Friday, October 25, 2013

Synthetic Biology Paradigms, Part I

Is any one conceptual framework is sufficient for advancing the field of Synthetic Biology? Will a new paradigm be needed eventually? We can apply a diverse set of paradigms to think about how biology works: Analog devices (clocks, gears, etc) or digital circuits (processor), biochemical pathways where the flow of substrates can resemble the flow of fluids through a complex network of pipes, or even complex mathematical constructs (like those developed by Wolfram) can all be used to represent or draw parallels with different aspects of molecular biology.

Synthetic Biology is sometimes described as molecular biology with an engineering perspective. Indeed, many leading researchers in the field have backgrounds in electrical engineering or some related discipline. Much of the earlier work done by pioneering synthetic biologists has been to formulate biological phenomenon in more familiar engineering terms. These efforts has enabled engineers to leverage their expertise during the design of complex artificial genetic networks and cell behaviors. Are the promises of synthetic biology within our grasp, limited only by a need for the import of more engineering knowledge?

Despite the success of an engineering paradigm in synthetic biology, I believe that harnessing the full potential of this field will require new concepts and perspective. Synthetic Biology can and should grow from conceptual frameworks borrowed from engineering disciplines, but also must not be constrained by them. 

Incorporating ideas and concepts from fields other than the digital logic of electrical engineering can only serve to strengthen the efforts of a synthetic biologist. In this series of articles, I will detail how perspectives from Molecular Biologists, as well as Economists and other Engineers (particularly those that deal with Analog systems) may prove to be an invaluable part of the synthetic biologist's conceptual toolkit. The remainder of this article (which you can view by selecting 'Read More') details a recent advance in synthetic biology: the design of an analog genetic circuit by the Sarpeshkar Laboratory Group. I also discuss how Synthetic Biology can mimic and learn from nature, and why keeping an open mind and a flexible vocabulary may be important.

Analog Versus Digital

Due to the complexity and sophistication of the cell, it is natural to try and draw parallels and analogies between the cell and other, better understood complex systems. The parallel most often used today is that of a computer and it's processor; these are arguably the most sophisticated man-made objects ever created. In many examples of synthetic biology to date, promoters and transcription factors have been crafted in a way to produce digital logic circuits, reminiscent (in an abstract, but not physical, way) of the mechanisms that operate on a circuit board. In fact, it is a dream of many in the field to eventually have enough standard parts to allow engineers to reliably design genetic circuits and achieve the desired cellular behaviors with minimal optimization. 

However, the road to sophisticated programmed cellular behaviors may be paved with more than just digital logic. A great of the behavior in natural biological system operates in an analog, not digital, fashion. The analog approach in this field is exemplified by the Analog Circuits and Biological Systems Group at MIT, lead by Dr. Rahul Sarpeshkar. 

The Sarpeshkar group has recently demonstrated that analog circuit design is possible in synthetic biology. These findings were published in Nature (Daniel, et al. 2013 "Synthetic Analog Computation in Living Cells." Nature Vol 497:7451 pp619-623). In the same issue of Nature, a summary of their research was published alongside the findings, both of which have been conveniently packaged into one PDF file.

Instead of creating a circuit that produces an 'all-or-nothing' digital-like response, the Sarpeshkar created a simple genetic circuit that produces a graded amount of the output molecule in response to increased input. Due to the pioneering nature of this accomplishment, several press releases and articles (in addition to the one in Nature mentioned above) have been written that summarize and comment on the research. Any efforts on my part to describe the nature and importance of the work done by the Sarpeshkar group is likely to be redundant, so I recommend you read the following summaries:

Press Release from the MITnews site

News Article from TheScientist

A post at the Temple Biomimetics and Bioinspiration blog

Analog circuits boost power in living computers
News in brief from ScienceNews

The creation of an analog genetic circuit itself is a notable achievement and important step in synthetic biology. Is is also striking how simple the circuit is: it requires only three components (transcription factors), and generates the desired behavior through manipulation of the relative number of copies and location of each component. The MIT group was able to achieve the desired affect by creating a positive feedback transcription loop on a low copy number plasmid. Alone, this positive feedback system will produce greater amounts of transcription factors in response to an inducer molecule. When paired with a high copy number 'shunt' plasmid, which siphons off excess inducer bound transcription factors from the feedback loop, the desired logarithmic graded response was achieved (of course, this require extensive modeling and optimization). 

Interestingly, the researchers attribute the effect of the shunt plasmid to more than a simple increase in binding sites for the transcription factor. Instead, the fact that these sites were on a plasmid separate from the positive feedback loop was suggested by their computer modeling to be partly responsible for aspects of the circuit's behavior. In this model, transcription factor binding of sites on the shunt plasmid required these factors to diffuse from their site of synthesis. The low copy number plasmid was not affected by transcription factor diffusion in the same manner, since the factors are synthesized from a gene on the low copy number plasmid. Remember, transcription and translation events are usually linked in bacteria, so the protein encoded for by a gene will usually be synthesized nearby the corresponding genetic information; there can be a physical link between them. Therefore, the transcription factors were synthesized in closer spatial proximity to their binding site in the low copy number plasmid. 


Years ago, I read Biomimicry, an interesting book by Janine M. Benyus from 1997. In this book, Mrs. Benyus gives several examples of nature serving as a model for solutions to our own problems, a guide to which solutions would stand the test of time, and as a mentor for how to structure other parts of our society.

Written before the advent of Synthetic Biology (at least as it is practiced today), Biomimicry may seem dated and out of place in a blog post in 2013. Indeed, the book does not focus on the field, and Mrs. Benyus probably did not have today's advancements in mind. However, the idea that we should look to nature for solutions and inspiration is still very relevant, perhaps nowhere more so than in the field of Synthetic Biology.

