Team:Thessaloniki/Engineering

☰







Engineering Success





Overview


For our engineering success, our main goal was the design, evaluation, and improvement of our biological system. Our designed system was based upon the technology of synthetic riboregulators called toehold switches. These molecules are designed to detect some specific miRNAs that are present during Pancreatic Ductal Adenocarcinoma, the most common type of pancreatic cancer. The process of engineering includes the detailed design of those synthetic riboregulators that are built and tested in the laboratory. Also, it includes the experiments for fluorescence measurement for their evaluation and improvement. The improvements concerned the replacement of the RBS for the second generation sequences and also, the selection of different reporter genes. In our case, the different reporter genes are different fluorescent proteins. Moreover, since it is crucial to store this sensitive synthetic system, we cloned our parts. Finally, we designed and tested an isothermal amplification protocol, the so-called EXPAR, to increase the sensitivity of our system. So, our engineering success is built upon these three main directions.



1st Direction



The main goal of our project is the early detection of adenocarcinoma in the pancreas. This certain type of cancer was chosen because its diagnosis is made in the last stages (III and IV) [1], and thus it is considered one of the most deadly forms of cancer [2]. Our goal is to diagnose PDAC in its early stages by detecting three upregulated miRs, miRNA 143-3p, 30e-5p, and 1246 in urine, which were selected from literature, and their importance was confirmed through extensive bioinformatics analysis. The detection was performed using synthetic RNA sequences called Toehold Switches.



| Engineering cycle 1




In the first engineering cycle, the RBS part in synthetic toehold switches sequences was modified. The first generation parts contained the RBS sequence used by Green et al. The second generation contained a different RBS, the RBS-DHFR, suggested by the manufacturer of the cell-free system we used. This modification was made to check whether the fluorescence levels measured were increased.



~Design

For our project, we chose the trigger RNAs hsa-miR-143-3e, hsa-miR-30e-5p, and hsa-miR-1246. Our synthetic sequences comprise of the complementary to the miRNA sequences, followed by the RBS, the AUG codon, and the Linker, which are mentioned in the article of Green [3]. The reporter gene selected was eGFP, which has a maximum absorption spectrum at 488 nm and an emission spectrum at 507 nm. The sequences were designed to avoid termination codons and restriction sites for the restriction enzymes we used.

From these sequences, 3 were selected. More specifically, we selected one sequence per miRNA. They showed the best and the most promising results since the thermodynamic parameter of Gibbs free energy from the RBS part to the end of the Linker was close to zero. This energy was calculated by ViennaRNA and NUPACK, software designed for studying nucleic acid interactions. This parameter is a thermodynamic property that is considered to have the highest correlation with the functionality of the toehold switch-trigger RNA system [3].

You can check our Model page for a more detailed description:

Model






~Build

In accordance with the design, we ordered specific sequences in the form of double-stranded, DNA (g –Blocks) from IDT.

Specifically, the following parts were used:

For these experiments, we used both a negative and a positive control sample. As a negative control, we used all the reagents from the cell-free system (NEB’s PURExpress® In Vitro Protein Synthesis Kit #E8600), while as a positive control we used the same volume of cell-free system reagents and the DHFR plasmid.



~Test

In order to test the functionality of Toehold switches designed, we executed a series of experiments using a cell-free system (NEB - PURExpress® In Vitro Protein Synthesis Kit #E8600). These experimental procedures aimed to prove the stability of our Toeholds; that is the unfolding of the hairpin as well as the production of eGFP protein, only in the presence of the trigger RNA. In our case, the trigger RNA is the miRNA. For this purpose, we used black 96-well-plates (Greiner #655077) and in each well, we added:
  • 1. The toehold switches, BBa_K3727017, BBa_K3727026, BBa_K3727029 respectively, with the reagents of NEB - PURExpress® “In Vitro Protein Synthesis Kit” (#E8600), in three separate wells. The toehold switches’ quantity was 30ng per 25 ul reaction.
  • 2. The toehold switches, as described above, with the complementary miRNA mimics (Thermofisher - mirVana® miRNA mimic #4464066). The miRNAs’ concentration was 10nM, in three separate wells.
  • 3. The reagents of the NEB’s cell-free system PURExpress® “In Vitro Protein Synthesis Kit” (#E8600), as a negative control.
  • 4. The reagents of the cell-free system NEB - PURExpress® and the DHFR plasmid as a positive control, in case of electrophoresis. The DHFR control template is supplied at 125 ng/μl. So, we used 2 ul for the positive control reaction.


The reagents used in the experiments are in Table 1:

Reagents Volume
Solution A 10 ul
Solution B 7.5 ul
RNAse inhibitor 0.5 ul
Template DNA 30 ng
Nuclease free water to 25 ul
Table 1. Reagents used in the fluorescence measurements.


