Advances in synthetic biology: on the path from prototypes to applications Ryan McDaniel and Ron Weiss Synthetic biology combines knowledge from various disciplines including molecular biology, engineering, mathematics and physics to design and build novel proteins, genetic circuits and metabolic networks. Early efforts aimed at altering the behavior of individual elements have now evolved to focus on the construction of complex networks in single-cell and multicellular systems. Recent achievements include the development of sophisticated non-native behaviors such as bi- stability, oscillations, proteins customized for biosensing, optimized drug synthesis and programmed spatial pattern formation. The de novo construction of such systems offers valuable quantitative insight into naturally occurring information processing activities. Furthermore, as the techniques for system design, synthesis and optimization mature, we will witness a rapid growth in the capabilities of synthetic systems with a wide-range of applications. Addresses Department of Molecular Biology, Princeton University, New Jersey 08544, USA Corresponding author: Weiss, Ron (rweiss@princeton.edu) Current Opinion in Biotechnology 2005, 16:476–483 This review comes from a themed issue on Protein technologies and commercial enzymes Edited by Bernhard Hauer and Brian K Kay Available online 12th July 2005 0958-1669/$ – see front matter # 2005 Elsevier Ltd. All rights reserved. DOI 10.1016/j.copbio.2005.07.002 Introduction Synthetic biology aims to create novel behaviors through the engineering of genetic elements and the integration of basic elements into circuits that implement more complex functions. Scientists have long viewed the beha- vior of complex biological systems as a function of the behavior of their constituent parts; for example, hybrid Boolean networks that consist of digital and analog logic elements [1–3]. Initial efforts in engineering synthetic elements focused on the development of novel transcrip- tional activators and repressors, often through the incor- poration of foreign transcription factor binding sites into promoter sequences [4,5]. Modifications to regulatory kinetics, such as transcription factor cooperativity and operator binding affinity, were used to develop elements with step-like digital responses that are robust to input noise [6]. In addition to having a collection of components that approximate digital behavior, synthetic biologists have also developed and utilized elements with fine- tuned analog characteristics (e.g. rheostat-like control). These engineering efforts have both helped to character- ize how specific residues or DNA-binding sites can alter biological activity and have created a library of useful components. To elicit more complex regulatory behaviors, elements can be combined into circuits, the simplest of which is the transcriptional cascade. Here, genes are arranged in series whereby each gene product regulates the expression of one downstream target (Figure 1a). In natural systems, evolution has already optimized the regulatory interac- tions between network elements in cascades and other network motifs to work cohesively towards achieving a particular behavior. In synthetic biology, networks are typically assembled from unrelated elements that have not been optimized by this evolutionary process. Hence, one of the main challenges in engineering synthetic circuits is altering the kinetics of individual elements until they are impedance-matched such that they func- tion correctly within the context of the new network [7,8  ]. Through the use of computational and directed evolution techniques that overcome such difficulties, several functional cascades have been built and analyzed as described below. Network motifs in synthetic systems Cascades are useful for studying the fundamental mechanisms of information flow in regulatory networks. They have a simple topology and, to a first approximation, their steady-state output is a direct monotonic function of the input. One useful property of certain signal transduc- tion cascades, such as the mitogen-activated protein kinase (MAPK) cascade in Saccharomyces cerevisiae, is that their steady-state behavior approximates digital logic with an ultrasensitive step-like dosage-response function [9]. Transcriptional cascades can also possess similar response properties. The analysis of synthetic transcriptional cas- cades of various lengths has shown that, under certain conditions, increasing the depth of the cascade sharpens the ultrasensitive response to a stimulus and makes the input/output function more digital [10](Figure 1b). Another important property of cascades is their dynamic behavior in response to both internal and environmental changes [11]. Through the construction of one- and two- step cascades, it was shown that the delayed response in transcriptional cascades is a function of cascade depth. In addition, long cascades can be robust to input noise by Current Opinion in Biotechnology 2005, 16:476–483 www.sciencedirect.com