Growth curve and cell growth dynamics

The OD730nm values of both strains (Synechocystis sp. PCC 6803 and Synechococcus sp. PCC 11901) were done at the same time for ten days. No lag phase is observed for either strain, which means the cultures that they were inoculated from were probably not at the right phase. By the OD730mn values that the original cultures had, they were probably at the stationary phase when it is recommended that they are at exponential growth phase for inoculation to generate a growth curve. The logarithmic growth phase is advisable since the cells actively divide and the growth rate is somewhat constant, which makes the increase in cell numbers more consistent and predictable, thus it is easier to observe growth dynamics and quantify growth patterns. Since the cell density from the original culture was too high to inoculate, the lag phase was 'skipped' and the culture started exponential phase directly (graph 1).

The cells were grown at 30ºC, under 300 µmol m/2s, the cells were grown at 180 rpm. (Note: Synechococcus sp. PCC 11901 strain doubling times and growth rate were not found in published literature since this strain is very new in cyanobacterial research. )

Graph 1 Growth curve of Synechocystis sp. PCC 6803 in BG-11 medium: The cells were monitored by optical density (OD) measurements at 730 nm. The y-axis represents the optical density readings which serve as an indicator of cell density. The x-axis shows the duration of the experiment in days.

Table 11. Growth rate and doubling time of Synechocystis sp. PCC 6803 and Synechococcus sp. PCC 11901. Calculations done using equations: Doubling Time (DT) = (t2 - t1) / log2(Off / ODi). Growth Rate (GR) = ln(2) / Doubling Time

Gene extraction

The MTR target genes: MtrA, MtrB, MtrC and CymA, were extracted and amplified from the genome of Shewanella Oneidensis MR-1 using OneTaq (NEB) polymerase. The PCR primers were designed to introduce CyanoGate overhangs in the extracted operons allowing their integration into Lv0 acceptor plasmid vectors. The MTR operons: MtrCAB, MtrCA and MtrAB, were extracted via PCR extraction (One-Taq polymerase) but with primers that did not introduce overhangs. A second PCR (Q5 polymerase) was then performed on the initial PCR product with overhang primers to produce MtrCAB, MtrCA and MtrAB g-blocks (gene with overhangs). All extracted genes and operons were analysed using agarose gel electrophoresis to determine the presence and size of the fragments. Appropriate negative controls were used for all gel electrophoresis runs (negative control was water).

The PCR extraction was successful, seen in figure 1 and 2, but only after troubleshooting (which involved using different enzymes, different annealing temperatures; and different extension times). The annealing temperature was determined used NEB Tm calculator. While One-Taq polymerase worked on the first try, Q5 and Platinum SuperFi II DNA Polymerase were preferred due to their very high fidelity. Finally, the bands from the gel were extracted, cleaned and taken forward into level 0 Cyanogate acceptor plasmid vector. All the extracted genes have been uploaded onto the IGEM registry.

Completed Level T Constructs:

LvT eYFP constructs:

LvT Synechocystis KS constructs:

LvT E. coli KS construct:

Biosafety Analysis

Level T eYFP constructs were made in order to characterize PcopM195-BCD zinc-inducible Synechocystis promoter by measuring eYFP fluorescence. Level T Synechocystis KS constructs were supposed to be transformed into Synechocystis to characterize kill switch activity. Level T eYFP and KS constructs were screened by colony PCR. For each construct there were colonies that gave a correct size amplicon (Figures 1 and 2).

Cultures from colonies that gave the correct cPCR band were set up, plasmids were purified and screened by restriction digestion for correct insertion (Figure 3 and 4). After digestion, while the correct number of bands was observed, they were all slightly larger than expected. Due to time constraints, decided to continue with Synechocystis transformations. Importantly, we later also sent the Level T constructs to Edinburgh Genome Foundry for sequencing by Oxford Nanopore technology and analysis. The sequencing report showed that most reads of the Level T Kill Switch constructs were of the approx. expected size, which suggested that larger than expected size bands observed after restriction digestion were likely an experimental error.

