Synthetic biology in the analysis and engineering of

来自 : 发布时间：2024-11-24

全球经销商斯洛文尼亚和克罗地亚

BIA doo

地址

Teslova 30Ljubljana, SI-1000

电话

386 1 425 47 63

传真

386 1 426 45 91< /p>电子邮件

sales@bia.si

网站

www.bia.si

过滤器：克罗地亚清除所有国家/地区

请从下面的菜单中选择一个国家/地区：

国家/地区阿富汗阿尔及利亚安哥拉亚美尼亚澳大利亚澳大利亚奥地利阿塞拜疆巴林孟加拉国比荷卢经济联盟贝宁博茨瓦纳巴西保加利亚布基纳法索法索布隆迪喀麦隆加拿大乍得智利中国哥伦比亚刚果克罗地亚塞浦路斯捷克共和国刚果丹麦吉布提埃及赤道几内亚爱沙尼亚埃塞俄比亚芬兰法国加蓬冈比亚德国加纳希腊几内亚几内亚比绍香港匈牙利印度印度尼西亚伊拉克爱尔兰以色列意大利象牙海岸日本约旦哈萨克斯坦肯尼亚韩国科威特吉尔吉斯斯坦拉脱维亚黎巴嫩莱索托利比亚立陶宛马达加斯加马拉维马来语西亚马里马耳他毛里求斯毛里求斯墨西哥摩洛哥莫桑比克纳米比亚尼泊尔新西兰尼日尔尼日利亚挪威阿曼巴基斯坦巴勒斯坦秘鲁菲律宾波兰葡萄牙波多黎各卡塔尔俄罗斯卢旺达沙特阿拉伯塞内加尔塞拉利昂利比里亚新加坡洛文尼亚西班牙斯里兰卡斯威士兰瑞典瑞士台湾塔吉克斯坦坦桑尼亚多哥突尼斯土耳其土库曼斯坦阿联酋乌克兰英国美国乌兹别克斯坦也门赞比亚津巴布韦产品

请选择产品：

产品MicroStationOdinOmniLogLinkedInTwitterFacebook

版权所有 2023 Biolog，保留所有权利用户隐私政策Site by Metropolis

产品产品实验室服务应用资源网络研讨会新闻与活动贸易展览新闻发布​​公司关于我们联系方式提交样品

全球经销商哥伦比亚

Sanitas Ltd.

地址

Calle 35 # 20-18Bogota

电话

57-12880177 ext 2010

传真

3132471021

电子邮件< p>molecular@sanitastec.com网站

www.sanitastec.com

过滤器：哥伦比亚清除所有国家

请从下面的菜单中选择一个国家/地区：

国家/地区阿富汗阿尔及利亚安哥拉亚美尼亚澳大利亚奥地利阿塞拜疆巴林孟加拉国比荷卢经济联盟贝宁博茨瓦纳巴西保加利亚布基纳法索布隆迪喀麦隆加拿大乍得智利中国哥伦比亚刚果克罗地亚塞浦路斯捷克共和国刚果丹麦吉布提埃及赤道几内亚爱沙尼亚埃塞俄比亚芬兰法国加蓬冈比亚德国加纳希腊几内亚几内亚比绍香港匈牙利印度印度尼西亚伊拉克爱尔兰以色列意大利象牙海岸日本约旦哈萨克斯坦肯尼亚韩国科威特吉尔吉斯斯坦拉脱维亚黎巴嫩莱索托利比亚立陶宛马达加斯加马拉维马来语西亚马里马耳他毛里塔尼亚毛里求斯墨西哥摩洛哥莫桑比克纳米比亚尼泊尔新西兰尼日尔尼日利亚挪威阿曼巴基斯坦巴勒斯坦秘鲁菲律宾波兰葡萄牙波多黎各卡塔尔俄罗斯卢旺达沙特阿拉伯塞内加尔塞拉利昂利比里亚新加坡矿石斯洛文尼亚西班牙斯里兰卡斯威士兰瑞典瑞士台湾塔吉克斯坦坦桑尼亚多哥突尼斯土耳其土库曼斯坦阿联酋乌克兰英国美国乌兹别克斯坦也门赞比亚津巴布韦产品

