We assemble some of our wet lab results from the Registry onto this page. Please note on our Inclusivity, Entrepreneurship, Education and Hardware pages, we documented interesting results from those activities.
Testing nickel enrichment module
Our nickel enrichment module is to maximize nickel bioaccumulation. For this, we have designed two approaches: enhancing nickel uptake and improving nickel retention.
Nickel Uptake: nikABCDE & NixA
To enable our Escherichia coli to efficiently absorb nickel ions, we focused on active nickel uptake, which involves a nickel transport protein known as nikABCDE (BBa_K4765129, and our ribozyme-connected version BBa_K5115082) and NixA (BBa_K5115071). We also created dimerizable versions for NixA: NixA-F1v (BBa_K5115086) and F1v-NixA (BBa_K5115087).
To measure the nickel uptake efficiency of our modified E. coli, we analyzed the nickel concentration in the bacterial supernatant after constructing a standard curve. A lower nickel concentration in the supernatant indicates greater absorption by E. coli, allowing us to calculate the percentage of nickel uptake during the culture.
The graph shows the percentage of Ni²⁺ absorbed by E. coli expressing different constructs after 5 hours of growth in a medium containing 30 mg/L Ni²⁺ (E. coli strain: BL21 DE3, induced with 1 mM IPTG). Ni²⁺ uptake was calculated based on the difference between initial and final concentrations in the supernatant, divided by 30 mg/L. The optical density (OD₆₀₀) of the initial bacterial suspension was adjusted to 0.5. Culture for 5 hours, at 37°C with a rotating speed at 220 rpm. Regarding NixA-F1v and F1v-NixA, AP20187 is a synthetic dimerizer that can be used to induce homodimerization of F1v domain. Three biological replicates were performed for each condition, and error bars represent the standard errors of the means (SEM) of these replicates. ANOVA test shows that all constructs increase Ni²⁺ uptake significantly compared to the control. Bacteria expressing NixA-F1v exhibit the highest Ni²⁺ uptake efficiency (p = 0.0052, Dunnett's post-test).
As shown in Figure 1, NixA generally outperforms nikABCDE. The ribozyme-connected version of nik demonstrates higher nickel uptake efficiency than the original nik operon. The dimerable versions of NixA exhibit varying nickel uptake efficiencies: F1v-NixA shows lower ability than NixA, while NixA-F1v surpasses NixA, making it the strongest nickel uptake protein among those tested. This difference may be attributed to the distinct structures of these two dimerable versions of NixA.
Orientations of NixA, F1v-NixA, and NixA-F1v are aligned. Plasma membrane is shown.
To investigate the difference between F1v-NixA and NixA-F1v, we used AlphaFold 3[1] to predict their structures. According to the computated models (Figure 2), the N-terminus of NixA is in the periplasm, while the C-terminus lies within the cytoplasm. Plasma membrane location marked based on an existing study[2]. F1v is on the periplasmic side when fused to the N-terminus of NixA and on the cytoplasmic side when fused to the C-terminus. The nickel uptake capability is higher when F1v is fused to the C-terminus (on the cytoplasmic side), possibly indicating a better dimerization effect, or the N-terminal F1v disturbs NixA's normal function.
We also conducted nickel absorption tests with different starting Ni²⁺ concentrations (20 mg/L and 50 mg/L), leading to the same result.
Nickel Retainment by RcnRC35L
Wild-type E. coli expresses RcnA, especially at high Ni²⁺ concentrations, to efflux nickel as a protective mechanism. Overexpressing a mutant RcnRC35L (BBa_K5115000) could limit the induction of rcnA, thereby preventing E. coli from effluxing nickel.
The part BBa_K5115000 has RcnR Cys residue at position 35 changed to Leu. This modification prevents RcnR in response to high nickel levels, thus limiting RcnA induction and ensuring that absorbed nickel remains within our engineered E. coli.
We combined RcnRC35L with nikABCDE and test their Ni²⁺ uptake capability.
The graph shows the percentage of Ni²⁺ absorbed by E. coli expressing different constructs after 5 hours of growth in a medium containing 20 mg/L Ni²⁺ (E. coli strain: BL21 DE3, induced with 1 mM IPTG). Ni²⁺ uptake was calculated based on the difference between initial and final concentrations in the supernatant, divided by 20 mg/L. The optical density (OD₆₀₀) of the initial bacterial suspension was adjusted to 0.5. Culture for 5 hours, at 37°C with a rotating speed at 220 rpm. Three biological replicates were performed for each condition, and error bars represent the standard errors of the means (SEM) of these replicates. RcnRC35L refers to a mutation in which cysteine (C) at position 35 in the RcnR protein was substituted with leucine (L). The results indicate that E. coli expressing RcnRC35L consistently has higher Ni²⁺ uptake efficiency compared to E. coli without RcnRC35L expression.
