Engineering

Engineering Overview

Welcome to the Engineering page of iGEM Athens 2024: e-PHAESTUS. Our journey to engineering success followed the foundational Design-Build-Test-Learn cycle. This approach helped us to improve our ideas and successfully get over the challenges we encountered.

Throughout the development and experimentation of our project , we applied these cycles to four key aspects of our project: Conceptualization, Gene Cloning Experiments, Protein Expression, and Chemical Engineering. We welcomed both success and failure as chances for development at every turn. This mentality was crucial to making breakthroughs and accomplishing our goals.

Cycle 1: Conceptualization

Design

The three months of constant research in e-waste management and bioleaching resulted in our final gene selection : HCN Synthase. More specifically, we are intrigued by the HCNabc operon that was found in pseudomonas putida and pseudomonas fluorescens. This operon enables the production of the valuable HCN from cheap amino acids like glycine and threonine. HCN is crucial because CN^- ions are the most traditional leachate in gold dissolution, yielding very high recovery rates. [1],[2],[3]

Further literature review indicated that glycine is the optimal amino acid and that a pH level of 9 would minimize the emission of the highly toxic HCN, while enhancing CN^- activity. 10 T o recover the metals, we consulted professors from the Chemistry Department specializing in Inorganic and Polymer Chemistry. Dr. Efthimiadou, Methenitis, and Pitsikalis guided us in synthesizing an amphiphilic copolymer. The first part is made from polystyrene (PS) ensures that the polymer is insoluble, while the second hydrophilic part from polyacrylic acid (PAA) is responsible for binding metallic ions. This membrane can interact with and bind metals depending on the solution's pH. Specifically, at pH < 3, the hydrophilic group exists in a molecular form (OH), whereas at pH > 9, the ionized group interacts as shown below. [1],[2],[3]

Overall, the PCBs (printed circuit boards) would be placed in a culture of transformed bacteria during their exponential phase. The pH of the growth medium would be regulated at 9, and the metals would be left for one week. Afterward, the solution would be filtered, retaining the metal residues and bacteria in the original container. The filtrate would contain soluble Au-CN complexes. T o recover the metals, the filtrate would pass into a vessel containing the aforementioned copolymer. Subsequently, the copolymer would be acidified in a new vessel, releasing the desired metallic ions.

Build

As the overall design of the project was over, our team began the “building” process. We started experimenting with the genes and we are one step before the purchase of the DNA from IDT.

Given the complexity of the overall gene design and the fact that this set of experiments did not actually proceed, we will only showcase the basic principles. The hcnABC operon contains three genes, hcnA which codes for a formate dehydrogenase and hcnB and hcnC which code for amino acid oxidases. As said before, these genes are part of the same regulatory unit, something we hoped to employ. In the paper, “Characterization of the hcnABC Gene Cluster Encoding Hydrogen Cyanide Synthase and Anaerobic Regulation by ANR in the Strictly Aerobic Biocontrol Agent Pseudomonas fluorescens CHA0” , the authors present the sequence of the hcnABC operon as can be seen in the picture bellow: [4]

In addition, they showcase several systems through which we could achieve expression of the above proteins, into a complete HCN-producing system, with the following one showing important potential: [4]

This particular system showcased good expression of all three proteins and was what we sought to replicate, based on the DNA sequence above. While several steps were undertaken to prepare the DNA sequence for a potential order from IDT , this plan did not materialize given the considerations showcased in the test part of this cycle.

Test

Rather than conducting immediate experiments, we rigorously evaluated our HCN Synthase prospect's theoretical and safety potential. Potential investors analyzed it first, followed by scientific committees from the Chemical and Biological Department of the National and Kapodistrian University of Athens. However, the most important test was the iGEM competition's safety form, which revealed serious dangers related to our proposed experiments.

When integrating the theoretical design into a lab-ready project, it was apparent that multiple safety and technical difficulties would arise:

Risk Level: Our lab lacked the necessary equipment, such as an appropriate modern fume hood, to safely conduct HCN synthesis.
Economic Viability: Competing with established HCN production technologies created substantial cost obstacles. Traditional pyrometallurgy and hydrometallurgy techniques were more cost-effective.
Limited Synthetic Biology Component: The project relied heavily on chemical engineering. Therefore, the synthetic biology aspect was relatively underdeveloped.
Toxic By-products: The by-products from synthetic HCN synthesis were as hazardous as those from traditional NaCN leaching methods.
Complex Stability: The stability of the Au(CN)₂^- complexes was much higher than anticipated, making dissolution difficult and expensive.
Membrane Protein Challenges: The purification process was complicated since the project included membranous proteins.

As Dr. Labrou thoroughly reviewed the whole project and it was ultimately discontinued Upon thorough review by Dr. Labrou, the project was ultimately discontinued due to these constraints.

