Team:Imperial College London/Results

The XylE gene, originating from the TOL plasmid of Pseudomonas putida, encodes the catechol 2,3-dioxygenase (C2,3O) enzyme, the final choice for the output signal mediator of our system. Colonies of cells that express XylE gene product become yellow within seconds after selection plates are sprayed with catechol, a colorless substrate, that is converted by C2,3O to the yellow product, 2-hydroxymuconic semialdehyde. This reporter system is ideal for our project as it has very fast kinetics, gives a visual output by naked eye and the substrate used is inexpensive.

The enzymatic reaction catalysed by C2,3O is an ideal output signal for our engineered bacterial detector and it can also serve as a very useful reporter gene. Its characteristics can be exploited in a wide range of fields across the biological sciences, so it was one of the team's first candidates for further characterization. Qualitative and quantitative data have been gathered and presented. All further tests involving XylE transformed cells were quantified by measuring absorbance at 380nm wavelenth.

Catechol,the substrate of C2,3O, is colourless. However within seconds of its addition, the colonies/liquid cultures of XylE-expressing cells become yellow, indicating production of a product which absorbs light in the visible spectrum. A spectrophotometric assay was prepared, where the spectra of two cultures of E.coli (one XylE gene transformed and the other not) were compared on addition of 0.1mM Catechol substrate. The spectra (figure 1) showed that in XylE transformed cells, a broad peak appears at about 380nm. The absorbance at this particular wavelength is by the product of the C2,3O reaction which is called 2-hydroxymuconic semialdehyde and is what causes the yellow output.

Spectra of cultures of cells expressing the XylE and cells negative for the XylE gene. Note the broad peak in the spectra of Xyle transformed cells, which is centered around 380nm.

References

• Chromogenic identification of genetic regulatory signals in Bacillus subtilis based on expression of a cloned Pseudomonas gene. Zukowski, M.M., Gaffney, D.F., Speck, D., Kauffmann, M., Findeli, A., Wisecup, A., Lecocq, J.P. Proc. Natl. Acad. Sci. U.S.A. (1983)

Threshold value experiment was conducted on a 96 well plates. Yellow wells are the positive tests, LB color and M9 color mediums wells are the negative tests.

Addition of catechol to a a culture of cells constitutively expressing C(2,3)0 enzyme results into production of yellow colonies on plate or yellow liquid cultures if E.coli is grown in liquid medium. In a lab where a spectrophotometer is readily available anyone can accurately quantify production of even the smallest quantity of the yellow 2-hydroxymuconic semialdehyde product. However, as our biological detector is intended for field use where sophisticated spectroscopic equipment is not available, it was obvious that we needed to establish the threshold value at which a positive result (yellow color) would be detectable by the naked eye.

The set up of this experiment involved preparation of XylE expressing E.coli cultures at an O.D. of 0.5 in LB and M9 medium. Then a gradient of catechol concentrations was prepared and added to the cell cultures. A group of 5 individuals, but not the lab member preparing the test samples, were given the test cultures and asked where on the gradient they were not able anymore to see the yellow color. The 2 concentrations where the transition from yellow to colorless occurred were identified and a second gradient of catechol concentrations was prepared. In this second gradient the gap between the 2 concentration was analysed. The highest catechol concentration was the least yellow from the first part of the experiment, while the lowest catechol concentration was the first "colourless" concentration from the first part of the experiment. Again the group of 5 testers identified the transition concentrations from yellow to colorless.

The threshold concentration of catechol needed for a person to identify a yellow positive test was found to be 30-40μM.

This result was fed back to the modelling team and allowed them to constrain the concentration values of catechol to be used. Furthermore, it facilitated the determination of conditions for 2-step amplification performing better than 1-step amplifier. If you want to know more on output amplification modelling, please follow the link.

Concentration of coloured compound.

Catechol converted to 2-hydroxymuconic semialdehyde by C2,3O

In our Parasight project the XylE gene product, C2,3O enzyme, is the means by which an output signal is generated. If the parasite is present is the water supply, the bacterial system is activated by an input signal which produces the yellow color, the positive output signal. Test assays in the lab on colonies of cells expressing XylE gene, become yellow only seconds after addition of catechol (100mM). From this it was realized our group that the kinetics of the C2,3O enzyme are such that make it an attractive candidate as a reporter enzyme. So, we decided that further characterization of the kinetics of the C2,3O might help identify a really efficient reporter for research centers.

Catechol assay on XylE-trasformed cells in a 96-well plate (A to H decreasing cell concentration, 1-10 decreasing catechol concentration, column 11 and 12 negative and control)

In this experiment we investigated how the velocity of the reaction (catechol converted to 2-hydroxymuconic semialdehyde by C2,3O) varies with increasing initial concentrations of catechol, the substrate. Overnight cultures of XylE transformed E.coli were plated on a 96 wells plate at an optical density (O.D.) of 0.5 units of absorbance. (The optimum initial O.D. of cells was determined in a previous experiment. This O.D. may vary across different labs as the sensitivity of the spectrophotometer/plate reader is a factor). Catechol concentrations ranging from 0.1-1mM were made from 100mM stock solution and added to the cells of the 96-well plate. The velocity of the reaction was monitored using the plate reader spectrophotometer as the reaction is directly proportional to the production of yellow color. The yellow output was measured at 380nm wavelength, while cell density at 600nm. The XylE gene expression was under the influence of J23101 promoter.