请选择产品：

产品MicroStationOdinOmniLogLinkedInTwitterFacebook

版权所有 2023 Biolog，保留所有权利用户隐私政策Site by Metropolis

产品产品实验室服务应用资源网络研讨会新闻与活动贸易展览新闻发布​​公司关于我们联系方式提交样品*abc abioss—Centre for Biological Signalling Studies, Albert-Ludwigs-Universität Freiburg, Engesserstrasse 4b, D-79108 Freiburg, Germany. E-mail: wilfried.weber@bioss.uni-freiburg.de; Fax: +49 761 203 97660; Tel: +49 761 203 97654 bFaculty of Biology, Albert-Ludwigs-Universität Freiburg, Schänzlestrasse 1, D-79104 Freiburg, Germany cDepartment of Biosystems Science and Engineering, ETH Zurich, Mattenstrasse 26, CH-4058 Basel, Switzerland Received 7th July 2009 , Accepted 14th October 2009First published on 16th November 2009 AbstractSynthetic biology as the discipline of reconstructing natural and designing novel biological systems is gaining increasing impact in signaling science. This review article provides insight into synthetic approaches for analyzing and synthesizing signaling processes starting with strategies into how natural and pathological signaling pathways can be reconstructed in an evolutionary distant host to study their topology and function while avoiding interference with the original host background. In the second part we integrate synthetic strategies in the rewiring of signaling systems at the nucleic acid and protein level to reprogram cellular functions for biotechnological applications. The last part focuses on synthetic inter-cell and inter-species signaling devices and their integration into synthetic ecosystems to study fundamental mechanisms governing the co-existence of species. We finally address current bottlenecks in the (re-)design of signaling pathways and discuss future directions in signaling-related synthetic biology. Michael Kämpf Michael Kämpf is currently pursuing his PhD at the Department of Biosystems Science and Engineering (ETH Zurich) in collaboration with the Centre for Biological Signalling Studies (bioss) at the Albert-Ludwigs-Universität in Freiburg, Germany. Prior to that he studied biology at the University of Regensburg and the University of Colorado at Boulder and completed his studies with his thesis about essential steps of N-glycosylation. After that he refocused his research interests to synthetic biology approaches and the construction of smart biomaterials. Wilfried Weber Wilfried Weber was appointed as Full Professor for Synthetic Biology at the Centre for Biological Signalling Studies (bioss) and the Faculty of Biology at the Albert-Ludwigs-Universität Freiburg, Germany. After completing his Masters thesis at Novartis Pharma AG, he joined the research group of Martin Fussenegger and James E. Bailey at ETH Zurich where he obtained his PhD on the development of mammalian cell-compatible inducible expression systems. After a Postdoc with Martin Fussenegger, he established his research group at the ETH Zurich Department of Biosystems Science and Engineering. His research focuses on synthetic biologic signaling devices for the design of synthetic cell-to-cell communication systems and the construction of smart biomaterials for biomedical engineering. Insight, innovation, integration The integration of synthetic engineering approaches with the analysis of biologic signaling processes gives insight into the molecular basis of disease and enables the rational design of artificial signaling systems to providing innovative solutions in biotechnology and drug discovery. This review article integrates recent synthetic approaches ranging from the transcriptional level by engineering genetic signaling networks to the design of synthetic ecosystems for an in vitro emulation of fundamental mechanisms governing the co-existence of species. We discuss current bottlenecks in merging signaling science with synthetic biology and provide an outlook on future developments of this integrated research field. Introduction The exchange of signals between nucleic acids, proteins, organelles, cells and whole organisms is fundamental to any living system, it enables the orchestration of the metabolic machinery, cell growth, differentiation and death, the survival in a dynamic environment as well as interactions between whole organisms. Consequently, defects in signaling processes are often the underlying basis of diseases like cancer, immunological disorders or metabolic diseases.1–4 Most of today’s insight into signaling phenomena has been achieved by analytical approaches, like taking the signaling pathways apart, knocking out individual components or coupling signaling proteins to reporter systems.5–7 These analytical techniques are now increasingly complemented by the emerging discipline of synthetic biology focusing on the re-design of natural biological systems and the construction of new biological parts and devices. Once the components of a signaling pathway have been identified by one of the \"omics” disciplines, they can be combined in a synthetic approach to reconstruct the signaling system, the functionality of which represents the proof, that all essential parts have been identified. Beyond the reconstruction of natural biologic signaling pathways, the thorough characterization of individual signaling elements has now enabled the de novo design of signaling systems for achieving defined and predictable cellular function at the nucleic acid, protein and cellular level.8–10 This review article integrates synthetic biology approaches in both areas, the analysis and the engineering of signaling processes. In the first part, synthetic approaches are described to study B-cell receptor signaling or to investigate the underlying mechanisms governing pattern formation during Drosophila development. The second part focuses on reprogrammed synthetic intracellular signaling pathways at the nucleic acid and protein level and the last part describes recent advances in intercellular and inter-species signaling devices that are at the basis of synthetically reconstructed ecosystems for studying medically and ecologically important multi-species interactions. Analysis through synthesis: reconstruction of natural and pathologic signaling processes The analysis of signaling pathways is more and more complemented by synthetic approaches following Richard Feynman’s (Nobel laureate in 1965) quote that \"what I cannot create, I do not understand”. Synthetic approaches in signaling research aim at the reconstruction of individual signaling processes or whole signaling networks in an evolutionary distant background for avoiding non-controlled interference with host components. In such an orthogonal environment, the successful reconstruction of a functional signaling pathway demonstrates that all its essential properties have been identified, in similar fashion to the classical approach in chemistry, where a new molecule is regarded as fully characterized as soon as its synthesis has been achieved.11 Furthermore, such a synthetic approach enables a close characterization of individual signaling proteins as the concentration of each partner can easily be modified or natural proteins can be substituted by mutant (pathologic) forms isolated from patients to elucidate molecular disease mechanisms. In the understanding of immunological processes, synthetic approaches have been pioneered by the group of Reth,12,13 who used evolutionary distant Drosophila S2 Schneider cells for the functional reconstruction and analysis of B cell antigen receptor (BCR) signaling. As S2 cells can easily be transfected with large amounts of vector DNA and show high protein synthesis rates, they provide an excellent tool for the simultaneous inducible expression of up to 20 mammalian genes.12 In an exemplary study the regulation of the protein tyrosine kinase Syk involved in BCR signaling was investigated by reconstructing all necessary BCR components in S2 cells.13 It was shown that after activation, the tyrosine kinases Lyn and Syk phosphorylated the ITAM (immunoreceptor tyrosine-based activation motif) consensus sequence located in the cytoplasmatic tails of the BCR components Ig-α and Ig-β. The resulting phosphorylated ITAM (ppITAM) stabilized Syk in an open and active conformation by sequestrating its autoinhibitory SH2 domains, leading to phosphorylation of neighboring ITAMs, recruitment of additional Syk protein and subsequent amplification of the BCR signal. These findings suggest that Syk is an allosteric enzyme that amplifies signal transduction from the BCR in a positive feedback loop (Fig. 1). In this system it was further shown that the positive feedback loop can rapidly be disrupted by co-expression of the phosphatase SHP-1 triggering dephosphorylation of Syk-activating ppITAM13 (Fig. 1). In a follow-up study to further characterize BCR secretion and assembly, μmHC, Ig-α and Ig-β were identified as essential elements, as the absence of any partner prevented shuttling of the nascent BCR complex through the endoplasmatic reticulum to the