If we can use the computer as a control, and if we can make a chemical device to act on the instructions as they come up—in fact, if we can make a D. N. A. synthesiser—then I think we can begin to build living tissue.

A for Andromeda (1962)

Related Topics
Synthetic biology

Synthetic Biology: a minimal function set

What would we need if we were to build a long-lived organism from scratch? This question is essential for the scaling up of synthetic biology processes (SynBio). It requires us to try to build a complete inventory of cellular functions, taking care not to forget those that are discrete but essential. To do this, we separate the cell into two independent components, the cell itself (the « chassis ») and its program. A general trend in synthetic biology has been to focus on the functions necessary for reproduction (i.e. growth). Now we need to look at what happens when cells just survive, and relate survival to longevity and death.

The most difficult part of synthetic biology is understanding the link between information and its material embodiment as a « chassis ». This is reflected in the two origins of biology: chemistry (as biochemistry) and genetics. But it is generally difficult to convey what genetics is and how it can be used as scientific evidence rather than biochemistry.

Here is a summary of the first workshop of an US-EU task force, focusing on synthetic genomes.

Report and conclusions of the US-EU Workshop on Synthetic Biology and Synthetic Genomes

Airlie House, April 24-25 (2006)

Summary of the US-EC Synthetic Biology workshop, Airlie House
Co-chairs: Maurice Lex (EU) and Marvin Stodolsky (DoE)


1) Biological research, empowerment by Synthetic Biology tools
2) The novel R&D arenas empowered by Synthetic Biology
3) Lessons already learned in the ELSI sector, with broad discussion on the novelties posed by Synthetic Biology
4) Novel gene constructs, genomes and their resource base
5) Discussions on issues of guarding against misuse of broadly available Synthetic Biology resources


Although the expression Synthetic Biology (SynBio) has been present in the scientific and technical literature as early as in 1904 (see La Biologie Synthétique*, by Stéphane Leduc), only in more recent times has it come to being as an umbrella concept to cover the whole of research developed at the interface between Molecular Biology and sensu stricto Engineering.  Synthetic Biology is becoming an increasingly inclusive concept, which [i] encompasses new theoretical frameworks that address biological systems with the conceptual tools and the descriptive language of Engineering,  [ii] addresses old questions and challenges with fresh approaches inspired in electric circuitry and mechanical manufacturing and [iii] pursues the creation of new materials with á la carte properties based on the rational combination of standardized biological parts decoupled from their natural context. In fact, standarization and detailed description of minimal biological parts and their interfaces, to the degree of reliability of the components of modern electronic circuits is one of the trademarks of the whole field.

The basic notion behind SynBio is that any biological system can be seen as a complex combination of functional, stand-alone elements not unlike those found in man-made devices, and can thus be–deconstructed in a limited number of components and reconstructed in an entirely different configuration for the sake of modifying existing properties or creating altogether new ones. In this context, Engineering as a discipline transits from being an analogy of the rational combination of genes made possible by modern Molecular Biology and Biotechnology to being a veritable methodology to construct complex systems and novel properties based on biological components. As any other domain of research, SB has an aspect of developing general-use technological and conceptual tools (biological parts, minimal genomes, artificial cells, DNA synthesis), addressing hitherto intractable problems (Biosynthesis of complex molecules, breakdown or recycling of toxic chemicals, biological detection of explosives, biological production of H2 and other fuels) and raising utterly novel challenges (DNA computing, design of biological pattern development, targeting bacteria to tumor cells, expanding the genetic code to non-natural amino acids).

Although many activities that would now qualify as Synthetic Biology have been going on for some time in Europe and the USA (protein design, modeling, metabolic engineering, biological nanomachines), it is only now that the immense potential of the field is recognized as one of the most promising pillars of the sustainable and competitive economy of the future. All in all, SynBio is not about understanding Biological Systems but about capitalizing such systems as a source of components for creating new devices and properties to solve a variety of problems. In that respect, SynBio maps altogether in the realm of Technology and thus clearly dissociates itself from basic Science –more interested to know and understand how existing Biological Systems work as they are.

