Team:TU-Eindhoven/Results
Results
“Once you see results, it becomes an addiction”
IBDetection is a time-efficient, non-invasive, cost-effective, specific, and easy-to-use method of monitoring IBD patients. To prove that this approach is feasible, we designed a series of characterization experiments as explained on the Proof of Concept page. The results for each of the three major characterizations (TtrS/R sensing system, ARG1 reporter system, and the entire sensor) will be discussed in this section.
Summary
In short, we were able to confirm the sensitivity of the tetrathionate (Ttr) sensor in various experiments, resulting in a full dose-response curve for the TtrS/R sensing system, which detects Ttr and produces sfGFP as output (n=1). Besides, for the ARG1 reporter system, we were able to detect gas vesicles using ultrasound equipment and showed that we can increase protein production with the induction of a promoter. At last, we were able to create the engineered plasmids we designed. Unfortunately, due to time restrictions and troubles during the performance of the experiments, we had no time to test the plasmids. Nevertheless, based on the current results and the established model, we can envision a proper functioning of IBDetection in bacteria in relevant context.
TtrS/R sensing system
As explained in our Proof of Concept, the TtrS/R sensing system is characterized by several experiments. The sensing system, consisting of purified pKD227 and pKD233.7-3 plasmids, was co-transformed into BL21 (DE3) cells (Figure 1a). The red color suggests the expression of the constitutive mCherry, a component of the pKD233.7-3 plasmid, confirming the successful transformation of the plasmid into the cells. The expression of mCherry was used to quickly determine which small cultures contained our plasmids (Figure 1b). Hereafter, large cultures were grown and used for the experimental measurements.
Figure 1: a) Agar plate with BL21 (DE3) containing the TtrR/S sensing plasmids. b) Small cultures of BL21 (DE3) in LB medium. Two cultures (left) do not contain the right cells and two cultures (right) contain the TtrS/R sensing system.
During the first sfGFP and mCherry measurements, results showed that the bacteria did not respond to varying Ttr concentrations (Figure 2). Since the co-transformations worked and the ordered plasmids had not been changed, we suspected the inducers to cause these unexpected results. Therefore, eight different combinations of Ttr, doxycycline (dox), and IPTG were tested for induction, by measuring sfGFP fluorescence. Through this experiment, it could be concluded that the sensor is working properly, however, the concentration of 250 ng/mL dox causes overexpression of the TtrR protein, resulting in high sfGFP expression (Figure 2b & c). To optimize the sensor, several concentrations of dox were tested, and it was discovered that the concentration of dox equal to or smaller than 10 ng/mL gives rise to a tetrathionate sensor (Figure 2d & e). Based on these results and results from the literature, we choose to further test the sensor system without dox. In addition, to see whether the purification step in the protocol was necessary, we tested the emission inside E. coli BL21 (DE3) (Figure 2d) and in lysate suspended in PBS buffer (data not shown). We concluded that lysis of the bacteria was not necessary and therefore, was no longer performed. For more details about our progress, please read our Notebook.
Figure 2: Normalized1 GFP emissions of the TtrS/R system induced with various concentrations doxycycline, IPTG and tetrathionate(Ttr). a) Dose-response curve measured in BL21 (DE3) lysate, induced with 0.1 mM IPTG, 250 ng/mL dox and different concentrations Ttr. b) Signal measured in BL21 (DE3) lysate, induced with 0.1 mM IPTG, with and without dox (250 ng/mL), and with and without Ttr (1000 µM). c) Signal measured in BL21 (DE3) lysate, induced with 0.01 mM IPTG, with and without dox (250 ng/mL), and with and without Ttr (1000 µM). d) Signal measured in BL21 (DE3), induced with 0.1 mM IPTG, 1000 µM Ttr and various concentrations dox. e) Signal measured in BL21 (DE3), induced with 0.1 mM IPTG, with and without 1000 µM Ttr and various concentrations dox.
With the optimal inducer concentrations for the bacteria to produce TtrR and TtrS, we performed one more characterization experiment with varying Ttr concentrations, which resulted in the dose-response curve visible in Figure 3.
