicon as ‘artificial cells’ capable of metabolism, programmable protein synthesis, and communication. We programmed gene expression cycles in separate compartments, as well as protein synthesis fronts propagating in a coupled 1D system of compartments. Gene expression in the DNA compartments reveals a rich, dynamic system that is controlled by geometry. The organization of matter in the compartment suggests conditions for controlled assembly of biological machines. This puts forth a man-made biological system with programmable information processing from the gene to a ‘cell’, and up to the ‘multicellular’ scale.


James U. Bowie, UCLA

Synthetic Biochemistry: Making Biofuels and Commodity Chemicals the Cell-Free Way

Considerable effort is currently directed to engineer micro-organisms to produce useful chemicals. The greatest potential environmental benefit of metabolic engineering will be the production of high volume commodity chemicals, such as biofuels. Yet the high yields required for the economic viability of low-value chemicals are particularly hard to achieve in microbes due to the myriad competing biochemical pathways. We are developing an alternative approach, which we call synthetic biochemistry. Synthetic biochemistry throws away the cells and builds biochemical pathways in reaction vessels using complex mixtures of isolated enzymes. As the only pathway in the vessel is the desired transformation, yields can approach 100%. The challenge for synthetic biochemistry is to replace the complex regulatory systems that exist in cells in a simplified form. We are designing and testing various ideas for building highly robust systems that can operate continuously for long periods of time.


Bruno Correia, EPFL

Computational Design of Functional Proteins for Biomedicine

Finely orchestrated protein activities are at the heart of the most fundamental cellular processes. The rational and structure-based design of novel functional proteins holds the promise to revolutionize many important aspects in biology, medicine and biotechnology. Computational protein design has led the way on rational protein engineering, however many of these designed proteins were solely focused on structural accuracy and completely impaired of function. I will describe a computational design strategy centered on the exploration of de novo protein topologies and the use of structural flexibility with the ultimate goal of designing functional proteins. This approach aims to solve a prevalent problem in computational design that relates to the lack of optimal design templates for the optimization of function. By expanding beyond the known protein structural space, our approaches represent new paradigms on the design of de novo functional proteins.


Tom de Greef, TU Eindhoven

Engineering Bioinspired Molecular Networks

Complex signalling networks enable living cells to process information from their environment using an intricate network of regulatory interactions. These biochemical circuits function by converting an input signal (stimulus) through spatiotemporal interplay of signalling molecules (transduction) to an output response (function). Inspired by biology, we engineer a wide range of minimalistic, artificial signalling circuits by employing a cell-free bottom-up strategy. Such simplified model systems composed of fewer species each with well-defined interactions could help isolate key molecular parameters and thus have the potential to uncover generalizable concepts. Examples will include autocatalytic, bistable and oscillatory systems.


Elisa Franco, UC Riverside

Programming circuits and materials with nucleic acids

Cells have unique abilities to sense, process, and actuate based on environmental stimuli: their molecular components are constantly running many parallel programs that ensure correct growth, motion, reshaping, and repair in response to external inputs. How can we harness such powerful toolkit of DNA, RNA, and proteins to create the next generation of molecular computers and smart biomaterials? I will describe our work in this area, which is centered on the combination of nucleic acids nanotechnology and dynamical systems theory. First, I will summarize our efforts in the design and synthesis of synthetic molecular clocks, essential devices to synchronize biochemical events. Specifically, I will describe the challenges arising in scaling up clock-driven circuits. Second, I will outline our progress in the creation of responsive biomaterials using DNA nanostructures powered by dynamic reaction networks.


Paul Freemont, Imperial College London

Exploring the potential of cell free system from different organisms: from prototyping to biosynthetic pathway discovery

Cell-free transcription/translation systems (known as CFPS or TX-TL) have recently been re-evaluated as a promising platform for enabling synthetic biology research and applications. In particular TX-TL has been shown to provide a reproducible protoyping platform for regulatory elements where measurements in vitro are in part consistent with similar measurements in vivo. The advantage of being non-GMO allows rapid automated assays for characterizing parts and genetic circuit designs for pathway engineering. My lab has been interested in exploring cell free extracts from different organisms and I will present our recent work on cell-free systems for Bacillus Subtilis Bacillus Megatarium andStreptomyces venezuelae and how general in vitro approaches can inform biosynthetic pathway design. I will also describe how cell free systems can be used as a platform for biosensor applications and will present our most recent work on developing a biosensor to detect bacterial infections in human respiratory samples.


