- I am currently an undergruate researcher in the Keasling Lab in Lawrence Berkeley National Laboratory. My current project is iGEM 2006.
- MIT Class of 2009; Majoring in Mathematics and Biological Engineering.
The Sortostat is a microfluidic chemostat integrated with a cell sorter. My project consists of demonstrating the sortostat's chemostat functionality, a technique rarely seen in microfluidics. Chemostasis must first be demonstrated before any meaningful demonstration of the sortostat's sorting capabilities can be shown. The sortostat's sorting ability is limited by the extent to which it can make optical descriminations between cells. Therefore, in order to demonstrate its sorting ability, two visually different populations of cells will be grown to steady state after which sorting is initiated. If successful the populations will clearly diverge in number, the one sorted against will diminish well below its steady state level and the one preserved should rise by the same amount that the other falls. The total population of cells in the chemostat should be conserved in the end if chemostasis was maintained throughout sorting.
- Debug a microfluidic chemostat (Sortostat) to improve the time-varying specific selection of cell populations.
- Currently troubleshooting problems: i) Cell Death after 3-4 days. This problem has not been an issue for the past few runs that lasted over 300 hours. We initially thought the cell death was due to oxygen depletion however if this was the case we'd expect to see this more consistantly. and ii) Inaccurate cell counts due to poor image processing. There is still much to be done to resolve this issue. Perhaps using software such as Cell Profiler will help to achieve more accurate cell counts.
- Evaluate the response of populations of E.Coli cells containing engineered genetic circuits (http://parts.mit.edu) to particular selective pressures using the Sortostat.
It is known that mutations are more likely to occur with higher cell division rates. Everytime a cell divides it runs the risk of making a mistake in the replication process and creating a mutant cell. Evolution is largely in debt to this phenomonon; without mutants an organism's genome would be nearly static. However, beneficial this might be to the survival of a species, it poses a problem for our genetically engineered cells. If a cell divides and mutates in the process it could "break" our engineered genetic device. The cell will likely go on living, however, it will cease to perform its intended function. This is an unfortunate reality of biological engineering, and, as such, we must learn the characteristics of our devices and their liklihoods of "breaking."
In order to obtain a quantitative measure of how "fast" our devices break we will first need a controlled environment. By using a chemostat we can control the rate at which cells divide enabling us to calculate the number of cell divisions that have occured over some time. Note that number of divisions is but one variable that can be used to characterize the longevity of a device. One might imagine testing other variables such as temperature or cell cycle position for example, both of which likely have an effect on the "breaking rate."
Detecting when a cell's device has broken can be a challenge in it of itself; however, this challenge becomes much easier with the correct choice of device. We are using a device such that it's success is fatal to the cell under certain conditions, however, when the device breaks the cell lives and will grow. This is known as a counter-selectable marker and allows us to ultimately find out the rate at which our device breaks.
Once a mutation rate is established, we can talk quantitatively about device longevity under prespecified conditions. We also, and more importantly, have a valid control in which to compare methods of prolonging a devices life which go beyond that of the counter-selectable marker. Our aim is to show that by remapping the cell's codon space we will be increasing the device's life.