E. coli results

Purification Results

Z17-EhaA overexpression

After transforming the pET plasmid containing the Z17 and the EhaA autotransporter construct (hereafter referred to as Z17) into E. coli BL21, we proceeded with expression and isolation of the protein. Figure 1 illustrates successful production of Z17 following IPTG induction at 37°C. After 2 hours of induction, the Z17 band was undemostrated, but high overexpression could be observed at both 3 h and 5 h. These results indicate that extending induction beyond 3 h does not significantly enhance Z17 yield (comparing the strength of the protein band from background), and future optimization (eg. to enhance solubility, comparing the strength of the protein band from background, and future optimization may therefore focus on induction conditions other than prolonged time, such as induction temperature or IPTG concentration.)

Optimizing Z17 solubility and purification

After expressing the Z17 construct (Figure 1), cells were lysed and the supernatant was subjected to His-tag affinity purification. Ten microliters of the elution fraction were loaded onto an SDS-PAGE gel. However, no detectable Z17 band was observed in the elution, and since the flowthrough fractions were not collected, the point of protein loss could not be determined (data not shown). We hypothesized that the construct either failed to bind to the His-tag column or was retained in the pellet due to the formation of inclusion bodies.

To address this issue, we first attempted protein expression at a lower temperature to reduce potential misfolding and aggregation. In parallel, we tested urea denaturation to solubilize inclusion body–associated protein and enhance His-tag accessibility for purification.

To evaluate the effect of temperature on Z17 expression, we compared protein production at 20 °C and 37 °C. SDS-PAGE analysis of whole-cell pellets revealed a distinct band at approximately 50 kDa, corresponding to the expected size of Z17. The band was visible at both temperatures (Figure 2, lane 2; Figure 1, lane 6) but appeared more intense relative to background proteins at 37 °C, consistent with the optimal growth temperature of E. coli.

These findings suggest that higher expression temperatures increase Z17 yield, though possibly at the cost of solubility. This emphasizes the importance of optimizing induction temperature to balance protein production and proper folding. Future experiments could include intermediate temperatures (25–30 °C) to identify the optimal compromise between yield and solubility.

Urea denaturation

After the unsuccessful His-tag purification of lysed Z17, we hypothesized that the His-tag might be buried within the protein’s folded structure or trapped inside inclusion bodies. To overcome this, we performed urea denaturation to solubilize the protein and expose the His-tag. This protocol was adapted from the Costa Rica iGEM 2019 team and optimized for our construct by treating Z17-containing cell pellets with 8 M urea (lanes 1–5 in Figure 3).

We tested varying incubation times and temperatures to assess their impact on solubilization efficiency. The results showed that temperature had minimal influence, whereas longer incubation times increased the solubilization of a wider range of background (non-target) proteins. Furthermore, the addition of glycine, as suggested by the Costa Rica 2019 team, did not improve extraction efficiency.

Based on these findings, we selected room temperature and a 1-hour incubation time as the optimal conditions for urea denaturation (see Urea Extraction protocol). This step allowed us to proceed with His-tag purification, with the goal of refolding and purifying Z17. The observed bands indicate the presence of our target protein (~50 kDa) along with other background proteins, confirming successful denaturation and solubilization (Figure 3).

Successful His-tag purification

After expressing Z17 in E. coli BL21 and denaturing the protein with 8 M urea, we performed His-tag purification as an additional purification step. The denatured lysate was loaded onto the His-tag column, and fractions were collected for analysis. A visible band in lane 9, Figure 4, corresponding to the first flowthrough, indicated that not all proteins were retained on the column. This partial loss may be attributed to the slow, manual application of the sample, which could have reduced binding efficiency compared to using a pump.

Despite this, Z17 was successfully eluted (Figure 4, lane 8). The target protein was primarily recovered within the first 5 mL of elution, suggesting that future purifications could be optimized by reducing the total elution volume from 15 mL to 5 mL, thereby increasing protein concentration and minimizing buffer use.