cell surface.12 The topology of the BCR complex was further elucidated in the S2 system by the insertion of cysteine to serine mutants in Ig-α and Ig-β subunits revealing inter- and intra-molecular disulfide bonds (between Ig-αC113 and Ig-βC135) promoting the formation of Ig-α/Ig-β heterodimers following BCR transport to the cell surface.14 This study on the essential dimerization of Ig-α and Ig-β to form a functional BCR was the basis for a subsequent work, in which the S2 cell system was used to reveal the molecular basis of agammaglobulinemia, a disease characterized by the absence or only low levels of serum immunoglobulins leading to a severely compromised immune system.15 The expression of an Ig-β variant derived from an agammaglubulinemic patient was expressed in S2 cells together with the other BCR components and analyzed for correct receptor assembly. It was shown that mutant Ig-β is no longer able to associate with Ig-α thereby abrogating BCR complex formation on the cell surface15 and preventing the emergence of immunoglobulin formation in response to infection. Fig. 1 B-cell receptor signaling. The B-cell receptor is comprised of the membrane-bound immunoglobulin (mIg) molecule and the Ig-α/Ig-β heterodimer harboring ITAM (immuno-receptor tyrosine-based activation) motifs in their cytoplasmatic tail. B-cell receptor signaling is initiated by ligand (antigen) binding, following the activation of the tyrosine kinases Lyn and Syk which results in phosphorylation of the ITAMs. Binding of autoinhibitory SH2 domains of Syk to ppITAMs promotes entire activation of Syk and phosphorylation of neighbouring ITAMs, recruitment of additional Syk and amplifying the signal by a positive feedback-loop. Activated Syk transduces the signal to its down-stream substrates like SLP-65. The positive feedback loop can rapidly be disrupted by phosphatases (PTP). Following this strategy to analyze signaling pathways by synthetically reconstructing them in an orthogonal host, the group of Luis Serrano emulated Drosophila embryonic pattern formation in a cell-free system by a spatially-defined genetic network.16 Development and pattern formation in the Drosophila embryo is triggered by the gap gene network, a complex interplay of morphogens with mutually repressing or activating function.17–19 A key morphogen is bicoid, the mRNA of which is transported to the anterior pole, where it is translated and forms a concentration gradient from the source. Bicoid acts as a transcriptional activator of downstream gap genes like hunchback, knirps, giant, and krüppel, the products of which interact in a mainly repressive way to control each other’s expression.17,20,21 In addition to its activating function, Bicoid acts as repressor of caudal mRNA translation thus generating a Caudal gradient inverse to the Bicoid gradient (Fig. 2A). The inverse Bicoid and Caudal gradients together with the interplay of the other genes belonging to the gap network result in the segmentation of the embryo determining its further development16 (Fig. 2A). Fig. 2 In vitro reconstruction of pattern-forming processes in Drosophila. (A) In Drosophila, mRNA encoding bicoid is transported to the anterior for translation and subsequent formation of a Bicoid protein gradient. Bicoid activates a number of downstream genes like hunchback, giant, krüppel and knirps the products of which interact in a mainly repressive network with the expression of each other resulting in a distinct pattern of the different proteins. Furthermore, Bicoid inhibits uniformly expressed caudal mRNA resulting in a Caudal gradient inverse to the Bicoid gradient. This distinct expression pattern of the specific factors represents the basis for later segmentation of the embryo. (B) Synthetic gene network for in vitro transcription and translation. The expression constructs are controlled by a T7 (PT7) or SP6 (PSP6) promoter recognized by the T7 and SP6 polymerases. Each promoter drives transcription of a DNA-binding repressor protein (A, B, C) that binds to its cognate operator (OA, OB, OC) and prevents transcription from the upstream promoter. (C) Initial distribution of system components in the agarose gel. Construct A was coupled to magnetic beads and localized to both ends of the agarose gel by external magnetic fields. T7 polymerase was pipetted to both ends of the agarose gel resulting in the diffusion-based formation of a gradient. The other constructs B and C as well as SP6 polymerase where uniformly distributed in the system. Following the reaction the gel was sliced and the concentration of the repressor proteins A, B and C were determined and plotted as a function of the position in the agarose gel. The mutual repression of the constructs resulted in expression patterns reminiscent to the situation in the developing Drosophila embryo. For the synthetic cell-free emulation of the pattern-forming gene network, Isalan and coworkers16 designed three expression constructs comprising a promoter, downstream operators, and repressors that bind to the operators (Fig. 2B). Construct A relied on a promoter driving expression of a transcriptional repressor for constructs B and C. Construct B produced a repressor which repressed the other two constructs as well as itself in a negative feedback loop while construct C drove expression of a repressor that inhibits transcription from constructs A and B. Construct A and B harbor a phage T7-derived promoter while construct C is controlled by an SP6 promoter. In order to emulate localized gene expression through mRNA transport, construct A was immobilized on magnetic beads and localized via magnetic fields on two opposing ends of a soft agarose gel emulating the embryo body (Fig. 2C). Localized transcription was achieved by adding T7 RNA polymerase to both ends of the gel resulting in diffusion-based gradient formation of this transcription-activating compound emulating the Drosophila activator Bicoid. The other two constructs B and C as well as SP6 RNA polymerase were evenly distributed throughout the gel which was further supplemented with an in vitro translation mix (Fig. 2C). In order to investigate patterning and segmentation over time, the embryo-emulating gel was sliced and the relative concentrations of the repressor proteins A, B and C were determined. These in vitro data accompanied by a diffusion-based mathematical model resulted in a defined pattern of transcription factors along the gel axis (Fig. 2C) reminiscent of the one observed in Drosophila. It was further shown in vitro and in silico that the sharpness of the patterns increased with increasing repressing interactions between the three constructs and that dynamic pattern stability could be adjusted by modulating repressor protein degradation via the addition of protease.16 This study impressively demonstrates how synthetic genetic networks combined with theoretical modeling can serve as tools for the elucidation and the testing of general design principles in fundamental developmental processes. The synthetic reconstruction of genetic networks in an orthogonal host has recently been applied for the discovery of small molecules for attenuating antibiotic resistance in Mycobacterium tuberculosis.22M. tuberculosis is inherently resistant to the thioamide ethionamide, a prodrug which must be activated in the bacterium by the Baeyer–Villiger monooxygenase EthA for the formation of a cytotoxic NADH adduct targeting InhA.23,24 However, ethionamide activation is rather poor as the promoter driving expression of ethA is repressed by the TetR/CamR-type repressor EthR.25 Based on the hypothesis that a small molecule inhibitor of EthR would increase ethA expression and trigger a more efficient conversion of ethionamide into the cytotoxic NADH adduct to kill the bacterium, a screening system to search for such molecules was designed in mammalian cells.22 Screening in mammalian cells offers the advantage that cytotoxic molecules are readily detected to be excluded from the hit list and that the intracellular target is only reached by compounds able to cross the mammalian membrane barrier, an essential feature for addressing intracellular pathogens like mycobacteria. The screening system was realized by fusing EthR to the Herpes simplex-derived transactivator VP16 for inducing a Drosophila-derived minimal heat-shock promoter Pmin fused to EthR-specific operator sites. A specific inhibition of EthR by a small molecule would thus abrogate DNA-binding and promoter activation that could be detected by lowered expression levels of a reporter gene under the control of Pmin. A screen using a focused chemical library identified the food additive 2-phenyl-ethyl butyrate as an EthR inhibitor that was able to cross the mammalian cell barrier and did not show obvious cytotoxic effects. Administration of 2-phenyl-ethyl butyrate to M. tuberculosis correlated with increasing ethA expression levels