Model

Z17/Z18

In our search for potential flocculation triggers, we came across two proteins, Z17 and Z18, which are reported to play a role in cell aggregation ^[1]. Because flocculation is central to our project, these proteins quickly became promising candidates. However, before moving into our final chassis, Chlamydomonas reinhardtii, we first needed to test their expression and function in a simpler and well-characterized organism, Escherichia coli.

Z17 and Z18 are adhesion proteins that naturally contribute to cell–cell binding. Their mechanism is based on recognizing and attaching to complementary structures, effectively acting as molecular “Velcro” that promotes aggregation. Importantly, both proteins already contain their own EhaA autotransporter domain, which enables them to be exported to and displayed on the cell surface in E. coli. This makes them highly attractive candidates for engineering synthetic flocculation systems, since they could be repurposed to bring cells together in a controlled way. ^[1]

One of the most interesting properties of Z17/Z18, described in the literature, is that flocculation is reversible. This reversibility relies on a competition mechanism:

1. When cells express Z17/Z18 on their surface, the proteins interact with one another, forming cell–cell bridges and aggregates.
2. If soluble Z17/Z18 proteins are introduced into the culture, they can bind to the same interaction sites.
3. These soluble proteins compete with surface-bound Z17/Z18, breaking the bridges between cells and leading to disaggregation.

Gene design

The design of the Z17/18 genes has gone through two different routes depending on which organism is to express them. This is because of the different ways the E. coli and the Chlamydomonas reinhardtii express the proteins to be able to dock against each other.

E. coli

Since there are only a few studies that exist on the Z17-Z18 pair, we decided to first check their reliability using AlphaFold^[2]. We first checked each protein on its own (Figure 1.A and 1.B) before simulating them together (Figure 1.C). Overall, the confidence levels are pretty good and really promising for the project.

For the expression and transportation of the Z17/18 proteins to the cell surface, it is necessary to attach an autotransporter to the protein. This would in our case be the EhaA. However, we wanted to be able to try and see if the protein was expressed in the cell, which led us to attach GFP to the Z17/18 (Figure 2). This would then hopefully lead to us being able to see the green fluorescence on the edges of the cell and thus providing us with an indicator of expression.

The application for the expression of Z17/18 in E. coli is for both testing the flocculation efficiency of the Z17/18 genes within E. coli, and thus trying to reproduce the results of the STUDY and to try and harvest the Z17 protein for dissociation by competition. We wanted to make our ordering of the gene as efficient as possible as we had to do two separate orders for each of the Z17/18 genes. We therefore chose to be able to use the ordered plasmid for flocculation and Z17 harvesting. We did this by adding a protease cleaving site (TEV) between the Z17 and GFP parts, resulting in the following design of the complex, see below:

The design developed over time to include linkers, TEV protease, XbaI and BamHI restriction site sequences. Further, codon optimization for E.coli was carried out using IDT Codon Optimization Tool^[5].

The restriction sites, XbaI and BamHI, Z18, a TEV protease cleavage site, GFP, and the autotransporter EhaA were integrated into the pET-blank-Kan plasmid. The restriction sites enabled substitution of Z18 with Z17. The pET-blank-Kan plasmid was selected for its kanamycin resistance, with kanamycin providing a viable and readily available selection marker. Further, expression can be induced using IPTG, this was of interest allowing us to control flocculation. Additionally, the plasmid was chosen for its blank backbone, which facilitates versatile genetic modifications and permits the use of specific restriction sites without interference from endogenous sites within the vector.

Chlamydomonas reinhardtii

For the expression of the Z17/18 in C. reinhardtii, we had to have a different approach for the expression, since the machinery is different than in E. coli. Two problems appeared to us: the correct transport of the protein to the cell wall and the anchoring of it.

In our research to answer those questions, we found a study by Molino et al. ^[7] presenting a cell surface display mechanism in Chlamydomonas reinhardtii. This system consists of the assembly, around the gene of interest, of the leader signal peptide SP7, that will ensure the correct translocation of the protein to the cell wall, and a native anchor protein, GP1. The anchor being native to Chlamydomonas, it faces no transcription or translation problems. The construct resulted in the successful expression of mCherry at the surface of the cell.
We decided to use that construct in the scope of our project and utilize it to surround our Z17/Z18 proteins. An AlphaFold^[2] prediction of the final construct was done to picture what to expect from it (Figure 5).

