Team:UPF Barcelona/Measurement
Results
Alexandria - Measurement
Introduction
Assembly of the detection tests
After lysing our cells we proceed with the detection (see lysis details in Biosensor library testing page). For being able to detect if our biosensors worked we used the DNaseAlert Substrate Nuclease Detection System from IDT. For each 50uL detection reaction, the components shown in Table 1 were added. For more detailed information about the detection protocol click here [1].
| Detection tube | Kit positive control | Kit negative control | |
|---|---|---|---|
| DNase Alert substrate (FQ-reporter) | 5uL | 5uL | 5uL |
| x10 DNase Alert Buffer | 5uL | 5uL | 5uL |
| gRNA-Cas12a | 36uL | - | - |
| Target DNA | 4uL | - | - |
| DNase I | - | 1uL | - |
| Nuclease-Free water | - | 40uL | 40uL |
| Total Volume | 50uL | 51uL | 50uL |
Table 1: Substrate Nuclease Detection System reaction.
It is important to note that the detection reactions were carried out in nuclease-free conditions, also all the materials to be used were cleaned with ethanol and nuclease decontamination solution. This ensured preservation of the FQ-reporter, since it is vulnerable to DNase degradation.
The reagents present in the DNase alert kit are fluorophore-quencher oligonucleotide probes that emit fluorescence after nuclease degradation. When the reporter is cleaved by a DNase, the substrate fluoresces pink. To know how the fluorescence signal works go to the Biosensor library testing page. The assays can be visualized qualitatively by eye (with the transilluminator) or quantitatively by the plate reader in RFU (Relative Fluorescent Units). An example of a qualitative fluorescence measurement is shown in Figure 1.
| Parameters | Value |
|---|---|
| Temperature | 37ºC / Room temperature |
| Duration | 2h |
| Use kinetic interval | Measure every 5 minutes |
| Excitation | 530 nm |
| Emission | 560 nm |
| Number of flashes | 20 |
| Measure from | Bottom |
| Gain | Optimal (100 RFA) + Use Gain Regulation |
Table 2: Plate reader configuration.
Innovative measurement tool
After measuring the fluorescence of our preparations, the IRIS mobile application comes into play. First, we take a picture with the mobile camera and after that, IRIS tells us if the sample presents resistance or not to the specific antibiotic. For more information about IRIS click here.
Biosensor testing
As a brief summary of all the conditions tested and measured, the following table compiles these specifications.
| Conditions tested: different combinations | |||
|---|---|---|---|
| Lysis type | with EDTA | without EDTA | |
| IPTG concentration for transcription induction | 10 uM IPTG | 100 uM IPTG | |
| Temperature | 37ºC | Room temperature (25ºC) | |
| gRNAs | Preliminary designs (gRNA construct, shorted gRNA for the figure legends) | Improved designs (Efficient gRNA construct, shorted Efficient gRNA for the figure legends) | |
| Additional specifications | |||
| Additional negative control (without Cas12a) | |||
| Biological replicates | |||
| Results measured over time | |||
| Sensitivity analysis | |||
Table 3: Summary of the tested conditions and further aspects considered.
Preliminary tests
All the preliminary experiments were carried out with the first gRNA constructs that we designed (see Part Collection page for a summary of all our parts): gRNA Chloramphenicol construct (BBa_K3791011), gRNA Erythromycin construct (BBa_K3791012), gRNA Kanamycin construct (BBa_K3791013) and gRNA Spectinomycin construct (BBa_K3791014). We did not perform any experiments with the gRNA Ampicillin construct (BBa_K3791010) nor Efficient gRNA Ampicillin construct (BBa_K3791020), because of an impairment of its design.
- It was observed that colonies and cultures for gRNA Ampicillin construct (BBa_K3791010) and Efficient gRNA Ampicillin construct (BBa_K3791020) were not viable. This led to a review of the design of these particular parts. It was hypothesized that the assembled gRNA-Cas12a system was targeting the plasmid’s selection resistance gene, in this case, Ampicillin. Cleavage of this particular sequence made the resistance gene unfunctional, provoking the cell’s death when cultured with the aforementioned antibiotic. This event actually proved that the gRNA-Cas12a was functional and able to cleave the targeted sequence, which is an important proof of our system viability. Our error in the gRNA design allowed us to verify biosensors functioning.
This way, we were able to test our detection system against 4 different antibiotic-resistant genes (chloramphenicol, erythromycin, kanamycin, spectinomycin). Also, it is important to keep in mind that the following reactions were carried out using a co-transformed cell lysate. Meaning that both LbCas12a and the corresponding gRNA plasmid were expressed in a single bacterial cell. Furthermore, in each reaction, measurements have been taken approximately after 30 minutes of incubation, as recommended by the commercial reporter kit [1].