It is in this context that an á la carte US-EC Workshop in Synthetic Biology and Synthetic Genomes was organized by the US Department of Energy in Airlie House (VA) in April 24-25 2006 under the auspices of the US-EC Task Force in Biotechnology. The meeting was attended by an even representation of EU and US scientists active in the field of SynBio, and counted on the valuable presence of observers from a range of US Agencies. While a more detailed account of the various sessions is presented below, the following four basic conclusions of the encounter follow as a plea to foster a vibrant transatlantic action in the field. 

• Synthesizing whole genomes. One of the persistent themes of SynBio is the possibility –not yet at hand, as it seems- to design genomes endowing cells with predetermined properties for a large number of applications. This endeavor first needs an operative description of a minimal but functional genome upon which DNA encoding the desired functions could be later inserted. But then, such a minimal genome has to be made operational by implantation in an existing biochemical milieu or by stepwise replacement of the genetic complement of a surrogate cell by the new chromosome and its further propagation. While these crucial issues are dealt with later in this Report, there is one upfront challenge that crossects all SynBio: the requirement for fast and accurate production and assembly of long segments of DNA, even to the size of a complete chromosome. The capacity of synthesizing DNA in the range of various megabases will make difference between those able to capitalize the amazing promise of SB and those that will remain as simple observers and mere users of the expertise.

• Biological parts. A second recurrent premise in SynBio is the vital need to standardize biological components and their interfaces, in a fashion detached from its natural circumstances. Context-independent behaviour of components is clearly a pre-requisite for the robust engineering new devices and properties. While the need and the opportunity of such a formatting have been clearly identified, the success of the endeavor has been quite limited so far. Despite the long list of biological parts entered in the open registry kindly supported by the MIT, only a few of them fulfill the requirements that would raise them to an standard barely comparable to e. g. transistors in electronic circuitry. This is in part due to the intrinsic qualities of biological functions to co-evolve as wholes (and thus behave in a very context-dependent manner), and as a result cannot be described in an easy manner. There is thus a need to develop more robust concepts and a dedicated language to deal and categorize such biological parts, which is based not only in their possible similarity to electronic counterparts, but also on a better comprehension of minimal biological functions—mostly related to regulation of gene expression. A community decision will have to be made between having a long list of ill-defined components or a shorter collection quite of robust and well characterized assets. These refinements can lay the basis of a future international agreement on the formatting of such parts, their availability and the registry of their users.

• Controlling noise. Even in the best-case scenario, building complex properties with biological components will not be simply a case of cutting and pasting DNA encoding parts. Unlike electronic circuits, where signals (i.e. current) can be faithfully channeled from one site of a given network to the other, biological circuits intrinsically rely on diffusible signal transmitters (proteins, nucleic acids, small molecules). This unavoidably produces noise (i.e., emission and reception of illegitimate signals) that, if not controlled, may end up destroying any artificial network. In fact, it is amazing how existing biological systems resist noise, surely the result of an extraordinary selective pressure. There is thus an extraordinary necessity to come up with ways of entering noise suppression devices in artificial biological circuits, an issue that has not been sufficiently addressed. Obviously, this may require some effort to better understand how noise is controlled in existing biological setups—as it may involve both the layout of the circuits as well as the physical structure and the possible compartmentalization of the intracellular milieu.