For the measurement of the fluorescence, we used the device “Victory-X 3000” from Perkin Elmer company.



~Learn

The measurements conducted adhere to the excitation/emission protocol of 485/535 nm.

From the results of these experiments, we concluded that in order to increase the ON/OFF ratio, it is essential to increase the fluorescence that is detected. The ON/OF ratio parameter is crucial for the quantification and hence, for the formation of the standard curve of fluorescence.

The improvement of the system can be achieved through a different design of toehold switches. To achieve higher expression of the reporter gene of eGFP, we tried incorporating a different RBS. Since we utilized NEB’s cell-free system during fluorescence measurements, we chose the DHFR plasmid RBS. This way we achieved a higher affinity between the RBS and the ribosome and improved the expression of the reporter gene.



Figure 1. This graph presents the fluorescence produced by Toehold switches in the presence of the complementary miR mimic (Thermofisher - mirVana® miRNA mimic #4464066) in comparison with the fluorescence produced by the Toehold switch in the absence of miRNA (Negative Control).




| Engineering cycle 2



In this cycle, the second generation of toehold switches was designed and tested experimentally. From the results obtained, it seemed necessary to optimize the conditions of the experiments and use another fluorescent protein to avoid the signal received due to background fluorescence.



~Design

The second generation of toehold switches was designed to be more compatible with the ribosomes of the cell-free system. For this reason, the sequences of the second generation of toeholds followed the same structure as the previous ones with one exception. The RBS used for this generation was the RBS from the DHFR plasmid, which is supplied with the cell-free system we used.





~Build

In accordance with the design, we ordered specific sequences in the form of double-stranded DNA (g –Blocks) from IDT.

Specifically, the following parts were used:





~Test

In order to test the functionality of Toehold switches designed, we executed a series of experiments using a cell-free system NEB’s PURExpress® “In Vitro Protein Synthesis Kit” (#E8600). These experimental procedures aimed to prove the stability of our Toeholds; that is the unfolding of the hairpin as well as the production of eGFP protein, only in the presence of the trigger RNA. In our case, the trigger RNA is the miRNA. For this purpose, we used black, 96-well-plates (Greiner #655077) and in each well, we added:
  • 1. The toehold switches BBa_K3727016, BBa_K3727027, BBa_K3727028 respectively, with the reagents of NEB’s PURExpress® “In Vitro Protein Synthesis Kit” (#E8600), in three separate wells. The toehold switches’ quantity was 30ng per 25ul reaction.
  • 2.The toehold switches, as described above, with the complementary miRNA mimics (Thermofisher - mirVana® miRNA mimic #4464066). The miRNAs’ concentration was 10nM, in three separate wells.
  • 3. The reagents of the NEB’s cell-free system PURExpress® “In Vitro Protein Synthesis Kit” (#E8600), as a negative control.
  • 4. The reagents of the NEB’s cell-free system PURExpress® and the DHFR plasmid as a positive control, in case of electrophoresis. The DHFR control template is supplied at 125 ng/μl. So, we used 2 ul for the positive control reaction.


The reagents used in the experiments are in Table 2:

Reagents Volume
Solution A 10 ul
Solution B 7.5 ul
RNAse inhibitor 0.5 ul
Template DNA 30 ng
Nuclease free water to 25 ul
Table 2. Reagents used in the fluorescence measurements.


For the measurement of the fluorescence, we used the device “Victory-X 3000” from Perkin Elmer company.



~Learn

The measurements conducted adhere to the excitation/emission protocol of 485/535 nm.



Figure 2. This graph represents the results of fluorescence experiments of Toehold switches BBa_K3727016 with miRNA 143-3p, BBa_K3727027 with miRNA 30e-5p, and BBa_K3727028 with miRNA 1246. The control bar shows the fluorescence of the well, which includes only the reagents of the “PURExpress In vitro Protein Synthesis kit” (#E6800) as a negative control. The second bar shows the fluorescence produced by Toehold switches in the presence of the complementary miR mimic (ON) (Thermofisher - mirVana® miRNA mimic #4464066) and the third one, the fluorescence produced by the Toehold switch in the absence of miRNA (OFF).




