APh/BE 162 - DNA Science Lab

The study of genetic switches is based on the quantitative relationship between transcription factor concentrations and the rate of protein production from downstream genes. This relationship can be explored using fluorescent reporter genes and quantified using image analysis of fluorescent microscopy.

The following genetic switch was established by transforming the lacI, tetR, and araC genes into MC4100Z1 strain of E. Coli, as described in Week 2 to produce a genetic switch. In particular, we are interested in the simple transcription unit where the P_LtetO-1 promoter (which constrains a tightly regulated TetR repressor) is fused to cI-gfp gene such that GFP is induced by the presence of anhydrotetracycline (aTc) as shown in Figure 1a.

Figure 1a. Synthetic transcription circuit: a.) Simple transcription unit (open loop, MC4100 + pZS12-tetR + pZE21-gfp). Cells expressing TetR can be induced, by adding aTc to the medium, to produce GFP (from Rosenfeld et al. JMB 2002). b.) Topography and sequence of promoter region (from Lutz et al. NCR 1997).

MC4100Z1 cells were grown at different concentrations of aTc (1-50 mg/mL) in order to characterize inducible expression. Cells of MC4100Z1 and pZE21-LacI were obtained as negative controls and to provide a measurement of auto-fluorescence. In addition, MC4100 was also transformed with pZE21-GFP where all cells will express GFP as a positive control.

In order to be viewed at 100x resolution, special preparation must be made of the slides. Slides were prepared by placing a 22x50mm coverslip over a base and adding 2.5 uL of 2% molten agarose gel and then gently placing another coverslip over it. The second coverslip should be exactly parallel to the coverslip underneath. Rotate the base if necessary to adjust the plane of the gel. Let the gel harden for 5-10 minutes then slip off the top coverslip. Cut an approximately 15x15mm portion of the gel and place it onto a slide. Pipette cells onto the gel and place a coverslip over it.

Slides were examined at 100x phase using both brightfield and fluorescence under a Zeiss Axio microscope. AxioVision 4.4 software was used to automate brightfield and fluorescent data acquisition to avoid differences in the two subsequent images. Exposure time was optimized for each sample and 10 images were taken of different locations throughout each slide in order to produce an accurate representation of the E. coli cell population.

Photobleaching was measured by acquiring time-lapse frames 500ms apart for 60 frames.

Images were analyzed using a custom software using MATLAB (Mathworks, Inc.). Analysis proceeded in several stages: Cells were identified in the brightfield image using an edge detection algorithm. Each detected area was labeled and identified as an individual cell and counted. In the fluorescent image, exposure time was accounted for by an intensity transformation that increasing the sensitivity of lower-value pixels. Background and cellular auto-fluorescence was subtracted from the channel by setting a lower limit boundary to the threshold. After image reprocessing, cells were identified using and edge-detection algorithm, labeled, identified, and counted.

Unfortunately photobleaching calculations could not be made because focus drift occurred over the time period of the data acquistion such that blurring of the images occurred faster than the photobleaching effect.

In the overall experiment, we would extract the lacZ gene from wild type E. Coli (MG1655, GenBank U00096), insert it into a pZE21-GFP vector with kanamycin resistance, and finally transform the E. Coli cell so that it can express lacZ upon the induction of tetracycline promoter. As the first series of this experiment, cloning vectors were obtained by double digesting pZE21-GFP with KpnI and HindIII. lacZ gene from wild type E.Coli (MG1655) was then amplified using PCR. The results were analyzed via gel eletrophoresis.

The blue color showed in petri dish containing X-gal indicated the presence of B-gal. Our desired DNA insert has been successfully incorporated and expressed.

Although steady-state was not achieved, it could be concluded that the the kinetics of the simple transcription unit follows Michealis-Menton induction rates.