Parts image

Results

Results are like a box of chocolates, you never know what you are gonna get.

Overview

For our project CAP’siRNA, we tried to build plasmids containing lhRNAs (long hairpin RNAs) targeting the Beet Yellows Virus (BYV) causing the virus Yellows disease to sugar beets. We also aimed to encapsidate the lhRNAs into the viral capsid of another virus, the Tobacco Mosaic Virus (TMV) to protect them from the external environment. This approach sought to enhance the stability of the lhRNAs and enable their successful delivery into the plant. Finally, we aimed to detect the BYV in sugar beet leaves to verify the efficacy of our solution.

Our results were organised in five different parts:

  • The lhRNA plasmid assembly, aiming to obtain plasmids containing our inserts of interest to create lhRNAs. We managed to ligate the sense and antisense sequence of each insert, but we did not obtain the expected constructs.

  • The capsid engineering, aiming to create a plasmid containing the Coat Protein (CP) and the parts necessary (promoter, terminator, RBS…). First, we amplified all necessary inserts by PCR, which were the promoter, terminator, RBS from their host plasmids and coat protein. We successfully amplified and extracted the desired fragments and proceeded to attempt the Golden Gate assembly.

  • The capsid production, aiming to extract and amplify the coat proteins of the TMV form leaves of infected plants.

  • The BYV detection, aiming to detect by qRT-PCR the BYV in infected leaves of sugar beets. We managed to detect it.

  • The expected results, although we ran out of time to successfully complete the experiments we planned, we outlined the following experiments along with their expected results.

lhRNA plasmid Assembly

Plasmid Amplification

To eliminate the BYV thanks to the RNAi pathway, we needed to produce our lhRNAs. These lhRNAs were first ordered as linearized DNA inserts. As the synthesis of long DNA strand was difficult with such a high similarity rate, we had to separate it in 2 parts. To see more about the design of the lhRNA, see in Engineering part.

To produce the RNAi precursors, we assembled them in a plasmid, so that the coding sequence for the lhRNAs were integrated into the genome of bacteria and could be transcribed into RNA, see in Design part.

Firstly, the aim was to amplify a plasmid (pET21a+) in DH5α bacteria to get higher quantities for the planned experiments. As we observed in Figure 1, many colonies of bacteria were obtained. To isolate and purify the plasmid DNA derived from DH5α bacteria, we performed some miniculture in LB medium with ampicillin, followed by minipreps and gel purification.

To evaluate the quality of the purified plasmid, we quantified the plasmid using a NanoDrop. The yields obtained were very low between 0 and 10ng/μL. So, we tried other purification methods by using magnetic beads. This method allowed us to get better results about 100ng/μL with correct ratios around 2 for the A260/A230 and around 1.8 for the ratio A260/A280. These indicators showed that the isolated plasmid p21 was well purified.

Figure 1

Figure 1: A petri dish containing ampicillin on which we can see clones of the pET21a+.

Digestion of p21 inserts, RNA-dependent RNA polymerase, and PDS

The aim was to digest our inserts with the chosen restriction enzymes to ligate them inside the receiving plasmid. We digested the sense sequences with BamHI and NdeI and the antisense sequences with BlpI and NdeI. As this digestion consisted only of the removing of a few bases on each sides of the inserts, we could not reveal the result on a gel.

Digestion of the pET 21a+ plasmid

The aim was to digest the receiving plasmid with BamHI and BlpI, so that it could have the restriction sites ready for the ligation of the inserts of interest. We made the digested plasmid migrate on an agarose gel 1% with a non-digested control. We expected to see a band at 5325bp corresponding to the digested plasmid. On Figure 2, we can see on the wells 3 and 5 bands around 6000bp corresponding to our digested plasmid. In the well 7, for the non-digested control, we saw a band around 4000bp. So, we suggested that the non-digested plasmid had a supercoiled shape. Overall, this result meant that we digested the plasmid.

Figure 2


Figure 2: Electrophoresis gel 1% representing the digested plasmid pET21a+ with BamHI and BlpI and a control with the plasmid non-digested.

Ligation of the inserts inside the backbone plasmid

The aim is to ligate the digested inserts sense and antisense at the same time in the digested plasmid. We then transformed the ligation products into DH5α competent bacteria and made them grow on plate overnight. The experiment was not conclusive as we did not see any colonies on the plates.

