From OpenWetWare

Revision as of 01:08, 15 March 2011 by Smolke (Talk | contribs)
Jump to: navigation, search

Home        Contact        Internal        Protocols        Lab Members        Publications        Research       


Research Overview

Engineered biological systems that process information, materials, and energy hold great promise for developing solutions to many global challenges, including energy and food production, materials synthesis, and medical advancement. Our ability to engineer biological systems is limited by the foundational tools available for programming cellular behavior (i.e., getting information into and acting on information in living systems) and our understanding of how such systems should be constructed (i.e., probing and accessing information in living systems). By developing genetically encoded technologies for reporting on, responding to, and controlling intracellular components in living systems, we are addressing prominent challenges faced in basic and applied biological research. Our research efforts focus on the design of new molecular tools for performing information processing, computation, and control functions in living systems and the application of these tools to programming biological systems. The resulting advances in our ability to transmit information to and from living systems and implement control within cells themselves, will broadly transform how we interact with and program biology, providing access to otherwise inaccessible information on cellular state and allowing sophisticated exogenous and embedded control over cellular functions.

Component and Device Engineering

Engineering RNA-based cellular information processing, communication, and control devices

Researchers: Ryan Bloom, Drew Kennedy, Jay Vowles, Josh Wolf

We have developed generalizable design strategies for a broad class of RNA molecules, called RNA devices, that process and transmit user-specified molecular input signals to targeted protein outputs, thereby linking molecular computation to gene expression. We proposed a first-generation framework for the construction of single input-single output RNA devices based on the modular assembly of three components exhibiting basic functions (Win et al., Chem Biol 2009): a sensor component, made of an RNA aptamer or binding element; an actuator component, made of an RNA gene regulatory element; and a transmitter component, made of a sequence that couples the sensor and actuator components. The transmitter component regulates the activity of the actuator component based on the binding state of the sensor component, providing genetic insulation and a standardized communication interface. We initially demonstrated RNA devices that function as single-input Buffer and Inverter gates that convert a molecular signal to increased and decreased gene expression output, respectively (Win et al., PNAS 2007).

One of the key contributions of our work to the field of molecular design is the development of composition frameworks supporting forward engineering and modular assembly approaches. Our design approach thus enables a modular device platform that supports the tailoring of device function (sensing, actuation, computation) through the direct swapping of component modules without time-intensive redesign of the device. Using our design strategy, we have demonstrated novel cis-acting RNA devices that function through a variety of regulatory mechanisms, including ribozyme cleavage (Win et al., PNAS 2007; Win et al., Science 2008), alternative splicing (Culler et al., Science 2010), and RNase III cleavage (Babiskin et al., Mol Sys Biol 2011; Babiskin et al., Nuc Acids Res 2011). We have also demonstrated novel trans-acting RNA devices that function through the RNA interference pathway (Beisel et al., Mol Sys Biol 2008; Beisel et al., Nuc Acids Res 2010) and antisense silencing (Bayer et al., Nat Biotech 2006). By accessing a variety of gene regulatory mechanisms, we have built tailored RNA devices that function in diverse cell types and organisms and that respond to endogenous and exogenous small molecule inputs (Win et al., PNAS 2007; Beisel et al., Mol Sys Biol 2008) and endogenous and heterologous protein inputs (Culler et al., Science 2010). Ongoing research is extending the capabilities of these design platforms by optimizing device architectures and integrating new actuation mechanisms.

Back to Top

Engineering higher-order cellular information processing devices

Researchers: Leo d’Espaux, Yen-Hsiang Wang

We have demonstrated the extension of the single-input RNA device frameworks to the construction of higher-order devices that perform multi-input signal processing in living systems (Win et al., Science 2008; Culler et al., Science 2010), further supporting the power of the developed modular assembly strategy. As an example, my laboratory described extended architectures for rationally assembling RNA components (sensors, actuators, transmitters) into multi-input signal integration devices and built genetic devices that function as logic gates, signal and bandpass filters, and exhibited cooperativity. Our research has demonstrated that the developed design frameworks provide a general approach for the forward engineering of multi-input devices, supporting the combinatorial assembly of many information processing, transduction, and control devices from a smaller number of components. More generally, this work has demonstrated the application of synthetic biology design strategies to scalable platforms for genetic device design and that advances in engineering design can transform the scale, efficiency, and speed with which we can engineer cellular behaviors.

