A Tail of Two Fluorescent Proteins
We are excited to give our first update to our supporters on the progress in developing the Open Yeast EduKit. In short, we have fully designed and built all of the Golden Gate Assembly (GGA) receiving vectors (also known as destination or entry vectors) as the foundation for our future work. From these designs we can build our vectors for our Open Yeast EduKit project. This is going to be a deep dive into the basics of recombinant DNA technology which you can gloss over or ignore completely.
Receiving Vectors as a Foundation for GGA
Receiving vectors are plasmids (also known as backbones) in which we assemble all of our Golden Gate parts for a given goal. They have different antibiotic selection markers, such as chloramphenicol (Cm) or kanamycin (Km), from the backbone that contains the library of DNA parts. Backbones, in general, are required to propagate and manipulate DNA in lab strains of E. coli, such as NEB5-alpha. We build all Open Yeast vectors in E. coli, but deploy in yeast. Open Yeast currently targets either Saccharomyces cerevisiae or Komagataella phaffii (previously called Pichia pastoris). All of the parts in the Open Yeast Collection (OYC) and the new Yeast Protein Expression Toolkit (yPET) use a backbone plasmid (pOpen_v3) with a selection marker that is resistant to the ampicillin (Ap) antibiotic. For Open Yeast, we are using the common Type IIS enzyme BsaI as the GGA enzyme for assembly. OYC does have some parts to build a receiving vector but that involves conventional cloning rather than GGA.
For the backbones of our receiving vectors, we have shifted to using the open SEVA vector system. These are ready made plasmids released without a Material Transfer Agreement (MTA), meaning they are in essence public domain vectors. They have a standardized architecture, using conventional DNA cutting (restriction) enzyme sites. Of importance to us are the PacI and SpeI restriction sites that flank the SEVA cargo area. The SEVA cargo area is also flanked by strong terminators, which helps to reduce spurious transcription of cargo elements. Our goal was to replace the SEVA cargo area with our two dropout parts.
For our Edukit, we have opted to use a variant of the SEVA vectors called BASIC_SEVA, which lacks an origin of transfer (oriT). The main benefits of not having an oriT is that it reduces the risk of horizontal gene transfer in educational settings and beyond, although if someone requires the oriT then they could use the regular SEVA vectors. The BASIC_SEVA vectors are released under the OpenMTA, and were kindly provided to us by Dr. Matthew Haines from Imperial College London. We are using the BASIC_SEVA19.10 (Ap resistant), BASIC_SEVA29.10 (Km resistant), and BASIC_SEVA39.10 (Cm resistant) high copy origin variant plasmids.
To facilitate GGA, we need a second selection strategy unrelated to antibiotics to distinguish between fully assembled final plasmids and the unmodified input receiving vector. This is where the dropout parts come into play. The dropout acts as a kind of placeholder that can be replaced as needed with all of the parts needed for a final yeast vector. A dropout contains a full transcription unit to express the secondary selection marker. The dropout is removed (drops out) during GGA and hence fluorescence is lost in our assembled product.
As with all of the parts in our library, a dropout part is flanked by two Type IIS restriction sites. However, unlike the parts in the library, the receiving vector has its cargo’s restriction sites in reverse orientation as shown in Figure 1. While there are a number of options for secondary selection markers, we chose to use green and red fluorescent proteins (GFP and RFP) to aid in the discrimination of bacterial clones on an agar plate which helps us to visually distinguish fully assembled vectors (no fluorescence) and uncut receiving vectors (fluorescence).
We designed two dropouts with the first having the core BsaI overhangs (S1 and S5 in Figure 2). The purpose of this part when cloned into our BASIC_SEVA vectors is to provide a backbone that has a different antibiotic selection marker from the ampicillin resistance gene used for all of the parts in the library that are in pOpen_v3 as well as a color selection for use in GGA. These core BsaI overhangs will receive the users full assembly of all of the parts in construct design during GGA. For this core dropout we opted to use the commonly used bright monomeric RFP called mScarlet, whose coding DNA sequence is identical to that used in the BASIC_SEVA vectors as well as others.
Our second dropout part uses a specifically engineered GFP known as the super-folding GFP (suGFP). This dropout spans a different part of the Open Yeast syntax with different BsaI overhangs (S3:ATGA and S4:AGAC) as shown in the figure above. The primary purpose of this dropout is to facilitate assembly of two or more GGA level I yeast transcription units to permit the building of metabolic pathways in yeast using Level II assembly. As with the first dropout, we want this one to also be cloned into the high copy BASIC_SEVA plasmids using PacI and SpeI restriction enzymes. The GFP dropout part was synthesized and cloned into pOpen_v3 at Twist Biosciences as part of the 2024 iGEMs distribution. We were able to easily transfer the GFP dropout from pOpen_v3 (ampicillin resistant) to the BASIC_SEVA29.10 (kanamycin resistant), and BASIC_SEVA39.10 (chloramphenicol resistant) high copy plasmids. To confirm these clones we performed whole plasmid sequencing using the amazing Plasmidsaurus. Having sequenced plasmid using old-school Sanger sequencing technique over the past thirty years I am in awe of their speed and quality of sequencing - next morning results from our local drop off point!
