IDR Team Summary 8
What is the role of evolution and evolvability in synthetic biology?
To circumvent the time-consuming, ad hoc nature of constructing new biological systems, some investigators have advocated efforts to “standardize” biological parts in such a way that their behavior in novel assemblies or environments becomes more predictable. The notorious complexity and context-dependency of the behavior of biological parts and systems, however, makes such standardization extremely challenging. For example, a biological device that is functional in one cell type may not exhibit the same behavior in another, even closely-related cell type. The stochastic nature of biochemical systems also presents a hurdle for prediction and standardization. It is unlikely in fact that biological parts can ever be fully standardized, and engineering methods that enable rapid optimization of synthetic biological systems will be needed. Nature’s optimization algorithm is evolution: evolution fine-tunes the functions of parts in new contexts and optimizes their assemblies in nature. Can directed evolution be used to do the same in synthetic biology? Evolution is also the source of all biological parts—can directed evolution reliably generate useful parts, especially those unlikely to be found in Nature?
All biological systems evolve under the pressure of mutation and natural selection. Natural selection, however, leads to the destruction of synthetic systems that place the organism at a selective disadvantage relative to dysfunctional mutants. Synthetic biology will have to confront this ubiquitous feature of living systems.
A hallmark of biological systems is their ability to adapt to changing
environments and challenges. Modularity appears to be a useful feature of evolvable, rapidly-adapting systems—some biological systems and even components are highly modular, such that components and sub-components can be rapidly swapped in and out to generate new functions. Eukaryotic signaling systems are a good example, but prokaryotes rely on much less modular systems that nonetheless serve them very well. Are there costs of evolvability in terms of system performance?
When and how can evolutionary methods contribute to design of synthetic systems?
How can evolutionary methods be best integrated with “rational” design, including computational design? What is the role of modeling?
Are there design objectives that can be addressed only through evolutionary strategies? Are there objectives for which evolutionary strategies are unnecessary?
What are the best targets for evolutionary optimization? Molecules? Circuits? Organisms?
What technologies and tools will be needed for rapid, efficient evolutionary optimization?
What strategies can we use to overcome the tendency of synthetic biological systems to mutate and escape programmed control?
How do we design systems and host organisms to ensure genetic stability?
How can we best understand mechanisms and consequences of mutation and develop routes for repair that enable designed functionality to be maintained?
To what extent is it important to pursue strategies for designing evolvable systems? What are the key features?
Haseltine EL, Arnold FH. Synthetic gene circuits: design with directed evolution. Annu Rev Biophys Biomol Struct 2007;36:1-19: http://arjournals.annualreviews.org/doi/full/10.1146/annurev.biophys.36.040306.132600?amp;searchHistoryKey=%24%7BsearchHistoryKey%7D. Accessed online 28 July 2009.
Bhattacharyya RP, Remenyi A, Yeh BJ, Lim WA. Domains, motifs, and scaffolds: The role of modular interactions in the evolution and wiring of cell signaling circuits. Ann Rev Biochem 2006;75:655-680: http://arjournals.annualreviews.org/doi/full/10.1146/annurev.biochem.75.103004.142710?amp;searchHistoryKey=%24%7BsearchHistoryKey%7D. Accessed online 28 July 2009.
Marguet P, Balagadde F, Tan C, You L. Biology by design: reduction and synthesis of cellular components and behaviour. J R Soc Interface 2007;4: 607-623: http://220.127.116.11/search?q=cache:IO9IXc-3zRAJ:www.duke.edu/~you/publications/marguet_etal.pdf+%22Biology+by+design:+reduction+and+synthesis+of+cellular+components+and+behaviour%22&cd=1&hl=en&ct=clnk&gl=us. Accessed online 28 July 2009.
Yokobayashi Y, Weiss R, Arnold FH. Directed evolution of a genetic circuit. Proc Natl Acad Sci USA2002;99:16587-16591: http://18.104.22.168/search?q=cache:t3xmv0D8FMwJ:www.princeton.edu/~rweiss/papers/yokobayashi-pnas-2002.pdf+%22Directed+evolution+of+a+genetic+circuit%22&cd=1&hl=en&ct=clnk&gl=us. Accessed online 28 July 2009.
IDR TEAM MEMBERS
Kirstie Bellman, The Aerospace Corporation, FFRDC
Jef D. Boeke, Johns Hopkins University
Eric Gaucher, Georgia Institute of Technology
Farren Isaacs, Harvard University
Rob Knight, University of Colorado at Boulder
Tanja Kortemme, UCSF & California Institute for Quantitative Biosciences
Norman Packard, ProtoLife Inc.
Casim Sarkar, University of Pennsylvania
Michael Sismour, Harvard University
H. Tom Soh, University of California, Santa Barbara
Narayan Srinivasa, HRL Laboratories
James E. Hataway, University of Georgia
IDR TEAM SUMMARY
By James E. Hataway, Graduate Science Writing Student, University of Georgia
It is widely accepted that organisms in the natural world evolve in order to adapt to changing environments and challenges. Evolution is a process that may take long periods to produce any observable change, and there is
no easy way of predicting which organisms will evolve when faced with a challenge.