Ligation of sense and anti-sense inserts independently

We decided then to work only with the PDS inserts sense and antisense, which were the ones for the lhRNA control to save quantities of other inserts. This time, we reproduced the experiment, but we first ligated the two inserts sense and antisense together, before ligating them in the plasmid. As it worked, we reproduced the experiment with the inserts for the other lhRNA constructions (Polymerase and p21).

On Figures 3 and 4, we saw that we obtained the expected bands. In fact, there are bands at around 1000bp corresponding to the two inserts ligated together and bands at around 500bp corresponding to the insert alone and not ligated. We also noticed that the bands for the polymerase (poly) inserts were a bit higher than the ones of the PDS and the p21, which was coherent with the expected bands (1020bp for the p21, 1022bp for the PDS and 1342bp for the polymerase), see constructions in the Parts page.

Figure 3


Figure 3: Electrophoresis gel 1.5% representing the ligated inserts sense and antisense for each insert couple, the digested plasmid pET21a+ with BamHI and BlpI, the non-digested plasmid pET21a+.

Figure 4


Figure 4: Electrophoresis gel 1.5% representing the ligated inserts sense and antisense for the polymerase and the p21.

Ligation of the inserts for the PDS inside the pET21 backbone

Then, we ligated the products of the ligation insert-insert into the pET 21. We tried many different ratios for the ligations: 3:1, 5:1, 7:1 and we transformed 30μL of DH5α bacteria with different volumes of ligation product (2 μL, 5 μL, 10 μL and 20 μL). We obtained clones for the ratio 7:1 when we transformed bacteria with 20 μL of ligation product, see in Figure 5. After mini culture, miniprep, digestion with BamHI and BlpI and gel electrophoresis migration, we saw only one band corresponding to the size of the digested plasmid alone meaning that the plasmid was ligated on itself.

Figure 4


Figure 5: A petri dish containing ampicillin on which we can see clones of the DH5α bacteria transformed with our ligation product (pET21a+ and PDS sense and antisense).

Ligation with a dephosphorylated plasmid

We tried to dephosphorylate the plasmid using phosphatase alkaline before performing the ligation with the two inserts. We saw again some colonies, but the results were not conclusive.

Ligation of the inserts in a plasmid containing mCherry

At some point, we decided to amplify the inserts in a plasmid (pSB1A3) containing mCherry to amplify them. We used the non-digested inserts having blunt ends and inserted them into a plasmid containing mCherry, which had been digested with the HpaI enzyme, also creating blunt ends. When the inserts were in the plasmid, the site corresponding to mCherry was cut. Therefore, we expected to observe on the plates: pink colonies corresponding to the bacteria containing only mCherry site, and white colonies corresponding to bacteria containing the inserts. On the Figure 6, we saw pink and white colonies, suggesting that the white colonies might have contained the construction of interest. To isolate and purify the inserts derived from DH5α bacteria, we performed minicultures and minipreps. After digestions, we did not retrieve the expected bands. We may conclude that the white colonies might be containing the mCherry and not the inserts. We suggested that bacteria were not incubated enough to express the mCherry and show a pink phenotype.

Figure 5


Figure 6: Petri dishes containing ampicillin on which we can see pink and white clones of the DH5α bacteria transformed with our ligation products. Bacteria were transformed with a plasmid containing mCherry with the PDS sense insert on the left and the PDS antisense insert on the right.

Ligation of our insert in other plasmids

After those several attempts to obtain our constructs, we wanted to test the insertion capacity of our fragments in other plasmids. To do so, we used other plasmids pET22b+ and pET11a containing the restriction sites of interest (BamHI and BlpI). We succeeded to get some colonies and did the minicultures, minipreps and appropriate digestions. However, after migration on electrophoresis gels, we did not see any relevant bands but a band around 5000bp. We hypothesized that the colonies were just containing the plasmid alone and so that the ligations did not work.

Capsid engineering

To obtain the construct of interest aiming to produce the TMV coat protein to encapsulate our RNAi precursors, we wanted to execute a Golden Gate. To do so, the construct required 5 components: a backbone pGGa select Destination Plasmid, a T7 promoter, a T7 terminator, a RBS and the coat protein (CP), see Design part.