and resulted in killing of the bacterium in the presence of otherwise non-inhibiting concentrations of the antibiotic ethionamide thereby demonstrating its potential to attenuate intrinsic drug resistance.22 Transcription-based synthetic signaling processes Engineering of transcriptional signaling pathways has pioneered two directions, the de novo design of modular genetic signaling circuits as well as the reprogramming of global transcriptional signaling systems. Modular genetic signaling circuits were pioneered in bacteria resulting in the construction of bi-stable memory devices for signal storage,27 signal counting28 or oscillating signal generation29–31 and were soon transferred to mammalian cells resulting in the design of coupled transcriptional and translational signaling switches,32 (semi-)synthetic signaling cascades33,34 or tunable oscillating signal generation35 as recently reviewed in detail elsewhere.9,36 In order to reprogram global transcriptional signals, central transcription factors were reprogrammed by mutagenesis,37 by the connection to alterative signaling partners38 or by overexpression to modulate signal intensity.39 In a pioneering study the group of Greg Stephanopoulos introduced global transcription machinery engineering (gTME) in yeast37 by mutating the TATA-binding protein (SPT15) or the TATA-binding protein-associated factor (TAF25) into haploid Saccharomyces cerevisiae thereby enabling the identification of dominant mutations leading to novel functions in the presence of the native chromosomal gene. Selection of mutations in the presence of high ethanol and high glucose resulted in an SPT15 triple mutant showing up to 13-fold improved growth yield at high glucose and significantly improved cellular viability at high ethanol concentrations. Expression profiling in the unstressed conditions revealed that the SPT15 mutations changed the overall transcriptional signaling network resulting in the differential expression of hundreds of genes most of which were up-regulated, however, no particular pathway or genetic network predominantly responsible for the observed signaling reprogramming could be identified. Attempts to obtain a similar tolerance to high ethanol and glucose concentrations by individually overexpressing differentially expressed genes were not successful, indicating that the identified genes are part of a bigger signaling network that must simultaneously be engineered in order to obtain the desired phenotype.37 In a similar approach, the same group introduced in Escherichia coli a library of mutated sigma factor 70 (rpoD, σ70) and performed selection either for ethanol tolerance or increased lycopene production.40 Upon performing multiple gTME rounds in the presence of high ethanol, the differentially expressed genes converged to a subset of around 40 suggesting a consensus-like signaling network optimized for coping with a changed environment.40 In order to globally reprogram transcriptional signaling pathways, the group of Luis Serrano performed a combinatorial approach by reconnecting the promoters and ORFs for seven master transcription factors, seven σ-factors and eight downstream transcription factors in all possible combinations.38 Each such combination of a promoter and an ORF creates a novel signaling path receiving input from the promoter-specific transcription factor and providing a reconnected output determined by the downstream targets of the ORF product. An E. coli library containing all rewired constructs was selected for serial passaging in liquid culture, for longevity at extended periods at 37 °C or survival after a 50 °C heat shock. Serial passaging selected repeatedly clones in which the promoter normally controlling flagellar genes was reconnected to other ORFs. This finding suggests that the unnatural connection of the flagellar gene promoter to other ORFs might suppress flagellar biosynthesis, a feature, which has previously been described to be beneficial for selection in serial subculturing.38 In longevity as well as in heat shock experiments, all surviving clones harbored the stationary phase and heat shock-activated rpoS promoter rewired to osmoregulation-controlling ompR. Transcriptional profiling showed that the rpoS–ompR rewired clones showed upregulated chaperones and heat shock genes and down-regulated permease explaining the new resistant phenotype. These rewiring experiments together with the gTME studies described above, show that cells can both tolerate and exploit radical changes in their overall transcriptional signaling circuitry and that such global perturbations can serve as substrate for evolution in nature as well as in the test tube.38 While the above-described studies mutated or rewired global transcription signals in a rather random way followed by selection for desired phenotypes, a recent study by Collins applied a targeted global engineering approach to attenuate antibiotic resistance in bacteria.39 Based on the observation that bactericidal antibiotics elicit the formation of radicals finally exerting a cytotoxic effect,41 it was shown that antibiotic-mediated bacterial killing can be enhanced by overexpression of lexA3, a repressor of the bacterial SOS defense program.42 Rather than inactivating the SOS response by a classic pharmacologic approach through the design of small-molecule inhibitors (e.g. to recA), Lu and Collins used a synthetic biology approach by designing a therapeutic M13-derived bacteriophage infecting the bacteria and overexpressing lexA3.39 It was shown that the phage-mediated introduction of the synthetic lexA3 expression cassette enhanced the killing of E. coli by 2.7 orders of magnitude in the presence of the quinolone antibiotic loxacin compared to the wild-type phage. The engineered phage showed a similar adjuvant effect on antibiotic-mediated killing of E. coli in infected mice resulting in an 80% survival rate of the animals compared to only 20% and 50% in the absence of the phage or the presence of the wild-type phage, respectively. The phage-based approach is not only suitable to overexpress one central regulator of a transcriptional signaling network, it can as well be applied for simultaneously activating or inactivating multiple pathways as exemplified by the coordinated phage-mediated expression of csrA, a global repressor of biofilm formation and ompF, increasing drug penetration into the cell. Simultaneous targeting of both genes in the presence of loxacin showed an enhanced bactericidal effect compared to phages targeting only a single gene. Such phages engineered for interference with host signaling networks might be used in industrial, agricultural, and food processing, where bacterial biofilms or other difficult-to-clear bacteria are present and these applications might be a first step in a future evaluation towards clinical use.39 Synthetic signaling processes based on protein–protein interactions Extracellular stimuli control intracellular signal transduction pathways that trigger cell growth, function, differentiation and death (see recent review articles8,43). Intracellular signal transduction is commonly mediated by protein phosphorylation or dephosphorylation in a highly specific and regulated way. Many signal transduction proteins share a modular design comprising an output domain with kinase or phosphatase activity as well as an input domain which links the activity of the output domain to upstream signals. The input domain can, for example, recruit activators for the output domain, determine the subcellular localization of the signaling protein or act as inhibitors of the output domain, where inhibition is only relieved upon binding to an input signal.8 The modular design of the signal transduction proteins enables the synthetic reconnection of one input domain to another output domain for redirecting signals from one pathway to another. In an exemplary study, the Lim group44 constructed a chimeric guanine nucleotide exchange factor (GEF) where a protein kinase A (PKA)-sensitive input module was fused to a Cdc42-derived GTPase output module. Therefore, the Dbl homology-pleckstrin homology (DH-PH) catalytic core from intersectin, a Cdc42-specific Dbl family member, was fused to the carboxy-terminus of the synthrophin PDZ domain and to the amino-terminus of a short peptide which binds to the PDZ domain and is close to the consensus sequence of the PKA substrate (Fig. 3A). It was shown that binding of PDZ to the peptide sequence repressed the DH-PH output module whereas PKA-mediated phosphorylation of the peptide inhibited binding of PDZ and triggered activation of the Cdc42-specific pathway, finally resulting in the formation of filopodia in the REF 52 fibroblast cell line44 (Fig. 3A). Fig. 3 Engineered signaling pathways. (A) Rewired pathway connectivity. The Cdc42 GTPase output module that triggers morphological changes was reconnected to the protein kinase