Team:TU Dresden/Results
Beaming BioBricks from Space...
Results
Project Achievements - Overview
- We successfully designed our own fusion protein in silico, ordered all the parts as separated BioBricks or used old ones from the registry and assembled them to our Full Construct
-
We proved that every single part of our Full Construct is working
- MBP → verified via amylose resin purification
- dCas9 → verified via Electrophoretic Mobility Shift Assay (EMSA)
- HRP → verified via TMB-conversion activity assay
- The only part of our Full Construct that is not working was the Strep-tag. It was designed for Western Blot analysis and not for Strep-column purification. (Read more about it here)
- We successfully designed our own tailored DNA extraction method from bacteria in which nucleic acids of high purity are immobilized in a cellulose paper strip.
- With this simple, fast and easy method we proved that dCas9 can bind to DNA that is immobilized on a paper strip
- We proved that dCas9 is not binding unspecifically to the paper
1. Cloning Full Construct
We successfully cloned the Full Construct in pSB1C3. We checked that using PCR induced mutagenesis we were successfully able to remove the illegal restriction site. Full Construct i.e. MBP-dCas9- Linker-HRP-Strep is 6 kb fragment (Figure 1).
Figure 1 - Restriction digest of pSB1C3 with EcoRI and PstI showing Full Construct at 6 kb.
We checked in silico the best restriction enzymes for detecting the correct insertion of our Full Construct (Figure 2). It can be seen in the first lane (digested with XbaI and PstI) that the difference in size between the plasmid backbone (5 kb) and the Full Construct (6 kb) is not sufficient to be properly separated on an agarose gel. Therefore, a new restriction digest had to be prepared in which a triple digest with XbaI, PstI and PmlI was set up. The fragments of different size make it easy to distinguish a successful insertion of the final construct in the expression plasmid (Figure 3).
Figure 2 - In silico restriction digest of Full Construct. Digestion with XbaI and PstI in the lane 1. Digestion with XbaI, PstI and PmlI in the lane 2.
Figure 3 - Restriction digest of full construct (BBa_K3037003) in pOCC97 (BBa_K3037000) with XbaI, PstI and PmlI. The lanes marked with the red X correspond to the clones with the correct insert. The background bands shown in the lanes correspond to the T7 expression plasmid transformed into the E. coli pRARE strain from NEB.
2. Expression and Purification of the Full Construct
We used the ExPASy ProtParam tool (https://web.expasy.org/protparam/) to calculate physical and chemical parameters for our protein:
Number of amino acids: 2106
Molecular weight: 238793.94
Theoretical pI: 8.39
Ext. coefficient*: 205750
Abs 0.1% (=1 g/l) 0.862, assuming all pairs of Cys residues form cystines
*Extinction coefficients are in units of M-1 cm-1, at 280 nm measured in water.
We tried expression of our construct in our expression plasmid BBa_K3037000 under different IPTG concentrations and two temperatures after induction (Figure 4 and 5). By comparing the expression levels (Figure 5) we determined that the optimal conditions for the expression of BBa_K3037003 in pOCC97 (BBa_K3037000) are 18 ºC and 0.5 mM IPTG. The expression seems to be more stable over time for the optimized plasmid than for the non-optimized.
Figure 4 - Expression of the Full Construct in pOCC97 optimized at different temperatures and IPTG concentrations
Figure 5 - Expression of the Full Construct in optimized pOCC97 under different conditions.
After proving that the Full Construct was expressed properly in our plasmid and improved its expression conditions, it was purified by using amylose Resin to bind its MBP affinity tag. Our Full Construct was present in all the elution fractions after purification with the amylose resin (Figure 6). The concentration of the purified protein was 4.56 mg/mL.
Figure 6 - MBP affinity tag chromatography purification of Full Construct using amylose resin.
After purification with the N-terminus MBP affinity tag, we tried to purify our Full Construct using our C-terminus Strep-tag. However, our protein was not able to bind to the column resin and all of our protein fraction is found in the column flowthrough and none in the eluates (Figure 7). Apparently, the Strep-tag BioBrick BBa_K823038 can be only used for Western Blot detection of the tagged protein.
Figure 7 - Strep-tag affinity chromatography of eluates from amylose resin purification.
We used the CRISPR Guide RNA Design from Benchling (https://www.benchling.com/crispr/) to design six gRNA which target the sry locus and one gRNA that targets our eGFP BioBrick. The gRNA with the PAM sequence and the context are specified in the following table 1.