Chlamydomonas results

Growth Curve of Chlamydomonas reinhardtii

We began by establishing growth curves for Chlamydomonas reinhardtii under our specific lab conditions for the wild type strain CC-125 and a starch over-accumulating strain dp120 (Figure 5). Cultures were maintained at a light intensity of 50 µmol photons m^-2s^-1, shaken at 140 rpm, and kept at approximately 20°C. The cell concentrations were measured using a cell counter. As the cell counter has a large error rate, 4 samples were measured for each of the algae strains. The unexpected peak in cell concentration observed on day 4 suggests that the measurements may not accurately reflect true cell concentration. We suspect that rapid cell sedimentation was the main cause of these unexpected results. Although the resulting growth curve does not provide a reliable model of C. reinhardtii growth, this experiment was an important step in learning how to culture this organism, that all of the team members but one were previously unfamiliar to, and troubleshoot cell count measurements. For future growth curve experiments, we emphasize the importance of thoroughly homogenizing culture samples prior to measurement.

Flocculation Efficiency

To test our project concept, we compared natural sedimentation of C. reinhardtii cells with chemically induced flocculation. By adding NaOH and CaCl₂ to the growth medium, we observed aggregation of algal cells followed by accelerated sinking. Flocculated cells settled at the bottom of the container significantly faster than untreated cultures. These results validated our choice of flocculation as a strategy to separate algae from culture media and provided a proof-of-principle experiment for our planned genetic approach.

We repeated the flocculation efficiency experiment four times, varying the conditions and observing different results. We also confirmed the occurrence of flocculation visually, by microscopy and a time lapse video (Figure 6). When growth medium containing NaOH and CaCl₂ was prepared in advance and added all at once at T=0, flocculation and rapid sedimentation occurred almost immediately (Figure 6, 7) and cell clumping was clearly visible under the microscope (Figure 8a-d). In contrast, gradual addition of base directly to cells followed by adding CaCl₂ at T=0 produced much slower and visually weak flocculation (Figure 9, 10a). By switching out the wild type CC-125 for the starch over-accumulating strain dp120, flocculation was barely visible (Figure 10b). This suggests that the mutant may respond differently to pH changes and metal cation addition compared to the wild type.

Electroporation

We optimized electroporation poring pulse voltages for C. reinhardtii using the linearized PLM005 empty vector in two strains: the wild type CC-125 and the cell wall-deficient CC-3403. Electroporations were performed with a 1 mm cuvette, applying two pulses for cell wall-intact cells and one pulse for cell wall-deficient cells. In the first trial, CC-125 cells were electroporated with two poring pulses at 125 V. No transformants were observed. In the second trial, CC-3403 cells were electroporated with a single poring pulse at 125 V. Cell growth was observed for one of the two replicates (Figure 11). In the third trial, we tested additional voltages (90 V, 105 V, and 120 V) on both CC-125 and CC-3403. No transformants were detected for either strain at these voltages.

Although plasmid integration could not be confirmed within the timeframe of the project, these results suggest that 125 V may represent a suitable electroporation condition for the cell wall-deficient strain CC-3403.

Summary

During this project we decided on working with two different model organisms, E. coli and the green alga C. reinhardtii . The aim was however the same, trying to express flocculation proteins in the organisms. We were working with different protein types to see which would work better: FLO1 and Z17/Z18 pairs. However, the idea was that these proteins would be expressed by the same mechanism, with a salt induced protomer. Firstly, we tried making a cassette for the FLO1 protein, however, there were problems with ordering the gene, since it is highly repetitive and we were unable to find a company that could synthesise the whole cassette of the salt promoter together with the FLO1 protein. We therefore ordered the gene as three different parts that we would assemble with Gibson assembly, however we did not have enough time to obtain the results. We therefore could not proceed with investigating its abilities to flocculate C. reinhardtii .

We also explored another set of proteins, the Z17/Z18 pair. We first started experimenting with transforming these proteins into E. coli , which is a simpler organism to study expression and overexpression in. This cassette only contained the proteins, without the salt protomer. We thereafter worked on isolating the protein pair, to study its interactions. The proteins need a membrane anchor to be expressed on the surface of E. coli , but since we wanted to study the purified protein interaction, and not cause flocculation of E. coli , we did not include an anchor in the cassette. We managed to isolate and purify Z17 from the bacteria, however, this was unsuccessful for Z18.