Learn

Although we didn't proceed with actual experimentation, this stage gave us invaluable knowledge and guided the future of our project. Here's what we learned:

Upcycling Over Recycling: Significant financial barriers were incurred when competing with well-established leaching technologies. Conventional methods of hydrometallurgy and pyrometallurgy were more economical.
Prioritizing Green Chemistry: Safety and sustainability became the main point of our approach. A project in order to be characterized as successful should use safe, green leachates that avoid toxic by-products. This is the way to ensure not only compliance with safety regulations but also strengthen the environmental aspect of the project.
Chemical Viability and Balance: We learned that the chemical processes behind metal complex formation must be both kinetically and thermodynamically viable. However, overly stable complexes can impede the dissolution of metals and burden the project financially.
Scalability and Enzyme Purification: To ensure practical industry application, we recognized the importance of enzyme immobilization. This method simplifies purification, improves scalability and aligns with the legislation of the EU making it easier to integrate our solution into industrial settings.

These crucial realizations enabled us to improve our idea and switch to glutathione, a more bio-friendly and viable solution in the long run.

Cycle 2: Gene Cloning Experiments

After the selection of the appropriate genes, namely GshFss and PPKT e, alongside the plasmid vector pET28a(+), the system needed to be designed to be fit for purpose. The vector pET28a was selected as it featured: 1) a 6xHis-tag which could be used for the isolation of the genes through affinity chromatography and 2) a T7-promoter which ensured a controlled but also high expression of the protein. In the picture below these two regions are highlighted on the pET28a(+) plasmid:

We sought to replicate successful experiments done by Zhang, X. et al (1) and so we chose to use the restriction enzymes BamHI and HindIII for the cloning of GshFss and EcoRI and SalI for the cloning of PPKT e. As can be seen in the picture above, the restriction sites for these enzymes are between the T7 promoter and the 6xHis-tag.

Designing the genes also posed some challenges:

The GshFss gene was sourced from the paper of Zhang, X. et al (1) (supplementary material) and was already optimized for E.coli expression. Therefore, after adding the restriction sites at the end of the gene, it was noted that the gene that resulted violated both the Biobrick standard and the Type IIS standard as it contained “illegal” restriction sites. In order to cause the least silent mutations on the gene, it was selected to make the gene compatible with the Type IIS standard, which required several silent mutations (please refer to part BBa_K5406000):

i) A SapI site was removed by changing A at position 1065 to a G

ii) A BsaI site was removed by changing G at position 891 to an A

The PPKT e gene was sourced from the NCBI database. It was not optimized for E.coli expression and therefore had to be optimized, which was done using Snapgene(R) software. After adding the restriction sites at the end of the gene, it was noted that the gene that resulted violated the Biobrick standard as it contained “illegal” restriction sites. As EcoRI was to be used for cloning, the gene had to be compatible with the Type IIS standard. When it was time to order the gene from Integrated DNA T echnologies, our sequence was too problematic to be synthesized by IDT . Therefore, using IDT software, the gene was once again optimized to be compatible with IDT synthesis procedures. This optimization produced an internal EcoRI site which was promptly removed (change of T at position 911 to an A, please refer to part BBa_K5406001).

The following pictures showcase the final position of the GshFss and PPKT e genes respectively in the final plasmid:

Build

To acquire the recombinant plasmids, we performed double digestion on both of the genes. The desired digested products were visualized through loading on an agarose gel (outlined on Figure 4), and then extracted using a “Gel Extraction kit” . Through electrophoresis of the extracted product, clear bands were seen at ~2250 bps, as expected for the genes and at ~5300 bps for the plasmid.

Test

In order to test if the cloning was successful, we proceeded to chemical transformation of DH5a cells with the ligation product. A positive (pET28a plasmid) and negative (only cells) transformation control were added. Some results of our attempts were the following:

Given the above results, we were hopeful that we had successfully cloned our genes into pET28a. But for that to be ensured, we needed to isolate the recombinant plasmids and visualize them through agarose gel electrophoresis. This was done with a plasmid isolation kit and results were the following:

Therefore, the bands serve as a clear indication of having successfully created the recombinant plasmid.

Learn

Given the above, we showcase that we have managed to create the recombinant plasmids, something that is further verified through the expression experiments (see further engineering cycles). These experiments have showcased the following: 1) That a 1:1 molar vector:insert ratio was enough to result in successful recombination for our system, and not a 2:1 or 3:1 as it has been suggested, 2) The pET28a(+) plasmid is an appropriate expression vector for genes of about 2200 bp. For a substantial confirmation of our molecular cloning experiments success, an examination of the plasmid DNA and gene constructs is yet to be completed, something we aim for in our future plans.