As highlighted at the excellent Scientific American blog the Oscillator (which focuses on the field), recent research has indicated that plants such as Arabidopsis thaliana can perform arithmetic division that might rival the calculations carried out by any artificial genetic circuit to date. The featured study of this article, conducted by Dr. Martin Howard, Dr. Alison Smith, and several other researchers at the John Innes Center in the UK, was published in the Journal eLife (Scialdone et al, "Arabidopsis plants perform arithmetic division to prevent starvation at night" June 25th, 2013). In this work, researchers uncover an analog genetic circuit plants utilize to calculate an appropriate storage and consumption of starch. Arabidopsis produces and stores starch during the day, and consumes the starch during the night. The amount and consumption of starch appears optimized so that just enough is made to last through the night. Most important, the plant can 'calculate' the correct amount, and adjust to perturbations to both starch production and length of the night. 

While we can develop ever more sophisticated genetic networks and designs, it is important to remember that nature has a 3.8 billion year head start over our efforts. It makes sense to leverage as much of these natural circuits as possible; there are likely many more that haven't been characterized (or perhaps some that even lie outside the imagination of most biologists.) 

Sometimes, attempting the design or programming of some cellular behavior can reveal dynamics to gene regulation previously unappreciated. This was the case encountered by a research team at Duke University, led by Dr. Lingchong You. In their study, recently published in Molecular Systems Biology (Stephen Payne et. al. "Temporal control of self-organized pattern formation without morphogen gradients in bacteria"), the researchers were observing the formation of patterns in a bacteria colony, based upon diffusion of an inducer and expression of a fluorescent reporter. They discovered that the patterns that were formed (which didn't match their predictions) were influenced not by inducer concentration, but rather by the timing of gene expression. In their system, the inducer concentration served as a timing mechanism; coupled with the different growth rate and metabolism of bacteria across the colony, a threshold concentration trigged production of a fluorescence protein. These findings were also summarized in a press release one of which you can read at (Growing bacteria keep time, know their place.)

As Mrs. Benyus argued more than a decade ago, nature can provide a model for crafting solutions to the problems in a diverse set of industries. Looking to nature, in addition to digital circuits, will also help inform what will be possible in Synthetic Biology. In turn, efforts to create these solutions and design complex cell behaviors can lead to a deeper understanding of what is possible, and what works, at the scale of molecular biology. In this respect, Synthetic Biology can not only provide novel technologies but also act as a basic science. In both cases, a flexible paradigm that can embrace diverse ideas is indicated.

Semantics and Solutions in Synthetic Biology: 
A Larger Point

As pointed out by many already, the addition of analog genetic circuits to the synthetic biologist's toolkit is a significant advance. The ability to have such a small circuit carry out calculations and generate graded responses could conceivably  be used to fine tune parts of a biochemical pathway that has been transplanted from one organism to another (as is often done by Synthetic Biologists, in order to more efficiently create some commercially useful product.) Moreover, this development is important because it illustrates how stepping outside of a digital engineering paradigm can produce notable improvements over existing designs.

The influence a particular paradigm can have on how synthetic genetic circuits are designed, or on any scientific endeavor, may be explicit and clear (as it is in the examples mentioned above). It is also possible that the influence can be more subtle, and that only the terminology and semantics from a paradigm are being employed. Borrowing language from a particular discipline or paradigm can be very useful, but may also steer a researcher in a particular direction, and can potentially blind them from considering alternative options or hypotheses. For example, the constant use of the term 'genetic circuit' draws a parallel with electrical circuits which in turn indicates that transcription factors can and should be arranged and programmed according to produce digital responses. Perhaps this served to put the design of analog genetic circuits in the blind spot of many Synthetic Biologists (This is speculative and may not be true, but I included it to illustrate the point).

The abstract point in the preceding paragraph was captured well by a reader to the Oscillator's post about the calculation of time by plants (mentioned above). They commented:

"Lest we forget: Hours are a human invention, as are calendars. Plants don't use that method to adjust their lives. Over the eons, they have develop their own methods to regulate their lives."

To rephrase this comment, it is important to remain aware how the language we use may confine us to thinking about particular concepts or to a particular paradigm. While certain terminology and parallels is necessary to communicate scientific ideas, they should not end up defining the ideas themselves. Thus we must remember that a genetic circuit is not really a circuit in the sense that the word is normally used. A genetic circuit is really a collection of genes, transcription factors, and small molecule inducers that interact to produce a particular pattern of gene expression. The components of a genetic circuit perform these actions in a complex environment in which interaction with many other proteins and genes is possible (although not always desirable). Most importantly, the means by which these patterns and behaviors are achieved by the cell in a natural setting arose through evolution, according to the laws of physics and chemistry and the demands of biology. 

It is also helpful to remember, as illustrated by the shunt plasmid in the analog circuit, that these transcription factors must operate not only alongside other proteins but in the confined and crowded three dimensional space that is the cell. Thus, movement and organization of the components in time and space must also be considered, creating a further disconnect between the concept of genes as a circuit and the reality of gene expression.

Of course, I am not the first to acknowledge the reality and difficulties in Synthetic Biology, nor do I think they are insurmountable challenges. Biological systems display remarkable diversity, sophistication, and (most of all) an incredible ability to respond to and effect their environment. To harness the power of Synthetic Biology, researchers should endeavor to be as diverse and sophisticated in their approach. In the remaining articles in this series, I will discuss even more radical ways in which the paradigm for Synthetic Biology can be altered, as well as expand upon how Synthetic Biology can serve as a powerful scientific approach. (Stay Tuned!)

How do you see Synthetic Biology? Is there a certain paradigm or field that you think Synthetic Biologists can borrow useful concepts from? Feel free to share by leaving a comment below.

No comments:

Post a Comment