Similarly, for Z17/Z18 to be expressed on the surface of the microalgae, it needed a membrane anchor, which is a problematic sequence to synthesize when ordering a cassette. This anchor can be found naturally in the genome of C. reinhardtii and we therefore attempted to retrieve it by colony PCR, but this was unsuccessful. Despite these challenges, we were able to make progress in other areas. Specifically, we optimized electroporation conditions in C. reinhardtii , showing that poring pulse voltages between 90–120 V were insufficient compared to 125 V in the cell wall-deficient strain CC-3403. We additionally worked on developing an assay to evaluate flocculation efficiency compared to natural sedimentation in C. reinhardtii .

We also considered using another surface protein pair to induce flocculation - FUS1-MAR1. Unlike FLO1 and Z17-Z18, FUS1-MAR1 are endogenous to C. reinhardtii , allowing us to clone the gene sequences from the genomic DNA of the algae by PCR, circumventing the issues we were facing with getting the parts synthesised de novo. FUS-MAR1 are adhesion proteins, which play a pivotal role in gamete fusion, however the pair has never before been used to induce flocculation in vegetative cells. We planned to amplify the endogenous gene sequences and to place them under a strong promoter. Since FUS1-MAR1 functions as a heterodimer, a population of algae, expressing either one of the proteins on its surface, would not flocculate together. Unfortunately, we were not able to amplify the coding sequences for the pair, as cloning C. reinhardtii genes turned out to be much more complex than expected. However, through trial and error, we were able to optimize the protocol for cloning C. reinhardtii genes, which could be used by future iGEM teams.

In conclusion, even though the main goal could not be reached due to the lack of time, we succeeded with several important steps along the way. We additionally worked on developing a method on how to measure flocculation, for future studies of flocculation efficiency. For further research it would be interesting to continue our research and fulfill our goals of inserting the different proteins into both C. reinhardtii to measure the flocculation, and to E. coli to study their interaction more closely.

Outlook

Our original aim was to transform constructed plasmids with flocculation genes (FLO1 respective Z17/18) into C. reinhardtii and quantify their effect by comparing genetically engineered flocculation with natural sedimentation. While we were unable to complete these steps within our timeframe of eight weeks, our experiments provided a strong foundation for future work. We learned how to culture C. reinhardtii, demonstrated flocculation efficiency through chemical induction, developed an assay method for comparing flocculation to sedimentation, and identified electroporation conditions for the cell wall-deficient strain CC-3403 that warrant further exploration.

Appendix

We aimed to evaluate the compatibility of the BamHI and XbaI restriction sites when used in combination. The PLM005 plasmid, which contains both recognition sequences, served as the template for this analysis. Digestion was expected to generate four fragments, although one fragment is only 6 bp in length and was therefore unlikely to be visible on an agarose gel. The remaining fragments were expected to appear as three distinct bands, see accompanying table.

Table 1. Amount of cuts and expected basepair length of pLM005 digested using BamHI and XbaI, single enzyme and double enzyme digestion.

A total of four digestion reactions were performed:

• Single-enzyme controls: The PLM005 plasmid was digested separately with BamHI (Thermo Fisher Scientific) and XbaI (New England Biolabs, NEB). Each enzyme was used with its recommended buffer: FastDigest Buffer for BamHI (Thermo Fisher Scientific) and rCut Smart Buffer for XbaI (NEB).
• Double digestions: Two additional reactions included both BamHI and XbaI simultaneously. Each double digest was performed using one of the individual enzyme buffers (either FastDigest Buffer or rCut Smart Buffer) to evaluate buffer compatibility and enzyme efficiency under different conditions.

The successful digestions are represented in Figure 12. The biggest band is the undigested supercoiled pLM005 plasmid (lane 6). Looking at the gel, all the expected bands were obtained. What is more, to see is that the buffer composition did not affect the functionality of the enzyme. Both bands (Figure 12, lane 4-5) are of same intensity, implying that both buffers are as effective as the other.

We aimed to digest the Z17-TEV-GFP-EhaA plasmid (hereafter referred to as pET plasmid) and the Z18 fragment with BamHI and XbaI to facilitate the replacement of Z17 with Z18.