Daniel Gibson, J. Craig Venter Institute / Synthetic Genomics

Design and synthesis of cells from the bottom up

The American physicist Richard Feynman once said, ‘What I cannot create, I do not understand’. With that inspiration, we set out to attain a deeper understanding of life by assembling it ourselves. Over the past 15 years, our teams at the J. Craig Venter Institute (JCVI) and Synthetic Genomics, Inc. (SGI) have been developing tools to design whole genomes, synthesize and assemble them in the lab, and install them into a living cell. Our goal was not just to elucidate the genetic components required for life, but also to establish the capacity to create organisms custom-tailored to specific industrial applications. Progress towards these goals will be discussed.


Jorgen Kjems, J. Aarhus University

Self-assembled nucleic acid-protein-lipid devices for biosensing, bioimaging and delivery of biologics

The application of self-assembled chemically modified nucleic acid structures driven by predictable Watson and Crick base pairing has enabled construction of highly defined nanodevices that in a robotic fashion can sense the environment, compute the signal and adjust a functional responds. Spatial nanometer control of oligonucleotides fused to peptides, sugars, lipids or small molecules has enabled us to study multivalency and pattern recognition for improved delivery. In particular, we have constructed and studied self-assembled structures capable of sensing complex mixtures of microRNA, delivering functional nucleic acids to specific cells, and compartments and control activity of enzymes. Three projects will be presented: Using the DNA-origami technique we have created a 35 nm nanopore with an inner pore diameter of 10 nm and with 3 dynamic flaps at the sides that can be unlocked in response to external signals, revealing hidden hydrophopic groups. The pore thereby becomes addressable to specified liposomes or cells and will allow multiplexed sensing of single biomolecules in solution.
Also, using the DNA origami technique we have demonstrated specific external control of single enzyme activity by shielding the enzyme in a dynamic cage in order to gate its access to free substrate molecules.
Hybridization of nucleic acids can also direct assembly of multifunctional drug delivery devices in Lego brick fashion. To achieve clinical relevance in animals and humans we have introduced chemical modifications to improved stability and pharmacokinetics. Based on these principles we have designed and successfully demonstrated improved targeted delivery of biologics to organs and tumors and diseased tissue.
Combining some of these devices opens up for “intelligent” diagnostic robots that for instance can diagnose complex diseases at cellular level and impose a tailor made treatment at the cellular level.


Heinz Koeppl, TU Darmstadt

Modeling and characterization of cell-free implementations of RNA-based logic gates

RNAs are involved in a wide range of cellular regulatory functions and were recently shown to be a promising tool for controlling gene expression in synthetic circuits. Escherichia coli-based cell-free transcription translation (TX-TL) systems offers an appealing and robust alternative environment to the traditional lengthy in vivo test cycles for implementing such synthetic circuits of increasing complexity. We show first that certain parts can be studied and decoupled in a TX-only environment. Here, we use this method for demonstrating the use of a dual system mRNA reporter, allowing our subsequent RNA circuits to be fluorescently tracked with the iSpinach along with the Malachite green aptamer. We also investigate the use of the well characterized theophylline riboswitch in a translational regulated cascade. A down-scaled version of a microfluidic chemostat based on previously published designs are used for applying different signal patterning over a sustained period of time. The reactions are thus performed in continuous mode, and a systematic comparison is made against their counterpart ran in batch from circular or linear templates. Finally, the challenges of integrating other RNA regulators such as Small Transcripton Activators or Repressors (STAR) RNA to such synthetic circuits are presented and strategies to measure their combinatorial effects through the TX-TL environment are discussed. We use a derivative matching approach based on Gaussian processes and Bayesian inference to calibrate ODE models to the measure cell-free circuit dynamics.


Tanja Kortemme, UCSF

Computational design of reprogrammed and new protein functions

There has been exciting progress in the computational design of proteins with new structures and functions, highlighting the potential to advance many applications in biological engineering, as well as to provide insights into the design principles of function. Many significant challenges remain, both in the accuracy of current computational approaches, and in the complexity of functions that can be designed at present. I will discuss our recent progress with computational methods and describe new approaches and their applications. Most recently, we utilized computational design to engineer new small-molecule binding sites into protein-protein interfaces. The designed proteins function as sensor/actuators that detect and respond to new small molecule signals in living cells. Cell-free systems provide a valuable platform to prototype these designed sensor / actuators and to assess their modularity, i.e. the ability to connect small molecule signal inputs to different output responses.