Figure 3: Normalized1 GFP emissions of the TtrS/R system induced with various concentrations doxycycline, IPTG and tetrathionate.1 a) Dose-response curve measured in BL21 (DE3), induced with 0.1 mM IPTG, 0 ng/mL dox and different concentrations Ttr. In addition a positive control (0.1 mM IPTG, 250 ng/mL dox and 1000 µM Ttr) and a negative control (no IPTG, no dox, no Ttr) were added. b) The dose-response from figure 3a plotted in a line-plot with logarithmic scale (EC50 = 49.1 ± 2.3 µM, n = 1).
Conclusion
Our results show a dose-response between 10 and 100 µM, and a EC50 of 49.1 ± 2.3 µM (n=1). The dose-response results correspond to the results of Daeffler et al. [1], who characterized the TtrS/R system and found a dose-response roughly between 10 and 100 µM Ttr with a EC50 of 50 ± 3 µM [1]. Hereby, we confirmed the reproduction of the sensing system of Daeffler et al. [1].
ARG1 reporter system
We conducted various experiments to characterize the ARG1 reporter system, as explained in detail in the Proof of Concept. The reporter system was transformed into BL21 (DE3) cells using the purified pET28a T7-ARG1 plasmid DNA strand (Figure 4). A colony was picked and the cells were used to grow large cultures, which were induced with different concentrations of IPTG. These large cultures have been used to purify the proteins, which were analyzed using SDS-PAGE for the expression of gas vesicle proteins, and also used to create a phantom that can be analyzed using ultrasound equipment.
Figuur 4: Agar plate with BL21 (DE3) containing the ARG reporter plasmid.
The ultrasound measurement appeared to be more difficult to perform than anticipated as described in Engineering Success. After receiving advice from experts and editing of the scripts, we performed a successful ultrasound measurement in which a signal was acquired that corresponds to the gas vesicles. The signal showed a decreasing intensity, while decreasing the amount of IPTG concentration used for the induction (Figure 5). In addition to the ultrasound images, we purified the gas vesicles by centrifuging the lysed cells (in PBS buffer) and discarded the pellet. The resulting solution was used for an SDS-PAGE to check whether the gas vesicle proteins (all < 35 kDa) would be visible. A band around 30 kDa, indicating the proteins GvpN, GvpL, GvpT [2], supported the results from the ultrasound. Furthermore, there is a visible gradient, which corresponds to the decreasing concentration of (at least one of the) gas vesicle proteins (Figure 6).
Figure 5: Ultrasound images of phantoms containing large culture with BL21 (DE3) induced with different concentrations IPTG. The images were made with the transducer at an angle with respect to the phantom containing 15 mL of 1% agar agar in PBS mixed with 15 mL large culture. Left) raw images before (pre) and after (post) running the collapse script (white signal is a reflection of the ultrasound signal). Right) processed difference between the pre and post collapse images. The white signal relates to the concentration gas vesicles.
Figure 6: SDS-PAGE of the centrifuge purified ARG proteins. On the left a Precision Plus ProteinTM All Blue Standards was used. Than there are three slots loaded with purified protein from wild type ARG large culture induced with 1000 µM, 10 µM and 0.1 µM IPTG respectively. Just as in the ultrasound a gradiënt is visible in intensity of a band around 30 kDa (red rectangle), suggesting successful control over the production of gas vesicle proteins.
The experiment was repeated two more times, though these results were not as expected. Both experiments failed due to different reasons resulting in ultrasound images without gas vesicles visible and with SDS-PAGE gels containing a lot of bands that might correspond to the gas vesicles, however, not showing any gradiënt. Therefore, these bands might also be just regular constitutive proteins from the bacteria which haven’t been purified during the process.
Conclusion
Nevertheless, we have been able to visualize gas vesicles and the proteins by ultrasound and by SDS-PAGE. Additionally, we were able to show a gradiënt in the gas vesicle (protein) concentration by both methods, which outlines the potential of the use of ARG1 as an output system for IBDetection.
Sensor
For the introduction of the tetrathionate sensor with pTtrB activated ARG1 (design A), and the tetrathionate sensor with transferred TtrR and pTtrB activated ARG1 (design B) into BL21 DE3), we first had to synthesize the plasmids that we explained in the Proof of Concept. The plasmids were created by the restriction of the pET28a_T7-ARG1 plasmid and were ligated with the required inserts, containing SgrAI and HF-NheI restriction sites. The restricted products were purified by gel electrophoresis followed by ligation. Since the distances between two bands of restricted and unrestricted DNA were small, the gel extraction was difficult. Therefore, the restricted plasmid (negative control) was transformed parallel to the transformation of the ligation product. This was done, to check if the colonies are actually the ligated product or if the colonies contain unrestricted plasmid contamination. Fortunately we had serval successful and pure gel extractions resulting in proper ligation products. For more details check Experiments and Notebook.