The first catechol assay data analysis. The assay involved addition of catechol concentrations of 0.2, 0.4, 0,6 0.8 and 1mM to 0.5 O.D. of XylE transformed E.coli cell cultures. The XylE gene was under the influence of J23101 promoter.

The results of the experiment confirmed previous hints of the huge catalytic turnover rate of the C(2,3)0 enzyme. The graph on the left illustrates production of yellow product per min. The slope at the initial stages of the reaction (first minutes) would represent the rate of the reaction, that is the initial Velocity. By calculating initial velocities of the reaction at different substrate concentrations, one can plot a Michaelis-Menten curve to extract kinetic parameters such as Vmax, Km and catalytic turnover number.

Several attempts were made to determine these parameters on whole cells. Unfortunately due to the actual nature of the enzyme, velocity of reaction and technical limitations of lab equipement this has been proven impossible. To start with one of the limitation was that the color output and the extinction coefficient of the yellow product is so large, that the system saturates within the first minutes after catechol addition to cell cultures. The spectrophotometer and plate reader reaches its detection maximum reading value from the start of the reaction. Also due to the high velocity of the reaction, in the time frame of about 10 seconds that are needed to load the plate into the plate reader after catechol addition, the reaction has already proceeded too far. So initial velocity measurements of the reactions needed for construction of Michaelis-Menten curve are inaccurate. These problem could be overcome by decreasing total enzyme concentration. In whole cell cultures this would be possible by decreasing cell density in each well as each cell contain an average C(2,3)O enzyme concentration. However, previous experiments showed that cell cultures with optical densities lower than 0.5 O.D. introduce serious perturbations in the readings of the spectrophotometer and plate reader detectors.

Despite these limitations some useful data were extracted out of these assays. More specifically we acquired data that delineate the course of the reaction in terms of yellow product production over time at various catechol concentrations. The results of one of these assays is presented in the figure below. In order to extract data that will allow characterization of the kinetic parameters of catechol dioxygenase enzyme our lab team proposes purification of the protein from cell lysate, several fold dilution, and in vitro characterization.

Graph shows production of HMS (yellow product) over time after catechol addition at time 0 minutes. Different curves represent different catechol concentration added to the cell cultures.

Assaying the growth behaviour of E.coli Top10 cells transformed with either the XylE gene promoted by Pveg or alternatively as control, transformed with a CMR vector of identical size, we were able to determine if and what effect either Catechol itself, or the breakdown product 2-hydroxymuconic semialdehyde had the expected growth-inhibiting effects on the cells.

Top10 cells were transformed with XylE and CMR respectively contained on the PSB1C3 vector. Experiments were performed both in LB and M9 medium. The 100ml wells were prepared to display an array of 0 to 2mM [Catechol] exposure. To determine cell-growth behavior we measured absorbance at 600nm using a POLARstar Omega plate-reader set to take measurements every 3 minutes over one hour, without catechol. Catechol was added to all wells after the pre-growth assay was completed and measurements of optical density were taken in 3 minute intervals over 3 hours.

Growth without Catechol

Pre-growth assays showed growth behavior as expected for cells growing normally.

Growth with Catechol

(LB) The addition of catechol had distinctive effects on the XylE expressing cells growing in LB medium. While at 0% catechol growth-behavior did not show a significant change (dark blue), even the lowest concentration of 0.25% catechol appeared to drastically reduce cell-survival (red). In contrast, CMR-control cells did not change their growing behavior in the presence of catechol.

From this we conclude that in LB medium, the breakdown product of catechol, 2-hydroxymuconic semialdehyde, has a lethal effect on E. coli.

(M9) Cells growing in M9 medium appeared more resistant to the effects of catechol. Even though absorbance at 380 nm increased significantly in well containing XylE expressing cells, indicating strong turnover of Catechol by C2,3O (4), Catechol did not appear to influence growing behavior in generalizable fashion: while a significant increase in the variance in the sample could be determined. Such was not observed for CMR expressing cells exposed to the same conditions. However, we were not able to establish a clear trend of growing-behavior in the presence of catechol as in case of the LB – XylE samples.

We conclude that the breakdown product of Catechol has a strong, deleterious effect on XylE expressing cells in an M9 medium, however cells appear more able to resist the effects of 2-hydroxymuconic semialdehyde in M9 medium.

These results support the hypothesis that the breakdown product and not catechol itself has an intense deleterious effect on cell survival. This contributes to the safety of our system, as the output module is inherently coupled to the introduction to a substance lethal to the cells.

As part of the part of the project, the team needed to identify a reporter gene that would give us the visual output. However the means of getting this visual output needed to be inducible, and not constitutively active, when the appropriate input is present. Another specification for this reporter was that it's activation did not to go through the usual route of transcription/translation, which is a slow process. We had little success of finding something suitable for our project so we decided to build our own novel system. We decided to use the gene product of XylE, catechol 2,3 dioxygenase (C2,3O). By looking at three dimensional diagrams of C2,3O and using software tools we proposed an inducible model of this enzyme.

