2D Biosensor

With our 2D biosensor technology we are able to detect the pathogen Pseudomonas aeruginosa on solid surfaces. The sensor system is comprised of two distinct but inseparable modules, a biological and a technical part:

• Sensing chips containing Cellocks, our engineered detective cells that fluoresce in the presence of the pathogen, make up the biological part of Cellock Holmes.
• Our measurement device WatsOn and the complementary software Measurarty complete our sensing technology on the technical side.

Principle of Operation

Cellock Holmes is designed upon a SynBio approach comprising a two-dimensional biosensor and a measurement unit. The two-dimensional biosensor is designed to recognize quorum sensing molecules secreted by the pathogen cells and to generate a distinct fluorescence signal; while the measurement device recognizes and analyzes the produced signal.
On the molecular side, we use the REACh construct to transform regular E. coli cells into Cellocks.

The basics of quorum sensing. Some microorganisms can sense the presence of their own kind based on quorum sensing which is a form of chemical communication. Quorum sensing allows these cells to behave differently depending on their cell density, e.g. by activation or deactivation of certain gene expression (Waters and Bassler, 2005).

Our sensor cells, Cellocks, are immobilized in agar chips. To make the chips, we mix the Cellocks with liquid LB agar. In the course of our project, we designed a casting mold specifically for the production of our agar chips. When the agar has cooled down, the chips are cut out of the mold and are ready to use. Storage of the readily usable sensor chips is possible for up to 2 days at 4 °C when using LB medium or up to 5 days if TB medium is used. A detailed description of the sensor chip manufacturing can be found in our Protocols section.

Assay to detect P. aeruginosa using Cellock Holmes This flow sheet shows the procedure to sample and detect P. aeruginosa: A sampling chip is briefly put onto the potentially contaminated surface, added onto one of our sensor chips and inserted into WatsOn.

Using Cellock Holmes, we developed a simple assay to detect P. aeruginosa.

• First, a sampling chip is placed on a solid surface that is potentially contaminated with the pathogen.
• Second, the sampling chip is removed from the surface and put onto one of our sensor chips. (Theorectically, the sensor chips could be directly used for sampling, however, this was avoided in our project to match biosafety regulations and to prevent the spread of GMOs into the environment.)
• Third, the two layered chip-stack is put into a petri dish which is inserted into our measurement device WatsOn for evalutation.

Mode of action inside WatsOn. Chips are incubated at 37 °C to stimulate cell growth and then illuminated with blue light to excite fluorescence. A picture is taken and analyzed for fluorescence signals using the software Measurarty.

Inside WatsOn, the chips are incubated at 37 °C and the sampled populations of microorganisms attached on the sampling chip start to grow and multiply. During incubation the chips can be illuminated with blue light at any time, and a photo of the chips is taken. The software Measurarty then analyzes any fluorescent signal. P. aeruginosa secrets an increasing number of quorum sensing molecules that are recognized by Cellocks, thereby producing a fluorescence signal. For detection of P. aeruginosa, we focused on a quorum sensing molecule called N-3-oxo-dodecanoyl-L-homoserine lactone (for short: 3-oxo-C₁₂-HSL), which is involved in virulence regulation of P. aeruginosa (Jimenez, Koch, Thompson et al., 2012). The incorporation of the 3-oxo-C₁₂-HSL detection system into the Cellocks is explained in detail in the REACh Construct section.

Development & Optimization

Medium

Prior to using our own device for detection of fluorescence emitted by the sensor chips we used equipment readily available in the lab. A Molecular Imager® Gel Doc^TM XR+ from BIO-RAD was available which uses UV and white light illuminators. However, only two different filters were available for the excitation ligth wavelength, which resulted in very limitted possibilities for the excitation of fluorescent molecules. For example, it was possible to detect the expression of iLOV in our sensor chips but the detection of GFP was not possible. It was thus determined that the Gel Doc^TM was not suitable for our project.

iLOV and GFP in the Gel DocTM Sensor cells producing iLOV (A) and GFP (B) 1.5 h after induction.

Regarding the medium used for our sensor chips, LB medium showed a high background fluorescence when exposed to UV light in the Gel Doc. Surprisingly, the background fluorescence resulting from the LB medium was too high to detect a signal emitted by our sensor cells. Hence, minimal media (NA, M9, Hartman (HM)) was used to minimize background fluorescence, but this approach resulted in less to no growth of our sensor cells.

Differend medium in the Gel DocTM complex media (LB) exhibited high background fluorescence while less background fluorescence was observed with the minimal media (HM, M9, NA).

Testing our chips' shelf-life. [http://parts.igem.org/Part:BBa_K131026 K131026] in NEB induced after 5 days of storage at 4 °C. The right chip was induced with 0.2 µL of 500 µg/mL HSL, and the image was taken after 1.5 h.