We wanted to check if the best gRNA combination for binding dCas9 to the target DNA sequence. We used our eGFP-dCas9 fusion protein ( BBa_K3037005)) for testing the enzyme ability of binding DNA. The mobility shift works very well with gRNA 1, and it increases when combinations of guide RNAs are used. The best gRNA combination is 1,4 and 2 (Figure 8). We also included negative controls loading only the sry amplicon with the gRNA's to test if the mobility shift that we saw in Figure 9 was influenced by a secondary structure formation of the gRNA and not by the dCas9 binding to the target sequence.
Figure 8 - Electrophoretic Mobility Shift Assay of sry targeted with dCas9 and different gRNA combinations
Lane 1: 1 kb DNA ladder
Lane 2: sry gene
Lane 3: sry gene + dcas9
Lane 4: sry gene + dcas9 + guide RNA 1
Lane 5: sry gene + dcas9 + guide RNA 4, 2 and 6
Lane 6: sry gene + dcas9 + guide RNA 1, 5 and 6
Lane 7: sry gene + dcas9 + guide RNA 1, 4, 3 and 6
Lane 8: sry gene + dcas9 + guide RNA 4, 3 and 7
Lane 9: sry gene + dcas9 + guide RNA 1, 4 and 2
Figure 9 - Electrophoretic Mobility Shift Assay of sry targeted with dCas9 and different gRNA combinations
We also tested if our eGFP-dCas9 can bind plasmid DNA targeted with the gRNA’s. In Figure 9a lane 8, we could see that the band from the digested plasmid DNA (lane 9) is completely prevented from entering the gel. This shows a high binding affinity of our eGFP-dCas9 with our gRNA to large plasmid fragments.
Figure 9a - EMSA showing mobility shift of plasmid DNA targeted with gRNA’s.
Lane 1 - Marker
Lane 2- 100 ng of Sry gene
Lane 3- Sry + dCas9-HRP
Lane 4 - Sry + dCas9-HRP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2
Lane 5 - Sry + dCas9-HRP+Guide RNA 1
Lane 6 - Sry + dCas9-GFP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2
Lane 7 - eGFP+dCas9-HRP+Guide RNA 7
Lane 8 - eGFP+dCas9-HRP+Guide RNA 7
Lane 9 - eGFP
Lane 10 - Sry + dCas9-HRP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2 (Here Sry gene was immobilized on cellulose strip and addition of dCas9 proteins and guide RNA was all done in cellulose strip).
Lane 11 - Sry + dCas9-GFP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2
Lane 12 - Sry gene loaded on cellulose strip.
With all of the previous experiments we were ready to test the activity of our full construct with the best working gRNA combination. In Figure 9b lane 2 we see the sry gene as a control, when we add a high concentration of our full construct dCas9-HRP (BBa_K3037003) we can see that the dCas9-HRP fusion protein completely binds all the sry PCR amplicon, preventing it from entering the TBE-PAGE. We can confirm that the shift that we see in lane 2 is produced by the binding of our full construct when we serially dilute our protein. Starting in lane 9, and then from lanes 6 to 8 we see that the intensity of the shifted bands decreases with lower dCas9-HRP concentration.
Figure 9b - EMSA performed with the full construct (BBa_K3037003) targeting the sry gene.
Lane 1 - 1 kb plus ladder
Lane 2 - sry gene
Lane 3 - 4.56 ug of dcas9-HRP + guide RNA (1,4 and 2) + sry gene
Lane 4 - 4.56 ug of dcas9-eGFP + guide RNA (1,4 and 2 ) + sry gene
Lane 5 - 600 ng of dcas9-HRP + guide RNA (1, 4 and 2 ) + sry gene
Lane 6 - 2.56 ug of dcas9-HRP + guide RNA (1,4 and 2) + sry gene
Lane 7 - 1.28 ug of dcas9-HRP + guide RNA (1,4 and 2) + sry gene
Lane 8 - 0.64 ug of dcas9-HRP + guide RNA (1,4 and 2) + sry gene
Lane 9 - 8.56 ug of dcas9-HRP + guide RNA (1,4 and 2) + sry gene
We were able to successfully express the HRP protein which was adapted to the RFC25 BioBrick standard from the BioBrick BBa_K1800002, expression was performed using our plasmid backbone BioBrick pOCC97 ( BBa_K3037000). The HRP expression increased over time only after induction of the cells with IPTG (Figure 10).