Synthetic biologists, however, are exploiting evolutionary principles in the laboratory to create new biological systems that may one day lead to breakthroughs in renewable energy, material synthesis and medicine.
The mission of synthetic biology is twofold: it involves the design and construction of new biological parts, devices and systems as well as the redesign of existing, natural biological systems for useful purposes.
To help facilitate the growth of the field, an Interdisciplinary Research (IDR) team of eleven scientists representing a variety of disciplines gathered at the 2009 National Academies Keck Futures Initiative Conference on Synthetic Biology to discuss the question: What is the role of evolution and evolvability in synthetic biology?
Using evolution in synthetic biology involves a combination of techniques and perspectives from both engineering and biology. Engineering principles provide methods for evaluating processes and how to monitor processes to achieve a desired outcome. For example, mechanical engineers design and implement sophisticated systems using machinery to create a specific outcome or product. Biology adds to this an understanding of the processes found in molecules, circuits and organisms within the natural world.
Through a process known as directed evolution, it is possible to introduce specific stresses that force components (molecules, for example) to evolve rapidly, eventually producing a biological system that is unique in both form and function.
But the attempts to marry engineering and biology are also fraught with difficulties. Members of the interdisciplinary research team observed that engineering generally relies upon consistency and predictability of processes, while biology is characterized by variation and diversification.
This disparity extends to the relationship between evolution and synthetic biology, because the results generated through directed evolution are sometimes difficult to replicate, and the components that evolve may continue to do so when placed in a new environment. The difficulties associated can lead to elevated lab cost, while continued evolution may result in an unstable system that behaves in ways that are unpredictable. Systems that function well in one cell type may not work the same way in others, even if the cells are closely related. Thus, systems that rely on directed evolution are not always the most stable.
With this in mind, the IDR team posed two hypothetical questions.
1) Is synthetic biology successful when evolution is no longer needed and systems are created by rational (rule-based) design, or 2) is synthetic biology successful when scientists can effectively harness the unique features of evolution, making it a central tool through which systems are crafted?
Ultimately, there is not enough evidence to determine which of these outcomes is more likely. Several team members emphasized that the selection processes for synthetic biology and evolutionary biology are more of a craft than an industrial process, although to achieve some of synthetic biology’s goals, creating organisms on an industrial scale will be necessary. Scientists must learn more before they use evolutionary principles for large-scale projects.
Indeed, directed evolution is such a boutique practice, proposals for advancement tend toward the conceptual rather than the concrete. As such, the group proposed a series of model problems identifying areas requiring additional research. Some of their model problems were:
To obtain a system that optimizes the output in financial terms (including the cost of setting up the system).
Obtain a robust enzyme circuit to do X, and to get the same behavior under various conditions (e.g. compounds, temperatures, media types and genetic background). That is, for directed evolution to have any broad application, we must create circuits that are not restricted to one function.
How does one initiate research when one cannot see the pathway from where we are to the final result? For example, how might one develop a bacterial population that spontaneously spells the word “HELP,” or an E. coli that can play music? These unusual examples emphasize the point that we do not yet know how to begin research with a specific application in mind. If we are to design systems that fight cancer or enhance the immune system, we must develop ways to initiate research even if the exact process is unknown.
How do you make a system that is robust and can therefore search functional space more easily? Biologists often refer to the functionality of a particular agent in terms of a “fitness landscape,” a graphical representation often conceived of as a series of peaks and valleys in which peaks represent the best outcomes for a given function. The group suggested a need to “smooth” the fitness landscape, meaning that we must find ways to reduce the number of “valleys,” or poor functions, and make the overall selection process more consistent and predictable.
Fundamentally, the team said it is necessary to develop methods to accelerate evolution to get to a desired result faster, while also developing ways to decelerate or stop evolutionary processes once the experiment reaches an end point.
To do so, biologists and engineers must develop more stable strains of bacteria into which end products of synthetic biology can be transferred or a strain where the mutation rate can be controlled. Ideally, this would prevent the over-mutation or under-mutation (i.e., evolution) of a system, thus making it significantly more reliable and malleable.
In addition scientists must create more robust systems in which swapping of components is seamless. That is, researchers must find a way to share evolved components without reengineering new components for each individual project.
The IDR team also suggested the creation of a universal fitness landscape readout from small molecules that applies across heterogeneous systems. This readout would apply to an overall fitness landscape. A universal fitness readout would simplify matters by allowing researchers to compare evolutionary processes for a variety of applications.
In order to create this kind of generalizability, the team argued that scientists must develop ways to predict and screen for sequences that are consistent with multiple objectives, what they called “multi-objective massively parallel optimization.” This would require the creation of libraries from which scientists could choose components that they know would act in specific, predicable ways in a multitude of conditions.
These suggestions are merely the first steps toward the creation of a more unified practice of directed evolution. Members of the group recognize that many of the processes used in directed evolution experiments are in their technological infancy, but they maintained that additional research might generate the requisite knowledge to create robust yet flexible systems that work in harmony with biological circuits found in nature.