The standard protocol recommended using 500 ng of plasmid DNA for efficient digestion. However, 500 ng of the Z18 fragment was not available. During previous restriction enzyme tests, 500 ng of the PLM005 plasmid was successfully digested. To approximate equivalent molar amounts for the Z18 fragment, we calculated the number of DNA copies using the following equation:

After initial visualization under UV light, neither the DNA ladder nor the samples were detectable. This observation was attributed to the accidental omission of SYBR Safe during gel preparation. To remediate, 300 mL of 1X TAE buffer was combined with 30 µL SYBR Safe (10,000X), and the gel was submerged in this solution. The gel was gently agitated at ~50 RPM and covered with aluminum foil for 30 min to allow DNA staining. Total exposure of the gel and DNA to UV light during imaging was approximately 1–2 min.

Figure 13 shows an analysis of Z17-TEV-GFP-EhA plasmid (pET plasmid) and Z18 fragment following digestion with XbaI and BamHI. The digested samples were intended for subsequent excision and purification using the Thermo Scientific GeneJET Gel Extraction Kit. Since Z18 is present only as a DNA fragment, it is not possible to directly confirm a successful digestion as it was first intended to create sticky ends on both sides. However, the plasmids in lanes 4–15 appear linearized, as indicated by their uniform band size. This contrasts with the untreated pET plasmid, which typically appears in multiple conformations, including supercoiled and circular forms. This is then a good indicator of the successful digestion of our plasmid.

NOTE: The subsequent transformation (see Transformation protocol) failed leading to troubleshooting of the ligation protocol. See the results of a new digestion and ligation using pLM005 below.

The Z18 DNA fragment was excised from an agarose gel and purified (see Gel Extraction Purification), and the pET, TEV-GFP-EhaA, plasmid backbone was similarly gel-purified following digestion. In parallel, an aliquot of digested Z18, that had not been run on a gel and therefore not undergone gel purification, was retained as an additional ligation substrate (referred to as Z18-not-pure).

Followed ligation protocol, see BMC protocols with the following modifications stated in the table below:

Additional modifications:

Due to the previously failed ligation of digested pET (Z17-TEV-GFP-EhaA) plasmid with Z18-fragment, a control experiment of digestion and ligation enzymes was done. We decided to do a lot of controls by changing a lot of parameters to try to assess the source of our previous failed experiments.

See Digestion protocol with the following modifications:

See Ligation protocol with the following modifications:

Table 2. 1% agarose gel electrophoresis of digested and ligated pLM005 and controls. Numbers representing lanes 1-20. GeneRuler 1kb DNA ladder from Thermo Fisher Scientific lanes 1 and 17. Status of pLM005 plasmid (digested, ligated or control) and which digestion enzymes are used and whether the samples are ligated are presented under corresponding lane.

In Figure 14. we tried to see why previous ligation of pET (Z17-TEV-GFP-EhaA) plasmid failed. To test the activity of the T4 DNA ligase (Thermo Scientific), ligation on pLM005 plasmid was performed. At first, we loaded the digested pLM005 plasmid. This digested pLM005 plasmid was then used for ligation.
However, it is good to point out that we had issues purifying the digested pLM005, which could explain the faint bands in Figure 7. Another explanation could be due to a potential low efficiency of T4 DNA ligase, which then could also explain the previous failed transformations. This hypothetical decrease of efficiency could be explained by the bad storage of the enzyme or by its age.
To conclude the failed transformation can be a result of multiple factors. Firstly, it was challenging to obtain high concentrations of digested plasmids. Combining this with the lower efficiency of T4 DNA ligase might have resulted in the unsuccessful transformations.

Z18 ligation
Varying ratios of digested and gel-purified Z18 insert to the digested TEV-GFP-EhaA (pET) plasmid, along with longer incubation times, were applied to the previously unsuccessful ligation protocol, resulting in a successful ligation confirmed by transformation into competent cells.

See Ligation protocol with the following modifications:

Table 3. The amount of copies of plasmid/insert relative to the length (bp) and total weight (ng) presented. 4 calculations, consisting of three Z18-insert and one pET plasmid calculation for the amount of copies needed to determine ratios, shown under sample type. Ratios are insert:plasmid. Calculations determined using formula in ligation protocol.

Transformation
DH5α and JM109 competent cells were used for ligation of Z18 with TEV-GFP-EhA plasmid (pEt plasmid) (see Transformation protocol for instructions).

Table 4. Transformation result of Z18-GFP-EhaA ligation (performed 30/7/2025). Ratios of Z18 to pEt plasmid, spunn/centrifuged prior to plating, bacterial strain, and amount of colonies formed are presented.