The goal of this project is to use innovative systems biology and synthetic biology approaches to quantitatively characterize and analyze bacterial gene regulatory networks underlying cellular responses to antibiotics, the formation of persisters and the emergence of resistance. With the alarming spread of antibiotic-resistant strains of bacteria, a better understanding of the specific sequences of events leading to cell death from bactericidal antibiotics is needed for future antibacterial drug development. Accordingly, there is a need for systems biology and synthetic biology approaches to discern the interplay between genes, proteins and pathways in furthering our understanding of how bacteria respond and defend themselves against antibiotics. The implications of the underlying logic of genetic networks are difficult to deduce through experimental techniques alone, and successful approaches will in many cases, involve the union of new experiments and computational modeling techniques. To address this problem, we have developed computational-experimental methods that enable construction of quantitative models of gene, protein and metabolite regulatory networks using expression measurements and no prior information on the network structure or function. In this project, we will use these approaches to reverse engineer bacterial gene regulatory networks underlying cellular responses to antibiotics, the formation of persisters and the emergence of resistance. The resulting networks and pathways will be analyzed to gain insight into the regulatory control of the associated biological processes, and the network models will be used to identify key regulators and mediators for a variety of phenotypic responses. This work could lead to new insights into the stress response of bacteria and the identification of novel targets for drug discovery, e.g., ones that overcome bacterial protective mechanisms or activate bacterial programmed cell death. This project may thus enable the development of novel classes of antibiotics that account for and utilize the complex regulatory properties of genetic networks. Project description: Remarkable progress in genomic research is leading to a complete map of the building blocks of biology. Knowledge of this map is, in turn, fueling the study of gene regulation, where proteins often regulate their own production or that of other proteins in a complex web of interactions. An important challenge in post-genomic research will be the dissection and analysis of the complex dynamical interactions involved in gene regulation, and the deduction of phenotypic cellular responses from the structure and behavior of such networks. While the notions of protein-DNA feedback loops and network complexity are not new, experimental advances are inducing a resurgence of interest in the quantitative description of gene regulation. These advances are beginning to set the stage for a modular description of the regulatory processes involved in basic cellular function. The implications of the underlying logic of genetic networks are difficult to deduce through experimental techniques alone, and successful approaches will in many cases, involve the union of new experiments and computational modeling techniques. We have been approaching this exciting area from two complementary perspectives ? systems biology and synthetic biology. In systems biology, we have been taking a top-down approach to gene regulation, and developing and applying integrated experimental-computational techniques to reverse engineer and analyze naturally occurring gene regulatory networks. In synthetic biology, we have been taking a bottom-up approach to gene regulation, and using tools from nonlinear dynamics and molecular biology to model, design and construct synthetic gene networks. For this project, we propose to use innovative systems biology and synthetic biology approaches to quantitatively characterize and analyze bacterial gene regulatory networks underlying cellular responses to antibiotics, the formation of persisters and the emergence of resistance. With the alarming spread of antibiotic-resistant strains of bacteria, a better understanding of the specific sequences of events leading to cell death from the wide range of bactericidal antibiotics is needed for future antibacterial drug development. Accordingly, there is a need for systems biology and synthetic biology approaches to discern the interplay between genes, proteins and pathways in furthering our understanding of how bacteria respond and defend themselves against antibiotics. To help address this problem, we have developed rapid and scalable methods that enable construction of quantitative models of gene, protein and metabolite regulatory networks using steady-state or time series expression measurements and no prior information on the network structure or function (Science 301: 102, 2003; Nature Biotechnology 23: 377, 2005). These methods are derived from a branch of engineering called system identification, in which a model of the connections and functional relationships between elements in a network is inferred from measurements of system dynamics (e.g., the response of genes and proteins to external perturbations). We tested our reverse engineering methods on the SOS pathway in E. coli. The SOS pathway regulates the cell's response to DNA damage and involves more than 100 genes. In a pilot study, we showed that our methods can identify an accurate model of functional connections between genes in the SOS pathway, and that the model can be used to identify the major regulatory control points in the network. We also demonstrated that the model can be applied to the RNA expression profiles obtained from chemical perturbations to identify the direct biochemical mediators of a compound's bioactivity in the cell. Recently, we developed and implemented a genome-scale systems biology approach for mapping transcriptional regulatory networks in microbes (PLoS Biology 5: e8, 2007). Our approach is analogous to shotgun genome sequencing in that we use statistical algorithms to piece together a high quality, genome-wide map of transcriptional regulatory interactions from random snapshots of diverse microbial gene expression responses. Our most recent algorithm, the CLR algorithm, applies information theoretic analysis to a compendium of gene expression profiles to identify the gene targets of transcription factors. We applied the CLR algorithm to 445 Affymetrix gene expression profiles of E. coli, and validated the predictions using 3216 known regulatory interactions from the E. coli RegulonDB. We identified 1079 regulatory interactions among 328 transcription factors with a 60% true positive rate. Of these interactions, 426 were identified with a higher confidence level (TP rate) of 80%. The targets of many transcription factors in this network are significantly enriched for one or more biological functions, including DNA damage response, central metabolic control, iron import, amino acid biosynthesis, and osmotic stress response. This work expanded the known E. coli regulatory network by more than 700 interactions, and demonstrated the feasibility of genome-scale mapping of transcriptional regulatory pathways in microbes using a compendium of gene expression data. As part of this proposed project, we plan to extend these efforts to reverse engineer and analyze bacterial gene regulatory networks underlying cellular responses to antibiotics, the formation of persisters and the emergence of resistance. We will collect expression profiles from E. coli under a diverse set of conditions and apply our reverse engineering approaches to map transcriptional regulatory networks. Whenever appropriate, we will integrate these data with publicly available expression profiles to expand the capability of our systems biology platform, and we will integrate these efforts with comparative sequence analysis and metabolic models. The resulting networks and pathways will be analyzed to gain insight into the regulatory control of the associated biological processes, and the network models will be used to identify key regulators and mediators for a variety of phenotypic responses. Recently, we employed a systems biology approach to identify novel mechanisms that contribute to bacterial cell death upon DNA gyrase inhibition by the widely used quinolone antibiotic, norfloxacin. It is well-known that gyrase inhibitors induce cell death by stimulating DNA damage, impeding lesion repair and blocking replication processes. We performed phenotypic and microarray analyses on E. coli treated with norfloxacin to identify additional contributors to cell death resulting from gyrase poisoning. In the course of this work, we discovered a novel oxidative damage cell death pathway which involves reactive oxygen species and a breakdown in iron regulatory dynamics following DNA damage induction. Specifically, we found that iron misregulation, related to repetitious superoxide-mediated iron-sulfur cluster decomposition and repair, or redox cycling, plays a major role in gyrase inhibitor-induced cell death following DNA damage induction. We demonstrated in vivo that these events promote the Fenton reaction-catalyzed generation of highly destructive hydroxyl radicals, resulting in oxidative damage to a wide array of biomolecules and ultimately, cell death. Importantly we showed that preventing the formation of hydroxyl radicals reduces the efficiency of norfloxacin, establishing the critical role that reactive oxygen species play in determining gyrase inhibitor-induced terminal cell fate. Formation of hydroxyl radicals following DNA damage has been demonstrated in higher-order systems, including mammalian cells, and is considered one of the hallmark features of apoptosis. We discovered that DNA damage in bacteria induced by a synthetic antibiotic, similarly leads to the formation of hydroxyl radicals. As part of this proposed project, we plan to use our reverse engineering approaches to reconstruct and analyze our proposed oxidative damage cell death pathway. These efforts will allow us to develop a mechanistic model for norfloxacin-stimulated cell death that involves iron and reactive oxygen species. Current antimicrobial therapies, which cover a wide array of targets, fall into two general