~Improved results using the protocol at 488/507 nm

According to the literature, the eGFP excitation spectrum has a peak at 488 nm, while its emission spectrum has a peak at 507 nm. To achieve a more accurate measurement, we used the device of Promega “GloMax-Multi Detection System” which has a special protocol and filter for the measurement of fluorescence of this protein. The spectrum of emission and excitation of eGFP is demonstrated in the next diagram.
~Data was taken from FPBase~


Figure 3.The emission and absorption spectrum of eGFP




Figure 4. This graph represents the results of fluorescence experiments of Toehold switches BBa_K3727016 with miRNA 143-3p, BBa_K3727027 with miRNA 30e-5p, and BBa_K3727028 with miRNA 1246. The control bar shows the fluorescence of the well, which includes only the reagents of the “PURExpress In vitro Protein Synthesis kit” (#E6800) as a control. The second bar shows the fluorescence produced by the Toehold switch in the presence of the complementary miR mimic (ON) (Thermofisher - mirVana® miRNA mimic #4464066) and the third one, the fluorescence produced by the Toehold switch in absence of miRNA (OFF).




From the results of this protocol, we concluded that the background fluorescence of the cell-free system is high. That means this specific protocol is not capable of detecting the low fluorescence produced only by the eGFP. Since the detection of small amounts of fluorescence is not achieved, then the detection of small amounts of miRNA can not be detected. To improve the ON/OFF ratio, it was important to find an appropriate protocol in which the cell-free system did not produce fluorescence. For that reason, an experiment to measure the background fluorescence in a different protocol was performed. The protocol chosen was designed for the fluorescent protein EBFP. The spectrum of emission and excitation of EBFP is shown in the next diagram.~The data used was taken from FPBase.~


Figure 5.The emission and absorption spectrum of eBFP.





Figure 6. Background fluorescence of NEB’s cell-free system(#E6800) for 2 different protocols : 1) at 488/507 nm, 2) at 380/420 nm.




The excitation/emission protocol used for the measurement of the background fluorescence of the cell-free system was 488/507 nm and 380/420 nm. These specific nanometer wavelengths correspond to the peaks of eGFP and EBFP spectra, as shown in Figures 3 and 5. From the results of this experiment, we can conclude that the detected fluorescence for the EBFP protocol is decreased in comparison with the fluorescence detected from the eGFP protocol.

Moreover, EBFP has several advantages compared to eGFP. The overlap of the two EBFP spectra is smaller compared to the overlap of eGFP’s spectra. This can be validated by calculating the ratio of the overlap area to the whole area of the spectra for each protein. Specifically for the protein eGFP, this ratio is calculated at 0.139 and for the protein EBFP, this ratio can be calculated at 0.075. The decrease in that ratio means that the interference effect is minimized and the process of fluorescence detection is more accurate. So, the use of this specific protein as a reporter gene is likely to improve our system’s functionality for detecting small amounts of miRNAs and in general to increase the ON/OFF ratio.






2nd Direction


After designing and ordering our parts we decided to clone them into an appropriate vector to be able to amplify and store them while also avoiding possible mutations so that they could be intact for future use.

Our attempted cloning procedures include 2 separate cloning methods; BioBrick Assembly and PCR Cloning. Our efforts can be categorized into 2 engineering cycles which are analyzed below. In the first one we utilized the BioBrick Assembly method and in the second one the PCR Cloning procedure.




| Engineering cycle 1



~Design

We designed our parts by combining the toehold switch sequence and the downstream gene eGFP while also adding the BioBrick prefix and suffix so that they could be cloned. We also added two leader sequences upstream of the BioBrick prefix and downstream of the BioBrick suffix, so that the restriction enzymes would be able to bind to the DNA sequence and cut at the appropriate site. These sequences would be cut off after the digestion of the parts and are not inserted into the vector - hence, they are not considered a section of the part.

The main layout of our components is the following:

Leader Sequence --> BioBrick Prefix --> T7 Promoter --> Toehold Switch --> eGFP --> T7 Terminator --> Biobrick suffix --> Leader Sequence

We designed our experiment protocols on the Benchling platform and simulated all the procedures with the Assembly Wizard tool.



Figure 7: This is the plasmid map of one of the Toehold Switches we cloned, and more specifically T-30e-5p-1, created by Assembly Wizard. For this simulation we indicated the restriction enzymes we decided to use, that is EcoRI and PstI and the software automatically created the map and designated possible mistakes. We got no error messages, which means that our part can be cloned in the desired vector. Here we provide only this particular example of our simulations because all the parts have the same layout.



We also simulated the restriction digestions using the virtual digestion tool to ensure that the restriction sites were appropriate for this cloning method.



Figure 8: In this figure the sizes of the linearized vector and the digested g-Block are shown. This tool helped us in understanding the sizes and the position of the bands we expected to see on the gel after electrophoresis and as a result the appropriate time to run the gel. The sample on the left is the digested g-Block and the bands at 75 bp correspond to the edges of the g-Block. These fragments are very small -approximately 15 bp each- and can not be visualized through electrophoresis.