The next step was to perform the Golden Gate assembly. To do this, we assembled the 4 PCR products (backbone pGGa select Destination Plasmid, a T7 promoter, a T7 terminator, a RBS and CP) and ligate them into the pGGA select destination plasmid.

Golden Gate Assembly

To observe if the transformed DH5α bacteria had integrated the construct by Golden Gate Assembly, we plated them and incubated them overnight. The following day, we observed many bacteria colonies, see Figure 10.

Figure 10


Figure 10: A petri dish containing chloramphenicol on which we can see clones of the DH5α bacteria transformed with our Golden Gate product (pGGa select Destination Plasmid with T7 promoter, I21 RBS, the coat protein sequence and the T7 terminator).

To verify that the colonies were containing our plasmids of interest, we did minicultures and minipreps of different clones from diverse plates, containing purified or non-purified mix. Purified mix was composed of specific gel extracted samples of the bands corresponding to each fragment (plasmid, RBS, CP, Promoter T7, Terminator T7). Non purified mix was composed of non-specific bands of the same gel extracted samples.
We then digested them with EcoRI as this enzyme cut at the edge of the plasmid around our insert site. We were expecting to see bands at 2000bp corresponding to the plasmid and bands at 773bp corresponding to the inserts.

We saw the bands at 2000bp corresponding to the backbone alone. For the second well on this gel, see Figure 11, we saw a second band at 500bp which was a bit low compared to our prediction. As the sample in this well was composed of a mix containing non-purified samples of the promoter and the terminator, we were not sure that the final product was the expected construction or if it contained a non-expected DNA. Moreover, the samples containing purified mix were presenting only one band associated to the backbone plasmid.
Therefore, we could not conclude that this experiment was a success, and we should repeat it another time.

Figure 11


Figure 11: Electrophoresis gel 1.5% of Non-purified mix (plasmid, RBS, CP, Promoter T7, Terminator T7) and Purified mix. Purified mix is composed of specific gel extracted samples of the bands corresponding to each fragment ; non Purified mix is composed of non-specific bands of the same gel extracted samples.

Capsid production

In parallel of Golden Gate Assembly method, an alternative way of producing the coat protein to encapsulate our lhRNA was conducted. This section aimed to present this new method, using wild type TMV, see Lab Notebook part. To undergo such process, we first needed to cultivate our plant of interest: Beta vulgaris and Nicotiana benthamiana. We built a greenhouse for their growth and protection, see Figure 12.

Then, we inoculated the TMV wild type virus to Beta vulgaris and Nicotiana benthamiana plants. We waited for 2 weeks before retrieving the wild type virus. We purified the leaves to obtain wild-type TMV, which will be used for a new inoculation in plants to produce more viruses or for capsid separation.
We separated the capsids from the viral RNA of the TMV and purified them by using a dialysis cassette. This enabled us to possess forming disks of capsids that were not degraded thanks to the dialysis.
To be able to adequately quantify the proteins obtained, we performed a BCA assay.

Figure 12


Figure 12: Greenhouse of Nicotiana benthamiana and Beta vulgaris.

We obtained an absorbance of 0.134 on average, and decided to find the working concentration by doing a standard curve, see in Figure 13. We obtained a protein concentration of our solution of 36ug/mL.

Figure 13


Figure 13: Standard curve of BCA assay of Coat Protein. Concentration in ug/mL in function of the Absorbance.

After knowing that we had proteins in our solution, we needed to make sure that the proteins we have were the coat proteins. For that, we needed to do an SDS-Page, to confirm the presence of our coat protein. We knew that the TMV coat protein had a molecular weight of 17.6 kDa. So thanks to the SDS-page, we were able to detect the presence of the coat protein. To perform this experiment, we used a control protein called BSA (Bovine Serum Albumin), that had a molecular weight of 66 kDa.

On the gel in Figure 14, we observed that the BSA samples have migrated at the expected size, nonetheless, the bands were diffused, and the ladder was overflowing on the right. Moreover, we detected a band at around 17kDa which might corresponds to the CP. However, as the controls were not presenting clear bands and the migration was not really straight we cannot conclude positively on this experiment.

Figure 14


Figure 14: SDS-PAGE gel to detect the TMV coat protein with a BSA control.

BYV detection

Before inhibiting the BYV, one important parameter to consider was to assess and detect the virus in contaminated sugar beets. To verify the presence or absence of the virus we wanted to quantify it. A method to do it as to use a qRT-PCR.