categories: bactericidal drugs which kill bacteria with an efficiency of >99.9% and bacteriostatic drugs which merely inhibit growth. Antibiotic drug-target interactions are well-studied and fall into three classes: inhibition of DNA replication and repair (e.g., quinolones), inhibition of protein synthesis, and inhibition of cell-wall turnover. However, our understanding of many of the bacterial responses that occur as a consequence of the primary drug-target interaction resulting in cell death remains incomplete. Building upon our pilot work with quinolones, we chose to investigate whether hydroxyl radical formation also underlies some fraction of the killing associated with the diverse class of bactericidal drugs. Indeed, we found in both gram-negative and gram-positive bacteria that all classes of bactericidal antibiotics regardless of drug-target interaction stimulate hydroxyl radical formation, which contributes to the lethal effects of these drugs. We also found, in contrast, that bacteriostatic drugs, which merely inhibit growth, do not produce hydroxyl radicals. Hydroxyl radicals are extremely toxic, and will readily damage membrane lipids, proteins and DNA. In this pilot study, we showed, by knocking out recA and disabling the DNA damage response system (SOS response), that hydroxyl radical production potentiates bactericidal antibiotics. We also showed, using an iron chelator, a hydroxyl radical quencher, and a metabolic mutant, respectively, that the killing efficiency of bactericidal antibiotics can be reduced by attenuating hydroxyl radical production, thus establishing hydroxyl radicals as a key player underlying the lethal effects of bactericidal drugs. These results provide for the development of novel therapeutics that exploit and enhance hydroxyl radical formation. Antibacterial drug design has focused on blocking essential cellular functions. This has yielded significant advances in antibacterial therapy; however, the ever-increasing prevalence of antibiotic-resistant strains has made it critical that we develop novel, more effective means of killing bacteria. Our preliminary results indicate that all three major classes of bactericidal drugs can be potentiated by targeting bacterial systems that remediate hydroxyl radical damage, including proteins involved in triggering the SOS response. As part of this proposed project, we plan to build on these early efforts and use our systems biology approaches to map out the regulatory networks and pathways underlying bacterial responses to bactericidal antibiotics. These efforts could provide insights into common cellular death mechanisms and triggers for the different classes of bactericidal antibiotics. Our pathway analyses and systems biology approaches may also uncover drugable targets for stimulating hydroxyl radical formation, which could result in new classes of bactericidal antibiotics. Recently, our research group has also become interested in bacterial persisters. The phenomenon of bacterial persistence is generally viewed as a reversible state during which a small sub-population of cells enters dormancy, enabling the population to gain survivability against diverse environmental stresses, including antibiotic treatment. Persisters are thought to be involved in the resistance of biofilms to antibiotic treatment, leading to chronic infections. Over the last few years, there has been growing interest in persisters in the microbiology and systems biology communities. Despite this interest and the emerging importance of this problem given the increasing prevalence of antibiotic resistance, the mechanisms underlying persister formation remain unknown. We recently uncovered novel mechanisms underlying persister formation. We investigated the functions of genes around the high persistence hip locus, and identified mechanisms for persister formation that rely upon two forms of cell-to-cell communication. The first involves cell-to-cell contact which affects the levels of persister cells formed and is mediated by the previously uncharacterized cell surface protein, YdeU, located immediately upstream of hip. The second communication system involves the autoinducer-2 (AI-2) quorum sensing system, including the lsr (luxS regulated) operon of E. coli, located upstream of ydeU. Perturbation of these quorum- sensing components significantly affected the formation of persisters. Both of these communication systems display media-specific effects on persister formation, suggesting that these interfacial systems modulate persister formation by sensing of the external environment. The cluster of cell-to-cell communication genes surrounding the hip locus appear to interact with each other in a complex fashion such that information regarding the external environment, e.g., cell density and nutrient conditions, are relayed among the cells in the bacterial population to regulate and tune the probability of switching into the persistent state. Cell-to-cell communication appears to provide an excellent means of controlling population heterogeneity, as local and global changes in environmental conditions can lead to fluctuations in survival benefits of