A (PKA) signaling pathway. Therefore, the Dbl homology-pleckstrin homology (DH-PH) catalytic core from intersectin, a Cdc42-specific Dbl family member, was fused to PDZ and to a short PDZ-binding peptide (circle) that resembles the PKA substrate consensus sequence. The interaction of PDZ with its binding partner inactivates DH-PH. Upon PKA activation (e.g. by forskolin), the short peptide is phosphorylated, PDZ-binding is inactivated and DH-PH becomes active thereby triggering the Cdc42-specific morphological response. (B) Engineered circuit topology in mitogen activated protein kinase signaling. In the native circuit, α-factor triggers activation of the Ste11-Ste7-Fus3 MAP kinase pathway via the pheromone receptor (R) and intermediate Ste factors. The positive feedback loop was engineered by placing a fusion of ste50 to a leucine zipper (Leu-zip) under the control of the mating-dependent promoter (P) and by fusing a second leucine zipper to the adaptor protein Ste5. α-factor-triggered pathway activation resulted in production of Leu-zip-Ste50, which then binds to Ste5 and mediates pathway activation by promoting interaction of Ste20 with Ste11. Ultrasensitivity was achieved by constitutively expressing a fusion of the pathway inhibitor Msg5 with a leucine zipper domain for inhibiting signal propagation. Initial pathway activation produces Leu-zip-Ste50, which then competes for Ste5-binding with the inhibitor Msg5. Combination of the Ste50-based positive feedback loop with the inhibitor/activator competition trigger an ultrasensitive signal output. In a similar approach, the group of Lim engineered the circuit topology of the α-factor-triggered mating pathway in yeast relying on mitogen activated protein (MAP) kinase signaling45 (Fig. 3B). Proper connectivity of the MAP kinase pathway requires the protein Ste5 serving as scaffold for anchoring the MAP kinase kinase kinase Ste11, the MAP kinase kinase Ste7 and the MAP kinase Fus3 for ensuring successive phosphorylation and activation along the signal transduction chain. Significantly higher pathway output was observed by attaching the scaffold Ste5 via a leucine zipper pair to Ste50 that promotes activation of Ste11. By placing expression of leucine zipper-tagged ste50 under the control of a mating-dependent promoter, a positive feedback loop was introduced as pathway activation results in Ste50 production that then further activates the pathway via Ste11 (Fig. 3B). While the original pathway showed a rheostat-like output signal in response to increasing α-factor concentrations (Hill coefficient ∼1), the pathway engineered for the positive feedback loop displayed switch-like characteristics with a Hill coefficient of ∼2.4. This configuration was the basis for the construction of an ultrasensitive pathway that was obtained by constitutively expressing Msg5 fused to a leucine zipper to bind to the scaffold Ste5 and inhibit signal propagation. Upon initial pathway activation by α-factor, a leucine-zipper-Ste50 fusion was produced under the control of the mating promoter (see above). The leucine-zipper-Ste50 fusion competitively replaced the inhibitor Msg5 and further activated the pathway via Ste11. This relief of Msg5-mediated inhibition together with the concomitant Ste50-based pathway activation increased the Hill factor to 2.8 characterizing the ultrasensitive signal response45 (Fig. 3B). While most natural signaling processes rely on phosphorylation or dephosphorylation to control signal amplification and propagation, an orthogonal signaling cascade has recently been constructed in mammalian cells, where covalent biotinylation of a transcription factor mediated signal transduction.46 In this configuration, the biotin concentration in the medium served as input which was perceived by E. coli biotin ligase BirA to biotinylate the Herpes simplex viral protein 16-derived transactivation domain (VP16) coupled to a synthetic biotinylation peptide (Avitag). Similar to the dimerization of phosphorylated proteins with SH2 domains,47 biotinylated VP16 dimerized with streptavidin which was fused to the tetracycline repressor TetR. Dimerization of VP16 with TetR via the biotin–streptavidin linkage triggered activation of the tetracycline-responsive promoter PTET harboring TetR-specific operator sites (Fig. 4). In the absence of biotin, gene expression remained silent (Fig. 4A). Upon administration of a short biotin pulse, gene expression was induced (Fig. 4B) and remained induced even after withdrawal of the input signal biotin from the medium (time delayed expression, Fig. 4C). Biotinylated VP16 was slowly degraded by the proteasome thereby limiting the duration of time delayed gene expression and resulting in a final decline of transgene expression (Fig. 4D). The duration of time delayed gene expression (Fig. 4E) was adjusted by engineering the stability of biotinylated VP16 via coupling to stabilizing (fluorescent proteins) or destabilizing (PEST sequence) protein domains (Fig. 4F) or by varying the concentration of biotinylated VP16 in the cell (Fig. 4G). Such synthetic time-delayed expression kinetics might be applied in the reconstruction and reprogramming of natural circuits involving time-delayed signaling like NF-kB activation48 or the circadian clock.49 Fig. 4 Biotinylation-based signal transduction at the example of a biotin-triggered time-delay circuit. In the absence of biotin (A) gene expression from the tetracycline-dependent promoter (PTET) is silent. Upon biotin addition (B) the biotinylation signal-containing transactivation domain VP16 is biotinylated by E. coli biotin ligase BirA and dimerizes with streptavidin (SA). Streptavidin was fused to the tetracycline-repressor (TetR) which binds its cognate promoter PTET. VP16 attracts components of the transcription initiation complex and activates the promoter resulting in transcription of the gene of interest (GOI). Upon withdrawal of biotin from the medium (C), biotin remains attached to VP16 and time delayed gene expression is sustained until the pool of biotinylated VP16 has been degraded by the proteasome (P). As soon as all biotinylated VP16 is degraded (D) gene expression is de-activated again. The time delay (time span of induced gene expression after biotin withdrawal (E)) is a function of the stability (F) as well as of the concentration (G) of biotinylated VP16. Synthetic inter-cell signaling processes Signaling systems for exchanging information between different cells is fundamental for the development of any higher multicellular organism, for coordinating the metabolic activity within a bacterial population by quorum-sensing as well as for establishing symbiotic or parasitic interactions across kingdoms.50–53 Individual bacterial cells synchronize their physiology with each other by quorum-sensing signals thereby emulating a multicellular entity and increasing the chances of survival in a complex environment.54 So far, three quorum-sensing mechanisms have been described in bacteria relying on oligopeptides, acyl homoserine lactones (AHLs) and autoinducer-2. Oligopeptides are mainly used by Gram-positive bacteria, AHLs represent species–specific signaling molecules for Gram-negative bacteria and autoinducer-2 is universally used for intra- and interspecies signaling.54 For the construction of synthetic signaling systems in bacteria mainly AHLs have been employed. AHLs can be synthesized by one enzyme (e.g. LuxI), they diffuse freely between the cells and modulate DNA-binding activity of a transcription factor (e.g. LuxR) for inducing a transcriptional response. The first synthetic communication systems were developed in bacteria for population control, where increasing population densities translated into high AHL levels that triggered the activation of a suicide gene,55 which resulted in oscillating population densities as revealed by single cell analysis in a microfluidic system.56 In an alternative study, a gradient of AHL produced by a colony of sender cells was perceived by a genetic band-pass filter and translated into a ring-shaped gene expression pattern around the sender cell population.57 As these signaling systems are reviewed in detail elsewhere,58–60 we focus here on recent developments in synthetic cell-to-cell signaling systems relying on multiple orthogonal communication channels in E. coli or signaling between higher organisms like yeast and mammalian cells. In order to enable cell-to-cell signaling in eukaryotes, Chen and Weiss adapted Arabidopsis thaliana signaling elements to function in S. cerevisiae.61 Yeast sender cells were engineered to express adenylate isopentenyl transferase (IPT) catalyzing the isopentenylation of ATP and ADP resulting in isopentenyladenine, a plant cytokinin involved in growth and development, which diffuses through the cell membrane into the surrounding environment. The cytokinin signal was perceived by receiver cells engineered