Even though the confidence interval of the construct is not as promising as the one for E. coli, the study seemed to be pretty hopeful concerning the use of this anchor system for protein surface display in Chlamydomonas.

With the use of the SP7 leader peptide, we hope to by-pass the degrading machinery of the cytoplasm and to get a correct translocation of the complex. All genes were codon optimized for Chlamydomonas using the Intronserter tool^[8].

The plasmid used as a suggested model plasmid for work with C. reinhardtii was pLM005. It includes resistance for antibiotics in both prokaryotes and algae. This is the optimal application for us as this means we can transform our plasmid into E. coli for replication to later be used for electroporation into algae.

The strategy involved inserting SP7, GP1, linkers, and Z17/18 into the pLM005 plasmid via the HpaI restriction site using Gibson assembly. HpaI was selected because it minimally disrupts adjacent sequences compared to alternative sites and the enzyme was a viable and available option to the lab. As a second restriction site, we chose to use PacI as it is not naturally present in pLM005. pLM005 is compatible with both C. reinhardtii and E. coli. Further, with permission from and courtesy to the Petroutsos lab, pLM005 provided a readily available vector for the experiments.

Extracting GP1 from wild-type

The GP1 protein is a 400 aa long peptide that is used for the surface expression of a particular gene within C. reinhardtii. The protein consists of a beta-barrel connected to a small alpha-helix with a large linker consisting of repeats of proline and glycine. This linker is about 340 aa long. When trying to order the GP1 gene from multiple different gene synthesisers, we got rejected due to the large amount of repeats. This is due to repeats being extremely hard to synthesise correctly and quality check with the instruments available.

Instead of ordering the GP1 gene, we tried to find the gene already inside C. renhardtii. We did a BLASTN search on NCBIs^[10] whole genome shotgun sequencing database with limitations for C. renhardtii. This resulted in 7 unique strains, whereas one, cc-1690, was one of the strains we were actively culturing. This in turn resulted in us starting to design primers to be able to extract the GP1.

The GP1 gene in cc-1690 was split into 3 exons, resulting in a total length of ~1.9kb. We planned to incorporate the GP1 in a cassette including Z17/18 and SP7. It was decided that we were to be using the PacI restriction site between the GP1 and SP7-Z17/18 cassette. To be able to insert the full cassette into the pLM005, we would need to add the 5’ end of the HpaI restriction site to the 3’ end of the GP1 so we can use it in later stages if we want.

The designs of the primers were hard to get done with precision. The best forward primer resulted in a melting temp of 61.3°C and with a very high GC content of 58.3%. The reverse primers were not much better. It was decided to order two different reverse primers. The first one was designed to incorporate the full GP1 protein. This however resulted in GC content of 65% and a melting temp of 59.6°C. The second one was further in on the protein which meant loosing 7 aa, but resulted in a GC content of 51.85% and a melting temp of 62.5°C.

We then produced two suggestions on how to incorporate the SP7-Z17/18-GP1 cassette into pLM005:

Gibson Assembly
The first idea was to use gibson assembly to incorporate the cassette into pLM005. For this, we used the NEBuilder^[11] tool from New England Biolabs to create optimal Gibson primers. For this, we created primers for both cassettes, one including the Z17 and one for the Z18.

When trying the Gibson kit we had on hand, the positive control provided in the kit failed multiple times. We believe that the product might not have been properly stored or had reached its expiry date.

Blunt end
As the digestion of pLM005 using HpaI resulted in blunt ends, we had to use blunt end ligation. As this was for both the 5’ and 3’ ends of the cassette, we would have problems knowing which of the cassettes were inserted the right way and which would become backwards.

As pLM005 has a lot of different restriction sites, we could use this to our advantage in sorting out the correctly ligated plasmid together with our PacI site we used for the ligation of GP1 and SP7-Z17/18. This is due to the GP1 gene being much larger than the SP7-Z17/18 complex.

We therefore planned to check the ligation using the BamHI and PacI restriction sites. We would do a digestion with these restriction sites, run an electrophoresis gel and do gel purification on the parts corresponding to the right lengths. We could thereafter do a simple ligation on these as both are sticky end restriction sites.