The results of our simulations proved that our parts are compatible with the BioBrick Assembly method as specified by the iGEM Competition.



~Build

We used 2 separate resources to decide on our protocol, [4] and [5]. The cloning pipeline we followed consisted of a series of experimental techniques which are shown in chronological order:
  • 1. g-Block Resuspension
  • 2. g-Block Digestion
  • 3. Vector Digestion
  • 4. Ligation of insert and vector
  • 5. Transformation in competent cells
  • 6. Inoculation of liquid bacterial cultures
  • 7. Plasmid DNA isolation with minipreps
  • 8. Diagnostic Restriction Digestion
  • 9. Sanger Sequencing


In the first step, we also need to mention that we aliquoted the resuspended g-Blocks to avoid contamination and many unwanted freeze-thaw cycles.

You can find these protocols in our Experiments Page in the section "Cloning":

Experiments





~Test

In this part of our engineering cycle, we tested our protocols as defined above, and we tried different conditions to achieve the best possible results. Unfortunately, we were not able to obtain any colonies out of our transformation efforts.

As mentioned above the first steps of our pipeline concern the digestion of the g-blocks and the vector. We conducted these experiments as described in the corresponding protocols and ran electrophoresis of the samples. We got the same image as expected from our simulations, which means that there were no mistakes in these procedures that could be responsible for our results.

Initially, we used a ratio of 1:3 insert:vector in our ligation procedure, but did not get promising results, so after troubleshooting and consulting both the supervisors in our lab, our mentors, and our partners from iGEM Patras, we changed the ratio to 1:5 and even later on, to 1:7. Also, we used a combination of the above, with different ligation conditions. At first, we used an incubation at room temperature for a short period of time - 15 to 20 minutes - and then heat inactivation at 80°C or we incubated the ligated product at 16°C overnight. Regrettably, all of the above did not bear any encouraging results.

In every transformation, we used one plate for a positive control to see whether the problem was the competent cells we were using but still, this was not the issue. Our positive control had a lot of colonies whereas our samples were empty. So, we gathered that our competent cells were not ineffective, as to our ligation which was indeed problematic.

After many unsuccessful efforts to clone our parts, we decided to test if we can digest the vector only and re-ligate the linearized part and its insert. We used the same protocols as mentioned above, but still, we got no colonies. This way we concluded that there was no problem with the design of our parts, but with the ligation protocol itself.

Furthermore, during these experiments, we found out that the aliquots of the resuspended g-Blocks had different concentrations and in many of them it was very low. This could also be a problem that held back our cloning.




~Learn

Although we got no results, through troubleshooting, we came to a conclusion of what was wrong and found possible solutions. Unfortunately, due to the pressuring timeline of the competition, we were not able to test all the suggested alternatives from the troubleshooting of the first engineering cycle so we decided to proceed with the PCR cloning method to see whether we would receive better results.

Below, we are citing the solutions we came up with and the factors we believe they would affect.

Troubleshooting Possible explanations
Different T4 ligase Increase the ligation efficiency
Extended ligation time Increase the possibility of a successful ligated product
Different growth medium Increase the growth of transformed cells during the outgrowth step of transformation.
Reorder the parts and repeat the resuspension Achieve a better concentration of our toeholds

Table 3: Troubleshooting for cloning procedure






| Engineering Cycle 2



~Design

We used the same parts but this time we used PCR to amplify our sequences. For the design of our PCR, we used the Tm Calculator from NEB to ensure the best annealing temperature with regard to our primers’ nucleotide number, Tm, and GC content.

We also used the Benchling platform to cross-check the primers’ Tm which showed a slight difference so we tried both of the suggested annealing temperatures.



~Build

For this procedure, we used NEB's “PCR Cloning Kit” and we organized our experiments based on the protocol supplied by the manufacturer. The pipeline of these experiments is the following:
  • 1. Primer preparation
  • 2. Amplification of g-Blocks with PCR
  • 3. Gel extraction
  • 4. Ligation of insert and vector
  • 5. Transformation in competent cells
  • 6. Inoculation of liquid bacterial cultures
  • 7. Plasmid DNA isolation with minipreps
  • 8. Diagnostic Restriction Digestion
  • 9. Electrophoresis
  • 10. Sanger Sequencing



You can also find these protocols on our Experiments page:


Experiments




For the amplification of the parts with PCR, we used the “Q5-High Fidelity Master Mix” from NEB and we tried different settings, based on the corresponding protocol.