Sebastian Maerkl, EPFL

Microfluidic cell-free systems for the rapid implementation and characterization of genetic networks

Cell-free synthetic biology emerged recently as a viable in vitro alternative for biological network engineering. We recently implemented cell-free systems on a microfluidic microchemostat array, which performs transcription-translation reactions at steady state for 30+hours, allowing us to run and characterize complex genetic networks in vitro. We successfully transferred these networks to a cellular host showing that the in vitro system emulated the cellular host quite well. Two three-node genetic oscillators designed and tested as part of this work showed surprising robustness and high “Q factors” in vivo, indicating that network robustness is not solely defined by network topology. Our current work revolves around the design and characterization of Zn-finger transcription factors and their use as repressors in cell-free systems, having the potential to provide a resource of novel transcription regulators for constructing increasingly complex genetic networks. To keep up with the ever-increasing demand for higher throughput we are also currently developing a microfluidic platform capable of characterizing hundreds of parts and devices in vitro. We expect that this work will form part of the technological and biological foundation required for the creation of artificial cells or cell-like entities as well as provide a simpler system than cells to study complex regulatory mechanisms in what one could think of as “systems biochemistry”.


Igor Medintz, U.S. Naval Research Lab

Catalytic enhancement of enzyme cascades co-localized on colloidal nanoparticles

Enzymes and especially multienzyme pathways are of tremendous interest for the production of industrial chemicals and in the development of metabolic sensors. One primary focus of synthetic biology is to design enzymatic production capabilities in a “plug and play” format within cellular systems. Living cellular systems, however, can suffer from toxicity, competing pathways and sometimes an inability to mix enzymes from different species. Application of enzymes for industrial catalysis is often achieved by immobilization on a surface since this often provides stability and facilitates purification and reuse of the enzymes from the reaction mixture. Unfortunately, immobilization of enzymes on large planar surfaces often results in loss of enzymatic activity. We seek to create cell-free enzyme systems that can circumvent these issues in a “plug and play” format where enzymes are assembled on nanoparticle surfaces but still overcome diffusion and stability issues. We and others have demonstrated that immobilization of enzymes on nanoparticles often results in enhanced enzymatic activity relative to the free enzyme in solution. Our recent studies have involved the immobilization of multienzyme cascades on nanoparticle surfaces in an attempt to harness the enhanced activity of nanoparticle-bound enzymes and potentially access substrate channeling phenomena. We have demonstrated the ability of nanoparticle immobilization to greatly enhance the kinetics and product yield of multiple pathways containing up to 7 enzymes. Additionally, we observe enhanced kinetics from single bound enzymes (>50 fold increases) within the context of the immobilized pathways and use mechanistic analyses to uncover the mechanism of these enhancements. Furthermore, we are able to provide strong evidence that the pathway enzymes are stabilized by nanoparticle binding and able to effectively "channel" substrates between the co-localized enzymes.


Richard M. Murray, California Institute of Technology

Towards Genetically-Programmed Artificial Cells and Multi-Cellular Machines

By leveraging work in the synthetic biology and molecular programming communities over the past decade, we are plausibly within 10-15 years of being able to produce genetically-programmed artificial cells and multi-cellular machines that can carry out useful engineering operations. Applications range from synthesis of bio-compatible materials, to environmental monitoring and remediation, to self-assembly of complex multi-cellular machines that mimic some capabilities of natural systems. Pursuing this vision will require new approaches to biomolecular systems engineering, focused on moving from creation and characterization of devices and simple circuits to systematic specification, design, integration, and verification of circuits, subsystems, cells, and multi-component systems. In this talk I will describe some of the work on individual technologies that might be used in designing and implementing artificial cells, as well as some of the challenges that remain (and possible approaches to address them).


Vincent Noireaux, University of Minnesota

Quantitative biology with a cell-free TXTL toolbox

The analysis and characterization of complex dynamical interactions involved in gene regulation is a major theme in post-genomic research. Novel multidisciplinary approaches are being developed to dissect gene networks and to determine their basic principles. In this talk I will present an all E. coli cell-free system that my lab has engineered to construct and characterize information-based dynamical processes by executing gene circuits in vitro. This cell-free platform is unique at three levels. First, it is one the most powerful E. coli expression systems on the market, with up to 2 mg/ml of protein synthesized in batch mode. Second, the transcription, based on the entire E. coli sigma factor family, is currently the largest for in vitro protein systems. Third, this platform was engineered so as to be highly versatile and modular, as demonstrated, for example, by our recent CRISPR extension. This system is now used in many applications that I will discuss in my talk, ranging from prototyping single regulatory elements to the complete synthesis of phages and the assembly of synthetic cell analogs.


