Research

 Environmental perturbations lift the degeneracy of the genetic code

genetic code


The genetic code is degenerate in that most amino acids are encoded by multiple codons. We found that an environmental perturbation, starvation, lifts the degeneracy of the genetic code. This work was published in PNAS.

Summary: We are using both population and single-cell studies in E. coli to understand the dynamic and state-dependent aspects of degeneracy in the genetic code. At the population level, we found that limitation of the cognate amino acid splits codon families into a hierarchy of associated protein synthesis rates. Some codons maintain their associated protein synthesis rate during this starvation, thus, they are robust to perturbation (e.g CUG, black, in above figure.) Other codons show a substantial decrease in associated protein synthesis rate; they are sensitive to perturbation (e.g. CUA, red, in above figure.) This hierarchy of robust and sensitive codons, established in a synthetic reporter library, explained the measured robustness and sensitivities of synthesis for endogenous proteins. The fitness cost of synonymous mutations in amino acid biosynthesis genes and transcriptional control of sigma factor genes also reflected degeneracy lifting. This work suggests that organisms may use degeneracy lifting as a general strategy to adapt protein synthesis in a fluctuating environment.

Former postdoc Arvind Subramaniam led the population-level study, which was a collaboration with Tao Pan's lab at the University of Chicago and published in PNAS.  Postdoc Lisa Marshall is working on the single-cell aspect of the study.
 
  Unifying principles of multi-drug resistance

image multi drug
Drugs combinations are commonly employed in the treatment of severe infections. Unfortunately, it is generally impossible to infer the net effect of a multi-drug combination directly from the effects of individual drugs. We are combining experiments with tools from statistical physics to explore how drug interactions accumulate as the number of drugs increases. Surprisingly, we have found that the effects of drug pairs are sufficient to predict the effects of larger drug combinations on growth in E. coli and S. aureus. This work was published in PNAS. We are currently extending this approach to multiple types of human cancer cells and plan to explore its utility for the systematic development of potent multi-drug therapies.

This project is the work of former postdoctoral fellow Kevin Wood, in collaboration with Satoshi Nishida. Rotation student Bryan Weinstein is extending this general approach to multi-cellular tumors.
 
 
  The single-cell chemostat
image tools to study

We have developed a microfluidic platform for high-throughput measurement of growth and gene expression in single, living bacterial cells. For further detail about the device, please see our paper in Lab on a Chip. If you are interested in using the device in your lab, please contact us.

Summary: Time-lapse microscopy of growing bacteria has been an extremely successful technique, revealing the natural heterogeneity that underlies growth and gene expression in single cells. However, the exponential growth of bacteria depletes the local nutrient environment and crowds cells, ultimately limiting measurement duration. With our device, we are able to circumvent the low throughput of such measurements by combining microfluidic techniques, microarray technology, and automated image analysis to create a massively parallel, high-throughput, single-cell chip: a living analog of the DNA microarray. The agarose-based chemostat allows us to cultivate and image hundreds of thousands of individual, living bacterial cells over 30-40 generations. The entire microfluidic assembly fits on standard microscope slides and permits high quality images with DIC, phase contrast, and fluorescence.

This project is the work of former postdoctoral fellow Jeff Moffitt and former undergraduate student Jeff Lee. Collectively, current lab members are working on improving the device, including optimizing buffer exchange and the analysis software, and developing methods to print multiple strains of cells.
 
 
  Temporal dynamics of flagellar promoters
research image

We study the temporal dynamics within the flagellum assembly cascade in E. coli at the single-cell level. The assembly of the flagellum’s basal body, hook, and chemotaxis system is governed by an intricate but closed system of feedback loops. The cell uses the incomplete flagellum as a pump to extricate the anti-sigma factor flgM, which blocks completion of the flagellum, at the appropriate time. Population-level studies have revealed potential ordering schemes, but single-cell experiments monitoring multiple promoters should reveal the relevant time scales and role of molecular memory within the system.

This project is the work of postdoc Mark Kim.