• Monitoring misuse. Finally, as any other new conceptual framework involving live elements or using biological materials with a commercial potential, SynBio might be the subject of a considerable public focus in which the fascination for the possible achievements is blended with the concerns for a possible misuse. While the predominant worry in the US is about the exploitation of SynBio for terrorist actions, the Europeans will more probably experience some anxieties about creating non-natural life forms, the possible effect of novel biological materials if released into the environment and—not the least—the contribution of SynBio to the economic globalization agenda.  Fortunately, the GMO debate has taught us some lessons on the correct handling of new technologies regarding the public perception. The bottom line in this case is to keep the debate and the public exposure of the technology to reasonable terms and staying away from false alarms. It is tempting for the SynBio community to bring public attention (and thus increased funding) to the field by using, inter alia, a provocative language (Engineering Life, Artificial Cells) which, while good for advertisement, will surely ring dreads—if not fantasies on possible wrongdoings. A re-enactment of the GMO controversy must be deliberately avoided for SynBio. While cultural sensitivities translate in different societal demands for action re. governance of SynBio activities, the one early action for some control on SynBio activities must be the obligatory monitoring of orders for long segments of DNA.

These four aspects of SynBio (DNA synthesis, Formatting Biological Parts, Noise Suppression and Misuse Control) have thus been identified as the bases of an ambitious UE-US agenda to encourage a balanced development of the field at both sides of the Atlantic.

SUMMATION OF SESSION 1: Empowerment of Biological Research by Synthetic Biology

This leading session reviewed, in four talks, some of the major challenges that are being addressed by the conceptual and technical framework brought about by SynBio.

1. Antoine Danchin (Institut Pasteur, Paris) emphasized the challenge of defining a minimal genome on the mere basis of comparing genes conserved in all known and sequenced bacterial chromosomes. As it happens, individual genes and regulatory circuits hardly evolve separated from a complex context, which heavily determines its behavior. In order to proceed, there is a need to define the basic processes that undepin biological phenomena: information transfer, metabolism and compartmentalisation. Remarkably, prokaryotes have seemingly random genome sequences, while higher organisms have highly repeated genome sequences. The very fact that these properties look uninteresting immediately shows to us that there is something deep to be seen in the genome sequence, which will escape superficial analysis. Groups of genes such as operons or pathogenicity islands tend to cluster in specific places, and they code for proteins with common functional relationships. Also, some “flexible” DNA motifs are ubiquitously present, suggesting general rules constraining genome organisation. Going deeper in the gene function, Danchin has documented a few universals of bacterial genomes, at the level of DNA and of proteins. He has first shown that the distribution of essential genes (genes essential for growth under laboratory conditions) with respect to the origin of replication is not random, with those genes located in majority in the leading replication strand. This allowed creation of a further class of ubiquitous genes, persistent genes, that are present in the vast majority of bacterial genomes, and systematically locate in the leading strand. Danchin has also shown that bacterial genomes are organised into clusters of genes with a common codon usage bias, indicating that the very process of translation is organising the genome structure. A subsequent analysis of proteomes allowed him to propose one possible function to an intriguing class of proteins, that of orphan proteins which are systematically present in proteomes, each time a new genome is sequenced. These orphans are species-specific, and can be said to label the self of the species. They are considerably enriched in aromatic amino acids (and, conversely, aromatic amino acids-rich proteins are often orphans), and it is proposed that many are gluons selected because they stabilise complexes inside the cell. Aromatic residues are prone to perform that interaction task. If the functions thus discovered when new genes are created become of strong selective interest, the protein will slowly evolve to loose its aromatic residues, because of the corresponding high metabolic cost. If we wish to progress into Synthetic Biology, it may be important to input at least some of the constraints imbedded in extant organisms to improve our chances to be successful.