to express the A. thaliana-derived cytokinin receptor AtCRE1 which phosphorylates endogenous yeast YPD1, a signaling kinase phosphorylating SKN7, which finally activates a promoter harboring an SKN7 response element (PSSRE) resulting in the expression of the green fluorescent protein (GFP) as reporter. The signaling system was validated by placing sender cells on a paper disk on top of a lawn of receiver cells followed by fluorescence imaging. A gradient of GFP expression was observed with highest expression close to the sender cells and gradually decreasing expression with increasing distance between senders and receivers. In an alternative configuration, the sender and the receiver modules were placed into the same cell yielding GFP expression only at high cell densities thus representing the first synthetic yeast quorum-sensing system.61 While bacteria-derived quorum-sensing molecules containing a lactone group represent ideal tools for the construction of orthogonal communication systems in prokaryotes, their transfer to mammalian cells to construct cell-to-cell communication systems has so far been hampered possibly by the relative instability of the lactone group in the cell culture environment. This limitation was overcome by the design of a cell-to-cell signaling system (a \"cell phone”) based on the messenger molecule acetaldehyde showing highly suitable characteristics for constructing synthetic signaling devices (Fig. 5A). The production of the messenger molecule could easily be performed by alcohol dehydrogenase in a one-step enzymatic reaction in a wide range of organisms like mammalian cells, plants, yeasts or bacteria62 with signaling-effective acetaldehyde concentrations below the no observable effect level (NOEL63) thereby reducing the risk of pleitropic side effects. Acetaldehyde-based signaling can be performed between cells in the same culture vessel, but it can also be used to connect cells in separated cultures via the gas phase (boiling point of acetaldehyde: 21 °C) thereby enabling the design of communication systems between cells requiring different culture media or culture conditions. Conversion of an acetaldehyde signal into a transcription readout is performed by the Aspergillus nidulans-derived transcription factor AlcR, which, in the presence of acetaldehyde, activates the synthetic promoter PAIR assembled from an AlcR operator site and a viral minimal promoter.62 Fig. 5 Synthetic cell-to-cell signaling systems. (A) Acetaldehyde-based cell-to-cell signaling. Sender cells were metabolically engineered to produce alcohol dehydrogenase which converts ethanol in the medium into acetaldehyde. The signal molecule acetaldehyde with a boiling point of 21 °C diffuses via gas or liquid phase to the receiver cells where it dose-dependently interacts with the Aspergillus nidulans-derived transcription factor AlcR to activate its cognate promoter PAIR controlling expression of the gene of interest (GOI). (B) Electro-genetic signaling system. A microelectrolysis chamber serves as sender unit where ethanol is electrochemically oxidized to the signal molecule acetaldehyde. Acetaldehyde is subsequently perceived by the receiver cells engineered for acetaldehyde-inducible gene expression (see (A)). (C) Ethylene-based signaling. The sender device (an apple in this example) produces ethylene that diffuses as signal molecule via the gas phase to a palladium chloride-based Wacker–Tsuji catalyst, where it is oxidized to acetaldehyde. Acetaldehyde dose-dependently triggers gene expression in the receiver cells (see (A)). (D)L-arginine-based signaling. Sender cells were genetically engineered to degrade the signal molecule L-arginine in the medium by arginase. The L-arginine concentration in the medium is dose-dependently translated into GOI expression by the receiver cells. Receiver cells harbor the transcription factor ARG2 which, in the presence of L-arginine, binds to and activates its cognate promoter PART1 to induce the GOI. The combination of a mammalian sender cell engineered for alcohol dehydrogenase expression and a receiver cell harboring an acetaldehyde-inducible expression system can be used to design cell density-, time- and distance-dependent synthetic multicellular systems which could for example be applied in the design of cell density-dependent bioprocesses64 or the synthetic reconstruction of gradient-based differentiation processes.62 An extension to the acetaldehyde-based cell phone was the recent establishment of electro-genetic devices, where gene expression is controlled by an electric signal65 (Fig. 5B). As input interface, a microelectrolysis chamber was designed, where ethanol was electrochemically oxidized into acetaldehyde that activated gene expression in cells engineered for the synthetic acetaldehyde-inducible promoter. This input interface was applied to design an electro-genetic frequency generator, where the intensity of the input current correlated with the frequency of the electric output signal. Using this configuration, neonatal rat cardiomyocytes were transduced with an acetaldehyde-inducible expression unit for bone morphogenetic protein 2 (BMP-2) and cultivated onto a microelectrode array for recording electrogenic cell beating. The cell-containing microelectrode array was placed adjacent to the microelectrolysis chamber. Increasing input currents in the electrolysis chamber correlated with increasing acetaldehyde production that triggered a dose-dependent increase in BMP-2 expression. Increasing BMP-2 levels dose-dependently increased the beating frequency of the cardiomyocytes, an electrogenic phenomenon that was read out and amplified by the microelectrode array to generate an oscillating electric output signal. This example demonstrates how an appropriate input (microelectrolysis chamber) and output (microelectrode array) interface can be used to plug mammalian cells engineered for a synthetic signaling network directly onto an electronic switchboard, where they can perform basic computational functions as a \"cellular processing unit” (CPU).65 The acetaldehyde-based signaling system was further connected to alternative input signals by the use of appropriate signal converters.66 In a recent study, ethylene was used as trigger signal, which was dose-dependently converted into acetaldehyde by a Wacker–Tsuji catalyst (Fig. 5C). In this configuration mammalian cells engineered for acetaldehyde-inducible reporter gene expression were used to genetically readout the ripening state of apples by differentially measuring the ethylene (via Wacker–Tsuji catalyst, as indicator for the ripening process) and acetaldehyde (without Wacker–Tsuji catalyst, as indicator for microorganism-based apple spoilage) emanating from the fruit. High ethylene-to-acetaldehyde ratios where shown to correlate with ripening apples whereas high acetaldehyde-to-ethylene ratios indicated that the apples were spoiled.66 While this signaling cascade from apples to mammalian cells represents a proof-of-concept for the design of synthetic multi-step signaling cascades for reading out the physiologic status of a sender organism, such concepts could potentially be used to design autonomous control units triggering a counter response to a pathologic state (e.g. spoilage by microorganisms) for example by the activation of appropriate corrective genes (e.g. antimicrobial peptides or enzymes). Recently, a second mammalian-cell compatible signaling system has been described based on the signal-transducing molecule L-arginine, the concentration of which can be modulated by sender cells genetically engineered for arginase production67 (Fig. 5D). The L-arginine level in the medium is detected by an L-arginine-inducible synthetic promoter and converted into a transcriptional readout. The construction of sender cell lines showing differential arginase expression levels and of receiver cells engineered for different output genes resulted in a palette of modular \"cytobricks” (a whole-cell biobrick) which were connected to each other via the L-arginine messenger and enabled the construction of multicellular circuits for example for adjustable time-delayed signal processing by simply exchanging the modular cytobrick components.67 Ecosystem-wide synthetic signaling processes Synthetic cell-to-cell signaling devices (cell phones) enable the assembly of synthetic ecosystems to study fundamental ecologic mechanisms in the co-existence of species. Several examples of synthetic ecosystems have recently been described,62,68–70 two of which are exemplarily illustrated below. Interactions between a predator and a prey are one of the most heavily investigated species interactions.71 A recent study reconstructed such predator–prey interactions in a chemostat reactor populated by predators and prey engineered from E. coli.69 In the synthetic predator–prey system (Fig. 6A), the predator produced the quorum-sensing