The GFP-XylE gene construct

In our construct, GFP was added to the N-terminus of C(2,3)O monomer. The C(2,3)O enzyme is an obligate homotetramer where all 4 subunits are required for its activity. We hoped that by adding GFP at the N-terminus of the each monomer, we would block the tetramerization of the enzyme, thus rendering it inactive. This modification is important as it makes our system inducible. The cleavage of the GFP by TEV protease, which is synthesised on arrival of the input signal, means that Catechol dioxygenase becomes active only when the parasite is detected.

To investigate this, an experiment was designed where catechol was added to GFP-XylE transformed E.coli. The progress of the reaction was monitored by measuring production of the 2-hydroxymuconic semialdehyde (yellow), which arbsorbs light at 380nm. The product concentraton correlates well to the velocity of the reaction, if divided by time, and by using various initial catechol concentrations as a substrate, a kinetic profile of the reaction could be constructed. The experimental set up is essentially the same as described above for determining the kinetics of C(2,3)O. The only difference is that the E.coli cells used were the ones transformed with GFP-XylE gene construct instead of XylE gene.

The results of the assay show conclusively that the N-terminal fusion of GFP gene to the XylE gene affects activity of the wild type C2,3O. Although the activity is leaky, producing very small amounts of the colored product of the reaction in the presence of catechol, the alteration in the reaction profile of the reaction is impressively large. The two graphs below are equivalent and comparable in the sense that the only different between these experiments is the E.coli cells used. The left graph represents cells expressing the original XylE gene, while the graph on the right hand side represents cultures expressing the GFP-XylE gene. These genes are both under the influence of the Pveg promoter so that we can assume equal production of protein product.

Graph shows production of HMS (yellow product) over time after catechol addition at time 0 minutes. Different curves represent different catechol concentration added to the cell cultures.

Graph shows production of HMS over time after catechol addition at time 0 minutes. Different colored curves represent different substrate concentrations added at time 0min in cell cultures

If we take 3 minutes as the reference point on the graphs, as this is where the XylE expressing cells saturate the system with the colored product, we can see that there is about a 10-fold reduction in the rate at which the protein product of GFP-XylE gene construct catalyses the reaction in comparison to the wild type XylE gene. From this we can deduce that the attached GFP protein influences tetramerisation of the Catechol dioxygenase enzyme, so that the equilibrium: 4 monomers (inactive) <-----> tetramer (active) is shifted is shifted towards protein remaining as monomers and thus inactive.

The results of this experiment were really encouraging for the team, as it shows that we have managed to inactivate the normally constitutively active reporter protein of our output module. The lab team now proposes an experiment where the inactive fusion protein (GFP-catechol dioxygenase) is exposed to TEV protease. Reversion of the reaction profile back to the kinetics of the wild type catechol dioxygenase would mean that we have successfully constructed an inducible reporter system.

In a previous experiment investigating cultures of XylE expressing E.coli, the addition of catechol and monitoring the production of HMS product in the plate reader allowed us to obtain a reaction profile of the reaction. However due to technical limitations on measuring velocity of the reaction in whole cell cultures, we were not able to obtain kinetic parameters of the catechol dioxygenase enzyme. In this experiment cell cultures of XylE expressing cells were lysed to obtain cell lysate with catechol dioxygenase activity for determination of kinetic parameters of catechol dioxygenase in vitro.

In this experiment cell lysate was assayed with increasing catechol concentrations. The reaction was monitored by measuring color output of the reaction in the plate reader. The rate at which the yellow HMS product appears is directl proportional to the velocity of the reaction. The cell lysate was prepared from cell culures of XylE expressing cells. Obtaining cell lysate with catechol assay activity allowed us to overcome the technical limitations of over saturating the system too quickly in previous experiment. Diluting the cell lysate will result in lower concentration of total enzyme in the assay, thus less substrate is converted to product per unit of time. This allowed us to take more readings at various time points before the system saturates the detector of the reader. In order to determine what the appropriate dilutions of the cell lysate for the assay, a pre-test was essential where various cell lysate dilutions were tested on their dioxygenase activity in respect to the ability of the detector to receive the signal.

Data collected from the assay allowed us to construct a Michaelis-Menten curve for the in vitro kinetics of C2,3O in cell lysate. The cell lysate was obtained from a 100ml overnight culture and the lysate was diluted by a factor of 20-fold in order to obtain a suitable concentration of total enzyme for the plate reader assay. The concentrations of catechol used to derive Michaelis-Menten curve were 1, 2, 5, 10, 25, 50 mM, prepared from 1M stock catechol.

Michaelis-Menten curve was drawn using velocity values calculated from data collected from the catechol assay. The velocity values were calculated from the rate slope at the initial stages of the reaction, as this is the only time when substrate concentration values are accurate. The plot was delineated by non-linear regression analysis using GraFit software tool. The calculated Km is 0.71mM catechol (with a Vmax of 3.37 in O.D. arbitrary units for this dilution of cell lysate).