Timothy S. Gardner

Construction of a genetic toggle switch in Escherichia coli.

It has been proposed that gene-regulatory circuits with virtually any desired property can be constructed from networks of simple regulatory elements. These properties, which include multistability and oscillations, have been found in specialized gene circuits such as the bacteriophage λ switch and the Cyanobacteria circadian oscillator. However, these behaviours have not been demonstrated in networks of non-specialized regulatory components. Here we present the construction of a genetic toggle switch—a synthetic, bistable gene-regulatory network—in Escherichia coli and provide a simple theory that predicts the conditions necessary for bistability. The toggle is constructed from any two repressible promoters arranged in a mutually inhibitory network. It is flipped between stable states using transient chemical or thermal induction and exhibits a nearly ideal switching threshold. As a practical device, the toggle switch forms a synthetic, addressable cellular memory unit and has implications for biotechnology, biocomputing and gene therapy.

Towards Environmental Systems Biology of Shewanella

Bacteria of the genus Shewanella are known for their versatile electron-accepting capacities, which allow them to couple the decomposition of organic matter to the reduction of the various terminal electron acceptors that they encounter in their stratified environments. Owing to their diverse metabolic capabilities, shewanellae are important for carbon cycling and have considerable potential for the remediation of contaminated environments and use in microbial fuel cells. Systems-level analysis of the model species Shewanella oneidensis MR-1 and other members of this genus has provided new insights into the signal-transduction proteins, regulators, and metabolic and respiratory subsystems that govern the remarkable versatility of the shewanellae.

Chemogenomic profiling on a genome-wide scale using reverse-engineered gene networks

A major challenge in drug discovery is to distinguish the molecular targets of a bioactive compound from the hundreds to thousands of additional gene products that respond indirectly to changes in the activity of the targets. Here, we present an integrated computational-experimental approach for computing the likelihood that gene products and associated pathways are targets of a compound. This is achieved by filtering the mRNA expression profile of compound-exposed cells using a reverse-engineered model of the cells gene regulatory network. We apply the method to a set of 515 whole-genome yeast expression profiles resulting from a variety of treatments (compounds, knockouts and induced expression), and correctly enrich for the known targets and associated pathways in the majority of compounds examined. We demonstrate our approach with PTSB, a growth inhibitory compound with a previously unknown mode of action, by predicting and validating thioredoxin and thioredoxin reductase as its target.

Reverse-engineering transcription control networks

Microarray technologies, which enable the simultaneous measurement of all RNA transcripts in a cell, have spawned the development of algorithms for reverse-engineering transcription control networks. In this article, we classify the algorithms into two general strategies: physical modeling and influence modeling. We discuss the biological and computational principles underlying each strategy, and provide leading examples of each. We also discuss the practical considerations for developing and applying the various methods.

Many Microbe Microarrays Database: uniformly normalized Affymetrix compendia with structured experimental metadata

Many Microbe Microarrays Database (M3D) is designed to facilitate the analysis and visualization of expression data in compendia compiled from multiple laboratories. M3D contains over a thousand Affymetrix microarrays for Escherichia coli, Saccharomyces cerevisiae and Shewanella oneidensis. The expression data is uniformly normalized to make the data generated by different laboratories and researchers more comparable. To facilitate computational analyses, M3D provides raw data (CEL file) and normalized data downloads of each compendium. In addition, web-based construction, visualization and download of custom datasets are provided to facilitate efficient interrogation of the compendium for more focused analyses. The experimental condition metadata in M3D is human curated with each chemical and growth attribute stored as a structured and computable set of experimental features with consistent naming conventions and units. All versions of the normalized compendia constructed for each species are maintained and accessible in perpetuity to facilitate the future interpretation and comparison of results published on M3D data. M3D is accessible at http://m3d.bu.edu/.

Rewriting yeast central carbon metabolism for industrial isoprenoid production

A bio-based economy has the potential to provide sustainable substitutes for petroleum-based products and new chemical building blocks for advanced materials. We previously engineered Saccharomyces cerevisiae for industrial production of the isoprenoid artemisinic acid for use in antimalarial treatments. Adapting these strains for biosynthesis of other isoprenoids such as β-farnesene (C15H24), a plant sesquiterpene with versatile industrial applications, is straightforward. However, S. cerevisiae uses a chemically inefficient pathway for isoprenoid biosynthesis, resulting in yield and productivity limitations incompatible with commodity-scale production. Here we use four non-native metabolic reactions to rewire central carbon metabolism in S. cerevisiae, enabling biosynthesis of cytosolic acetyl coenzyme A (acetyl-CoA, the two-carbon isoprenoid precursor) with a reduced ATP requirement, reduced loss of carbon to CO2-emitting reactions, and improved pathway redox balance. We show that strains with rewired central metabolism can devote an identical quantity of sugar to farnesene production as control strains, yet produce 25% more farnesene with that sugar while requiring 75% less oxygen. These changes lower feedstock costs and dramatically increase productivity in industrial fermentations which are by necessity oxygen-constrained. Despite altering key regulatory nodes, engineered strains grow robustly under taxing industrial conditions, maintaining stable yield for two weeks in broth that reaches >15% farnesene by volume. This illustrates that rewiring yeast central metabolism is a viable strategy for cost-effective, large-scale production of acetyl-CoA-derived molecules.