FLO1

FLO1 encodes a lectin-like, GPI-anchored cell-wall protein in Saccharomyces cerevisiae that mediates mannose-dependent cell–cell adhesion and floc formation. FLO1-expressing cells preferentially adhere to one another (a “green-beard” effect), which benefits the group under stress but carries a measurable growth cost to individual expressers. At the protein level, FLO1 comprises an N-terminal lectin domain, a central repeat-rich Ser/Thr region, and a C-terminal GPI-anchor signal; surface localisation can be detected by immunofluorescence and ELISA.^{[15][16][17][18]}

Natural and engineered variation within internal tandem repeats of FLO1 tunes adhesion strength: more (or specific types of) repeats generally increase flocculation. Genome-wide repeat mining (e.g., with EMBOSS ETANDEM) has been used to identify such repeat-containing ORFs, and FLO1 emerges as a canonical example of repeat-driven phenotypic modulation [2,4,6,7]^{[16][18][19][20]}. In iGEM, BBa_K2935073 captures a core FLO1 region (S288C Chr I: 205,985–206,661) shared across alleles—useful as a design scaffold when exploring repeat dosage and codon/intron optimisation for heterologous hosts [5]^[21].

Assembly of Chlamydomonas reinhardtii gene FLO1

The objective of this part of a project was to design genes and subsequently primers for cloning the Chlamydomonas reinhardtii gene FLO1 into a plasmid vector using Gibson Assembly. The primers were designed to add approximately 20 base pair homology arms that match the ends of the gene fragments and the corresponding vector regions. Restriction sites (XhoI, BsaI, HpaI, and XbaI) were also included where useful to allow for downstream cloning applications. We chose to prepare three different plasmid designs in order to compare the strength of the salt promoter with the native promoter of pLM005. Since FLO1 has been documented by the previous iGEM team (FAFU-CHINA 2019^[12]), this sequence was modified to contain introns, as introns should improve the expression of proteins in eukaryotes.

Further on in the research, we came across the salt-inducible promoter CrGPDH3. Our plan was to use this promoter to induce the expression of flo1 and thereby control flocculation. To evaluate this strategy, we designed a series of experiments comparing three constructs: the original sequence obtained from FAFU-CHINA 2019^[12], an adjusted sequence with introns introduced to improve stability and expression, and a construct in which FLO1 expression is driven by the salt promoter to allow external regulation.

To support this design, we used the target gene sequence in FASTA format, the vector sequence and map, and software such as Benchling^[9] (for overlap design), SnapGene^[13], and NCBI BLAST^[14] or Primer-BLAST (for specificity checks). All notes were kept in the lab notebook.

The cloning strategy relied on Gibson Assembly, which requires 20–25 base pair overlaps between adjacent fragments. Because the FLO1 gene contains repeats, it had to be split into three parts by introducing introns. This resulted in the design of four primer pairs that create compatible overhangs for assembly. The three assemblies were constructed as follows: one using the constitutive promoter PSAD with FLO1 containing introns, one using the salt-inducible promoter CrGPDH3(salt promoter) with FLO1 containing introns, and one using PSAD with FLO1 lacking introns. To insert promoters, we cut out the region between BamHI restriction sites.

The primer design requirements included a GC content of 40–60%, a melting temperature between 58–62°C, and minimal potential for hairpins or dimers. The reverse primers were designed so that their 5′ ends included 20–25 base pairs matching the downstream vector flanks in reverse complement.

For the salt promoter assembly, the following primers were used:

• Forward Primer 1:
  ACAGTGTGGACGTTGGGATCGGGATCCCCGCTCCGTGTAA (connects the vector to fragment 1 of FLO1, includes upstream vector homology).
• Forward Primer 2:
  GACCCGCTACGACGGCATTAACCTGGTCACCGTCCGTGAG (connects fragment 1 to fragment 2, includes overlap with the second fragment).
• Reverse Primer 1:
  CTCACGGACGGTGACCAGGTTAATGCCGTCGTAGCGGGTC (connects fragment 1 to fragment 2, includes downstream overlap with fragment 2).
• Forward Primer 3:
  ACGGACCCGCTCATGGAGAAGTTCAACGAGTCGCTGCCCT (connects fragment 2 to fragment 3, includes upstream overlap with fragment 3).
• Reverse Primer 2:
  AGGGCAGCGACTCGTTGAACTTCTCCATGAGCGGGTCCGT (connects fragment 2 to fragment 3, includes downstream overlap with fragment 3).
• Reverse Primer 3:
  CAGGTCGACTCTAGAGGATCAACCTAGTGCTGTCCCTCCGC (connects the vector to fragment 3 of FLO1, includes downstream vector overlap).