~Test

The bottleneck for this cycle was to achieve a satisfying amplification of the toeholds with PCR so we tried different approaches to succeed it. In this part of the engineering cycle, we used different temperature conditions for each PCR step. Specifically, we used both of the suggested Tms from Benchling and NEB Tm Calculator and we also changed the time of every step after troubleshooting and many consultations. As a way to find the best conditions, we changed the PCR cycles from 30 to 35 to receive a better final concentration.

After determining the appropriate annealing temperature, we experimented with different concentrations of the components of the reaction. Since our g-Blocks are resuspended in TE buffer, which contains EDTA that chelates Mg2+, we tried different concentrations of Mg2+. The concentration spectrum was 2-4 mM. Higher Mg2+ concentration increased the yield but also increased the byproducts of the reaction.



Figure 9: This is a photo of three PCR samples. For the first one, we initially added half the quantity of DNA of the other two samples and the higher Mg2+ quantity. The other two samples had the same quantity of DNA, but the first one had extra Mg2+. This experiment confirms the above, that the Mg2+ increases both the efficiency of the amplification but also, reduces the purity of the product.



After we succeeded to achieve a satisfying PCR result we were able to continue the procedure with the next steps as indicated in the Build section.



~Learn

In the end, we were finally able to clone 15 out of 16 of our toehold switches. The 16th toehold and specifically the part BBa_K3727030 that corresponds to the Toehold_Switch-1246 (3) was not cloned. We troubleshot to find out what could be the issue. Even though we received colonies on the plates there was no bacterial growth in our liquid cultures. For this reason, we tried preparing a new antibiotic stock, new plates with the proper antibiotic, and a new LB growth medium. However, the results were still disappointing. Since we didn’t have a lot of time to repeat the cloning procedures for this toehold we had the cloning of it at a stalemate until the end of the presentation.






3rd Direction


After we ensured that our toeholds functioned properly, we proceeded to the determination of the isothermal amplification method that would precede the detection of miRNAs by the Toehold Switches. To increase our method’s sensitivity and affordability, we decided to use the Exponential Amplification Reaction (EXPAR) method.

Our EXPAR protocols can be divided into 3 engineering cycles:
  • 1. Our attempts to establish an isothermal amplification protocol high specificity at 65 oC with end product DNA
  • 2. Our attempts to establish a low-temperature isothermal amplification protocol at 37 oC with end product DNA
  • 3. Our attempts to establish a low-temperature amplification protocol at 37 oC with end product RNA and the Template DNA needed for this reaction





| Engineering cycle 1


~Design

The EXPAR technique requires instead of primers, a very specific single-stranded DNA sequence, the so-called “Template DNA”, composed of three main regions (Figure 10).

Two complementary with the microRNA sequences in both 3' and 5' end (black color) and a nicking site in the reverse orientation in the middle (green color). The reaction starts when the miRNA binds in the first region (3' end) and the DNA polymerase starts the replication. The nicking enzyme cuts the upper clone of the dsDNA in the nicking site and because the polymerase lacks the exonuclease activity a DNA sequence of the miRNA is released.





Figure 10: EXPAR Template DNA. The red color indicates the microRNA, the black areas that are complementary with the microRNA regions and the green color indicates the nicking site.





More specifically, we designed three Template DNA sequences, one for each of the selected microRNAs (Table 4).

microRNA Template DNA sequence Size
miR-30e-5p 5'-CTTCCAGTCAAGGATGTTTACAGCAGTGCTTCCAGTCAAGGATGTTTACA-3P 50 b
miR-143-3p 5'-GAGCTACAGTGCTTCATCTCAGCAGTGGAGCTACAGTGCTTCATCTCA-3P 48 b
miR-1246 5'-CCTGCTCCAAAAATCCATTGCAGTGCCTGCTCCAAAAATCCATT-3P 44 b

Table 4: The Template DNA sequences that we designed for each microRNA




We ordered the sequences from the Lab Supplies Company with phosphate-modified 3’ ends, for increased specificity [6].

As a nicking enzyme, we chose Nb. BtsI because its recognition sequence is not present in any of the other sequences used (plasmids, toeholds, eGFP gene). Moreover, according to NEB, the supplier of the enzyme, Nb. BtsI has good activity both at low and high temperatures, such as 37 oC and 65 oC, respectively [7].

Figure 11 shows the nicking site of Nb. BtsI. Inevitably, when the enzyme cuts the upper clone of the template DNA, the miRNA released has 5 extra bases “5’-CACTGC-3’” on its 3’ end that might affect its detection by the toeholds.




Figure 11: Nicking site of Nb.BtsI enzyme




Thus, we decided to check the thermodynamic characteristics of the reaction before we use them in the lab. The results based on Gibbs free energy showed that these extra bases on miRNAs increase the affinity with the toehold switches. Specifically, we calculated the binding energy of the two molecules, the miRNAs with the extra bases and the toehold switches, with the RNAcofold.exe of ViennaRNA software and we observed that the binding energy increased significantly in comparison with the miRNAs without the extra bases.