As shown in Figure 3, RcnRC35L improved nickel absorption in E. coli, both with the nik-ribozyme and NixA. This result validates our design that by inhibiting nickel efflux could effectively trap Ni²⁺ within the cells.
For more details, please checkout our nickel module BBa_K5115068.
Testing survival modules
By employing survival modules, engineered E. coli could be better prepared to withstand environmental pressures while effectively absorbing and reducing nickel. The challenges E. coli may face in practical applications include high concentrations of heavey metal ions, freezing conditions, threats from phages, etc. Two new survial modules were developed this year: enhancing heavy metal tolerance and providing anti-phage protection.
Heavy Metal Tolerance by Hpn
Heavy metal ions are cytotoxic, so while we aim for E. coli to absorb nickel ions, we must also ensure its safe growth by achieving heavy metal tolerance. To accomplish this, we chose to express a nickel-binding protein, Hpn (BBa_K5115036), which is capable of storing Ni2+ ions.
The graph illustrates the effect of Ni²⁺ on the growth of E. coli expressing Hpn compared to E. coli without Hpn expression in a medium containing 50 mg/L Ni²⁺ (E. coli strain: BL21 DE3, induced with 1 mM IPTG). The optical density (OD₆₀₀) of the initial bacterial suspension was adjusted to 0.5. E. coli growth was measured by OD₆₀₀, and the bacterial counts were calculated using a standard conversion, where OD₆₀₀ = 1 corresponds to 5.39 × 10⁸ cells. The results indicate that E. coli expressing Hpn has greater tolerance to Ni²⁺, exhibiting higher growth rates than E. coli without Hpn expression under the same conditions.
To evaluate the growth of our modified E. coli in a nickel-rich environment, we generated growth curves by measuring OD600. As shown in Figure 4, E. coli expressing Hpn consistently outperformed the controls in a medium with a nickel concentration of 50 mg/L. This demonstrates that Hpn effectively protects E. coli in high nickel environments. We also repeated the experiments at 20 mg/L and 100 mg/L nickel concentrations and obtained similar results.
Our primary goal in expressing Hpn is to alleviate the survival pressure on E. coli in high nickel environments, allowing it to absorb nickel more efficiently. To assess this, we measured nickel uptake in E. coli with and without Hpn, enabling us to compare the differences in nickel absorption.
The graph shows the percentage of Ni²⁺ absorbed by E. coli expressing different constructs after 5 hours of growth in a medium containing 50 mg/L Ni²⁺ (E. coli strain: BL21 DE3, induced with 1 mM IPTG). Ni²⁺ uptake was calculated based on the difference between initial and final concentrations in the supernatant, divided by 50 mg/L. The optical density (OD₆₀₀) of the initial bacterial suspension was adjusted to 0.5. Culture for 5 hours, at 37°C with a rotating speed at 220 rpm. Our best nickel uptaker NixA-F1v was used, and AP20187 was added to induce NixA dimerization.
As shown in Figure 5, the nickel absorption ability of the Hpn + NixA-F1v bacteria was significantly higher than that of NixA-F1v alone, achieving 43% nickel uptake. This indicates that Hpn effectively relieves the survival pressure on E. coli and enhances its nickel uptake capacity.
Anti-phage by YejM
Phages in the environment pose a significant threat to bacterial survival, and bacterial lipopolysaccharides (LPS) can act as a physical barrier to prevent phage infection. YejM (BBa_K5115070) inhibits LpxC degradation, which is the enzyme critical for LPS production.
We infected E. coli TG1 with kanamycin-resistant M13KO7 phage and plated a 200-fold dilution on kanamycin plates. This allowed us to assess the phage infection efficiency by counting the number of colonies.
E. coli TG1 carrying either stayGold fluorescent protein and YejM under the J23107 promoter were infected with M13KO7 phage at different MOIs. Colonies were counted after incubating at 37°C for 16 hours, on selection LB plates. "*" indicates a p-value less than 0.01. Under any MOI condition, the colony count of E. coli TG1 carrying YejM was significantly lower than that of stayGold, indicating that YejM expression confers resistance to phage infection.
According to Figure 6, it's clear that TG1-YejM significantly exhibits better growth than TG1-stayGold. YejM enables E. coli to thrive in the presence of phages, demonstrating its effectiveness in providing anti-phage capability.
Testing nickel microparticle module
Our nickel microparticle module is designed to efficiently absorb nickel ions and convert them into less toxic microparticles. The microparticles consist of hydrogenases, a carboxysome shell, and reduced nickel, with diameters around 50-80 nm. We have constructed two composite parts: the U module (BBa_K5115066) and the F module (BBa_K5115067).