We performed the Synechocystis transformation protocol received from ABOA-Turku’s iGEM team with some modifications. First, the main Synechocystis sp. PCC 6803 culture used for transformations was a mixture of two cultures that were at OD630=0.8, and not at OD750=0.8 (mid logarithmic growth phase) as suggested in the original protocol. Second, the recommendation was to keep plates at 30°C and 50 µE until slight green film arose (ca. 2-4 days), however, we only had a shaking incubator with a light source at our disposal, so we kept the plates there. Synechocystis transformations failed - there was no green bacterial film on the plates after a week, the plates started drying out and had to be disposed of. Importantly, even if we were able to grow the film of cyanobacteria it would not have been possible for us to select for transformed cells because it requires multiple rounds of addition of antibiotic by increasing its concentration and waiting for colonies to grow, which is a slow process for Synechocystis and can take up to a couple of months, which in the case of iGEM timelines is unfeasible.

Because Synechocystis transformations failed we decided to try and characterise NucA-NuiA kill switch in E. coli. For this purpose, we ordered a G-block encoding the whole Kill Switch cassette with the same NucA and NuiA sequences, but with Synechocystis promoters substituted with corresponding E. coli promoters, zinc-inducible PzntA promoter for antitoxin and constitutive Anderson promoter (BBa_J23114) for toxin. The G-block also contained the required overhangs to allow direct assembly into a LvT acceptor vector. We ran cPCR of level T E. coli KS constructs (Figure 5). Positive control failed because no template DNA was added, but three colonies, 2, 3, and 8, gave the expected amplicon size (1930 bp), so we decided to proceed to Kill Switch characterization with colony 2.

We set up two overnight cultures, one from colony 2 that gave the correct cPCR band size for level T E. coli KS construct, and the other one from a blue colony containing the empty vector (WT+vector). We used these initial cultures to make new cultures in triplicates with or without added 400 µM Zn2+ for both KS and WT+vector control. For KS characterisation, we grew these cultures in a shaking incubator at 37C, 160 rpm, and took OD600 measurements at 1 hour intervals to produce growth curves (Figure 6). From growth curves, cells containing Kill Switch plasmid or empty vector grew better without added zinc, indicating that the addition of 400 μM ZnSO4 slows down cell growth. Importantly, there was no apparent decrease in growth of cells containing the Kill Switch in media without added zinc compared to those grown with added zinc. This could be due to nuclease not being expressed, thus unable to kill the cells. Notably, NucA requires divalent metal ions like Mn2+ or Mg2+ as cofactors, the optimal concentration for these being around 5 mM, while NucA activity was shown to decrease with increasing concentration of monovalent salt. LB media contains a high concentration of NaCl but no divalent salts, meaning that NucA might not function due to lack of cofactors. Alternatively, zinc-inducible promoter might not function as expected, leading to constitutive expression of antitoxin, thus inhibiting nuclease. In future experiments, NucA and NuiA expression should be verified by Western Blotting before characterizing the Kill Switch. Also, LB media should be supplemented with appropriate concentrations of divalent salts to ensure the presence of cofactors for NucA.

For checking the viability of cells containing the KS in conditions with and without added zinc compared to WT+vector control we performed colony forming unit (CFU) counting assay (Figure 7). In general, the experiment was unsuccessful since the numbers of colonies observed after plating 10 µL of culture diluted by the factor of 107 were too low to be able to accurately deduce the CFU number in original undiluted samples. Thus, in the future the experiment should be redone by plating lower dilution or higher volume of cells. Moreover, as noted previously, since NucA requires divalent metal ions like Mn2+ or Mg2+ as cofactors, ideally the plates should be supplemented with them. From our results, at 6 hours, no significant difference in CFU/mL numbers was observed between KS-containing cells grown with or without added zinc (2 sample t-test, p-value > 0.5).

A comprehensive list of the parts designed can be found on our contributions page.

Salt growth curve of BL21 (DE3) cells

To determine the level of growth and salt tolerance in Escherichia coli BL21 (DE3), a growth curve was made, incubating the E. coli in various concentrations of NaCl in LB media. 29ml LB media was inoculated with 1ml overnight culture of E.coli and the initial OD600 was measured. This was then followed by measurements taking 1ml of media every hour for 6 hours. 1% salt has been removed as the initial OD600 was significantly higher than those of the others, which may have affected the results. The results showed a significant decrease in growth after 6 hours of the E.coli BL21 after 3% NaCl (w/v) LB (Figure 1). The doubling time of the E.coli can be seen to steadily increase until 3.5% where the doubling time begins to increase significantly as the concentration of salt in the medium rises (Table 1). The E.coli grown in 0.5% salt (no additional salt), grew at the fastest rate for the first 5 hours with a doubling time of 1.05 hours in comparison to 1.5% salt with a doubling time of 1.09 hours. However, more testing would be necessary to determine if this difference is due to increased salt in the medium. The E. coli grew at 0.5% then saw a decline between hours 5 and 6, which was not mirrored in other cultures at different concentrations. However, a decline before recovery had been seen in other cultures, such as 3.5% in hour 5.