Cycle 3: Protein Expression and Isolation

This section is dedicated to the cycle of experimental procedures aiming at the heterologous overexpression of the enzymatic system of GshF/PPK through bacterial systems.

Design

As it was mentioned before, we opted for the enzymatic biosynthesis in vitro for the production of GSH, since it offers a multitude of advantages, adapting well to our project’s effort. Therefore, it was required to produce the recombinant proteins whose gene constructs were previously designed, and then isolate both enzymes through protein purification.

Having concluded on the enzymes to be studied, namely GshFSS (Streptococcus sanguinis SK36) and PPKTe (Thermosynechococcus vestitus (nee. elongatus) BP-1), sequences encoding the respective enzymes were inserted in suggested plasmid DNA vectors, pET28a(+). Besides the expression cells transformed with pET28a-GshF and pET28a-PPK recombinant vectors, we also made use of Control samples, transformed with an empty plasmid vector (pET-28b), in order to investigate the native expression of the proteins GshF and PPK in E. coli strains. It is also important to highlight that so far there is no indication of a systemic impact of the heterologous expression of GshF in E. coli bacterial cells.

Build

Our goal was to transform each of the two target - genes, gshf and ppk, in different competent expression cell strains of Escherichia coli, mainly characterized as DE3 according to research papers of Zhang et. al (2017)^[4] among others. Based on the availability of biological materials in our laboratory, we opted for the use of SHuffle T7, Rosetta2/PlysS (DE3) and later with BL21/PlysS (DE3), in order to achieve and further optimize protein overexpression.

Test

During the first experimental trial of protein overexpression (Weeks 05/08 - 09/08), the transformation of E. coli Rosetta2/PlysS with either the pET28a-GshF or pET28a-PPK plasmids resulted in no observable cell growth. This lack of growth was attributed to potential toxicity effects caused by the expression of these recombinant proteins, as control cultures transformed with the empty pET28b vector demonstrated healthy growth during the same incubation period. In contrast, E. coli SHuffle T7 cells that were transformed with the same plasmids exhibited significant growth. Consequently, these SHuffle T7 cultures were selected for further development into liquid bacterial cell cultures aimed at induced protein overexpression. A portion of these successful cultures was preserved at -80°C in sterilized YT liquid culture medium supplemented with glycerol to ensure viability for future experiments.

In order to evaluate the results at the end of our first protein expression trial, samples from both the supernatant and cell pellet collected were subjected to Bradford protein quantification as well as SDS-PAGE Gel Electrophoresis.^[9]

In the second trial (Weeks 09/09 - 13/09), both gshF and ppk genes were again transformed into the SHuffle T7 strain, as well as into BL21/PlysS(DE3) cells. However, no growth was observed in the BL21/PlysS(DE3) cultures transformed with the recombinant plasmids, which echoed the previous findings from the Rosetta2/PlysS strain. This lack of growth was similarly attributed to toxicity effects on the culture. In contrast, the SHuffle T7 strain again demonstrated significant growth with both recombinant plasmids.

We then decided to conduct a liquid bacterial cell culture in a larger volume (Weeks 09/09 - 13/09) to obtain a substantial quantity of protein samples heterologously expressed. The culture was grown in YT medium using IPTG as well, a common practice for the induction of protein expression in in vitro bacterial systems. Given the aforementioned trials, competent E. coli SHuffle T7 cells previously transformed with pET28a-GshFSs and pET28a-PPKT e (Weeks 05/08 - 09/08), in order to proceed with isolating the enzymes through Ni-NTA affinity chromatography. In order to evaluate the conclusive results, samples from the supernatant of SHuffle T7 competent expression cells collected were subjected to Bradford protein quantification, using both crude enzyme samples as well as the samples after the purification.

Protein electrophoresis results were assessed, by detecting the corresponding zones of each sample before and after the purification. More specifically, significant overexpression was noted especially for GshFSs, as a zone of high intensity at 85 kDa, compared to the crude sample of the expression cells protein content. In the case of PPKT e, overexpression was achieved and purification was successful, detecting a distinct zone at approximately 85 kDa based on the protein marker, however it could have been improved compared to the GshFSs samples success.

Learn

The results of both studies highlighted the importance of selecting the appropriate competent E. coli strain for protein expression. Cell growth was not observed in either Rosetta2/PlysS or BL21/PlysS, suggesting that these strains may not be capable of supporting the overexpression and protein folding modifications of GshF and PPK. In contrast, the SHuffle T7 strain yielded excellent results, with considerable growth observed in both experiments. This implies that SHuffle T7 competent cells were more compatible with the expression of the GshF and PPK respective genes, presumably because of its capacity of facilitating the appropriate folding as well as post-translational modifications of proteins that contain disulfide bonds, which is a frequently occurring issue in E. coli systems.