According to recent bibliography, while EXPAR is a cheap technique with very good amplification results, it has a lot of by-products mainly due to the ab initio DNA synthesis by the enzymes used [6].

Ab initio DNA synthesis is the synthesis of dsDNA from free dNTPs in the absence of template or miRNA. The dsDNA produced by this procedure consists of short (4–36 bp) tandem repeats of AT with a low GC-content [8]. But, when the reaction takes place at a high temperature, by-products seem to be reduced and the reaction specificity increases [8]. That’s why we chose the EXPAR protocol at 65oC to ensure higher specificity as it is higher than the melting temperature of our microRNAs (Table 5).

miR Sequence Size Tm*
30e-5p UGUAAACAUCCUUGACUGGAAG 22 45.9 °C
143-3p GGUGCAGUGCUGCAUCUCUGGU 22 55.6 °C
1246 AAUGGAUUUUUGGAGCAGG 19 42 °C
Table 5: Melting temperature of the microRNAs used
* The melting temperature was calculated with the type: Tm= (wA+xT) * 2 + (yG+zC) * 4




For our reaction, we chose Bst 2.0 DNA polymerase because according to the bibliography it has low ab initio synthesis and when used with Nb. BtsI works perfectly at 65oC and it has been used in EXPAR protocols with good results [9].

SYBR Green I was used to monitor the reaction in the qPCR machine (Applied Biosystems 7500 Fast Real-Time PCR System). SYBR Green I binding to dsDNA absorbs blue light (497 nm) and emits green light (520 nm) [10].



~Build

We followed the protocol of Ellie Mok et al. (2016) [11] and the reagents used are the following:
  • ~Template DNA (instead of primers) for each microRNA
  • ~microRNAs
  • ~Nb.BtsI (nicking enzyme)
  • ~rCutSmart™ Buffer (buffer for Nb.BtsI)
  • ~Bst 2.0 DNA Polymerase
  • ~Isothermal Amplification Buffer (buffer for Bst 2.0 DNA Polymerase)
  • ~Deoxynucleotide (dNTP) Solution Mix
  • ~SYBR Green I
  • ~PCR H2O


Briefly, the procedure we completed is the following:

  • 1. Design of the protocol in the qPCR machine
  • 2. Template DNA and microRNAs preparation and aliquots
  • 3. Incubation of Template DNA and miRNA to achieve hybridization
  • 4. Adding of the enzymes
  • 5. Isothermal amplification and reaction monitoring in the qPCR machine
  • 6. Electrophoresis of the results


You can find the analytical protocol in our Experiments Page in the section "EXPAR":

Experiments







~Test

When we tested the above-mentioned protocol we noticed that while our reaction was successful the amplification ratio was not as high as in the bibliography and that we had a lot of by-products even in the negative controls (samples with all the reagents except the miRNA).

We were very careful when we were creating the negative control and no miRNA was added into these samples, which means that ab initio DNA synthesis happened.



~Learn

Although initially, we wanted to test more conditions of the above-mentioned protocol, we decided not to proceed with an isothermal amplification method at 65 oC, because the high temperature used creates the need for expensive equipment. The rest of our reactions and mainly the toehold switches reactions are conducted at 37oC, which means that if the final product of our project needs 65oC for the amplification, it also needs a machine such as a PCR and that increases the cost.

So based on the first results of this isothermal amplification method, we chose to change our DNA polymerase with Klenow which works perfectly at 37 oC.




| Engineering Cycle 2



~Design

We used the same template DNA and the only thing that changed in our design was the DNA polymerase; instead of Bst 2.0 DNA polymerase, we used Klenow Fragment (3'→5' exo-) which also lacks the exonuclease activity [12].



~Build

We followed again the same protocol of Ellie Mok et al. (2016) [11] and the reagents used are the following:
  • ~Template DNA (instead of primers) for each microRNA
  • ~microRNAs
  • ~Nb.BtsI (nicking enzyme)
  • ~rCutSmart™ Buffer (buffer for Nb.BtsI)
  • ~Klenow Fragment (3'→5' exo-) (DNA polymerase)
  • ~NEBuffer 2 (buffer for Klenow Fragment (3'→5' exo-))
  • ~Deoxynucleotide (dNTP) Solution Mix
  • ~SYBR Green I
  • ~PCR H2O


The procedure we followed was the same, too. The only difference was the temperature of the reaction that in all cases, both in the hybridization and amplification step, was at 37 oC.