Table 1 - Doubling Times of Salty E.coli Cultures at 6 hours. Doubling times were calculated using the formula Dt = hours/ log2( final OD600/ Initial OD600)

stpA level T

Due to being unsuccessful in extracting purified stpA gene from genomic DNA, a gene block was designed to be inserted into a cyanogate level T vector with correct cut sites and overhangs for an assembly reaction wherein the restriction enzyme BbsI would cut the gene block revealing sticky ends which could then be ligated into the level T vector. A level T reaction was conducted and the reaction product was transformed into TOP10 cells which were then spread onto XGAL-IPTG plates for blue/white selection. White colonies were then picked and colony PCR was conducted to confirm successful transformation. The PCR product was then run on 1% agarose gel and the gel was analysed using a UV imager (Figure 2). The results showed bands of the expected size across the white colonies indicating that the plasmid had been successfully constructed and transformed.

Figure 2: Imaged Agarose gel electrophoresis. Colony PCR was performed generating bands at the expected positions between 1.5 and 2kb.

The ggpS gene was extracted from synechocystis sp. PCC 6803 by PCR from genomic DNA. The gene was then purified by gel extraction and ligated in an assembly reaction into the CyanoGate level 0 backbone. A further digestion and ligation was performed to assemble the gene along with the constitutive promoter ptrc10, which functions in both synechocystis as well as E.coli and the terminator pC0.082. This level 1 assembly was then transformed into E. coli TOP10 cells and confirmed by colony PCR and digestion. However, the final plasmid was found to be around 1kb larger than anticipated (Figure 3). Nevertheless, the correct band sizes indicating that ggpS has been successfully ligated into the level 1 vector and transformed were present. The decision was then made to proceed with error-prone PCR using ggpS in this backbone.

Figure 3: Agarose Gel Electrophoresis image of undigested Level 1 construct and digested with NotI and BbsI. Plasmid miniprep was digested with NotI and bbsI and run on 1% agarose gel in gel electrophoresis and imaged. Repeats can be found on the right.

However, after repeated attempts, this error-prone PCR was unsuccessful as when run with dye on 1% Agarose gel and imaged, the bands indicated two PCR products, one at 4.5kb and one at 2.5kb consistently. In silico there was considered no reason to believe that primer mis-annealing should take place and so it was hypothesised that either the plCH47732 level 1 acceptor vector, the PTRC10 promoter, or pC0.082 terminator in level 0 was incorrect. A PCR reaction was performed using the level 0, level 1, and genomic PCR extract as template DNA and all templates except the level 1 were successful (Figure 4).

To circumvent the plCH47732 backbone, epPCR of ggpS was performed again using the level 0 assembly as a template and NEB Taq polymerase. The genomic Forward and Genomic Reverse primers were used. These primers when digested, would generate the overhangs necessary to assemble into a pJUMP29-1A acceptor vector with a constitutive PJ23100 promoter, pET ribosome binding site, and L3SAP51 terminator. The resulting library was digested with BbsI, creating 1.5kb double-stranded linear DNA fragments. These digested fragments were then used in a level 1 JUMP assembly reaction to generate level 1 constructs with a GFP selection marker. Fluorescent colonies were picked for colony PCR to check for successful assembly and transformation. However, the first attempt was unsuccessful.

After a subsequent attempt utilising longer extension times, the colony PCR indicated successful assembly and transformation of the epPCR library (Figure 5). The bands were at approximately 1.8kb as expected in colonies 1, 2, and 4. These results were not demonstrated in our final report. The error-prone PCR library was then transferred to the Edinburgh Genome Foundry, along with chemically competent cells for heat shock containing the stpA level T plasmid for transformation and subsequent performance of the assay.

Figure 4: Agarose gel electrophoresis image: Q5 PCR of ggpS from ggpS, lvl0 and lvl1P1 templates, and Q5 of pICH47732 backbone from lvl1 template.

Sanger sequencing of the digested ggpS epPCR library showed that the consensus sequence from the sequencing matched the expected sequence from in-silico testing.

Figure 5: UV Imaged Agarose Gel electrophoresis. 4 Colonies were picked for colony PCR using P1 and P2 primers. Results show colony 1,2 and 4 with the expected band size.