molecule 3OC12HSL (3-oxo-C12-homoserine lactone) that diffused to the prey cells, bound to the transcriptional activator LasR and finally induced expression of the ccdB toxine gene. The prey cells produced 3OC6HSL (3-oxo-C6-homoserine lactone), which diffused to the predator population and bound to the activator LuxR thereby inducing expression of the antitoxin ccdA which neutralized the cytotoxic activity of the toxin ccdB expressed from a separate promoter. Thus, predator cells can only survive in the presence of 3OC6HSL-producing prey cells while prey cells die in the presence of 3OC12HSL-producing predator E. coli. In a chemostat co-culture, prey cells first expanded while predator cell numbers declined, however, at higher prey cell concentrations, predators started to multiply thereby triggering a decline in the prey population. Decimated prey populations entailed a decrease in the predator density which started the cycle from the beginning. Such oscillating population densities are also commonly observed in nature thereby indicating, that the synthetic ecosystem reflects well the kinetic properties of its natural archetype.69 Fig. 6 Synthetic ecosystems. (A) Synthetic E. coli-based predator–prey ecosystem. The predator cell produces the signal molecule 3OC12HSL that diffuses to the prey cells and binds to the transcription factor LasR which subsequently induces activation of its cognate promoter PLuxI. Activated PLuxI triggers expression of the toxin ccdB resulting in cell death. The prey cells produce the signal molecule 3OC6HSL which diffuses to the predator cells where it binds LuxR and triggers activation of the promoter PLuxI. Activated PLuxI triggers production of the antitoxin ccdA which binds to constitutively expressed ccdB and neutralizes its cytotoxic action thereby enabling survival of the predator cells. The synthetic predator and prey E. coli were cultured in a chemostat bioreactor and showed oscillating population densities reminiscent of naturally occurring predator and prey populations. (B) Synthetic parasite–host ecosystem. Host cells were constructed by engineering Chinese hamster ovary cells for constitutive expression of beta-lactamase fused to a mammalian cell-compatible Igκ-derived secretion signal (secreted beta-lactamase, sBLA). sBLA degraded ampicillin in the medium thereby enabling growth of parasite E. coli cells which ultimately compromised host cell growth due to nutrient consumption and production of toxic by-products. When cultivating both cell types with a semi-continuous supply of fresh medium and ampicillin, host cells first grow, degrade ampicillin and subsequently enable growth of the parasite. High parasite populations trigger a decrease in host cells, lower sBLA production and accumulation of ampicillin. High ampicillin concentrations decimate the parasites, which enables recovery of the host and the cycle starts again from the beginning. Such oscillating population densities are reminiscent of naturally occurring parasite–host interactions. In a similar approach, a parasite–host ecosystem has been constructed between an E. coli parasite and a mammalian Chinese hamster ovary (CHO) host cell engineered for the production of a secreted beta-lactamase62 (Fig. 6B). In a semi-continuous co-culture in the presence of ampicillin, CHO cells were first growing, producing beta-lactamase and degrading ampicillin, which subsequently enabled growth of the parasite E. coli. Growing E. coli consumed nutrients and produced toxic by-products that compromised growth of CHO host cells and triggered a decline in the mammalian cell population. Washing out of beta-lactamase and addition of ampicillin with the fresh medium triggered a decline in the E. coli population, thereby enabling the recovery of CHO cells and a restart of the cycle62 (Fig. 6B). These examples demonstrate how synthetic cell-to-cell signaling systems can be used to reconstruct the interactions between multiple species in a test tube in order to investigate fundamental ecological mechanisms. Conclusions and outlook Despite its only recent emergence, synthetic biology has already become an integral part in the analysis and engineering of signaling systems. On the one side, synthetic biology strategies serve in elucidating the functionality and pathology of signaling phenomena to understand the underpinnings of living systems as well as the molecular basis of disease. On the other side, synthetic biology strategies have been implemented in the design of synthetic signaling elements at the transcriptional, translational, protein and cellular level thereby enabling the de novo design of synthetic signaling circuits for achieving desired cell phenotypes. Extrapolation of the previous impact of synthetic biology on signaling science to the future suggests that significant advances will be achieved in the following areas: (i) The elaboration of synthetic biology tools in different host organisms will provide a palette of orthogonal backgrounds for the reconstruction and analysis of natural or pathologic signaling circuits as already demonstrated today for reconstructed B cell signaling in Drosophila cells13 or the discovery of antituberculosis compounds.22 Another suitable host for studying mammalian signaling systems might be the moss Physcomitrella patens which is evolutionary even more distant than insect cells. In a recent study72 it has been shown that mammalian cell-optimized transcriptional and translational elements like promoters (inducible and constitutive), internal ribosome entry sites as well as secretion signals are generally functional in the moss suggesting that it could be possible to directly transfer complex mammalian signaling circuits into the orthogonal moss host. (ii) The design of synthetic signaling circuits is expected to be significantly facilitated in the near future. While biobrick components for designing signaling circuits at the transcriptional, translational and inter-cell level in the different host types are available now (see for example http://partsregistry.org), it still remains a challenge to rationally design signaling cascades with predictable kinetic behavior. In order to make the assembly of biologic signaling networks more predictive, synthetic biologists apply mathematical tools and models to characterize individual biologic parts and to simulate their functional integration into the overall signaling network. A number of software tools (reviewed recently elsewhere73) have been developed which are directly connected to the biobrick databases and enable the design of genetic circuits in a \"drag and drop” manner reminiscent to design software in electronic engineering, where complex electronic circuits can be assembled and modeled in silico. As the precision of such predictions is a direct function of the quality of the input data, it is of key importance to have available well characterized biobricks with tunable activity for seamless integration into network topology. Two recent studies point out how the design of synthetic networks could be made more predictable, either by combining libraries of diversified components with in silico modeling to guide predictable gene network construction without the need for post hoc tweaking74 or by benchmarking reverse-engineering and modeling approaches.75 In an alternative approach the Voigt group developed a thermodynamic model of bacterial translation initiation and used this model to predict the efficiency of translation initiation as a function of the ribosome binding site motif.76 The experimental validation of more than 100 predictions in E. coli demonstrated accuracy within a factor of 2.3 over a total range of 100000-fold. This thermodynamic approach together with the library-based strategies74 should be broadly applicable to the thorough characterization of biobricks and thus significantly facilitate the software-based forward engineering of large genetic systems. (iii) The analysis of reconstructed or de novo designed signaling pathways requires sensors for visualizing signaling events. As most signals are transmitted via transient spatial proximity of two partners, sensors for protein heterodimerization have gained momentum in analyzing signaling events. Examples include FRET- and BRET-based techniques as well as split-protein approaches,5,6,77 where the heterodimerization of two partners reconstituted a functional protein (e.g. luciferase, beta-galactosidase, fluorescent proteins) that could easily be localized and quantified. While such dimerization-based approaches are routinely used today, other signal detecting devices, for example real-time assays for the in vivo localization and quantification of the activity of specific kinases or phosphatases, still need to be developed and would represent highly valuable tools for the time- and space-resolved analysis of signaling processes. These future directions, the