The F module includes BBa_K5115063 (hox and hyp, with EP targeted to hoxF) and BBa_K5115060 (ribozyme + RBS + cso without csoS3 + stem-loop). We expected this design to accumulate hydrogenase with the carboxysome, facilitating a stable environment for nickel reduction.
The U module functions similarly to the F module, with the main difference being that, in the U module, the EP is fused with hoxU instead of hoxF. The hoxF subunit is essential for electron transport, while hoxU subunit conducts electrons between hoxH and hoxF. This design allows us to compare the effectiveness of different EP subunit fusions.
Carboxysome linkage: EP sequence
To create an efficient reduction reaction vessel, we linked hydrogenase with the carboxysome-targeting sequence to build a more focused and effective reaction space. To test the feasibility of this approach, we first fused Staygold with the EP (BBa_K5115002) sequence to validate the concept.
Images were captured using spinning disk confocal with a 150x objective lens, as described on our Experiments page. Bacteria in A-C only express stayGold fused with EP BBa_K5115057, while bacteria in D simultaneously express BBa_K5115057 and BBa_K5115065. 1 mM IPTG was added to A,B only. Images without scale bar are 5x5 µm square, unless specifically indicated below.
(A) The entire image field is shown (41.27x41.27 µm square), with brightfield image on the left, and green fluorescence image on the right.
(B) Four regions in (A) are enlarged, showing uniform distribution of green fluorescence.
(C) Although no IPTG was added, leaky expression from the promoter is sufficient to fill bacteria with green.
(D) With all carboxysome components expressed, stayGold fused with EP concentrated to the carboxysome. Leaky expression from the promoter is sufficient to drive 1 or 2 carboxysome formed within each bacteria.
In Figure 7, in (A)-(C), we can observe that due to the lack of carboxysome expression, Staygold-EP is evenly distributed throughout the cell. However, in (D), Staygold-EP is directed into the carboxysome. This confirms the feasibility of our approach to assemble hydrogenase into the carboxysome.
Nickel Reduction by F Module
The F module (BBa K5115067), composed of hydrogenase and carboxysome, reduces nickel ions into nickel microparticles using hydrogen as an electron donor. We cultured the F module E. coli in LB medium containing 100 mg/L nickel ions, inside a seal bottle, after bubbling with 10x volumes of 5.6% hydrogen gas.
To visualize nickel microparticles in F module expressing bacteria, osmium tetroxide and uranyl acetate were used to negative stain the fixed samples, and further examined by TEM. Osmium tetroxide binds to lipid membranes to enhance contrast, while uranyl acetate binds to nucleic acids and proteins, to improve the visibility of cellular structures. Nickel microparticles are visible due to their higher electron density.
Osmium tetroxide and uranyl acetate were used for the staining A-E. Scale bar shown on the image. (A) Overview of E. coli cells. (B) Sections of bacteria, filled with carboxysome-sized regions (CSR) surrounded by electron dense dots. In one cell, all visual CSR are circled by yellow dash lines. (C) The size of CSR are various, with two examples circled. (D,E) For cells with less electron dense dots, CSR are clear, with the cell in C fully packed with CSR and the cell in D sparsely packed. (F) No uranyl acetate staining. The image confirms that the electron dense dots throughout A-E are not salt crystals but actual metallic particles, which we believe are Ni particles. Three 80-nm square regions are enlarged, showing polyhedral outline of CSR, with Ni particles surrounded.
As shown in Figure 8, the F module converts nickel ion into particles inside bacteria.
Nickel Uptake: F module vs. U module
If the intracellular nickel ion is successfully reduced, would bacteria absorb more Ni2+? To determine which module better facilitates the reduction of nickel to produce microparticles, we measured the nickel absorption efficiency of the U module and F module.
The graph shows the percentage of Ni²⁺ absorbed by E. coli expressing indicated modules (E. coli strain: BL21 DE3). Ni²⁺ uptake was calculated based on the difference between initial and final concentrations in the supernatant, divided by 100 mg/L. The single bacteria colony was picked and grown overnight to reach optical density (OD₆₀₀) > 1. Prepare a sealed 25-mL LB culture in a 250-mL bottle, with: 100 µL overnight bacteria liquid culture, 25 µg/mL Kan, 1 mM methyl viologen dichloride, 100 mg/L NiCl2, bubbled with ~250 mL 5.6% hydrogen gas (slowly, with hand-shaking, about 5 minutes). Culture for 30 hours, at 37°C with a rotating speed at 220 rpm. Four biological replicates were performed for each condition, and error bars represent the standard errors of the means (SEM) of these replicates. Plain BL21 DE3 was used as control. None of the four bacteria with U module was able to grow during overnight culture if induced with 1 mM IPTG, only 1 F module grow, which was further examined by TEM. Additional 1 mM IPTG was added into the 25-mL culture of "F induced". ANOVA test shows that all constructs increase Ni²⁺ uptake significantly compared to the control (U module, p = 0.0045; F module, p < 0.0001). P value was calculated using Dunnett's post-test.