For the assembly without the salt promoter, the following primers were used:

• Forward Primer 1:
  TACTCACAACAAGCCCATCTAGAAACATGTTGAAGACGGCTGGC (connects the vector to fragment 1 of FLO1, includes upstream vector homology).
• Forward Primer 2:
  GACCCGCTACGACGGCATTAACCTGGTCACCGTCCGTGAG (connects fragment 1 to fragment 2, includes upstream overlap with fragment 2).
• Reverse Primer 1:
  CTCACGGACGGTGACCAGGTTAATGCCGTCGTAGCGGGTC (connects fragment 1 to fragment 2, includes downstream overlap with fragment 2).
• Forward Primer 3:
  ACGGACCCGCTCATGGAGAAGTTCAACGAGTCGCTGCCCT (connects fragment 2 to fragment 3, includes upstream overlap with fragment 3).
• Reverse Primer 2:
  AGGGCAGCGACTCGTTGAACTTCTCCATGAGCGGGTCCGT (connects fragment 2 to fragment 3, includes downstream overlap with fragment 3)
• Reverse Primer 3:
  GAGCCACCCAGATCTCCGTTAACCTAGTGCTGTCCCTCCGC (connects the vector to fragment 3 of FLO1, includes downstream vector overlap).

The same primers used for the “without salt promoter” assembly were also used in the third assembly, which involved PSAD with FLO1 lacking introns. In total, eight distinct primers were designed. Additional restriction sites (XhoI and XbaI) were added to facilitate potential subcloning in future work.

FUS1-MAR1

While Chlamydomonas reinhardtii are single cell organisms that do not flocculate, cell adhesion is still a vital part of their reproductive cycle. Under normal conditions, the haploid C.reinhardtii reproduce asexually through mitosis. However, low nitrogen levels can trigger cells of opposite mating types to fuse together, forming diploid zygospores. The surface receptor pair FUS1 (MT+) and MAR1 (MT-) has been found to be essential for gamete adhesion. We propose that the FUS1-MAR1 receptor pair could be utilized as a novel strategy for inducing cell flocculation in C.reinhardtii. Both proteins are native to C.reinhardtii and the coding sequence can be cloned from their genomic DNA, however the proteins are only expressed when cells enter the sexual cycle. This means that under normal laboratory conditions, the wildtype haploid cells of both mating types would not clump together with cells in which FUS1/MAR1 expression had been induced. Secondly, like Z17-Z18, it is a heterodimer pair, meaning that cells expressing only one of the surface receptors will not flocculate with each other. Additionally, FUS1/MAR1 are both transmembrane proteins, meaning that the receptors would not need to be attached to a cell surface anchor, such as GP1.

Gene design

The plan was to amplify the full-length FUS1 and MAR1 coding sequences from C.reinhardtii and insert the genes into pLM005 plasmid using Gibson assembly. This would place the genes under the control of the constitutive PSAD promoter. Additionally, the plasmid also contains the coding sequence at the 3’ end of the gene of interest, coding for a crVenus reporter, as well as for a 3xFLAG tag. The plasmid would then be electroporated into C.reinhardtii cells to produce two separate transgenic strains – one constitutively expressing FUS1 and the other MAR1. Additionally, FUS1/MAR1 proteins would also have crVenus and 3xFLAG linked to the C terminus end of the protein, allowing for easier detection and quantification. This system would however not allow us to temporally control flocculation as the receptors would be always expressed. But since FUS1 and MAR1 cannot form pairs with each other, it would still be possible to control flocculation by containing the two strains separately and mixing them together to induce the process.

In order to amplify the genes from C.reinhardtii genomic DNA and to attach overlaps needed for Gibson assembly, we decided to employ a nested PCR approach, using two pairs of primers for both genes. The first primer pair was designed in a way in which the forward primer would bind downstream the GOI, while the reverse primer would bind just before the stop codon. The second pair would be used to amplify the GOI only and to attach ovarhangs needed for Gibson assembly. It is important to note that the genes would be cloned without the stop codon, so that it could be fused together with the reporter and tag.