Sven Panke, ETHZ

The perks of cell free – forward design of a complex in vitro system

Multienzyme systems are an attractive way to conduct complex multistep chemistry in one vessel, potentially facilitating operations and allowing overcoming thermodynamic obstacles by coupling to cofactor-dependent reactions. However, the non-linear dynamics of enzymes and the notoriously unreliable determinations of enzyme constants under process reaction conditions makes the behavior of such systems difficult to predict. We have developed an on-line MS method which allows measuring continuously the concentrations of multiple compounds in the effluent of a continuously operated enzyme membrane reactor. The enzyme system in the reactor can be challenged with different perturbation functions and the response of the system accurately recorded and then be used to parameterize an ODE-based model of the reaction system.


Keith Pardee, University of Toronto

Rapid, Low-cost Tools for Global Health: Using Cell-free Synthetic Biology for Diagnostics and the Portable Manufacture of Therapeutics.

The recent Zika virus outbreak highlights the need for low-cost diagnostics that can be rapidly developed for distribution and use in pandemic regions. In early 2016 we developed a pipeline for the rapid design, assembly, and validation of cell-free, paper-based sensors for the detection of the Zika virus RNA genome. This work was built upon a paper-based system that we originally published in 2014 that used freeze-dried cell-free reactions to deploy synthetic gene networks outside of the lab in a sterile and abiotic format. By linking isothermal RNA amplification to toehold switch RNA sensors in this paper-based system, we were able to detect clinically relevant concentrations of Zika virus sequences, discriminate between viral strains and detect Zika virus from the plasma of a viremic macaque.

We have also recently extended our cell-free approach to the portable manufacture of therapeutics. Using pellets of freeze-dried cell-free reactions we demonstrated the synthesis over 50 products. This included the manufacture and functional validation of antimicrobial peptides, vaccines, antibody conjugates and small molecules. Our freeze-dried biomolecular platform resolves important practical limitations to the deployment of molecular diagnostics and protein-based therapeutics to the field, and demonstrates how cell-free synthetic biology can be used to develop tools for confronting global health crises.


Yannick Rondelez, CNRS

Localizing synthetic dissipative DNA circuits.

It is possible to create test-tube simplified analogs of genetic circuits that rely on DNA-base pairing rules to enforce network topology, and a few purified enzymes as a dissipative (energy consuming) machinery. These artificial reaction networks can implement precisely programmed chemical dynamics in a well mixed closed reactor.

In this presentation, I will discuss our recent attempts at using such a synthetic molecular programming approach for the creation of synthetic communities, loosely inspired from embryogenesis or cooperating insects. This approach uses tethered DNA strands to program the local chemical behavior of microscopic solid “agents” distributed in a feeding solution. The DNA-programmed agents can sense the behavior of neighboring agents (through multiple orthogonal chemical communication channels), and use these stimuli to decide their own action. I will present collective behaviors involving thousands of agents of various types, for example retrieving information over long distances, or creating spatial patterns.


Yolanda Schaerli, UNIL

Synthetic gene regulatory networks to study evolutionary constraints

During development of multicellular organisms gene regulatory networks are crucial for patterning bodies. Comparisons of related species indicate that the same phenotype can be the result of quite different networks and mechanisms. With the recent advent in synthetic biology, we are no longer limited to perturbation of gene regulatory networks in a small number of model organisms. Instead, we can now build and study gene regulatory networks at will to improve our understanding of their mechanisms, properties and evolution. Synthetic biology also allows us to test different solutions (i.e., genotypes) for a phenotype of interest and compare them to each other. We previously explored the design space of 3-node networks for stripe formation in a morphogen gradient and built four synthetic networks displaying the same phenotype, but functioning with different dynamical mechanisms (Schaerli et al, 2014, Nat. Commun. 5:4905). In two of these circuits we now introduced random mutations into the regulatory regions and analysed the resulting phenotypes. We find that the phenotypes obtained are different for the two networks and can be explained by their underlying dynamical mechanisms. We thus provide empirical evidence that the dynamical mechanism of a gene regulatory network biases which novel phenotypes are accessible through mutations.


Rebecca Schulman, Johns Hopkins University

Far-From Equilibrium DNA Strand-Displacement Circuits for Controlling Materials

DNA strand displacement circuits enable scalable computation and manipulation under a wide variety of temperatures and chemical buffers. While DNA strand displacement circuits can autocatalytically amplify a signal, perform complex logic and emulate complex dynamics such as oscillations, they are traditionally used in a closed system, where they are capable of only a single calculation or preprogramed response to an input trigger. We have developed a “chemical battery,” which couples these circuits to a potential, enabling them to turn over and drive dynamic processes in materials. I’ll discuss these circuits and demonstrate their function in material systems.


