Current Research in the WetLab
Single cell trapping in a microfluidic device
Obtaining single cell data is a challenging task. For yeast cells, volumina of roughly 100 fL has to be measured as exact as possible. We choose the microscope in combination with microfluidic devices as best technique for our purposes. The microscope offers temporal and spartial resolution not reached by any other method in this combination. Microfluidic devices are ideal to control and change the experimental conditions for liquid cultured yeast cells. The device is designed to trap cells in a continuous flow of fresh media and wash away daughter cells. Using an analogon of pulse width modulation from electrical engineering, it is possible to adjust concentrations of inducers or ligands fast and accurate.
Observing transcription in live cells using tagged bacteriophage coat proteins has become popular in recent years. We utilize the PP7 system and its stem loops in an inducible background. Yeast cells are incubated and observed in a microfluidic chip designed for single cell studies. Upon induction with β-estradiol, transcription is initialized and transcription sites become visible through the accumulation of PP7-GFP fusion protein onto the synthesized mRNA. Drugs and varying levels of inducer are used to perturb cells in order to study cellular heterogeneity. Intensity time traces of the transcription sites are extracted via image analysis of the 4D data. Due to the non-stationarity resulting from induction, the established autocorrelation-based analysis is not applicable. Instead, we use a rigorous Bayesian approach to jointly analyze measurements from several cells in order to quantify heterogeneity and to calibrate and compare different models for transcription elongation. This also facilitates a unified treatment of photo-bleaching and measurement noise parameters. We show that elongation and termination rates and the number of polymerases involved can be estimated with reasonable confidence. We also investigate the evidence for the stalling of polymerases due to the simultaneous recruitment of several RNAPs.
Cell Free System
RNA-based regulators have recently emerged as a promising tool to control gene expression. Newly discovered transcriptional regulators have been shown to exhibit high robustness, reversibility, and in particular faster signal propagation than the traditional proteins-based circuits. This makes them a platform of choice for a wide range of applications, from molecular diagnostics to RNA circuit engineering. However, despite previous efforts to characterize certain types of transcriptional circuits, little work has been done on studying and prototyping the combinatorial effect of these new tools for matching the needs of higher complexity circuits. Here, we base our work on an Escherichia coli-based cell-free transcription-translation (TX-TL) system for rapidly implementing different synthetic circuits of increasing complexity. Cell-free systems offer an appealing and robust alternative environment to the classical lengthy design-build-test cycles associated with in vivo studies. Through the use of this platform, we investigate novels designs of logic and dynamical circuits based solely on the combination of varied RNA regulators, such as transcriptional and translational riboswitches as well as the recent Small transcription activating RNAs (STARs).
Collective motion occurs when a number of similar agents adopt the same interaction rules. These rules non-linearly determine the behavior of a group as a whole. The bacterial swarming, as one example, allows a lot of control over the experiments but since there are a lot of obscured factors influencing the dynamics, the rules determining a swarming pattern remain unclear. We conduct experiments with B. subtilis in microfluidic channels in different media, e.g. YEPD, Terrific Broth Sigma, and SOB, as well as on agar plates, in order to achieve highly-correlated bacterial swarming and to determine interaction rules in the system. Afterwards, we capture the bacterial behavior as a sequence of images. The images are then processed in order to perform the multi-target tracking and construct a set of trajectories per each bacterium. When the tracking is performed, we regard its output as observational data and use it for inferring models, e.g. a self-propelled rods model, and their parameters that conform to the data as close as possible.