Ongoing research is extending the complexity of information processing schemes that can be reliably implemented in cellular systems by building regulatory networks composed of distinct genetic devices. For example, we are examining device design schemes that will allow for ratiometric or differential sensing of multiple molecular signals, signal amplification, error detection, and signal restoration. We are also examining the implementation of single and multi-layered architectures in genetic pathways that will allow scaling of computational complexity (i.e., multi-input processing and multi-output control), while maintaining reliable information transmission between different layers (i.e., genetic constructs or transcripts).

Back to Top

Enabling Technology Development

Developing computer-aided design (CAD) tools for RNA device design and optimization

Researchers: Ryan Bloom

The integration of engineered information processing and control devices with cellular networks to effectively control systems-level functions often requires precise tuning of the response properties of the engineered molecules to match the quantitative signaling range of the system of interest. To address this challenge, we have been developing computational design tools that enable the rational tuning and optimization of RNA device function to streamline the process of integrating these genetic elements into biological systems. Specifically, we developed kinetic models of RNA device function based on detailed molecular descriptions of functional mechanisms (Beisel et al., PLoS Comp Biol 2009). We used these models to examine the effects of rate constant variation on the resulting response curve to elucidate strategies for rationally tuning the performance of an RNA device through modulation of rates associated with different mechanistic steps. Our modular assembly platform results in device functions (sensing, actuation, computation) being encoded in distinct and independent components, thereby resulting in individual model parameters mapping to individual components. We used this unique property of our molecular platform to map parameter tuning strategies predicted through the computational models to direct sequence changes of the appropriate components in the device to generate an early sequence-to-function framework for optimizing RNA switch function (Beisel et al., Mol Sys Biol 2008).

Ongoing work is focused on refining and improving the CAD tool for RNA device design. We are examining different RNA folding algorithms and determining the most appropriate way to process the structures and energies that are predicted from the programs to obtain an accurate predictive measure of conformational partitioning in vivo (Beisel et al., Mol Sys Biol 2008), which can then be fed into our computational models to predict the quantitative device response from a primary RNA sequence. Future work will integrate structural studies with our computational models and cellular systems to achieve an integrated understanding of device folding, dynamics, and activity. The longer-term goal of these efforts is to develop a robust sequence-to-function framework that supports the in silico design and optimization of the quantitative response properties of RNA devices from well-characterized component libraries.

Back to Top

Developing high-throughput strategies for the generation of new RNA component functions

Researchers: Andy Chang, Joe Liang, Brent Townsend, Jay Vowles

The implementation of genetic devices within biological systems is broadly limited by our ability to tailor the input-output activities of these devices to applications of interest. Libraries of well-characterized RNA components (encoding sensing, actuation, and computation functions) that are compatible with the device platforms are necessary to realize the full advantages of our design approach. As one example, the generation of RNA aptamers exhibiting desired in vivo affinities and specificities is a current challenge in device design due to inefficiencies and limitations in scaling with existing selection strategies. We are developing a high-throughout in vivo selection strategy in which RNA sensors are screened for desired binding activities within the device platform to enable a solution-based selection for aptamer sequences that are compatible with the device framework and exhibit desired binding properties in the cellular environment. We have established an efficient high-throughput screening strategy for device activity based on fluorescence activated cell sorting by linking the output of an RNA device to the expression of a fluorescent protein, where sequences within the RNA device can be randomized to generate device libraries to search for novel component functions (Liang et al., in preparation). We are applying this in vivo screen to the generation of RNA components exhibiting enhanced or novel activities.

The de novo generation of RNA aptamers will likely require searching a larger sequence space than can be directly achieved in vivo. As such, we are currently developing a high-throughput in vitro selection strategy for the RNA device platform where partitioning of active members will be conducted through PCR recovery, bead-based partitioning, or capillary electrophoresis. We will develop a streamlined in vitro-in vivo selection strategy that will be applied with high throughput characterization strategies, such as surface plasmon resonance (Win et al., Nuc Acids Res 2006), to develop libraries of novel sensing functions to diverse ligands of interest, including natural product families, central metabolites, clinically-relevant drug molecules, disease markers, and exogenous chemicals with properties appropriate for use as trigger molecules.