a growing versus a non-growing cell population. Our pilot work shows that persister formation is a dynamic and variable process influenced by both short-range (cell-to-cell contact) and long-range (AI-2 levels and AI-2 influenced genes) cell-to-cell communication. We propose to extend these studies and use our systems biology approaches to reconstruct and analyze the regulatory networks and pathways (e.g., involving the hip locus and the AI-2 quorum sensing system) underlying persister formation. Such efforts may provide mechanistic insight into the factors leading to the formation of persisters, which could be valuable for the development of therapeutic interventions to prevent and eliminate persisters. We also plan to build on this work and explore the function of bacterial toxin-antitoxin (TA) pairs, which are pairs of genes (e.g., ccdB and ccdA) that mediate the response of bacteria to stress conditions, such as antibiotics and nutritional stress. TA pairs exert their effects by inhibiting key cellular processes such as DNA replication and protein synthesis. However, it is unclear whether TA pairs are responsible for programmed cell death or whether they activate protective mechanisms (e.g., persister formation) until environmental conditions improve. It has been difficult to study TA pairs because of their toxic effect on cells, and consequently little is known about their associated pathways and regulatory networks. Accordingly, we plan to combine our efforts in synthetic biology and systems biology to study the function of TA pairs. Specifically, we plan to utilize engineered riboregulators that we have developed (Nature Biotechnology 22: 841, 2004; Nature Biotechnology 24: 545, 2006). These RNA switches enable tight and tunable control of gene expression (i.e., 98% repression and 19-fold increases in expression) at the post- transcriptional level, and are thus ideal for studying toxic genes. We plan to use these riboregulators in conjunction with our reverse engineering approaches to infer and analyze TA- pair pathways and regulatory networks. This work could lead to new insights into the stress response of bacteria and the identification of novel targets for drug discovery, e.g., ones that overcome bacterial protective mechanisms or activate bacterial programmed cell death. Evidence of innovativeness: Our research group has introduced innovative developments in diverse areas of biomedical research at all scales, ranging from whole-body dynamics to organs to gene networks. Our initial work at Boston University focused on nonlinear physiological dynamics with an emphasis on balance control, sensory function and cardiac dynamics. We were the first to recognize that noise-based techniques could be developed and utilized to improve neurosensory function (Nature 376: 236,1995; J Neurophysiology 76: 642,1996). Traditionally, noise has been viewed as a detriment to signal detection and information transmission. However, we realized that input noise could be utilized to lower sensory detection thresholds in humans (Nature 383: 769, 1996; Nature 383: 770, 1996). In a series of studies, we showed that reduced vibrotactile sensitivity in older adults, patients with stroke, and patients with diabetic neuropathy could be significantly improved with input mechanical noise (Arch Phys Med Rehab 83: 171, 2002). We also demonstrated that the application of sub-sensory mechanical noise to the soles of the feet via vibrating insoles could improve balance control in healthy young subjects, older adults, patients with diabetic neuropathy, and patients with stroke (Lancet 362: 1123, 2003; Annals of Neurology 59: 4, 2006). We co-founded Afferent Corporation to commercialize this novel neurostimulation technology. Afferent is using our noise-based sensory enhancement technology to create a new class of medical devices to address complications resulting from diabetic neuropathy, restoring brain function following stroke, and improving elderly balance. We have also pioneered the development and application of model-independent control techniques for eliminating cardiac arrhythmias (Phys Rev E 53: R49, 1996; Phys Rev Lett 78: 4518, 1997). We have used these techniques to suppress alternans rhythms (period-2 rhythms) in a cardiac model and in rabbit-heart preparations, respectively. This work has important clinical implications given that cardiac alternans rhythms often precede serious cardiac arrhythmias and are a harbinger of sudden cardiac death. It is possible that the suppression of alternans rhythms could curtail a route to a fatal arrhythmia. Along these lines, our techniques were recently successfully applied to human patients, and are currently under development at Cornell Medical School. Recently, we discovered the potential role of systems engineering at the genome level, and began working in synthetic biology and systems biology. In synthetic biology, we pioneered the use of techniques from nonlinear dynamics and molecular biology to model, design and construct synthetic gene networks. For example, our group has designed and constructed a genetic toggle switch ? a synthetic, bistable gene regulatory