Figure 10 - HRP expression after induction over time. HRP is marked with a black arrow pointing left.
The activity of HRP in the Full Construct was measured in a dynamic assay. We confirmed that our HRP from our Full Construct is active and is able to catalyze the conversion of TMB to its oxidized blue colored product (Figure 11). The absorbance measured at 650 nm over time confirmed the oxidation of TMB when comparing the clarified cell lysate from the E.coli cells that expressed the Full Construct with the HRP, versus the cells without our Full Construct plasmid (Figure 10).
Figure 11- Conversion of TMB substrate by HRP over time using clarified cell lysate from E.coli
expressing the HRP BioBrick BBa_K1800002 and our Full Construct BBa_K3037003. The HRP
activity of our Full Construct was compared to the HRP BioBrick BBa_K1800002 and E.coli pRARE cell lysate.
We quantified the concentration of DNA that we eluted from the Whatman cellulose disk using a NanoDrop spectrophotometer. We performed the reaction in duplicates and found out that 1 minute of cell lysis at room temperature works the best (Figure 12). Extended lysis times led to unnecessary waiting which did not improve the extracted DNA yield.
Figure 12 - HRP expression after induction over time. HRP is marked with a black arrow pointing left.
From Figure 13 we conclude that when we perform cell lysis at room temperature and extract DNA with cellulose disk, it yields a higher purity in the eluted sample. The absorbance ratio measured at 260/280 nm is between 1.8 - 2.0, which indicates that there is little to none protein contamination from the cell lysate.
Figure 13 - Absorbance ratio at 260/280 nm to measure protein contamination
We wanted to find the best matrix that would be able to selectively bind DNA from a cell lysate and that will allow us to perform our dCas9 gene targeting without nonspecific interactions. In order to understand which type of nucleic acids were extracted from the aforementioned cell lysates, we decided to load the cellulose disks into an agarose gel electrophoresis.
Since we were trying to develop an affordable method for diagnostics we chose Whatman filter paper because of it's low cost. As a control we used a nitrocellulose membrane that according to Zou et al. (2017) had a stronger binding to nucleic acids. Since they also mentioned that higher salt concentrations increase the electrostatic interactions between the cellulose and the DNA.
We found that by adding 100 mM NaCl to the cell lysates we were able to extract a good amount of genomic DNA using cellulose disks (Figure 14). The nitrocellulose membrane had a great binding capacity of DNA; however, as mentioned by Zou et al. (2017) the interactions between the nitrocellulose and the DNA were too strong and the nucleic acids were not leaving the membrane in the agarose gel electrophoresis (Figure 15). This high electrostatic interactions could interfere with the ability of our dCas9-HRP to bind DNA on top of the matrix.
For the aforementioned reason we decided to use the cellulose disks for DNA extraction with 100 mM NaCl added to the cell lysates.
Figure 14 - Varying salt concentration and the impact on DNA yield extracted from E.coli GB05 cell lysate using cellulose disc
Figure 15 - Varying salt concentration and the impact on DNA yield extracted from E.coli GB05 cell lysate using nitrocellulose disc
We can identify the extracted genomic DNA from the bacteria cell lysate from the bands shown in the agarose gel electrophoresis above the 10 kb band from the marker; in addition, there is a band smear at the expected plasmid size in Figure 16 lane 3. We could only extract plasmid DNA when no extra salts were added to the cell lysate.
Figure 16 - Varying salt concentration and the impact on plasmid DNA yield extracted from E.coliGB05 (pSB1C3 expressing RFP; BBa_J04450) cell lysate using nitrocellulose disc
In order to get genetic material from buccal swab samples to test sry - positive or sry - negative cells with our DipGene method, we tried to apply the same concept for DNA extraction that we proved works very well for bacterial cells. However, the amount of starting material that we got the buccal epithelial cells was not enough for visualizing nucleic acids in agarose gel electrophoresis (Figure 17).
Figure 17 - Agarose gel electrophoresis of cell lysate from buccal epithelial cells. The cells were lysed using NEB's Monarch gDNA Cell lysis buffer at room temperature and at 95 °C. The attempt to bind DNA from the cell lysate using the cellulose disks was not noticeable due to low DNA concentration in the lysate.
Following the idea of the fast DNA extraction method from Zou et al. (2017), we decided to create and optimize cell lysis buffers tailored for mammalian cells that can be used for PCR amplification.
