Design

Our first attempt on cloning the 64 repeats of resilin was to apply a classical restriction and ligation approach. Specifically, we aimed to apply directed PCR-free engineering to obtain these highly repetitive DNA sequences [1].
This method is based on a sophisticated arrangement of recognition sites for four type IIS restriction enzymes. In particular the enzymes NdeI, SacI, BsaI and BsmBI are used. It is an approach for a stepwise and seamless elongation of repetitive sequences. We would like to demonstrate that method using our project as an example.
To use this approach, we needed two parts.
The first part is the backbone. Here, we used the pMK1 backbone that was mutated to only have the restriction recognition sites for NdeI and SacI once at the desired position of the backbone and nowhere else. In figure 1 we depict a dummy vector. Note, that this figure is only for demonstration purposes and does not reflect the sequence for the actual pMK1 vector!

The second part is an initial sequence for the gene that is supposed to be repeated - in our case resilin. To be able to skip a few steps we modified the sequence of resilin using different bases for the same amino acid to get some variance between repeats. Thus, we were able to synthesize a gene fragment containing four repeats of resilin which were flanked with restriction recognition sites (RS) for NdeI, BsaI, BsmBI and SacI (Fig.2). This fragment will be referred to as R4 (Fig. 2).

Both the vector and R4 are being cut NdeI and SacI to create complementary overhangs for a following ligation. The predicted result is depicted in figure 3 and is from now on addressed as PR4 (plasmid with 4 repeats).

Now the first combination of repeats takes place. PR4 is now used in two different restriction reactions before combining them again. One part is cut in a double digest using BsmBI and SacI to open the plasmid directly behind repeat 4. This is the accepting vector. The other part is cut in a double digest using BsaI and SacI to cut out the four repeats including the BsmBI RS. This is the donor part provding the new insert. The insert now can be incorporated into the accepting vector by ligating the SacI overhangs and the overhangs from BsmBI and BsaI which conveniently have complementary overhangs (Fig. 4). Thus, now a new vector containing 8 repeats is obtained (PR8, Fig. 5).

This procedure is repeated for the PR8 vector to obtain PR16 containing 16 repeats. As seen, there is a linear increase in repeats. The procedure can be repeated to eventually get a vector containing all 64 repeats (Fig. 6).

To now express the resilin, the repeats can be cut using BsmBI and BsaI to be inserted into an expression vector. In our case we planned to use the pET-19b to include our 64 repeats to finally have the expression system at our hands (Fig. 7).

Using this method it would be easy to express different lengths of resilin (16 repeats, 32 repeats or 64 repeats).

Build

We ordered the R4 sequence and started the procedure by inserting it into the pMK1 plasmid was kindly provided by the group of Elke Deuerling. We then continued proceeding with trials to combine two pET-19b_R4 with each other to obtain the pET-19b_R8 plasmid.

Test

When attempting the insertion of R4 into pMK1 we promptly encountered difficulties. After transforming the supposedly pMK1_R4 into *Escherichia coli* and obtaining the plasmids afterwards, we conducted an analytical digest using BsaI and SacI. As expected, the lanes containing the alleged pMK1_R4, two bands were visible, representing the shorter insert (lower band) and the bigger vector backbone (higher band) (Fig. 8).

What was peculiar was the control lane with the empty pMK1 vector. According to the literature there was supposed to be no BsaI restriction recognition site in the empty pMK1 [1]. Thus, when cutting with BsaI and SacI we would only have expected SacI to cut, thus only one band to be seen in the gel. Since the ladder was not optimal and to investigate the abnormality we attempted another double digest.
For the second double digest we used SacI and NdeI. This time, we expected one band at around 3800 kb representing the backbone of pMK1. The fragment between the SacI site and the NdeI site should be too small to be seen on the gel. Against our expectations, two bands appeared for the empty pMK1 plasmid (Fig. 9).

We figured, there must be something inaccurate with the arrangement of restriction recognition sites in pMK1, so we sent the plasmid to sequencing. The sequencing report confirmed our suggestion. The plasmid contains two more additional restriction recognition sites, compromising our plan: one for BsmBI and one for BsaI (Fig. 10)

Learn

At this point of our story line we could include several more engineering cycles on the trials of removing the restriction sites by mutagenesis.
In summary, removing the unwanted sites of BsmBI and BsaI took longer than expected and the method itself, even if it would work at the first attempt, would have taken several weeks as well. Due to time constraints we decided that we should look around for new, faster methods and to sadly let this cloning approach down.