Robust identification of large genetic networks.

Temporal and spatial gene expression, together with the concentration of proteins and metabolites, is tightly controlled in the cell. This is possible thanks to complex regulatory networks between these different elements. The identification of these networks would be extremely valuable. We developed a novel algorithm to identify a large genetic network, as a set of linear differential equations, starting from measurements of gene expression at steady state following transcriptional perturbations. Experimentally, it is possible to overexpress each of the genes in the network using an episomal expression plasmid and measure the change in mRNA concentration of all the genes, following the perturbation. Computationally, we reduced the identification problem to a multiple linear regression, assuming that the network is sparse. We implemented a heuristic search method in order to apply the algorithm to large networks. The algorithm can correctly identify the network, even in the presence of large noise in the data, and can be used to predict the genes that directly mediate the action of a compound. Our novel approach is experimentally feasible and it is readily applicable to large genetic networks.

Neutralizing noise in gene networks.

Noise is inherent in gene expression, and can lead to sizeable fluctuations in the concentrations of expressed RNA and proteins. But many biological processes, such as circadian rhythms, are predictable and must be robust to internal noise. New simulations and studies of synthetic gene networks indicate that negative feedback may counteract noise.

Predicting gene targets of perturbations via network-based filtering of mRNA expression compendia

MOTIVATION DNA microarrays are routinely applied to study diseased or drug-treated cell populations. A critical challenge is distinguishing the genes directly affected by these perturbations from the hundreds of genes that are indirectly affected. Here, we developed a sparse simultaneous equation model (SSEM) of mRNA expression data and applied Lasso regression to estimate the model parameters, thus constructing a network model of gene interaction effects. This inferred network model was then used to filter data from a given experimental condition of interest and predict the genes directly targeted by that perturbation. RESULTS Our proposed SSEM-Lasso method demonstrated substantial improvement in sensitivity compared with other tested methods for predicting the targets of perturbations in both simulated datasets and microarray compendia. In simulated data, for two different network types, and over a wide range of signal-to-noise ratios, our algorithm demonstrated a 167% increase in sensitivity on average for the top 100 ranked genes, compared with the next best method. Our method also performed well in identifying targets of genetic perturbations in microarray compendia, with up to a 24% improvement in sensitivity on average for the top 100 ranked genes. The overall performance of our network-filtering method shows promise for identifying the direct targets of genetic dysregulation in cancer and disease from expression profiles. AVAILABILITY Microarray data are available at the Many Microbe Microarrays Database (M3D, http://m3d.bu.edu). Algorithm scripts are available at the Gardner Lab website (http://gardnerlab.bu.edu/SSEMLasso).

Quorum-Sensing Regulation of a Copper Toxicity System in Pseudomonas aeruginosa

The LasR/LasI quorum-sensing system in Pseudomonas aeruginosa influences global gene expression and mediates pathogenesis. In this study, we show that the quorum-sensing system activates, via the transcriptional regulator PA4778, a copper resistance system composed of 11 genes. The quorum-sensing global regulator LasR was recently shown to directly activate transcription of PA4778, a cueR homolog and a MerR-type transcriptional regulator. Using molecular genetic methods and bioinformatics, we verify the interaction of LasR with the PA4778 promoter and further demonstrate the LasR binding site. We also identify a putative PA4778 binding motif and show that the protein directly binds to and activates five promoters controlling the expression of 11 genes--PA3519 to -15, PA3520, mexPQ-opmE, PA3574.1, and cueA, a virulence factor in a murine model. Using gene disruptions, we show that PA4778, along with 7 of 11 gene targets of PA4778, increases the sensitivity of P. aeruginosa to elevated copper concentrations. This work identifies a cellular function for PA4778 and four other previously unannotated genes (PA3515, PA3516, PA3517, and PA3518) and suggests a potential role for copper in the quorum response. We propose to name PA4778 cueR.