2. Victor de Lorenzo (Ntl. Center of Biotechnology, Madrid) explained  the development of a discrete set of formatted effector-responsive bacterial promoters as building blocks for designing complex regulatory circuits. Considerable work in this respect has been made with XylR, the activator of the upper TOL operon for the biodegradation of toluene, m-xylene and p-xylene encoded by the pWW0 plasmid of Pseudomonas putida mt-2. The N-terminal, sensor A domain of XylR binds specifically pathway substrates and other related aromatic compounds. A DNA shuffling library of the A domains of homologous XylR, DmpR and TbuT, proteins followed of error-prone PCR amplification generated a large number of protein variants which are potential responders to combinations of different effector profiles and even to unrelated aromatic compound. In a different approach, antibody fragments, expressly the Fv moieties of immunoglobulins assembled as single-chains (scFvs), have been explored as attractive candidates for performing with mimetic regulators which interfere with the buildup of transcription initiation complexes.  To this end, a phage M13 library of scFvs was raised against XylR from immunized mice. The scFv pool was then expressed intracellularly in an E. coli host inserted with a reporter Pu-lacZ cassette and xylR∆A. This strain allowed the assembly of functional scFvs and the direct testing of their activity on the Pu promoter in vivo.  Specifically, genetic screening for lacZ-minus colonies yielded a number of scFvs able to down-regulate transcriptional output in live cells. Since assembly of initiation complexes is stimulated or inhibited by regulatory proteins it was argues argue that anti-XylR scFvs operate as bona fide transcriptional inhibitors of the Pu promoter.

3. Hamilton Smith (Venter Inst., Rockville, MD USA) reported on the plans and progress towards understanding much more fully, the organization and functioning of Mycoplasma genitalium.  With a genome of 580 kb M. genitalium is among the smallest bacteria that can be grown in a defined culture media. The chromosome has 482 protein coding genes and 43 genes encoding RNAs.   An estimated 383 genes are necessary for viability, as non-letal mutations have been generated in 99 genes.  Synthetic biology methods are core to the ongoing efforts.  The entire genome is to be synthesized in vitro as a series of overlapping DNA cassettes which can be combined into a full genome.   To provide assistance, the exceptional capabilities of Diplococcus radiodurans for fragmented genome reassembly are being analyzed in detail.  The purified D. radiodurans DNA repair protein set may be reconstituted, and then used to drive the union of M. genitalium DNA cassettes into a complete chromosome in vitro.  A replacement of the resident chromosome within a M. genitalium lipoprotein envelope by the reconstituted genome would be the final stop.  Following success of a viable genome construction as a control, the process could be repeated with designer mutations within any desired cassette.  From observing the mutations’ effects, a much more detailed understanding will progressively emerge of the genome as an Operating System (OS).  Using the analogue of a computer system, how do the codes of the OS specify both the major hardware of the cell, protein and RNA macromolecules, and the myriad interactions  between them and their responses to the external environment.?

4. Vladimir Larionov (National Cancer Inst., Bethesda, MD, USA) related his work with Natalia Kouprina, achieving the first de novo construction of a human artificial chromosome (HAC) from fully defined constituents.  The roles of the telomere tips of the chromosomes and overall chromosome replication is substantially understood. However the detailed functioning of the centromeres in governing faithful segregation of daughter chromosomes to daughter cells has long been obstructed. The centromeres contain families of repeating DNA sequence, designated alphoid DNA, with some sequence variation among them.  Large pieces of alphoid DNA with added telomeres have in the past been used to construct “minichromosomes” competent upon transfer into mammalian cells.  But further understanding of the DNA sequence to function relationships was blocked by the heterogeneity within the input alphoid DNA.  This problem has been solved in the new construction approach.  A short input alphoid DNA piece of know sequence is first amplified in vitro by a process or “rolling circle” DNA amplification. Long linear repeats of the defined input DNA are generated, which are further joined together achieving the lengths necessary for centromere functionality.  With further addition of telomeres and genes useful for function reporting, the first DNA sequence defined human mini-chromosomes thus been constructed.  Designer variants of the mini-chromosomes are easily generated by these processes.  They will serve in the elaboration of DNA structure-function relationships.  Genes useful for therapeutic purposes could also be added to such minichromosome constructs.


* When this page was created a pdf of the book was available on the WWW. A year later, for reasons that I do not understand, not only did the pdf disappear but I it could not be retrieved from standard caches. Using deep exploration of the Internet I succeeded in collecting a pdf copy of the text, which is now present at this site.
Here is a related text by the same author, that I took care to retrieve so that it should remain available during the time when this site will be alive.