Back to Top

Designing ‘intelligent’ therapeutic molecules

Researchers: Ryan Bloom, Leo d’Espaux, Megan Palmer, Jay Vowles, Kathy Wei

We are designing molecular switches to act as targeted or ‘intelligent’ therapeutic molecules. Projects in this area focus on the construction of ligand-regulated RNA-based regulators of gene expression that function in mammalian cells through diverse regulatory mechanisms such as the RNAi pathway or ribozyme-based cleavage. Research areas have been initiated in the design of molecular switches for advancing targeted breast cancer treatments and immunotherapy strategies. In the area of developing next-generation cancer therapies and detection strategies, RNA switches will be constructed to take different hormone and growth factor biomarkers identified for different breast cancers as input signals. In response to the presence of particular set of biomarker indicative of breast cancer, these molecules will regulate the expression of target output genes such as genes involved in regulating cellular behavior (apoptosis, cancer phenotype) or genes associated with a monitorable signal (detection/diagnosis strategies). In the latter area, RNA switches will be constructed to take target small molecules or biomarkers for different tumor cells as input signals. These molecular switches will be engineered into T cells and respond to the presence of these localized inputs by activating the T cell to kill the nearby tumor cells, thereby developing more effective and safe immunotherapy treatments. Cellular engineering projects are currently being conducted in model cell lines and will later effectively be transferred into animal models for these diseases. Both of these projects have translational clinical collaborators at the City of Hope (Professors Carlotta Glackin - breast cancer therapies; Mike Jensen - T cell engineering).

Back to Top

Metabolic engineering of Saccharomyces cerevisiae for alkaloid production

Researchers: Stephanie Galanie, Josh Michener, Michael Siddiqui, Isis Trenchard, Kate Thodey

We are engineering synthetic circuits in yeast for the production of different value-added compounds. Current research efforts are focused on the development of Saccharomyces cerevisiae as a microbial host for the total biosynthesis of diverse alkaloid compounds. Synthetic metabolic pathways are being assembled for the production of two different classes of alkaloids, the purine alkaloids and the benzylisoquinoline alkaloids (BIAs). The purine alkaloid pathway, resulting in the synthesis of caffeine and similar analogs, is being engineered in yeast largely as a model pathway through which to explore general design principles and strategies for integrating molecular switches and assembling signal processing schemes with synthetic metabolic pathways. Strategies will be developed for applying these engineered molecular switches for establishing rapid and generalizable pathway optimization screens and selections. Furthermore, control theory will be used to explore the design parameters for constructing dynamically regulated networks with switch-based control loops as a way to optimize pathway flux. The purine alkaloid pathway enables a more immediate demonstration of these strategies and design principles as aptamers to these metabolites are readily available.

The BIA pathway, resulting in the synthesis of codeine, morphine, and sanguinarine, is being explored for generating a microbial host that can (i) readily synthesize an array of BIA molecules with diverse pharmacological activities and (ii) be used to set up rapid functional genomics screens to effectively identify enzymes that can act on these molecules from EST libraries of native plant hosts. The BIA pathway is particularly appropriate to this type of metabolic engineering effort as they are a complex class of molecules that are not effectively synthesized through traditional chemical means. In addition, there is no source for many of the intermediate metabolites of pharmaceutical interest, as they do not accumulate in the native hosts and genetic engineering efforts remain challenging in plants. As the pathway including the early steps resulting in the synthesis of the BIA backbone (norcoclaurine) has not been entirely elucidated from plant hosts, a synthetic network composed of genes from bacteria, humans, and various plants is being assembled and optimized for BIA production in yeast. It is anticipated that the molecular tools developed in the purine alkaloid pathway will be readily transferable to the BIA pathway. This research effort has a plant biologist collaborator from the University of Calgary (Peter Facchini).

Back to Top

Personal tools