Bryan Hernandez/UROP Proposal
UROP Proposal Fall 2006
Context and Scope
Biological Engineering is a newly emerging field that applies an engineering discipline and approach to biological problems. Synthetic biology, a subset of biological engineering, is concerned with constructing and modifying biological “parts” for the use in more complex systems. Classical biologists have been perturbing biological systems for decades now, however, what makes the synthetic biologist different is the construction of novel biological devices for the use in new or existing systems. The use of newly fabricated biological parts, in the past, has never really been pursued at great length because of standardization issues. For example, electric circuit construction is relatively trivial in comparison to genetic circuit construction partially due to the fact that the electrical components that make up the circuit are standardized and hence well characterized; this is not exactly the case with genetic circuit components. As such, synthetic biology also has the added scope of creating a standard for building in this new biological space.
iGEM, the International Genetically Engineered Machine competition, is a competition with the aims of expanding the field of synthetic biology. Using advanced genetic components and technologies compiled in the Registry of Standard Biological Parts (www.parts.mit.edu), students from around the world design and assemble these new biological machines as a means of exploring and pushing the boundary of capabilities in this new science.
For my UROP, I plan on continuing the project started on UC Berkeley’s iGEM team over the summer. The project itself is to develop networks of bacteria capable of learning. Networks of interacting cells provide the basis for neural learning. We have developed the process of addressable conjugation for communication within a network of E. coli bacteria. Here, bacteria send messages to one another via conjugation of plasmid DNAs, but the message is only meaningful to cells with a matching address sequence. In this way, the Watson Crick base-pairing of addressing sequences replaces the spatial connectivity present in neural systems. To construct this system, we have adapted natural conjugation systems as the communication device. Information contained in the transferred plasmids is only accessible by "unlocking" the message using RNA based 'keys'. The resulting addressable conjugation process is being adapted to construct a network of NAND logic gates in bacterial cultures. Ultimately, this will allow us to develop networks of bacteria capable of trained learning.
The conjugative addressability of these bacterial networks is dependent on the orthogonal ‘lock’ and ‘key’ pairs; it is this part of the project with which I am primarily concerned. My goal is to construct several working orthogonal lock and key pairs. The lock and key together make what is known as a riboregulator. A riboregulator is a post-transcriptional regulator that prevents translation of a protein by blocking initiation of the ribosome.
Genes under riboregulatory control give the “user” selective control over its expression. For example, consider the simplified system that consists of three independent genes A, B, and C. If A and B are under orthogonal riboregulatory control then the user has the ability to choose the level of expression of A without affecting B or C; likewise, B can be controlled in a similar manner. This is achieved by first suppressing a gene, or ‘locking’ it, and then bringing it back to the desired level by introducing the key, hence ‘unlocking’ it. Suppression is, as mentioned, due to steric hindrance at the ribosome binding site which prevents the ribosome from translating the protein. This steric hindrance is created by the ‘lock’ which is only neutralized by its specific key. The reason this post-transcriptional regulatory method is so potentially powerful is the ability to make an arbitrary number of riboregulators that do not cross talk. This is what makes the riboregulatory control orthogonal; any key can only unlock locks for which it was designed. This orthogonality is essential for the propagation of any type of signal under riboregulatory control, without it signals might cross-talk. By using this binary functionality we can construct Boolean-like logic gates which are the basis for larger network functions which can give rise to learning.
This project makes use of modern biological techniques. Among these include, PCR, rtPCR, flow cytometry, flourometry, and cloning. Lab work will be performed in Professor Drew Endy’s synthetic biology lab in building 68. I plan on spending approximately 10-15 hours a week working on this project. At this rate I feel I can complete my portion of the project by the end of the year.
I really like this UROP. I think it is great exposure to a newly forming and potentially influential field. I have previous research experience in this field and plan on continuing it in the future. I am beginning to learn the importance of doing research in biological engineering as a means of learning outside the classroom, despite being a course 20 major, because this is where I have learned the most. At the end of this project I think I will have results worthy of publishing which I hope to do.
UROP Proposal Spring 2006
Due to our lack of understanding and the high degree of complexity in biological systems, engineering biology has proven a difficult and tedious process. Past biological engineers have largely used ad hoc approaches to build genetic devices. Much can be done to improve the approach to engineering complex biological systems. In particular, standardization and characterizaton of parts and devices will enable construction of complex biological systems with the same confidence found in more mature engineering fields.
To help ground biological device design on a solid foundation, the Endy and Knight Labs have started a registry of basic biological parts and devices (e.g. promoter, ribosome binding site, terminator, etc.) that are both characterized and able to be integrated with each other. With this grounding, devices are being engineered that are composed of the characterized and integratable parts contained in the registry. Genetic inverters, analogous to a Boolean NOT function, are an example of a device that can be constructed by integrating these parts. Characterization of parts and devices is necessary if these constructs are to be integrated in more complicated systems just as logic gates are used in complex circuitry.
My UROP is focused on developing a tool to characterize these genetic devices - a microfluidic chemostat with cell sorting capabilities, called a "Sortostat". This device contains a bioreactor composed of a 16nL ring etched into a PDMS microfluidic chip. Media and cells are mixed around the ring via a parastaltic pump, and scheduled dilution events give the chip its chemostat function. The chip is also capable of sorting cells, by flushing sub-populations of cells out as waste in a discriminatory manner.
The Sortostat is mounted on an inverted Nikon TE2000 microscope and can interface with software to analyze images from the microscope camera and make sorting decisions automatically. In this way we are able to make more specific and complicated selections in the cell population allowing for directed evolution. Additionally, we will be able to put evolutionary pressure on the cell population in ways that are pertinent to the characterization of the genetic device hosted by the cells.
Currently, the Sortostat is not capable of culturing cell populations for long periods of time; population falloff has been observed to begin after 3-4 days. Although not yet confirmed, population decline is presumed to be a result of oxygen depletion due to poor oxygen diffusion through the PDMS and/or an accelerated diffusion out of the reactor because of the pressure gradients created in the pressurized control lines that run throughout the chip. Specifically, I will be responsible for planning and conducting experiments to test and troubleshoot this problem in order to enable longer experimental runs in the device. I will also characterize the Sortostat so it can be considered a controlled environment that allows for selection to be consciously and intelligently applied to a cell population.
Already in my research I have been required to apply basic molecular biology techniques, quantitative microscopy analysis, and the operation of PDMS microfluidics in an effort to solve the problem of debugging and standardizing the Sortostat. I am interested in both the practical problems of this project concerned with the actual engineering of tools like the Sortostat as well as the more abstract and long-term goals of standardardizing biological parts and genetic circuitry; this is an exciting new approach to biology that is just beginning to make its way into the biological research and engineering communities.