Petra Schwille, Max Planck Institute of Biochemistry

Reconstituting a minimal divisome in cell-like compartments

The ability to self-replicate constitutes one of the most fundamental features of life. Although much progress has been made with respect to identifying minimal self-replicating information units based on nucleic acids, the faithful division of compartments that could serve as minimal protocells is still a great challenge. In the quest for a minimal protein-based machinery that could accomplish the self-organized division of a membraneous compartment, we assembled key components of the bacterial (E.coli) divisome in membrane-clad microcompartments and aqueous droplets enclosed by lipid monolayers. We show that the spatial cue for aligning a division ring can be established by oscllations and gradient formation of the MInCDE positioning machinery. The tubulin homologue FtsZ, main component of the bacterial Z ring, spontaneously aligns itself into dynamic protofilaments perpendicular to the long axis of a rod-like membrane compartment and assumes a ring shape at spatial scales that resemble the bacterial geometry. I will present the recent accomplishments and discuss remaining challenges on the way towards self-replicating protocells.


Georg Seelig, University of Washington

DNA strand displacement-systems for disease diagnostics

In this talk, I will cover two complementary approaches to using DNA strand displacement for disease diagnostics. First, I will propose a novel method for gene expression profiling that employs DNA-based molecular computation to classify samples with varying gene expression patterns. This approach results in a rapid, cost-efficient and reliable alternative to existing gene expression diagnostic methods. We demonstrate a systematic methodology for translating the architecture of a single-layer neural network for gene expression classification into a DNA-based molecular device. Classification takes place at the molecular level without the need of measuring each gene transcript. We demonstrate that our approach is generalizable to any gene expression pattern and it is potentially scalable to classification problems with tens or hundreds of gene transcripts. Second, I will show how we can use model-guided reaction pathway engineering to quantitatively improve the performance of selective hybridization probes in recognizing single nucleotide variants (SNVs). Specifically, we build a detection system that combines discrimination by competition with DNA strand displacement-based catalytic amplification. I will show, both mathematically and experimentally, that the single nucleotide selectivity of such a system in binding to single-stranded DNA and RNA is quadratically better than discrimination due to competitive hybridization alone.


Friedrich Simmel, TU München

Cell-free operation of RNA circuits

RNA-based gene regulatory mechanisms hold great promise for the rational design of gene circuits. RNA regulators can act both on the transcriptional and translational level and can thus either be used to control the production of other RNA molecules or of proteins. Correspondingly, the “wiring” of RNA-based gene circuits is relatively straighforward and “sequence-programmable”. Additional elements such as riboswitches can act as sensors and thus allow the modulation of gene circuit behavior using externally supplied chemical signals. In this talk, a variety of RNA regulatory circuits will be presented, which utilize simple transcriptional switches termed “genelets”, the CRISPR/dCas9 interference mechanism as well as toehold riboregulators. Their operation in a cell-free context will be discussed and compared with their performance in bacteria.


Gasper Tkacik, IST Austria

Biased partitioning generates large and long-lived phenotypic heterogeneity in Escherichia coli

Molecular mechanisms underlying phenotypic variation in isogenic bacterial populations remain poorly understood. Using a microfluidic platform and real-time observations at the single cell level, we studied the behavior of AcrAB-TolC, the main multi-drug efflux pump of Escherichia coli. We observed a strong partitioning bias of the pump for old cell poles. Mother cells inheriting old poles are phenotypically distinct and display increased drug efflux activity relative to daughters. Consequently, we find systematic and long-lived growth differences between mother and daughter cells in the presence of sub-inhibitory drug concentrations. A simple model for biased partitioning predicts a population structure of long-lived and highly heterogeneous phenotypes. This straightforward mechanism of generating sustained growth rate differences at sub-inhibitory antibiotic concentrations has implications for understanding the emergence of multi-drug resistance in bacteria.


Erik Winfree, Caltech

Enzyme-free nucleic acid dynamical systems

Chemistries exhibiting complex dynamics -- from inorganic oscillators to gene regulatory networks -- have been long known but cannot be reprogrammed at will because of a lack of control over their evolved or serendipitously found molecular building blocks. Here we show that information-rich DNA strand displacement cascades could be systematically constructed to realize complex temporal trajectories specified by an abstract chemical reaction network model. We codify critical design principles in a compiler that automates the design process, and demonstrate our approach by building a novel DNA-only oscillator. Unlike biological networks that rely on the sophisticated chemistry underlying the central dogma, our test tube realization suggests that simple Watson-Crick base pairing interactions alone suffice for arbitrarily complex dynamics. Our result establishes a basis for autonomous and programmable molecular systems that interact with and control their chemical environment. (Joint work with Niranjan Srinivas, James Parkin, Georg Seelig, and David Soloveichik.)