In our device WatsOn, optimized wavelengths of 450 nm and 480 nm were used for excitation of iLOV and GFP, respectively. When exposed to either excitation wavelength TB medium, which is basically an improved LB medium and highly supports cell growth, showed strong background fluorescence in our own device. High background fluorescence was also observed for TB medum when using the Gel Doc. In contrast to the Gel Doc LB medium showed minimal fluorescence in our device WatsOn and no difficulties in cultivation of our Cellocks were observed. Because of the reduced fluorescence compared to TB medium when using Watson for sensor chip evaluation and because of sufficient cultivation conditions for our 'Cellocks LB medium was chosen over TB mediium for sensor chip manufacturing.

Further experiments were conducted to test long-time storage of the sensor chips. Storage at -20 °C resulted in the loss of our sensor cells. Adding 5-10% (v/v) glycerol ensured survival of the sensor cells, but resulted in an expression stop of fluorescence proteins. Thus, the idea of long time storage of the sensor chips had to be passed on. However, it was possible to store ready-to-use sensor chips for 2 days at 4 °C when using LB medium and storage for 5 days was possible with chips made from TB medium.

Agar Concentration

For the sensor chip manufacturing, a concentration of 1.5% agarose was found to be optimal. When agarose concentrations below 1.5% (w/v) were used the sensor chips were easily damaged and were not transportable. Agar concentrations over 1.5% (w/v) had to be avoided, because the agarose started to solidify before it could be poured into the chip casting mold. Agarose was chosen over agar, because of a more even linkage between molecules resulting in a better chip homogenity. In addition, agarose reduced diffusion of inducer molecules through the chip. A reduction in diffusion was desired in order to achieve distinct fluorescent spots on the sensor chips.

Chip Form

Various approaches were tried for production of sensor chips with reproducable quality. The first approach was to cast every sensor chip individually. In order to achieve a plain chip surface, which was required for high quality images, we tried to cast the sensor chips between four microscope slides. This approach had to be rejected, because the agar was too liquid. In a second try, we produced a closed mold into which liquid agar was injected using a pipette, but we encountered a high number of bubbles in the chips when using this method. Bubbles in the sensor chips resulted in problems during fluorescence evalutaion.

Chips made using the closed mold With this method we encounter problems due to frequent bubble formation.

Finally, we used an open mold into which the agar was poured right after mixing with the sensor cells. When the agar had solidified the chips were cut out along precast indentations in the casting mold. An advantage of the open mold was the ability to simultaneously produce nine sensor chips while the surface tension of the liquid agar ensured a plane chip surface.

Final chip mold. We found the above shown casting mold to be ideal for our purposes.

Induction

For artificially induction of our molecular detection constructs we simluated the presence of P. aeruginosa by use of IPTG or 3-oxo-C12 HSL. A minimal pipetting volume was desired for induction, because initial experiments showed that diffusion of the inducers through the chip hindered formation of distinct spots on the chips. Due to the pipetts available the lowest volume we could pipett was limitted to 0.2  µL . Sensor cells based on E. coli BL21, which incorporated the K1319042 construct were able to detect IPTG concentrations down to 1 mM (0.2 µL), so were sensor cells based on E. coli BL21, which incorporated the REACh constructs. Sensor cells based on E. coli BL21, which incorporated the [http://parts.igem.org/Part:BBa_K131026 K131026] construct were able to detect HSL concentrations down to 500 µg/mL (0.2 µL). Further more, detection of growing Pseudomonas aeruginosa cells based on secreted HSLs was possible using the [http://parts.igem.org/Part:BBa_K131026 K131026] construct. A detailed description including pictures of the experiments leading to the just mentioned findings can be found in the Achievements section.

Achievements

We are able to detect IPTG, 3-oxo-C₁₂ HSL and Pseudomonas aeruginosa. To prove that the sensor constructs produce the flourescence signal and not the medium or E. coli in its own we have [http://parts.igem.org/Part:BBa_B0015 B0015] in NEB as a negativ control for IPTG, HSL and Pseudomonas aeruginosa induction.

Negativ control B0015 in NEB as negativ control induced with A) 0.2 µl of 100 mM IPTG, image taken after 2.5 h; B) 0.2 µl of 500 µg/ml HSL (3-oxo-C12), image after 2.5 h; C) with 5 spots of Pseudomonas aeruginosa on the left and one big spot on the right, image taken after 2 h

Testing our Sensor Chips in a Plate Reader

To establish a prove of principle we used our construct [http://parts.igem.org/Part:BBa_K1319042 K1319042] an IPTG inducible iLOV. They were introduced into our sensor chips and then fluorescence was measured every 15 minutes after an induction with 2 µl 100 mM IPTG.

There is a clear difference in fluorescence between the not induced chip (top) and the induced chip (bottom). It is distinctively visible that the middle of the bottom chip start to exhibit fluorescence and then the fluorescence increases over time and spreads outward. The top chip also shows a slight increase in measured fluorescence but it is nowhere near the level of the induced chip and is primarily attributable to a leaky promoter and the background fluorescence.

This demonstrates a general proof of principle of the sensor chip design. Therefore the next was testing the detection of 3-oxo-C₁₂ HSL.