establishment and validation of orthogonal backgrounds for analyzing natural signaling processes, the development of design strategies for the rational and predictive assembly of signaling cascades as well as the development of online signal sensors are expected to emerge within the near future and will provide significant options to understand and design signaling in order to provide novel opportunities in understanding and improving human health. Acknowledgements We thank Maria Karlsson and Erik Christen for critical comments on the manuscript. Work of MMK is supported by the GEBERT RÜF STIFTUNG (grant GRS-042/07) and work of WW is supported by the Excellence Initiative of the German Federal and State Governments (EXC 294). References T. C. He, A. B. Sparks, C. Rago, H. Hermeking, L. Zawel, L. T. da Costa, P. J. Morin, B. Vogelstein and K. W. Kinzler, Science, 1998, 281, 1509–1512 CrossRef CAS. J. Massague, S. W. Blain and R. S. Lo, Cell, 2000, 103, 295–309 CrossRef CAS. P. J. Morin, A. B. Sparks, V. Korinek, N. Barker, H. Clevers, B. Vogelstein and K. W. Kinzler, Science, 1997, 275, 1787–1790 CrossRef CAS. P. E. Kovanen and W. J. Leonard, Immunol. Rev., 2004, 202, 67–83 CrossRef CAS. C. D. Hu and T. K. Kerppola, Nat. Biotechnol., 2003, 21, 539–545 CrossRef CAS. M. Walter, C. Chaban, K. Schutze, O. Batistic, K. Weckermann, C. Nake, D. Blazevic, C. Grefen, K. Schumacher, C. Oecking, K. Harter and J. Kudla, Plant J., 2004, 40, 428–438 CrossRef CAS. J. C. Clemens, C. A. Worby, N. Simonson-Leff, M. Muda, T. Maehama, B. A. Hemmings and J. E. Dixon, Proc. Natl. Acad. Sci. U. S. A., 2000, 97, 6499–6503 CrossRef CAS. P. M. Pryciak, Chem. Biol., 2009, 16, 249–254 CrossRef CAS. W. Weber and M. Fussenegger, Chem. Biol., 2009, 16, 287–297 CrossRef CAS. M. N. Win, J. C. Liang and C. D. Smolke, Chem. Biol., 2009, 16, 298–310 CrossRef CAS. K. C. Nicolaou, D. Vourloumis, N. Winssinger and P. S. Baran, Angew. Chem., Int. Ed., 2000, 39, 44–122 CrossRef CAS. T. Wossning and M. Reth, Immunol. Lett., 2004, 92, 67–73 CrossRef CAS. V. Rolli, M. Gallwitz, T. Wossning, A. Flemming, W. W. Schamel, C. Zurn and M. Reth, Mol. Cell, 2002, 10, 1057–1069 CrossRef CAS. G. M. Siegers, J. Yang, C. U. Duerr, P. J. Nielsen, M. Reth and W. W. Schamel, Int. Immunol., 2006, 18, 1385–1396 CrossRef CAS. S. Ferrari, V. Lougaris, S. Caraffi, R. Zuntini, J. Yang, A. Soresina, A. Meini, G. Cazzola, C. Rossi, M. Reth and A. Plebani, J. Exp. Med., 2007, 204, 2047–2051 CrossRef CAS. M. Isalan, C. Lemerle and L. Serrano, PLoS Biol., 2005, 3, e64 CrossRef. R. Rivera-Pomar and H. Jackle, Trends Genet., 1996, 12, 478–483 CrossRef CAS. W. Driever and C. Nusslein-Volhard, Cell, 1988, 54, 95–104 CrossRef CAS. W. Driever and C. Nusslein-Volhard, Cell, 1988, 54, 83–93 CrossRef CAS. L. Sanchez and D. Thieffry, J. Theor. Biol., 2001, 211, 115–141 CrossRef CAS. J. Jaeger, M. Blagov, D. Kosman, K. N. Kozlov, Manu, E. Myasnikova, S. Surkova, C. E. Vanario-Alonso, M. Samsonova, D. H. Sharp and J. Reinitz, Genetics, 2004, 167, 1721–1737 CrossRef CAS. W. Weber, R. Schoenmakers, B. Keller, M. Gitzinger, T. Grau, M. Daoud-El Baba, P. Sander and M. Fussenegger, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 9994–9998 CrossRef CAS. J. Engohang-Ndong, D. Baillat, M. Aumercier, F. Bellefontaine, G. S. Besra, C. Locht and A. R. Baulard, Mol. Microbiol., 2004, 51, 175–188 CrossRef CAS. F. Wang, R. Langley, G. Gulten, L. G. Dover, G. S. Besra, W. R. Jacobs, Jr. and J. C. Sacchettini, J. Exp. Med., 2007, 204, 73–78 CrossRef CAS. A. R. Baulard, J. C. Betts, J. Engohang-Ndong, S. Quan, R. A. McAdam, P. J. Brennan, C. Locht and G. S. Besra, J. Biol. Chem., 2000, 275, 28326–28331 CAS. M. N. Win and C. D. Smolke, Science, 2008, 322, 456–460 CrossRef CAS. T. S. Gardner, C. R. Cantor and J. J. Collins, Nature, 2000, 403, 339–342 CrossRef CAS. A. E. Friedland, T. K. Lu, X. Wang, D. Shi, G. Church and J. J. Collins, Science, 2009, 324, 1199–1202 CrossRef CAS. M. R. Atkinson, M. A. Savageau, J. T. Myers and A. J. Ninfa, Cell, 2003, 113, 597–607 CrossRef CAS. M. B. Elowitz and S. Leibler, Nature, 2000, 403, 335–338 CrossRef CAS. J. Stricker, S. Cookson, M. R. Bennett, W. H. Mather, L. S. Tsimring and J. Hasty, Nature, 2008, 456, 516–519 CrossRef CAS. T. L. Deans, C. R. Cantor and J. J. Collins, Cell, 2007, 130, 363–372 CrossRef CAS. B. P. Kramer, M. Fischer and M. Fussenegger, Metab. Eng., 2005, 7, 241–250 CrossRef CAS. B. P. Kramer, W. Weber and M. Fussenegger, Biotechnol. Bioeng., 2003, 83, 810–820 CrossRef CAS. M. Tigges, T. T. Marquez-Lago, J. Stelling and M. Fussenegger, Nature, 2009, 457, 309–312 CrossRef CAS. D. Greber and M. Fussenegger, J. Biotechnol., 2007, 130, 329–345 CrossRef CAS. H. Alper, J. Moxley, E. Nevoigt, G. R. Fink and G. Stephanopoulos, Science, 2006, 314, 1565–1568 CrossRef CAS. M. Isalan, C. Lemerle, K. Michalodimitrakis, C. Horn, P. Beltrao, E. Raineri, M. Garriga-Canut and L. Serrano, Nature, 2008, 452, 840–845 CrossRef CAS. T. K. Lu and J. J. Collins, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 4629–4634 CrossRef CAS. H. Alper and G. Stephanopoulos, Metab. Eng., 2007, 9, 258–267 CrossRef CAS. M. A. Kohanski, D. J. Dwyer, B. Hayete, C. A. Lawrence and J. J. Collins, Cell, 2007, 130, 797–810 CrossRef CAS. R. T. Cirz, J. K. Chin, D. R. Andes, V. de Crecy-Lagard, W. A. Craig and F. E. Romesberg, PLoS Biol., 2005, 3, e176 CrossRef. T. Pawson, Cell, 2004, 116, 191–203 CrossRef CAS. B. J. Yeh, R. J. Rutigliano, A. Deb, D. Bar-Sagi and W. A. Lim, Nature, 2007, 447, 596–600 CrossRef CAS. C. J. Bashor, N. C. Helman, S. Yan and W. A. Lim, Science, 2008, 319, 1539–1543 CrossRef CAS. W. Weber, J. Stelling, M. Rimann, B. Keller, M. Daoud-El Baba, C. C. Weber, D. Aubel and M. Fussenegger, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 2643–2648 CrossRef CAS. T. Pawson and M. Kofler, Curr. Opin. Cell Biol., 2009, 21, 147–153 CrossRef CAS. M. W. Covert, T. H. Leung, J. E. Gaston and D. Baltimore, Science, 2005, 309, 1854–1857 CrossRef CAS. D. B. Forger and C. S. Peskin, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 14806–14811 CrossRef CAS. J. E. Cooper, J. Appl. Microbiol., 2007, 103, 1355–1365 CrossRef CAS. H. Li, L. Wang, L. Ye, Y. Mao, X. Xie, C. Xia, J. Chen, Z. Lu and J. Song, Med. Microbiol. Immunol., 2009, 198, 113–121 CrossRef CAS. C. T. Parker and V. Sperandio, Cell. Microbiol., 2009, 11, 363–369 CrossRef CAS. R. F. Fisher and S. R. Long, Nature, 1992, 357, 655–660 CrossRef CAS. S. Hooshangi and W. E. Bentley, Curr. Opin. Biotechnol., 2008, 19, 550–555 CrossRef CAS. L. You, R. S. Cox, 3rd, R. Weiss and F. H. Arnold, Nature, 2004, 428, 868–871 CrossRef CAS. F. K. Balagadde, L. You, C. L. Hansen, F. H. Arnold and S. R. Quake, Science, 2005, 309, 137–140 CrossRef. S. Basu, Y. Gerchman, C. H. Collins, F. H. Arnold and R. Weiss, Nature, 2005, 434, 1130–1134 CrossRef CAS. K. Brenner, L. You and F. H. Arnold, Trends Biotechnol., 2008, 26, 483–489 CrossRef CAS. P. E. Purnick and R. Weiss, Nat. Rev. Mol. Cell Biol., 2009, 10, 410–422 CrossRef CAS. D. Sprinzak and M. B. Elowitz, Nature, 2005, 438, 443–448 CrossRef CAS. M. T. Chen and R. Weiss, Nat. Biotechnol., 2005, 23, 1551–1555 CrossRef CAS. W. Weber, M. Daoud-El Baba and M. Fussenegger, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 10435–10440 CrossRef CAS. W. Weber, M. Rimann, M. Spielmann, B. Keller, M. Daoud-El Baba, D. Aubel, C. C. Weber and M. Fussenegger, Nat. Biotechnol., 2004, 22, 1440–1444 CrossRef CAS. W. Weber and M. Fussenegger, Curr. Opin. Biotechnol., 2007, 18, 399–410 CAS. W. Weber, S. Luzi, M. Karlsson, C. D. Sanchez-Bustamante, U. Frey, A. Hierlemann and M. Fussenegger, Nucleic Acids Res., 2009, 37, e33. W. Weber, S. Luzi, M. Karlsson and M. Fussenegger, J. Biotechnol., 2009, 139, 314–317 CrossRef CAS. W. Weber, M. Schuetz, N. Denervaud and M. Fussenegger, Mol. BioSyst., 2009, 5, 757–763 RSC. H. J. Kim, J. Q. Boedicker, J. W. Choi and R. F. Ismagilov, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 18188–18193 CrossRef CAS. F. K. Balagaddé, H. Song, J. Ozaki, C. H. Collins, M. Barnet, F. H. Arnold, S. R. Quake and L. You, Mol. Syst. Biol., 2008, 4, 187. K. Brenner, D. K. Karig, R. Weiss and F. H. Arnold, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 17300–17304 CrossRef CAS. P. Chesson and J. J. Kuang, Nature, 2008, 456, 235–238 CrossRef CAS. M. Gitzinger, J. Parsons, R. Reski and M. Fussenegger, Plant Biotechnol. J., 2009, 7, 73–86 Search PubMed. M. A. Marchisio and J. Stelling, Curr. Opin. Biotechnol., 2009, 20, 479–485 CrossRef CAS. T. Ellis, X. Wang and J. J. Collins, Nat. Biotechnol., 2009, 27, 465–471 CrossRef CAS. I. Cantone, L. Marucci, F. Iorio, M. A. Ricci, V. Belcastro, M. Bansal, S. Santini, M. di Bernardo, D. di Bernardo and M. P. Cosma, Cell, 2009, 137, 172–181 CrossRef CAS. H. M. Salis, E. A. Mirsky and C. A. Voigt, Nat. Biotechnol., 2009, 27, 946–950 CrossRef CAS. J. Müller and N. Johnsson, ChemBioChem, 2008, 9, 2029–2038 CrossRef.