Before performing the qRT-PCR, we needed to obtain some viruses to detect as a control. For this, we infected sugar beet plants with aphids Myzus Persicae infecting leaves of grown sugar beets. After a few weeks, we obtained yellow spots which are symptoms of the Yellows virus disease caused by the BYV, see Figure 15.

Figure 15


Figure 15: Sugar beet leaves, infected by the BYV with yellow spots.

We directly extracted and purified the RNA of the contaminated leaves of the sugar beets, as well as non-infected ones for a negative control. Then, we did a qRT-PCR to see the specific constructs of the different primers synthesized, see Design part.

First, we needed primers for the BYV, but also for reference genes that had a constant high expression inside the sugar beets. We wanted to synthetize the primers in our laboratory for these referent genes. We selected, at first, a gene called Actin1 that should always be expressed in high quantity.
Before setting up the machine, we needed to optimize the primers to know the best quantity to use for the reaction. We tried the following concentrations: 200nM, 300nM, 400nM and 500nM. We also tried them with three concentrations of RNA: 100ng, 10 ng and 1 ng per well.

Figure 16


Figure 16: Amplification plot for the optimization of the primer forward and reverse for the Actin1 in qRT-PCR.

On Figure 16, the red curve corresponded to the 200nM condition, the orange is the 300nM condition and the purple is the 400nM condition.

The goal of an amplification plot was to determine the relative gene expression of a gene by taking into consideration the reference gene. We wanted to know at which concentration our Actin1 primers worked the best.

After setting up the Fam threshold at 500 of fluorescence, we analyzed the Ct of each condition, and we saw that all of them appeared around 32 Cycle threshold (Ct). Usually, a reference gene is detected around 20 Ct. It showed that the gene Actin1 was detected too late.

After this many cycles of qRT-PCR, this showed that the gene Actin1 was detected in either the contaminated sugar beets or the healthy ones. We tried also at 500nm and 700nm of forward and reverse primer for the Actin1 gene, but the same results were obtained.
If we look at their melting curve on Figure 17, other information can be deducted.

Figure 17


Figure 17: Melting curve for the optimization of the primer forward and reverse for the Actin1 in qRT-PCR.

The goal of a melting curve was to verify the specificity of the reaction, while making sure to obtain only one amplicon. If one pic was obtained, this meant that we have one amplicon. In our case, we saw two pics, meaning that there were two amplicons of different characteristics. If we took into consideration the amplification plot, we could deduce that the bands corresponded to dimers of primers that got amplified or non-specific amplifications.

On Figure 18, by comparing the RT- to the other two, we saw that we had a specific amplification that was supposed to be 125 bp. There were also 2 other bands that were not specific. We could hypothesize that we had primer dimers or contaminant genomic DNA amplification.
Therefore, our Actin1 gene was not a relevant referent gene, as it was both not enough expressed and not constantly in sugar beets. There was also another possibility to have a primer that was not specific enough.
Furthermore, after multiple trials, we decided to use other reference genes that we found in the literature: the elongation factor 1-β (EF1- β) and the glyceraldehyde-3-phosphate dehydrogenase (GAPDH), see in Engineering part.

Figure 18


Figure 18: Electrophoresis gel 2% of a qRT-PCR for Actin1 in Beta vulgaris infected or not by BYV.

On Figure 19, the purple and red curves represented EF1-β promoter, and the orange and green represented ones the GAPDH promoter. The cyan curve was the negative control for the qRT-PCR (RT-), without putting the enzyme reverse transcriptase. This showed the threshold at which there was no amplification. In the amplification plot, we saw that the two amplicons EF1-β and GAPDH were detected at the same Ct overall, after 25 Ct. Regarding the melting curves, we saw one pic close to 81 °C regrouping the specific amplicons created by the two couples of primers. But we clearly saw two other pics: one pic for the EF1- β promoter closed to 72 °C and one closed to 75 °C for the GAPDH. This could mean that we obtained non-specific amplification maybe caused by dimers of primers.

Figure 19


Figure 19: Amplification and Melting curve of the promoter EF1-β and GAPDH in qRT-PCR.