network ? in E. coli (Nature 403: 339, 2000). As a practical device, the toggle switch forms a synthetic, addressable cellular memory unit and has implications for biotechnology, biocomputing and cell therapy. Our venture into synthetic biology represented an entirely new world for us. Upon discovering the potential role of systems engineering at the genome level, we were told by world-leading scientists (including a Nobel Prize winner), that the concept of constructing synthetic gene networks was infeasible. We were not dissuaded, and proceeded to design and build a functioning toggle switch in E. coli. This work helped to launch the field of synthetic biology, an emerging area of great excitement and promise in translational biomedicine and biotechnology. I have been fortunate to receive a number of awards in recognition of our innovative accomplishments. For example, in 1999, I was selected for Technology Review's inaugural TR100 ? 100 young innovators who will shape the future of technology, and in 2005, I was selected for the Scientific American 50 ? the top 50 outstanding leaders in science and technology. Additionally, in 2003, I received a MacArthur Foundation ?Genius Award"""""""", becoming the first bioengineer to receive this honor. The award citation noted, """"""""Throughout his research, Collins demonstrates a proclivity for identifying abstract principles that underlie complex biological phenomena and for using these concepts to solve concrete, practical problems."""""""" How the planned research differs from my past or current work: Our work to date in systems biology and synthetic biology has focused on developing novel techniques and approaches, and demonstrating the potential of these in proof-of-principle studies, e.g., showing that our reverse engineering approaches can be used to identify the mode of action of pharmacological compounds. The planned research differs from our past and current work in that we propose to use our systems biology and synthetic biology approaches to address a specific set of biological problems and to identify biological mechanisms. Namely, we propose to use our approaches to quantitatively characterize and analyze bacterial gene regulatory networks underlying cellular responses to antibiotics, the formation of persisters and the emergence of resistance. This new direction for our research program may enable the development of novel classes of antibiotics that account for and utilize the complex regulatory properties of genetic networks. Suitability for NDPA program: The planned research is uniquely suited for the NDPA program because it proposes pioneering systems biology and synthetic biology approaches to a major contemporary challenge in biomedical research, namely, the development of more effective antibiotics to address the ever-increasing prevalence of antibiotic resistance. This high-risk/high- impact work will be directed towards gaining an increased understanding of the networks and pathways underlying the responses of bacteria to antibiotics and other environmental stresses. The planned research would not be suitable for a traditional grant mechanism because it spans a diverse range of disciplines, it is quite broad in scope, and it is too novel, challenging conventional notions concerning bacterial responses to antibiotics. This innovative work would fall between the cracks and not fare well in traditional programs and study sections.
| Cohen, Nadia R; Ross, Christian A; Jain, Saloni et al. (2016) A role for the bacterial GATC methylome in antibiotic stress survival. Nat Genet 48:581-6 |
| Lobritz, Michael A; Belenky, Peter; Porter, Caroline B M et al. (2015) Antibiotic efficacy is linked to bacterial cellular respiration. Proc Natl Acad Sci U S A 112:8173-80 |
| Belenky, Peter; Ye, Jonathan D; Porter, Caroline B M et al. (2015) Bactericidal Antibiotics Induce Toxic Metabolic Perturbations that Lead to Cellular Damage. Cell Rep 13:968-80 |
| Cameron, D Ewen; Collins, James J (2014) Tunable protein degradation in bacteria. Nat Biotechnol 32:1276-81 |
| Pandey, Shree P; Winkler, Jonathan A; Li, Hu et al. (2014) Central role for RNase YbeY in Hfq-dependent and Hfq-independent small-RNA regulation in bacteria. BMC Genomics 15:121 |
| Dwyer, Daniel J; Belenky, Peter A; Yang, Jason H et al. (2014) Antibiotics induce redox-related physiological alterations as part of their lethality. Proc Natl Acad Sci U S A 111:E2100-9 |
| Brynildsen, Mark P; Winkler, Jonathan A; Spina, Catherine S et al. (2013) Potentiating antibacterial activity by predictably enhancing endogenous microbial ROS production. Nat Biotechnol 31:160-5 |
| Modi, Sheetal R; Lee, Henry H; Spina, Catherine S et al. (2013) Antibiotic treatment expands the resistance reservoir and ecological network of the phage metagenome. Nature 499:219-22 |
| Belenky, Peter; Camacho, Diogo; Collins, James J (2013) Fungicidal drugs induce a common oxidative-damage cellular death pathway. Cell Rep 3:350-8 |
| Cohen, Nadia R; Lobritz, Michael A; Collins, James J (2013) Microbial persistence and the road to drug resistance. Cell Host Microbe 13:632-42 |
Showing the most recent 10 out of 35 publications