Unexpected Speed Bumps Slows us Down
We had a completely different and baffling story whilst attempting to clone our RFP-based Dropout fragment. We were able to have the mScarlet dropout part synthesized okay at Twist Biosciences through a small grant from the iGEMs Engineering Committee but this proved difficult to clone into pOpen_v3 after three attempts in our lab. This same part sequence, along with the GFP part sequence, was also included in the new parts list that iGEMs used for synthesis of the 2024 iGEM Distribution. However, Twist was also unable to clone the fragment and hence it wasn’t available as starting material for moving to BASIC_SEVA as we did with the GFP Dropout part.
When faced with this problem we decided to try cloning the RFP dropout fragment into a BASIC_SEVA19.10 (ampicillin resistant) vector thinking the problem was due to the pOpen_v3 vector. Using the same PacI/SpeI strategy we obtained only a single RFP positive colony - a far cry from the hundreds of GFP positive colonies for the other dropout. We sent it for whole plasmid sequencing and indeed the full fragment did clone into BASIC_SEVA. However, there was an additional SpeI fragment derived from the parent BASIC_SEVA19.10 vector cloned into the SpeI side of the dropout.
"Okay, this should be an easy fix", we thought. "Just cut with SpeI, run the cut DNA on agarose gel electrophoresis, then cut out the larger fragment from the gel and purify its DNA followed by re-ligation to join the ends". Doing this we got more colonies but nowhere near the amount we got for the GFP dropout cloning steps and there were far more colonies that lost the RFP signal. In addition, for those red colonies the RFP signal was often very weak taking a few days to mature. We sent a clone with the strongest signal for sequencing. Interestingly, the plasmid had a rearrangement in the backbone - a small chunk of the ampicillin resistance gene joined with a small fragment of the origin and inserted in between the selection marker and the origin. Frustrating!
Next we cut the BASIC_SEVA19-RFP-DO dropout plasmid (with the extra SpeI) with PacI and SpeI and ligated that into the PacI/SpeI cut BASIC_SEVA29 and BASIC_SEVA39 fragments. This resulted in similar outcomes and transferring from one agar plate to another resulted in loss of the RFP signal. We were starting to get the picture that this RFP dropout is toxic to E. coli for some unknown reason. Very odd given that the GFP and RFP dropout fragments are identical except for the two BsaI overhangs (S5:CGAA & S1:AAAA vs. S3:ATGA & S4:AGAC) and the commonly used GFP and RFP genes! We decided to move on, rather than spend more time investigating this odd phenomenon.
Plan B- Moving from RFP to GFP for our Core Vectors
When we designed the two dropout parts we wanted them to be as universal as possible so we incorporated different pairs of conventional restriction sites flanking each of the two BsaI sites as shown in Figure 3. This allows anyone to independently swap out either or both BsaI sites for something else, perhaps a different Type IIS restriction enzyme that is better suited to their needs. These pairs of cut sites are identical between the GFP and RFP dropout. On the left (5’) side of the dropout the BsaI is flanked by XbaI and XhoI and on the right (3’) side the BsaI is flanked by MfeI and SpeI.
These conventional restriction sites also give us a strategy (Plan B) to replace the RFP transcription unit with our GFP one but leave the core BsaI site and overhangs as shown in Figure 3. By cutting BASIC_SEVA19-RFP-DO (Ap resistant) and BASIC_SEVA29-GFP-DO (kanamycin resistant) dropout plasmids with XhoI and MfeI then mixing and ligating we should end up with GFP positive colonies on ampicillin agar plates and this is what we got. Sending DNA of one mapped clone for whole plasmid sequencing confirmed our success. The next step was to move the core GFP dropout to BASIC_SEVA29 and BASIC_SEVA39 using PacI/SpeI as before. We finally have our core receiving plasmids albeit with GFP instead of RFP. This allows us to move on to the next step of GGA.
Much of the cloning work since receiving the funds was performed by Emi and Emily with oversight by Scott. Emily helped found Open Science Network Society back in 2015 and has been a long time active member. Emi joined our community lab a few years ago and, as a high school student, and has become quite skilled in synthetic biology through our Budding Scientist program. Super bright and highly curious, Emi started first year Science at the University of British Columbia this month. We wish her all the best with her studies and future endeavours. We also thank Nicholas for his working editing these lab notes.
Our next Lab Note will go into more detail on what our final Edukit plasmids will look like. Barring any difficulties this should be a straightforward step. If you have any questions or comments about this work please post them below.
1 comments