The first conditions to test were the enzymatic lysis conditions (whether the buffer should contain EDTA or not) as well as two different IPTG concentrations (transcriptional inducer): 100uM and 10uM. These were firstly performed with one of our gRNAs: the gRNA Chloramphenicol construct (BBa_K3791011). It was assumed that the remaining biosensors would share intrinsical properties and thus cleavage efficiency, since the only variable part would be the gRNAs sequence. That is why the conclusions drawn from these experiments were extrapolated and applied to the remaining gRNAs: gRNA Erythromycin construct (BBa_K3791012), gRNA Kanamycin construct (BBa_K3791013) and gRNA Spectinomycin construct (BBa_K3791014).
Discussion
It has been observed that in Figure 2a, fluorescence difference between positive and negative samples is not clear. However, it is much more evident in Figure 2b (without EDTA), where we can clearly differentiate between positive and negative samples, achieving a successful detection. This can be attributed to the fact that the Cas12 nuclease could be inhibited by the EDTA, which acts as an ion scavenger, and thus its activity is inefficient. Also, when comparing both IPTG concentrations, in Figure 2b, it can be seen that a bigger difference in fluorescence intensity between positives and negatives is achieved at an IPTG concentration of 100uM.
Nevertheless, it can be perceived that all measurements show quite a high fluorescence level. This can be attributed to the fact that, when working with cell lysates, there is an intrinsic abundance of nucleases, which could be contributing to the fluorescence signal. Contrary, the negative control provided by the kit (see Table 3 above), which does not contain cell lysates, shows a very low fluorescence signal. Regarding the positive control (see Table 3 above), it is expected to show high nuclease activity. Its measurements are not attributed to any biological significance, since the only information that it gives is to verify that the nuclease detection kit is working properly.
In conclusion, after having analyzed the previous statements, the lysis conditions without EDTA and inductions at 100uM IPTG seem to be optimal. These are then pre-established and tested with the other gRNAs: gRNA Erythromycin construct (BBa_K3791012), gRNA Kanamycin construct (BBa_K3791013) and gRNA Spectinomycin construct (BBa_K3791014).
Additional tests were carried out to verify the previous condition-related hypothesis. As expected, significant differences between the detection reaction (with target DNA) and the negative control (without target DNA) were observed in a 100uM IPTG cuLture, lysed without EDTA presence.
From now on, another additional negative control was taken into account, being a reaction only containing bacterial lysate with the corresponding gRNA (negative control for Cas12a). This additional test implies biological significance, as it aims to verify that high fluorescence is a consequence of the coupled activity of the gRNA-Cas12a complex.
Discussion
It has been observed in Figure 3a, that the detection reaction shows a peak at fluorescence measurement of 44158,5 RFU while in Figure 3b the peak is at 50149 RFU and in Figure 3c at 30874,5 RFU. A significant difference with all the negative controls is observed. It is also noticed that there is considerable background noise in Figure 3a and c (due to intrinsical nuclease activity of the lysates).
It can be assumed that in all three gRNAs: gRNA Erythromycin construct (BBa_K3791012), gRNA Kanamycin construct (BBa_K3791013) and gRNA Spectinomycin construct (BBa_K3791014) detection measurements were successful with the predetermined conditions. Consequently, these will be applied for all the next experiments.
Systematically, reactions were carried out at 37ºC as recommended in protocols for in vitro detection with Cas12a [3]. But additionally, room temperature (RT) detections were performed in order to verify if good results could be obtained without following the incubation requirements. This would make sense, since not in all contexts an incubator would be available for carrying out the detection of antibiotic-resistant infections. As our final aim is that ARIA can be applied as a fast detection system in clinical use, it was important to test if our biosensors were able to detect at room temperature.
Discussion
As it can be seen in Figures 4b and 4d, significantly different results were obtained for the positives versus the negatives, unlike Figures 4a and 4c. As it cannot be confirmed that this detection method will work in all cases, we decided to continue with our measurements at 37ºC. Further development and refinement of the detection protocol would be needed to perform it at room temperature.
Further improvements
To aim for higher efficiency in the final product, we improved our existing gRNAs (visit Design and Engineering pages). Another important addition was considering three biological replicates for more statistical accuracy and trustable results. With each of these measurements, a mean value was calculated as well as a standard deviation. This last parameter made the data more reliable.