You can find the analytical protocol in our Experiments Page in the section "EXPAR":

Experiments





~Test

Once again, when we tested our protocol we noticed that while our reaction was successful, the amplification ratio was ever lower compared to 65oC and that we had by-products even in the negative controls, too. We were very careful when we were creating the negative control samples, which means that ab initio DNA synthesis happened here, too.

We then noticed that the final concentration of Klenow Fragment DNA polymerase we used was lower than in the suggested protocol and that the total amount of Mg2+ in the final sample was high because both of the buffers used contained Mg2+ [13], [14].

Unfortunately, due to the small amount of Klenow Fragment DNA polymerase that we had in our possession and the large number of samples that we had to test, we couldn’t increase the final DNA polymerase concentration in the samples. Thus, we decided to lower the final concentration of NEBuffer 2 from 1x to 0.5x in an attempt to reduce the total concentration of Mg2+. We chose to reduce the buffer of the DNA polymerase and not the buffer of the nicking enzyme because according to NEB, Klenow Fragment (3'→5' exo-) has full functional activity when used with rCutSmart™ Buffer [15].

After these modifications, we noticed again ab initio DNA synthesis in the negative controls but the amplification ratio increased a little bit, and more copies of our miRNAs were produced.



~Learn

Although a small improvement of the protocol was achieved, more conditions should be tested. Unfortunately, due to the pressuring timeline of the competition, we didn’t have the time to order Klenow Fragment (3'→5' exo-) DNA polymerase to test our protocol with a bigger final concentration of the polymerase.

So, we decided to use the rest of the polymerase to test a slightly different EXPAR-like amplification protocol. The problem with EXPAR is that the final product produced is DNA while our toeholds are designed to detect RNA sequences. Thus we decided to modify a little bit the above-mentioned protocol to produce RNA sequences as a final product.



| Engineering Cycle 3



~Design

This time we designed different "Template DNA" sequences, composed of three main regions (Figure 12); instead of a nicking site, they have a T7 RNA polymerase promoter sequence in the reverse orientation (blue region). The 5' complementary region is designed to be double-stranded (green region) to ensure that the miRNA can only bind to the 3' end of the template (black region), which makes the technique more sensitive.




Figure 12: EXPAR-RNA Protocol Template DNA (the red color indicates the microRNA)





Only when the miRNA binds in the 3' end, a DNA polymerase can start the replication and a double-stranded DNA sequence of the T7 RNA polymerase promoter is produced. Then T7 RNA polymerase can start the transcription of the miRNA and a lot of RNA copies of the target miRNA are produced.

More specifically, we designed three Template DNA sequences, one for each of the selected microRNAs (with red color is the T7 promoter sequence):

microRNA Template DNA sequence Size
miR-30e-5p 3P-GAAGGTCAGTTCCTACAAATGT-5' 22 b
5'-CTTCCAGTCAAGGATGTTTACACTATAGTGAGTCGTATTACTTCCAGTCAAGGATGTTTACA-3P 62 b
miR-143-3p 3P-CTCGATGTCACGAAGTAGAGT-5' 21 b
5'-GAGCTACAGTGCTTCATCTCACTATAGTGAGTCGTATTAGAGCTACAGTGCTTCATCTCA-3P 60 b
miR-1246 3P-GGACGAGGTTTTTAGGTAA-5' 19 b
5'-CCTGCTCCAAAAATCCATTCTATAGTGAGTCGTATTACCTGCTCCAAAAATCCATT-3P 56 b



Template DNA part Modification of 3’ end Sequence (5’ end to 3’ end)
ER-143-3p-UP phosphorylated TGAGATGAAGCACTGTAGCTC
ER-143-3p-DOWN NONE GAGCTACAGTGCTTCAT
ER-143-3p-MAIN phosphorylated CTCACTATAGTGAGTCGTATTAGAGCTACAGTGCTTCATCTCA
ER-30e-5p-UP phosphorylated TGTAAACATCCTTGACTGGAAG
ER-30e-5p-DOWN NONE CTTCCAGTCAAGGATGTTT
ER-30e-5p-MAIN phosphorylated ACACTATAGTGAGTCGTATTACTTCCAGTCAAGGATGTTTACA
ER-1246-UP Phosphorylated AATGGATTTTTGGAGCAGG
ER-1246-DOWN NONE CCTGCTCCAAAAATCC
ER-1246-MAIN phosphorylated ATTCTATAGTGAGTCGTATTACCTGCTCCAAAAATCCATT




~Build:

This time before we started with the amplification protocol, we had to “build” our Template DNA sequences in the lab. For this reason, we implemented a T4 ligase protocol based on NEB’s [4].