Design

The Rolling Circle Amplification (RCA), was one of the recommendations provided by Daniel Wedemeyer (Link) which makes use of unique polymerases, capable of isothermal DNA replication of circular DNA (cDNA) molecules. Their products consist of multiple repeats of the given cDNA template through a continuous addition of nucleotides to the primer, making it an ideal candidate for our goal of producing a molecule with up to 64 repeats of the resilin exon [2].
Ideally this method would consist of two steps:

1. Circularization of our DNA template by using a ligation template and the T4-ligase followed by a purification.
2. Replication by the phi29 polymerase which adds complementary nucleotides continuously to one primer until aborted.

For the use in our project, we planned a double digestion with BsaI and BsmBI after the amplification of pMK1-RE4. The samples would be run on an 2 % agarose gel from which the 180 bp RE4 fragment would have been cut out and purified for further usage.

The T4-Ligase could then be used to ligate our fragment to form the cDNA template. A complementary sequence to the individual repeats was planned to be used as the primer for the phi29 polymerase, which would continuously add nucleotides until terminated.
With this approach, we would have to terminate the reaction after varying durations in our attempt to obtain molecules comprised of 16, 32, and 64 repeats.
Finally, to insert our fragment into an expression vector, we considered blunt end ligation and SLiCE cloning.

A complementary sequence to the individual repeats was planned to be used as the primer for the phi29 polymerase, which would continuously add nucleotides until terminated.

---

With this approach, we would have to terminate the reaction after varying durations in our attempt to obtain molecules comprised of 16, 32, and 64 repeats.
Finally, to insert our fragment into an expression vector, we considered blunt end ligation and SLiCE cloning. However, this method was abandoned due to expected complications and the further development of another more promising approach.

Build

The previously constructed pMK1_RE4 plasmid was used as our basis for this approach. The phi29 polymerase expression construct was included in the current distribution kit and the primer was ordered from IDT.

Test

We encountered issues as early as our agarose gels after digesting our pMK1_RE4 plasmid with BsaI and BsmBI. The 180 bp fragment including our four resilin repeats that we intended to isolate was not visible (Fig. 13).

Learn

Simultaneously to planning the RCA method we also started to work on the following Repeatigo method approach.
After comparing both methods, we decided to go for the Repeatigo method, since we were facing diffuculties with the RCA approach.
For Repeatigo we knew how to design our parts, how to build experiments in the lab, and who to contact in case of problems (see our Human Practices page).
Thus, we discarded the RCA approach and focused on the Repeatigo method.

Design

Our final approach for the cloning of the resilin repeats was a method based on the assembly of oligos. Thus, we developed the Repeatigo method. Read more about the concept of the method on our Project Description page (Link) or Parts Page (Link).
The concept of this method is to synthesize single stranded oligos with a 5’-phosphorylization, that will assemble while repetitively creating overhangs for further assembly. Finally, double-stranded DNA will be generated containing several consecutive repeats of the same sequence. The start and end oligos do not allow further assembly and stop the extension process.
The first ligation is seperated by size in agarosegel and the desired length of the insert can be cut out and cleaned up form the gel.
We chose to use pET28a(+) as a vector because it is a simple standard expression vector with a C-terminal His-tag which we wanted to use to purify our resilin proteinpolymer. The His-tag is removed with thrombin which we already had in the lab.
The start and end oligos incorporate a restriction recognition side. The start oligos have a NdeI restriction recognition side and the end oligos have a SacI restriction site. This allows specific restriction and ligation into a vector. The ligated resilin oligos and pET28a(+) are cutted with NdeI and SacI and ligated.

To express our vector we can not use normal E. coli strains. Standard strains do not like repetitive sequences and tend to recombine them, cut or even degrade them. Therefore we looked for a recombinase A deficient E. coli strain.

Build

We ordered the six needed oligos and the E. coli strain JM109 for expression as it is recA-deficient.

Test

We used different ratios of start- and middle-oligos (Fig. 18) and different amounts of middle-oligos during the ligation step (Fig. 19). The gel showed DNA hinting to the fact that at least some oligos

Learn

Our newly developed method did not work out at the first attempt. We decided to give it more chances and try different concentrations ratios, annealing and ligation times.

Design & Build

In a few more trials we had a lot of new realizations on how we could improve the Repeatigo method. Here a short overview on why and how we improved the process.

1. At first we did not see any bands in the gel. To try out a new approach we decided to change up the times for the annealing of the start and end oligos. The largest middle part is formed in this time, so we gave it 55 min to anneal instead of only 12. This made our repeats significantly longer.
2. We noticed that eventhough we clearly could see that the Repeatigo step worked, we got no PCR product visible on gels after the PCR amplification. We figured, this might be due to the fact that the primers of the PCR, which were exclusively binding to the start- and end-oligos, could not bind because these start and end oligos simply did not ligate to the middle part during the ligation step. Before, the start oligos were added and ligated to the middle part overnight before the end-oligos were added for a ligation over 2 h. Now we added the start-oligos and only waited 2 h for them to ligate before adding the end-oligos for a ligation overnight. This resulted in better products of the reaction as all oligos have enough time to be ligated and also the ends can bind.
3. The amount of DNA to obtain after the annealing and ligation is not very high. So after doing a gel clean up from the Repeatigo products we decided to run a PCR to amplify it.

Test

The annealing and ligation times were improved so longer resilin fragments can form. We used 55 min for complementary oligo annealing and ligated all oligos over night with a T4 ligase.
Different concentrations of mid oligos were tried and the differences are visible in the gel (Fig. 24).

Gel slices were extracted from 3, 1.5 and 1 kb and cleaned with the Quiagen gel clean up kit.
The following PCR amplification and gel (Fig. 25) worked as well after optimizing the ligation, DNA concentrations and the gel clean up, showing us a fragment at approximately 1400 bp. This means we were able to synthesize 30 resilin repeats.

After repeating the PCR some more times and saving the positive probes, we did a double digest of our resilin repeat sequence and pET28c(+) . We used pET28c(+) instead of pET28a(+) because it was available in our lab and has the same sites and vector properties.

The restricted pET28-c and resilin sequences were effectively ligated into the pET28-c_resilin plasmid. In comparison to the negative controls of just the undigested pET28-c plasmid lacking the resilin insert, the plasmids containing the resilin insert demonstrate a larger size, which corresponds to an approximate size of over 1 kb. As previously demonstrated, the size of the resilin fragment, synthesised using the repeatigo method, was approximately 1,4 kb.
Our plasmid pET-28c(+) was successfully transformed into E. coli DH5a strain. The following Colony PCR with primers binding in the resilin insert was positive for all testes colonies.

Our insert is still 1,4 kb long so we successfully cloned about 30 resilin repeats into E. coli.

Engineering Success!

By using the Repeatigo method we were finally able to clone consecutive repeats of resilin. We had positive results in the gel after the ligation of the oligos (Fig. 24), were able to amplify the product in a PCR (Fig. 25), and finally could clone the insert into pET28c(+) followed by a positive result in a colony PCR (Fig. 27), we were able to confirm the presence of the resilin repeats in the vector. Our plasmid is currently at sequencing and we can not wait to see the results!

Design

We relied on gel extraction kits for experiments on both resilin and hyaluronic acid. For example for resilin we used our Repeatigo method to generate many resilin fragments of different lengths, but only wanted to work with fragments that were approx. 1 kb in size. We cut out these bands at the appropriate height and extracted them up. In the course of the process, we used three different gel clean-up kits and encountered two significant problems with all of them: Despite working carefully and strictly following the protocol instructions, our samples were heavily contaminated with proteins, probably salts from the gel and TAE buffer, and organic solvents from the clean-up, as indicated by the 260/280 and 260/230 ratios. Furthermore, when we amplified the samples by PCR after the gel clean-up and then tried to load them onto a gel again, the samples did not sink to the bottom as usual but flowed out of the gel pocket. As a result, we didn't get any bands on the gel and could not progress with our experiments from this point on.

Build

We, therefore, thought about how we could optimize the Quiagen MinElute Gel Extraction Kit ourselves and came up with the following new approaches:

• additional washing step with 70 % ethanol
• let the alcoholic components evaporate by letting the tubes stand for 15-20 min with an open lid
• after the evaporation transfer the spin column in a new tube
• elute with 20 µL Elution Buffer
• run PCR as usual
• Before loading the PCR samples, let the samples again stand with an open lid for about 10-15 min
• add the loading dye pipette up and down and let it stand again for 5 min
• Pipette up and down before loading the samples until loading dye and sample are well mixed

The working group of Prof. Landau, also located at the CSSB, kindly provided us with 70 % ethanol, enabling us to test our new approaches to improving the gel clean-up. We had enough of the other necessary materials, such as tubes, which accelerated the implementation of our ideas.

Test

As we received the 70 % ethanol we tested our newly designed protocol. The instruction of Quiagen are marked in black and our additional steps are marked in green.

1. Excise the DNA fragment from the agarose gel with a clean, sharp scalpel.
2. Weigh the gel slice in a colorless tube. Add 3 volumes of Buffer QG to 1 volume of gel (100 mg gel ~ 100 µL). The maximum amount of gel slice per spin column is 400 mg. For >2% agarose gels, add 6 volumes Buffer QG.
3. Incubate at 50°C for 10 min (or until the gel slice has completely dissolved). Vortex the tube every 2-3 min during incubation to help dissolve the gel.
4. Alter the gel slice has dissolved completely, check that the color of the mixture is yellow (similar to Buffer QG without dissolved agarose). If the color of the mixture is orange or violet, add 10 pl 3 M sodium acetate, pH 5.0, and mix. The color of the mixture will turn to yellow.
5. Add 1 gel volume of isopropanol to the sample and mix by inverting
6. Place a MinElute spin column • in a provided 2 mL collection tube or A into a vacuum manifold. For information about set up, see the MinElute Handbook.
7. Apply sample to the MinElute column and • centrifuge for 1 min or A apply vacuum until the entire sample has passed through the column. • Discard flow-through and place the MinElute column back into the same collection tube. For sample volumes of more than 800 µL, simply load and spin again.
8. Add 500 µL Buffer QG to the MinElute column and • centrifuge for 1 min or A apply vacuum. • Discard flow-through and place the MinElute column back into the same collection tube.
9. Add 750 µL Buffer PE to MinElute column and centrifuge for 1 min or A apply vacuum. • Discard flow-through and place the MinElute column back into the same collection tube. Note: If the DNA will be used for salt-sensitive applications, such as direct sequencing and blunt-ended ligation
10. Add 750 µL 70% ethanol to MinElute column and centrifuge for 1 min
11. Let the alcoholic components evaporate by letting the tubes stand for 15-20 min with an open lid
12. Transfer the MinElute column after the evaporation in a new 2 mL tube
13. Centrifuge the column in a 2 mL collection tube (provided) for 1 min. Residual ethanol from Buffer PE will not be completely removed unless the flow-through is discarded before this additional centrifugation.
14. Place each MinElute column into a clean 1.5 ml microcentrifuge tube. To elute DNA, add 20 µL instead of 10 µL Buffer EB (10 mM Tris-Cl, pH 8.5) or water to the center of the MinElute membrane. (Ensure that the elution buffer is dispensed directly onto the membrane for complete elution of bound DNA.) Let the column stand for 1 min, and then centrifuge the column for 1 min.
15. Measure concentration and 260/280 and 260/230 ratios using a NanoDrop
16. run PCR as normal
17. Let the samples stand with an open lid for about 10-15 min before loading the PCR samples
18. Add the loading dye and pipette up and down to suspend
19. Let it stand again with an open lid for 5 min
20. Pipette up and down before loading the samples until loading dye and sample are well mixed

First, we tested the optimized clean-up protocol using pUC_HasA and pMK1 samples, which we amplified by PCR and then loaded onto a 1% agarose gel. We can show that we increased the purity regarding contamination with salts and organic solvent.

additional washing step with 70% ethanol	Sample	Concentration ng/µL	260/280	260/230
yes	pUC_HasA 1	53.6	1.77	1.25
no	pUC_HasA 2	64.8	1.72	0.13
yes	pMK1 1	16.4	1.47	0.75
no	pMK1 2	30.9	1.42	0.15

After following our optimized protocol, we were first able to load our beforehand extracted and then amplified resilin DNA sequence on an agarose gel (Fig. 25) and then move on to the next steps in cloning resilin, which was extracting the amplified resilin sequence and starting a restriction-ligation reaction with pET28c(+), followed by transformation into E. coli.

Engineering Success

We learned that the elongated, additional evaporation steps and the 5-min incubation with the loading dye before loading the samples on the agarose gel were significant in avoiding the samples floating out of the gel pockets. We are really proud of our engineering success because it allows us to move on with our project and especially bring the cloning of our resilin repeats forward.