To verify our amplification, we did an electrophoresis gel of the PCR product for the EF1- β and GAPDH, see Figure 21. For the two reference genes, by comparing them with the RT-, we can see a specific amplification close to 100 bp, which are absent in the RT-. We also see, mainly in the RT-, 2 non-specific bands that could be primer dimers. Indeed, we did another negative control with water and we obtain the same melting curve profile and band after electrophoresis migration in the RT- (data not shown). Then, we work on the BYV detection to assess its expression into the infected sugar beets. After this, we finally did the qRT-PCR for the BYV. We titrated the gene expression of the BYV in healthy sugar beets and in contaminated ones. Results are shown in Figure 20.

Figure 20


Figure 20: Amplification and Melting curve of the promoter BYV in healthy and infected sugar beets, with the reference genes EF1-β and GAPDH in qRT-PCR.

The blue curves represented infected sugar beets with the BYV primers. The red ones represented the healthy sugar beets with the BYV primers, and the green ones represented the healthy and infected sugar beets with the two references EF1- β and GAPDH primers.

By looking at the amplification curves, we saw an amplification at 17 Ct for the infected sugar beets, whereas it was only 26 Ct for the healthy ones. We saw the same amplification for the sugar beets healthy and infected with the two reference genes.

If we looked at the melting curve, we clearly saw 2 distinct pics: the blue one representing the BYV amplicons, and the red and green ones representing other amplifications from the reference genes, but of the same length. We could therefore clearly say that our primers against the BYV were more specific than the others, seeing that there is only one pic for them, whereas there were 2 pics for others, possibly representing dimers of primers.

Overall, we have designed primers that seems to be able to detect the BYV RNA genome into the infected sugar beets. Because we also found an unexpected amplification in the non-infected sugar beets, we performed an electrophoresis to control the specificity of this amplification, see Figure 21.

Figure 21


Figure 21: Amplification and Melting curve of the promoter BYV in healthy and infected sugar beets, with the reference genes EF1-β and GAPDH in qRT-PCR.

By comparing the H2O with the BYV primers condition, the non-infected sugar beets and the infected ones, we see a band at 122 bp only in the infected sugar beets. We could conclude that the qRT-PCR amplified the correct amplicon belonging to the BYV. These results and the amplification and melting curve results show that we are able to detect viral RNA after infection of sugar beets. In this section, we showed that the BYV can be detected, and we confirmed its presence. However, we still need to optimize our detection process by improving the specificity of the reference and BYV primers. When the optimization will be done, we could try to calculate the relative BYV RNA quantity at diverse concentrations, ranging from high to very low quantity. For the housekeeping genes, different parameters could be optimized, such as the freshness of the sugar beets leaves, as we extracted the RNA from old leaves. Indeed, we can see in the Figure 20 and Figure 21, that the infected sugar beets seem to have a different expression level that could be linked to the presence of the virus. It could be interesting to study the gene expression modification in sugar beets in an infection state.

Expected results

For the expected results, we wanted to build the plasmid with the inserts sense and antisense ligated in the backbone. We were expecting a band at around 5300bp (5325bp) corresponding to the digested backbone with BamHI and BlpI and bands at 1000bp for the plasmid containing the PDS inserts (1003bp) and for the plasmid containing the p21 inserts (1002bp), for the plasmid containing the polymerase we will see the second band at around 1300bp (1324bp).

Figure 22


Figure 22: Expected results 1% Electrohoresis gel of pET21 with PDS, p21 and Poly-ligation.

Furthermore, we aim to encapuslate our lhRNA in TMV coat proteins. To visualise our experiment, we will run an electrophoresis gel. We will migrate our lhRNAs previously envelopped with the CP and without to have a negative control. We aim to see no bands for our enveloped lhRNAs, as the proteins will inhibit the migration of the RNA. For the control condition we will see a band of the size of our naked lhRNAs (around 1000bp). If we see a band for the lhRNA with the CP it means that the encapsidation did not work. Finally, when we will have our encapsulated lhRNA, we would be able to really test our solution. This idea is to test 2 conditions: the curative aspect of our solution with the direct elimination of the BYV already infecting sugar beets, and the preventive aspect by eliminating the virus.

Figure 23


Figure 23: Expected qRT-PCR amplification of the BYV RNA in the plants with or without treatment. The red curve represents the positive control with a plant infected by the BYV. The green curve represents the quantity of virus in the plants infected with the virus and treated with the RNAi precursor. The blue curve represents the negative control with a non-infected plant.

Large Image Behind Footer