According to Figure 9, we conclude that the nickel absorption capacity of the F module is superior to that of the U module, especially after induction. Despite not having an engineered nickel uptaker (nik or NixA), the F module still achieves about 10% nickel absorption. This indicates that a significant amount of Ni2+ in the cells is reduced by hydrogenase, resulting in a relatively low intracellular Ni2+ concentration, which activates E. coli to actively absorb nickel from the environment.
Testing hydrogen supply module
Given that the production of hydrogen gas by cyanobacteria is well-established in the literature[3], we did not consider it is necessary to test hydrogen production in our study. Instead, we focused on confirming the adhesion between E. coli and cyanobacteria to ensure a consistent and sufficient supply of hydrogen.
E. coli-cyanobacteria Interaction
In February 2024, under the guidance of the Fudan 2023 team members, we repeated last year's E. coli / S. elongatus Aggregation Assay to ensure that E. coli could adhere to S. elongatus. For more details, please see BBa_K4765110.
To examine the adhesion between E. coli and cyanobacteria, the optical density (OD600 for E. coli and OD685 for S. elongatus) of the supernatant was measured at 0, 2, 6, and 24 hours after mixing the two organisms. A decrease in OD over time indicates that cells have aggregated and settled out of the suspension, reflecting the adhesion efficiency between the two species; only the 6-hour data is shown in Figure 10.
After aTc induction, E. coli bacteria express intimin-LCA, which specifically binds with Synechococcus elongatus. We used OD685 to plot the ratio of the remaining S. elongatus in the supernatant to the initial concentration after 6 hours of cultivation. The aTc-induced sample show a significant reduction in the bacteria remaining in the supernatant compared to the uninduced sample, indicating enhanced aggregation between E. coli and S. elongatus. "**" indicates statistical significance less than 0.01.
Figure 10 shows a significant reduction in the remaining S. elongatus in the supernatant of samples where E. coli was induced to express intimin-LCA, demonstrating the adsorption and co-precipitation of E. coli and cyanobacteria. This allows E. coli to utilize the hydrogen produced by cyanobacteria in close proximity.
What we have learned?
After testing our nickel enrichment module, we discovered that:
- NixA-F1v is our best nickel uptake protein, outperforming nik-operon, nik-ribozyme, NixA, and F1v-NixA (Figure 1)
- RcnRC35L improves nickel absorption by inhibiting nickel efflux (Figure 3)
After testing our survival modules, we found out that:
- Hpn effectively protects E. coli in high nickel environments, therefore improving nickel uptake (Figure 4, Figure 5)
- YejM enables E. coli to thrive in the presence of phages (Figure 6)
After testing our nickel microparticle module, we proved that:
- The absorbed nickel was successfully converted into nickel microparticles in the bacteria expressing F module (Figure 7)
- Surprisingly, without a engineered nickel uptaker, the nickel absorption capacity of the F module expressing bacteria is significant comparing to the control, and is superior to that of the U module (Figure 8)
After testing our hydrogen supply module, we made sure that:
- The adhesion between E. coli and cyanobacteria is possible, thus stable hydrogen supplement (Figure 9)
These results are just examples of our wet lab achievements. Other accomplishments are available on our Measurement, Software, Hardware, and Safety pages.
Abramson, J., Adler, J., Dunger, J., Evans, R., Green, T., Pritzel, A., Ronneberger, O., Willmore, L., Ballard, A. J., Bambrick, J., Bodenstein, S. W., Evans, D. A., Hung, C.-C., O'Neill, M., Reiman, D., Tunyasuvunakool, K., Wu, Z., Žemgulytė, A., Arvaniti, E., … Jumper, J. M. (2024). Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature, 630(8016), 493–500. https://doi.org/10.1038/s41586-024-07487-w ↩︎
Hernandez, J. A., Micus, P. S., Sunga, S. A. L., Mazzei, L., Ciurli, S., & Meloni, G. (2024). Metal selectivity and translocation mechanism characterization in proteoliposomes of the transmembrane NiCoT transporter NixA from Helicobacter pylori. Chemical Science, 15(2), 651–665. https://doi.org/10.1039/D3SC05135H ↩︎
Dutta, D., De, D., Chaudhuri, S., & Bhattacharya, S. K. (2005). Hydrogen production by Cyanobacteria. Microbial Cell Factories, 4(1), 36. https://doi.org/10.1186/1475-2859-4-36 ↩︎