We successfully amplified the sry gene from sry positive cells lysed with NEB's Monarch gDNA Cell lysis buffer (Figure 18).
Figure 18 - PCR of sry gene using the cellulose disk DNA extraction method.
We tried the same buffers that we used for bacterial cell lysis (E.coli GB05) to check if we could extract gDNA from Spirulina (cyanobacteria used in our other project, read more about it) and use it to include homology arms for Homologous recombination. The target region was successfully amplified via PCR with the cellulose disk method (Figure 19).
Figure 19 - PCR of locus with homology arms used for homologous recombination in Spirulina using the cellulose disk DNA extraction method.
A critical part of our project design was to develop an easy method to selectively immobilize DNA on a cellulose matrix, and tailor buffers that can be used to wash away proteins and debris that could interfere with our dCas9-HRP fusion protein in binding to the target DNA region. We optimized the wash buffer composition to selectively wash proteins out of our cellulose matrix, while leaving nucleic acids immobilized in the disks. Tests were performed using common western blot blocker proteins to check which buffers work best for removing proteins that adhere to the cellulose matrix.
We decided to saturate the cellulose disk with eGFP-dCas9 fusion protein (with gRNA’s targeting the sry gene) and incubate them at 37 °C for 30 minutes. The disks were washed with our tailored wash buffer and dipped into a solution containing the sry PCR amplicon, followed by another incubation at 37 °C for 30 minutes. The disks were loaded into a TBE-PAGE to check for a mobility shift (in an EMSA gel) which will only happen if the eGFP-dCas9 was not successfully washed away from the disk. In Figure 20 lane 4 and 5 we don’t see a band shift, which means that the eGFP-dCas9 was completely removed from the cellulose disk.
Figure 20 - https://2019.igem.org/Wiki/images/1/1b/T--TU_Dresden--results-fin1.pngFigure 20 - EMSA showing that the eGFP-dCas9 does not bind nonspecifically to the cellulose disks and can be washed away with the W1 buffer.
Lane 1 - Marker
Lane 2 - sry
Lane 3 - dCas9-HRP + (1,4,2 guide RNA) + sry
Lane 4 - dCas9 + gRNA (1,4 and 2) first incubated and then washed once and then sry gene.
Lane 5 - dCas9 + gRNA (1,4 and 2) first incubated and then washed twice and then sry gene.
Lane 6 - sry loaded on cellulose
Lane 7 - (200 ng) dCas9-eGFP + gRNA + sry - 1 minute incubation
Lane 8 - (200 ng) dCas9-eGFP + gRNA + sry - 5 minutes incubation
Lane 9 - (200 ng) dCas9-eGFP + gRNA + sry - 10 minutes incubation
Lane 10 - (200 ng) dCas9-eGFP + gRNA + sry - 30 minutes incubation
Lane 11 - (200 ng) dCas9-eGFP + gRNA + sry - 60 minutes incubation
Lane 12 - (400 ng) dCas9-eGFP + gRNA + sry - 60 minutes incubation
Moreover, when the eGFP-dCas9 is incubated in the cellulose disk in the presence of DNA, the fusion protein can bind to the DNA and remain bound after washes with the tailored buffers. This can be seen when we loaded the cellulose disks inside of the TBE-PAGE, there is a faint band shift (Figure 21, lane 11) at the same height as the control eGFP-dCas9 without the cellulose matrix (Figure 21, lane 6).
Figure 21 - EMSA showing that the eGFP-dCas9 binds the target DNA sequence on top of the cellulose disks and remains bound after the washes with the tailored W1 buffer.
Lane 1 - Marker
Lane 2- 100 ng of Sry gene
Lane 3- Sry + dCas9-HRP
Lane 4 - Sry + dCas9-HRP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2
Lane 5 - Sry + dCas9-HRP+Guide RNA 1
Lane 6 - Sry + dCas9-GFP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2
Lane 7 - eGFP + dCas9-HRP+Guide RNA 7
Lane 8 - eGFP + dCas9-HRP+Guide RNA 7
Lane 9 - eGFP
Lane 10 - Sry + dCas9-HRP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2 (Here Sry gene was immobilized on cellulose strip and addition of dCas9 proteins and guide RNA was all done in cellulose strip).
Lane 11 - Sry + dCas9-GFP+Guide RNA 1 + Guide RNA 4 +Guide RNA 2
Lane 12 - Sry gene loaded on cellulose strip.
our great sponsors
We are proudly sponsored by the following partners in science:
