Notebook

Nanobodies, Secretion System and Cell Assays

Week 1 (July 05 - July 08)

An introduction into FX cloning, sequencing orders and to the lab in general, our supervisors walked us through the FX cloning process with GFP. We cloned the GFP gene into the pSBinit expression vector and grew the cells overnight. We then performed a miniprep to send the results for sequencing to compare with the digital sequence.

Week 2 (July 11 - July 15)

Our nanobody fragments arrived on Monday, so we began the FX cloning process for all of these into pSB_init in order to later express and purify. All were miniprepped and sent for sequencing. Near the end of the week we performed large scale expression with the cells. Periplasmic extraction was performed on the cells to remove the nanobodies from the periplasm. This worked very well for the monovalent nanobodies, but the bivalent constructs may have been too large to enter the periplasmic space, so whole cell lysis may be necessary. The periplasmic extract was purified using batch binding columns. This was succesful for the monovalent nanobodies, giving a high yield.

The media for three different cell lines (THP-1, MM6 and CACO2) was prepared and the cells were thawed and seeded. Cells' well-being was controlled every day and media was exchanged if necessary.

Week 3 (July 18 - July 22)

In order to perform an ELISA, the target protein for the nanobodies (TNFα) was biotinylated. An ELISA was then performed with the purified nanobodies, but was unsuccessful. Towards the end of the week, cells were grown for expression once again, but this time bivalent nanobodies were harvested through whole cell lysis. The yield was much more adequate for the bivalent nanobodies.

Cell counting was performed twice to calculate the doubling time of the cells. The epithelial cell line (CACO2) was split once while the monocytic cell lines (THP-1 and MM6) required splitting twice a week due to their faster recovery and growth.

Week 4 (July 25 - July 29)

The newly harvested nanobodies were purified similarly to the first batch, and used to run a second ELISA to test their ability to bind to TNFα. This ELISA was successful in showing that all nanobody constructs had binding capabilities.

MM6 cells were prepared for the first cell assay by seeding them in a 96-well plate in starvation media. Stimulation was conducted the following day with TNFα and the purified monovalent nanobodies obtained in week 2 and 3 were tested. After 24 hours of stimulations cells were harvested, their RNA isolated and a reverse transcription perfomed. Afterwards, stimulation was analysed by qRT PCR at the end of the week. Analysis showed a failed stimulation, so cell assay needed to be repeated. The other two cell lines were regularly checked and split if necessary.

Week 5 (August 01 - August 05)

Double transformations of E. coli Nissle with the secretion system plasmid and nanobody expression plasmid were unsuccessful. Electroporation of Nissle with both plasmids was performed twice but failed as well. Further tests showed that Nissle was able to integrate the plasmid containing the nanobody but not the one containing secretion system. We assumed that the antibiotic resistance might cause an issue so we tried to transform Nissle with plasmids containing different resistance genes. Transformations with plasmids containing ampicillin or kanamycin resistances worked, while transformations with chloramphenicol or tetracyclin did not. Upon these results, primers were ordered to exchange the chloramphenicol resistance in the secretion system with a kanamycin resistance.

All cells were checked regularly and split if necessary. Splitting frequency of THP-1 cells was reduced in order to have more cells which can be used for the next cell assay (week 6).

Week 6 (August 08 - August 12)

Primers arrived and the Gibson assembly was performed to exchange the antibiotic resistance genes of the secretion system plasmid. Additionally, E. coli MC1061 was chemically transformed with both plasmids to see if this strain is able to take both plasmids up. Cultures grew and were confirmed to contain both plasmids. A glycerol stock was made of the successfully double transformed MC1061 culture. We contacted a research group from America and ask for help concerning our transformation issues with Nissle. We received two protocols to make Nissle chemically and electrocompetent.

A second cell assay was conducted with the THP-1 cell line and three monovalent nanobodies plus one bivalent nanobody were tested. Analysis contained the same steps as described in week 4 with the qRT PCR at the end of the week showing a successful inhibition of TNFα actions by our purified nanobodies. The other cells not used for cell assay were regularly checked and split.

Week 7 (August 15 - August 19)

Double transformations using the secretion system plasmid, as well as the plasmid containing the nanobody with its respective tags. These were unsuccessful in E. coli Nissle, but worked for E. coli MC1061. The transformed MC1061 expressing monovalent nanobodies was grown and induced using arabinose. The supernatant was separated for a Western blot, and the cells were lysed to use the lysate as a sample in the Western blot. This way we could see if the nanobodies had stayed inside the cell if secretion were to fail. Western blot will be performed on Monday (week 8) Nissle transformations with the kanamycin resistant plasmid were attempted but unsuccessful.

Fresh media was prepared for THP-1 cells and they were split in 2 seperate cultures less frequently in order to receive even more cells for a bigger cell assay in the following week. CACO2 cells were regularly checked and splitted while the MM6 monocytic cell line was eliminated.

Week 8 (August 22 - August 26)

A western blot was run on the supernatant and cell lysate generated the previous week, showing that the secretion of monovalent nanobdodies had been successful. The lysate showed no conclusive results. An additional Western blot was performed with secreted bivalent nanobodies, showing that some of the bivalent constructs did not get secreted or displayed the wrong fragment size. Nissle still showed issues with transformations. Therefore, a fresh new batch of competent E. coli Nissle cells was produced and tested, but with unsuccessful results again.

Third cell assay was performed with THP-1 cells. All monovalent and bivalent nanobodies were tested and more TNFα concentrations were used for the stimulation of the cells. Unfortunately, analysis at the end of the week revealed an unsuccessful stimulation and inhibition. Replicates were made, so the second patch of harvested cells is analysed the following week. THP-1 and CACO2 cells were regularly checked and split.

Week 9 (August 29 - September 02)

E. coli Nissle was still not complying with transformation, so different protocols were attempted to achieve competence. Finally, a method for electrocompetence yielded competent Nissle cells. These were transformed using a monovalent nanonbody (VHH#2B) and a bivalent nanobody (biv. VHH#12B) along with the secretion system. An ELISA was performed to test the TNFα-binding capabilities of the secreted nanobodies from MC1061. This ELISA was successful!

Replicates of the cell assay performed in the previous week were analyzed and confirmed the failure of the stimulation. THP-1 and CACO2 cells were regualarly checked and split.

Week 10 (September 05 - September 09)

E. coli Nissle was double transformed with secretion system and nanobody plasmids containing a monovalent and a bivalent nanobody resepectively. Additionally, MC1061 was transformed with all nanobody cadidates as a control to Nissle. The Nissle and MC1061 colonies were precultured and secretion was induced with arabinose. Supernatant was tested with a western blot similarly to that performed in week 7. Western blot showed successful secretion of nanobodies (monovalent and bivalent) in Nissle and MC1061!

THP-1 and CACO2 cells were regularly checked and split.

Week 11 (September 12 - September 16)

More competent Nissle cells were made with the protocol that yielded functional cells. Secreted nanobodies from the previous week were tested with an ELISA and showed that all nanobodies are able to bind TNFα. Since we were able to proof that MC1061 and Nissle are able to secrete functional nanobodies, we ordered primers to insert three nanobody candidates into the plasmid that contains the NO sensor and sfGFP.

CACO2 cells were regularly checked and split. THP-1 cells were again split into 2 flasks and less frequently splitted to receive more cells for the last cell assay.

Week 12 (September 19 - September 23)

The attempt to exchange the sfGFP with a nanobody by Gibson assembly failed for two constructs but one monovalent nanobody was successfully inserted into the plasmid that contains the NO sensor. MC1061 and Nissle were transformed with the secretion system plasmid and the NO-sensing plasmid containing a monovalent nanobody.

Fourth cell assay was conducted with the same conditions that were used in week 6. Only one monovalent and bivalent nanobody each were tested on THP-1 cells. Analysis was postponed to the next week. CACO2 cells were given to a colleague because there was no time left to start an additional cell assay with CACO2 cells.

Week 13 (September 26 - September 30)

Successfully transformed colonies of MC1061 and Nissle with the ability to secrete a monovalent nanobody upon nitric oxide induction were precultured and induced with DETA/NO (nitric oxide source) over night. A Western blot was perfomed and showed that both strains were able to secrete a nanobody after NO-induction! ELISA showed successful TNFα-binding abilities of the secreted nanobodies.

Analysis of the last cell assay showed an inhibitory effect of the bivalent nanobody but not with the monovalent one. The remaining THP-1 cells were eliminated.

NO-Sensor and Integration Lab

Week 2 (July 11 - July 15)

We revived E. coli Nissle 1917 (EcN) from Mutaflor pills and made the bacteria chemically competent. Competent cells were then aliquoted and stored for later use.

Week 3 (July 18 - July 22)

We amplified our plasmids piGEM1, piGEM2, and piGEM3 in the lab strain Mach1, miniprepped them, and stored them for later use. Samples were also sent for sequencing to check the sequence of our plasmids. Hereby, we named our plasmids as follows:

• piGEM1: control plasmid without any promoter, with sfGFP and NorR
• piGEM2: our actual construct with pNorVβ, sfGFP and NorR
• piGEM3: plasmid containing the promoter from the ETH, sfGFP and NorR

We then transformed EcN with these plasmids and ran the first assay of NO induction with technical duplicates of piGEM1,2,3 and non-transformed EcN used as a negative, non-fluorescent control.

Week 4 (July 25 - July 29)

We re-ran the same induction experiment as the week before but using the transparent minimal medium M9 and analyzed the data. We found similar patterns of expression of GFP as in the assay with LB.
We also prepared the first steps for working on integration: we linearised our nitric oxide sensing system for later Gibson assembly to the suicide integration vector. We made DB3.1 bacteria competent and transformed them with pOSIP-KO from the iGEM distribution kit. DB3.1 is a strain resistant to the toxin present in pOSIP-KO so it would be used for cloning purposes.

Week 5 (August 01 - August 05)

Since no transformant colonies had grown, we tried the transformation of pOSIP-KO in DB3.1 again under different conditions. In the meantime, we went on preparing the Gibson assembly of our suicide vector and successfully linearised the fragment with the secretion system.
After two days in the 27°C incubator, we found one colony on the pOSIP-KO transformants plate. We miniprepped the colony, linearized the plasmid, and ran the Gibson assembly of the suicide vector. Following this, we transformed Mach1 with the newly assembled vector. In parallel, we sent the miniprepped pOSIP-KO for sequencing.

Week 6 (August 08 - August 12)

On the first day of this week, the results from the sequencing of pOSIP-KO came back negative. We spent this week trying to find out what was wrong with this plasmid, sent several samples for sequencing, and set up several PCR reactions, all followed by gel electrophoresis to assess the purity and nature of our product. However, we were not able to get any sensible bands for pOSIP-KO. We then decided to transform the miniprepped plasmid into DB3.1 and Mach1. Our idea was that regular Mach1 bacteria should die after transformation with pOSIP-KO because they are not resistant to the toxin encoded on this plasmid. However, multiple colonies grew on the Kan-plates: This is how we could prove that the plasmid we thought to be pOSIP-KO was not.
When we realized this, we decided to try the integration with different versions of pOSIP, namely pOSIP-TT and pOSIP-CH from the iGEM kit. We tried to transform DB3.1 with these plasmids, but nothing worked.
At the same time, we continued to work on our NO sensing system and performed several PCRs to exchange the number of RBSs in front of sfGFP and circularized the obtained products using Gibson.

Week 7 (August 15 - August 19)

We transformed Mach1 with our new constructs (2 RBSs and 3 RBSs), miniprepped them, and sent them for sequencing to check the successful replacement of the original sequence by the additional RBSs. We then set up a NO-induction assay with piGEM2_1RBS, piGEM2_2RBS, and piGEM2_3RBS to see how the amount of RBSs in front of sfGFP affects the response to nitric oxide. On the next day, we ran several PCR reactions to remove NorR from our constructs to see how vital the feedback loop is. We then circularized the obtained products using Gibson, transformed the newly assembled plasmid into Mach1, miniprepped it, and sent it for sequencing.
This week, we also attempted to transform EcN with the pOSIP-KO plasmid from the UNIL team kit, but this was no more successful than the other trials. Fortunately, the pOSIP-KO we had ordered from Addgene had arrived, and the transformation into DB3.1 was successful. We miniprepped it the next day and sent it for sequencing. The results soon came back positive.

Week 8 (August 22 - August 26)

While rethinking our integration approach, we realized that the two primers we used for Gibson assembly of our integration vector were not optimally designed. Hence, we designed and ordered new ones. When these arrived, we retried the assembly process using the miniprepped pOSIP from Addgene and transformed EcN and MC1061 (as a control) with the fully assembled integration vectors. We hereby named the integration vectors as follows:

• pMonster1: assembled suicide vector allowing for the integration of the nitric oxide sensing genes (with sfGFP as a reporter) as well as the genes required for secretion.
• pMonster2: assembled suicide vector allowing only the integration of the NO sensing genes (with sfGFP as a reporter). This was thought of as a control in case pMonster1 would have been too big for integration.

In parallel, we set up a gradient PCR of the fragments and backbone of our integration vector to check whether a particular annealing temperature would lead to a purer product. However, we found that different temperatures did not make a difference. At the same time, we set up a new NO-induction assay in EcN with the plasmids piGEM2_1RBS, piGEM2_2RBS, and piGEM2_3RBS without NorR. Although this experiment was done using only technical triplicates, we could see that depletion of the NorR-loop substantially reduces the response to NO in all constructs tested.

Week 9 (August 29 - September 2)

Since we had already gained a clear overview of what parts were essential to our sensing device, we now wanted to characterize our system correctly. For this purpose, we set up a NO-induction assay with technical and biological triplicates of EcN to compare piGEM1, piGEM2_1 RBS, piGEM2_2RBS, piGEM2_3RBS and piGEM3 with each other. Since we were assessing technical and biological triplicates of all constructs, we had to distribute the samples between 3 well plates, each being read for 16 hours. This time, we changed the carbon source in our medium to glycerol instead of glucose, as done in our reference paper.
Furthermore, we tried a colony PCR to check for integration in pMonster1 and pMonster2 transformants. Unfortunately, the primers were not designed accordingly to the EcN strain genome, and the colony PCR did not work.

Week 10 (September 05 - September 09)

Data analysis of the experiments from Week 9 showed very unexpected patterns of NO detection. Therefore, we returned to M9 with glucose and reran the same experiment. This time, the curves looked familiar and were usable for proper data analysis.
Furthermore, the new primers for colony PCR had arrived, and we had transformed more EcN with the integration vectors, so we set up a colony PCR of new pMonster1 and pMonster2 transformants. Unfortunately, we could not prove successful integration because we later discovered that the integration plasmid would also have yielded a band of the expected size on the gel.

Week 11 (September 12 - September 16)

Retrying colony PCR with other colonies from Week 10, which had been stored in the fridge, showed that most colonies had lost the suicide plasmid over time, but we could not find any integrants. This is why we decided to try the whole integration process one more time under varying conditions. We incubated some transformants at 27°C, as specified by the protocol, and some at 37°C because we hypothesized that maybe the integrase had not been expressed long enough in precedent trials. The next day, we found several colonies on all plates, independently of the incubation temperature. We set up a colony PCR with most of these colonies, but the colony PCR reaction did not work. We also set up the last three plate reading assessments this week to characterize our NO sensing system: first, we compared the construct we had identified from previous assays as the most promising, piGEM2_2RBS with NorR, to piGEM2_2RBS without NorR. From this experiment, we could prove that NorR is essential in eliciting a strong response to NO. We then ran two further measurements of sfGFP expression at several different NO concentrations in order to determine the EC50 value of our system.

Week 12 (September 19 - September 23)

We wanted to try colony PCR of the integration vector transformants one last time, but the reaction failed again. Since university courses had started again, we, unfortunately, no longer had the time to retry colony PCR with different primers or send colonies for genome sequencing.

Week 14 (October 03 - October 07)

In order to have a better look at how our constructs behave, we did a flow cytometry for the following plasmids: piGEM1, piGEM2_1RBS, piGEM2_2RBS, piGEM2_3RBS, piGEM2_2RBS without NorR, and piGEM_3. The raw data showed us similar results to the GFP readouts we did weeks before.