We then followed an EXPAR-like amplification protocol inspired by the work of Nicholas Emery et al.; [6]. The reagents we used are the following:

  • ~Template DNA (instead of primers) for each microRNA (previously prepared with primer ligation T4 protocol)
  • ~microRNAs
  • ~T7 RNA Polymerase
  • ~T7 RNA Polymerase Buffer
  • ~Klenow Fragment (3'→5' exo-) (DNA polymerase)
  • ~NEBuffer 2 (buffer for Klenow Fragment (3'→5' exo-))
  • ~Deoxynucleotide (dNTP) Solution Mix
  • ~Ribonucleotide (rNTP) Solution Mix
  • ~SYBR Green I
  • ~PCR H2O


Briefly, the procedure we completed is the following:

  • 1. Incubation of Template DNA “-up” and “-down” to achieve hybridization of the double-stranded region of the Template DNA sequence (at 51oC). We used the PROMEGA ™ Calculator https://worldwide.promega.com/resources/tools/biomath/tm-calculator/ to find the best temperature for the Template DNA “-up” and “-down” hybridization
  • 2. Ligation of the double-stranded part with the “-main” part of the Template DNA sequence with T4 DNA ligase (at 37oC)
  • 3. Deactivation of T4 ligase
  • 4. Design of the protocol in the qPCR machine
  • 5. Incubation of Template DNA and miRNA to achieve hybridization (at 37oC)
  • 6. Adding of the enzymes
  • 7. Isothermal amplification and reaction monitoring in the qPCR machine (at 37oC)
  • 8. Electrophoresis of the results


You can find both of the protocols in our Experiments Page in the section "EXPAR":

Experiments






~Test

The first time we tried the ligation protocol was a total disaster. After a brainstorming session, we concluded that we should increase the incubation period for the T4 DNA ligase and that we should use a new T4 ligase buffer. T4 ligase buffer contains ATP that is sensitive to freeze-thaw cycles and without it, the T4 ligase has no activity.

Indeed, the use of a new buffer gave us better results and when we increased the incubation period at 37 oC we had an even better outcome.

Then we proceeded to the EXPAR - RNA amplification protocol. The first time was also a total disaster and no amplification happened.

Once again, we concluded that we should reduce the Mg2+ concentration because all of the buffers used (T4 ligase buffer, NEBuffer 2, and T7 RNA Polymerase Buffer) contained Mg2+, which in high amounts might inhibit the reaction [16].

Because the right Template DNA is necessary for the reaction, we decided not to change the amount of T4 ligase Buffer used.

Thus, we decided to lower the final concentration of NEBuffer 2 from 1x to 0.25x and the final concentration of T7 RNA polymerase buffer from 1x to 0.5x. We reduced the NEBuffer 2 more because the T7 RNA polymerase buffer contains spermidine which is very important for the reaction and is not present in the other two buffers.

After these modifications, our EXPAR-like RNA protocol succeeded, and as the continuously increased signal of SYBR Green I indicated. Nevertheless, in the reaction, there were again by-products as the diagram of the melting curve analysis indicates (Figure 13). If there weren’t any by-products we could only see one pick in the diagram.





Figure 13: Diagram of the melting curve analysis of the EXPAR - RNA Protocol products. With blue color, we have miR-30e-5p, with brown color miR-143-3p, and with grey color miR-1246.





~Learn

In the end, we managed to amplify our microRNAs at 37 oC with both EXPAR-DNA Protocol and EXPAR-RNA Protocol, too. So we were able to proceed with our Proof of Concept and to test the products of the two reactions with the best of our toehold switch sequences.

Nevertheless, more experiments need to be done to improve the amplification ratio so that we can achieve even bigger sensitivity.

We troubleshot to find out how to do that - e.g. different polymerase concentrations -, but unfortunately, we didn’t have the necessary time to test our hypothesis.

We are determined to continue our experimental procedures after the wiki freeze but we hope that some of the future iGEM teams continue experimenting with our innovative EXPAR-RNA protocol, too!






Summary


  • 1. 2 Engineering cycles for improving the Toehold Switches’ detection
    • i)Replacement of the RBS
    • ii)Change of the reporter gene
  • 2. 2 Engineering cycles for effective cloning of our system with 2 different cloning procedures:
    • i)Use of the Biobrick Assembly
    • ii)Use of PCR cloning
  • 3. 3 Engineering cycles for effective isothermal amplification of our microRNAs with 3 different cloning procedures:
    • i)EXPAR-DNA isothermal amplification protocol at 65 oC with end product DNA
    • ii)EXPAR-DNA isothermal amplification protocol at 37 oC with end product DNA
    • iii)EXPAR-RNA isothermal amplification protocol at 37 oC with end product RNA



References