Detecting 3-oxo-C₁₂ HSL with Sensor Chips

As a next step, we used [http://parts.igem.org/Part:BBa_K131026 K131026] from the 2008 iGEM Team Calgary in our sensor chips to detect 3-oxo-C₁₂ HSL which is produced by Pseudomonas aeruginosa during quorum sensing. First, we tested them by direct induction with purified 3-oxo-C₁₂ HSL (0.2 µL, 500 µg/mL). A fluorescence measurement was taken every 15 min with an excitation wavelength of 496 nm and an emission wavelength of 516 nm (for GFP).

The measured fluorescence again showed a distinct signal on the induced chip (bottom) compared to the uninduced chip (top). The fluorescence clearly starts in the middle of the chip (point of induction) and then extends outwards, still showing an ever increasing signal of fluorescence. The base level of fluorescence is attributed to leakiness of the promoter and general background fluorescence of growing E. coli cells. In the induced chip (bottom), the background fluorescence is a lot lower than in the uninduced chip (top) because the signal masks the noise. The difference between the induced and uninduced chips indicates a clear response to the HSL and a proof for the ability of our sensor chip design to detect the HSL produced by Pseudomonas  aeruginosa.

Detecting IPTG with Sensor Chips

This video shows the construct [http://parts.igem.org/Part:BBa_I746909 I746909] from the 2007 iGEM Team Cambridge. This BioBrick is a producer of superfolder GFP under the control of a T7 promoter. It was introduced into BL21(DE3) cells making the expression IPTG inducible through the T7 RNA Polymerase encoded in the genome of BL21(DE3) under the control of a lacI promoter.

The left chip does not show visible fluorescence and the right chip exhibits a strong fluorescence signal showing clearly the ability of the sensor chip technology to detect IPTG. On top of that, the fluorescence response is strong enough to be detected and analyzed by the measurement device WatsOn.

Detecting the 3-oxo-C₁₂ HSL with K131026 in our Sensor Chips with WatsOn

The next step towards the final goal to detect Pseudomonas aeruginosa was to replicate the detection of 3-oxo-C₁₂ HSL, which was established in the plate reader, in our own WatsOn device. Therefore, we again used K131026 as our construct in E. coli BL21(DE3) cells and induced with 0.2 µL 3-oxo-C₁₂ HSL with a concentration of 500 µg/mL. The right chip was induced and - as a negative control - the left chip was not induced. Pictures were taken every 4 min.

The result was a clear replication of the success of the plate reader experiment. The induced chip shows a clear fluorescence response eminating from the center where the induction with HSL took place. This demonstrates the ability of not only our sensor chips but also our measurement device WatsOn to successfully detect 3-oxo-C₁₂ HSL.

Detecting Pseudomonas aeruginosa with K131026 in our Sensor Chip with WatsOn

After establishing the successful detection of 3-oxo-C₁₂ with our sensor chips the next step was the detection of Pseudomonas aeruginosa with our measurement device WatsOn. Therefore sensor chips with K131026 were again prepared and the right chip was induced with 0.2 µl of a Pseudomonas aeruginosa culture while the left chip was not induced.

The results clearly demonstrate our ability to detect Pseudomonas aeruginosa with our measurement device WatsOn. On the induced chip a definite fluorescence response is visible in response to Pseudomonas aeruginosa. The fluorescence eminates outward from the induction point and shows a significant difference to the non induced chip. Therefore detection of Pseudomonas aeruginosa is possible with our sensor chip technology in our measurement device WatsOn!

Comparing the REACh Construct with K731520 and I746909

More information about the kinetic differences between these construct in our sensor chips, look under The REACh Construct Achievements.

Outlook

After successfully detecting P. aeruginosa, the next step in developing our sensor chip platform further is an improvement of the sampling chip. The current technique of using a simple agarose chip is not sufficient to collect all microorganisms from the sampled surface. Therefore, the aim is to improve the sampling chip by trying different, more adhesive material.

Furthermore, diffusion in the sensor chips will be reduced to limit the spread of the fluorescence signal. Currently, the fluorescence spot grows at lot beyond the point of detection and makes it difficult to successfully differentitate between multiple points of induction. By introducing different diffusion barriers into our chips, the growth of the fluorescence spots might be limited, thus enabling the detection of multiple sources of fluorescence lying close together.

Additionally, the application of our sensor chips in combination with our WatsOn device is currently being evaluated for the detection of signals other than fluorescence. Detecting bio- and chemiluminescence is identified as an area of potential future application.

References

• Waters, C. M., & Bassler, B. L. (2005). QUORUM SENSING: Cell-to-Cell Communication In Bacteria. Annual Review of Cell and Developmental Biology, 21(1), 319-346. Available online at http://www.annualreviews.org/doi/full/10.1146/annurev.cellbio.21.012704.131001.

• Jimenez, P. N., Koch, G., Thompson, J. A., Xavier, K. B., Cool, R. H., & Quax, W. J. (2012). The Multiple Signaling Systems Regulating Virulence in Pseudomonas aeruginosa. Microbiology and Molecular Biology Reviews, 76(1), 46-65. Available online at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3294424/#B63.