Results Design A
Results Design B
For design A, we were able to successfully transform the pET28a_pTtrB-ARG1 in XL10-Gold and purify the DNA. The next step was to introduce both (original) plasmids of the sensing system, containing the TtrS/R sensing system, together with this newly synthesized plasmid into a BL21 (DE3). After multiple failed attempts of introducing all the plasmids in a single co-transformation, we learned that it is also possible to do it in multiple steps. We made a culture of our BL21 (DE3) cells containing both plasmids for the TtrS/R sensing system and made them competent. We added the third plasmid required for design A to these competent cells, but this transformation failed as well. As described in the Human Practices segment the Science, we already considered this risk in the early stages of the project. Therefore, we were prepared to switch our research to testing design B and canceled experiments with design A, and all focus was set on design B.
For design B, we synthesized and transformed the pET28a_TetR-TtrR-pTtrB185-269-ARG1 plasmid into XL10-Gold. After some failures, we repeated the restriction ligations and tried to transform the product into competent E. coli TOP10 and XL10-Gold cells. In the third attempt, colonies were grown on the plates, so it can be concluded that the required plasmid was successfully transformed. This plasmid was then amplified, purified and successfully co-transformed into BL21 (DE3), XL10-Gold, and TOP10 cells together with the p15a_TtrS plasmid to create design B (Figure 7). This means that we were able to successfully create several types of bacterial cells, containing our engineered system.
Figure 7: Agar plates with bacteria containing both plasmids required for design B and thus IBDetection to work. a) BL21 (DE3) b) XL10-gold c) TOP10
Conclusion
Due to the troubles encountered during the synthesis of both design A and design B, no time was left to actually characterize the co-transformed cells containing the IBDetection sensor-reporter system. Nevertheless, we are able to predict that the system should be successful. If you want to know why, take a look at the Proof of Concept.
Project Achievements
Succes
As described above, we had several successes inside the lab.
-
Reproduce and characterize the TtrS/R sensing system
-
Create a full dose-response curve for the TtrS/R sensing system
-
Reproduce the ARG1 reporter system
-
Visualize gas vesicles with ultrasound and SDS-PAGE
-
Create a plasmid with ARG1 gene under the pTtrB185-269 promoter.
-
Create a plasmid with ARG1 gene under the pTtrB185-269 promoter with an additional TtrR and tetR protein upstream.
-
Co-transform IBDetection (design B) in BL21 (DE3), TOP10 and XL10-Gold.
Road to Succes
To reach the above successes, we had to test and adapt regularly. The most important developments were:
-
Doxycycline concentration optimization (Engineering Success)
-
Ultrasound set-up optimization (Engineering Success)
-
Restriction & ligation adjustments (Human Practices)
Repeated transformations in various types of E. coli competent cells (Notebook)
Future Goals
Due to time constraints, several experiments have not yet been done and other experiments could only be conducted once. Therefore we first have to test the complete sensor and than check for the reproducibility of our results. This reproducibility is of great importance for clinical application. Therefore, it is necessary to repeat the characterization experiments that have been performed, to see if the same intensity and dose-response are measured every time. After confirming this, further research should be done towards the final design of our project, including tests with constitutive promoters for TtrR and TtrS, experiments in Lactic Acid Bacteria (e.g. L. plantarum WCFS-1), and determination of the number of cells required for the pill. Moreover, we should test the safety of the GMO, various types of possible pills and differences between users. Details about this can be read on the Implementation page.
1. All normalized GFP emissions are signals measured at 512 nm and normalized on the mCherry emission at 610 nm.
-
Daeffler, K., Galley, J., Sheth, R., Ortiz‐Velez, L., Bibb, C., Shroyer, N., Britton, R. and Tabor, J., 2017. Engineering bacterial thiosulfate and tetrathionate sensors for detecting gut inflammation. Molecular Systems Biology, [online] 13(4), p.923.
-
Bourdeau R, Lee-Gosselin A, Lakshmanan A, Farhadi A, Kumar S, Nety S et al. Acoustic reporter genes for noninvasive imaging of microorganisms in mammalian hosts. Nature. 2018;553(7686):86-90.