The first gRNA to be tested following the previous guidelines was the Efficient gRNA Chloramphenicol construct (BBa_K3791013), which corresponds to our composite part (BBa_K3791021). It is with this part that additional features were measured (see Wetware proof of concept page).
Due to time constraints, the functionality of the method couLd only be verified with the Efficient gRNA Erythromycin construct (BBa_K3791012). Tests with the Efficient gRNA Kanamycin construct (BBa_K3791013) and the Efficient gRNA Spectinomycin construct (BBa_K3791014) are left for future work.
Testing the new design for the optimized gRNAs the detection was again successful.
Discussion
For both Efficient gRNA constructs a distinguishable difference between the positives and negatives can be seen. Peak values of 24578.67 RFU for Efficient gRNA Chloramphenicol, and 32677 RFU for Efficient gRNA Erythromycin are found in an optimal range. Even though in some of the tests with the previous gRNA constructs (not efficient ones) a distinguishable difference between the positives and negatives was also achieved (see Figure 2b and 3a), those are less reliable since no biological replicates were used. Further testing of Efficient gRNA constructs should aim to determine if efficiency was certainly increased.
Until this point, detection reactions were carried out using a cell lysate where both Cas12a and the gRNA plasmids were expressed (“joined” approach). This procedure aimed to ensure the biosensors assembly prior to the addition of the target DNA.
However, another constraint was tested. In this case, Cas12a and gRNA plasmids were transformed and expressed in bacteria separately (“separate” approach). Then, when preparing the detection reaction, both components were mixed in situ. This approach allowed us to test how long would it take for the Cas12a to couple with the gRNA, and consequently digest the target DNA sequence.
Discussion
While in the “joint” approach a clear detection is observed, in the case of the “separate” approach the results obtained are not reliable. More replicates and further experiments are needed to confirm under which conditions could we perform the “separate” approach and obtain reliable and coherent results. It is hypothesized that the necessary conditions for Cas12a and gRNA coupling have not been met. In order to carry out this approach, a better protocol sequence should be studied.
A detection maintained over time (2h) was performed to observe the stability of the fluorophore. See Wetware proof of concept page for all conditions and results.
Additionally, the biosensors’ sensibility was also tested. DNA template dilutions were carried out at different concentrations (90ng/uL, which was used in all previous tests, but also 25 ng/uL and 2.5 ng/uL).
Discussion
ResuLts shown in the graphic apparently demonstrate that lower template DNA concentrations could also be feasible for detection, since the 25ng/uL sample is recognized as a positive. The result is inconclusive for the concentration of 2.5ng/uL. A very large standard deviation (19587 RFU) can be observed, which is due to the fact that one of the values obtained is significantly lower than the mean value (787 RFU). This value could be discarded and then the result would turn out as a positive, anyway it was decided to leave this value so as not to omit information.
It is also observed that the negative control of the kit is very high, probably due to a possible contamination by DNAses. This is why further experiments are still pending to verify the sensitivity of the biosensor.
For this section, preliminary results which give important information were accomplished. We also realized some crucial issues for our further work in the future. Regarding the autolysis protein, we were able to perform the cloning of the fragments correctly (Figure 8). As can be seen, both the Digestion and PCR of the purified parts went well, and we assume that there was no problem in performing the Gibson Assembly.
However, we did not consider an important aspect. After performing a transformation on NZY5α competent cells, we cultured them on an Agar plate without D-Glucose. The result was that the next day we had not obtained any colony (Figure 9).
Discussion
We consider that the lack of colonies is due to the leakiness of the expression of the autolysis protein and that its minimal expression is already able to lyse the cells. Therefore, we know that in the future we will have to grow the cells in a medium rich in a high concentration of D-glucose to completely inhibit the PBAD promoter [4].
Further experiments remain to be performed, but we have proved one important feature about Protein E leakiness. For this reason, a composite part registry page containing this information and our design with pBad promoter and Protein E was created (BBa_K3791025) in order to improve the Protein E existing part (BBa_K2500006).
References
[1] (N.d.). Retrieved October 17, 2021, from: Windows.net website
[2] (N.d.-b). Retrieved October 5, 2021, from: Wisc.edu website
[3] New England Biolabs. (n.d.-b). In vitro digestion of DNA with EnGen® Lba Cas12a (Cpf1) (M0653). Retrieved October 17, 2021, from: Neb.com website
[4] Guzman, L. M., Belin, D., Carson, M. J., & Beckwith, J. (1995). Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. Journal of Bacteriology, 177(14), 4121–4130. Doi: 10.1128/jb.177.14.4121-4130.1995