Furthermore, the results of SDS-PAGE electrophoresis of enzyme elution samples before and after purification, compared to the crude samples, demonstrated that both enzymes were satisfyingly isolated. We received this as an indication of a successful first effort in the protein purification through affinity chromatography to be successful.

The characteristics of PPK and more importantly, the association of this enzyme with cell membranes is an area we find worth exploring further. While reviewing research papers, we found that members of the polyphosphate kinase family are partially insoluble and associated with the outer cell membrane^{[4] [10]}. More specifically, the PPK enzyme exhibits a soluble behavior, despite 90% or more of the enzyme, even when overproduced, being firmly bound in the membrane fraction. After purification, the PPK protein is further associated with the outer cell membrane. Nevertheless, according to researchers, the in vitro activity of the enzyme seems to remain unaffected by its contact with the membrane.^[10]

We speculated whether a cell-free expression system would have been a more suited approach towards the efficient production and isolation of PPK, besides heterologous overexpression through bacterial systems. Thus, our immediate future plans involve the use of a cell - free expression system, in order to compare, evaluate and amplify protein expression in vitro, specifically in the case of PPKT e, needed for the ATP regeneration in our multi-enzyme system.

Cycle 4: Chemical Engineering

In the chemical engineering aspect of the project our team wanted to test the results of the leaching between glutathione and metals since there is not any literature available.

Design

In designing the leaching experiments, modeling and computational chemistry were essential tools. For optimal results, Computational chemistry was utilized to establish the initial conditions, particularly stabilizing the pH at 11 for optimal results. This valuable step ensured that the reaction conditions were suited for effective metal solubilization.

To maximize the yield and performance of the glutathione leaching process, sodium thiosulfate was used. It plays a critical role in enhancing the solubilization of metals like gold (Au) during the initial phase of the leaching process. Later, sodium thiosulfate is replaced by glutathione, which takes over to further increase efficiency and support the formation of glutathione-coated gold nanoparticles—an essential component for the success of the project. Modeling studies demonstrated that glutathione-metal complexes are significantly more favorable than thiosulfate-metal complexes. This preference indicates that glutathione not only enhances metal recovery but also improves the stability of the resulting complexes, making it a superior choice for our leaching strategy.

Interaction between glutathione and gold, picture taken from Gaussian

Based on the research done by Konado et al. we decided to introduce common and cost-effective reducing agent, sodium sulfite (Na₂SO₃).This is crucial for preventing the oxidation of glutathione and the formation of disulfide bonds between glutathione molecules, ensuring the stability and continuity of the leaching process. ^[11]

Sodium Sulfite anion, picture taken from Gaussian

Build

For the leaching experiments aimed at extracting copper and silver, the following conditions and starting quantities are outlined:

Experimental Conditions:

Temperature: 37 °C
Pressure: 1 atm
Shaking Speed: 150 rpm
pH:Regulated at 11 with an appropriate solution.

Copper

Solution	Total Volume (mL)	Concentration of Cu (mg/mL)	Concentration of GSH (mg/mL)	Concentration of Na₂SO₃ (mg/mL)	Concentration of Na₂S₂O₃ (mg/mL)
A	5	4	1	0	0
B	5	4	1	2	0
C	5	4	1	2	1

Silver

Solution	Total Volume (mL)	Concentration of Ag (mg/mL)	Concentration of GSH (mg/mL)	Concentration of Na₂SO₃ (mg/mL)	Concentration of Na₂S₂O₃ (mg/mL)
D	5	4	1	0	0
E	5	4	1	2	0
F	5	4	1	2	1

Test

Investigation of Solutions: After the first 48 hours, the solubility and condition of the solutions should be visually assessed.

End of Leaching Period: The leaching should be stopped after 168 hours to quantify the solubility yield.

Quantification of Solubility Yield: The quantification process will be conducted under the guidance of Dr. Karavoltsos, a professor of Environmental Chemistry at the National and Kapodistrian University of Athens with the use of ICP-MS T echnique (Inductively Coupled Plasma Mass Spectrometry)

Due to the limited time and the delay of shipment of the metals, we did not have the time to effectively test our system. The metals arrived 5 days before the wiki freeze, and despite planning all the experiments, we were not able to gain results. However, supposing we had enough time, we would have constantly changed the concentrations of all above in order to find the optimized system. Modeling helped us to find the best conditions, but pH=7 should be tested as well.

Learn

In this segment we would include the finalized and optimized concentrations and conditions for the leaching experiments. The knowledge after these tests would be valuable since the optimized system would be transferred in the final system that will contain stabilized enzymes.

Engineering Cycle Conclusion

In conclusion, our project successfully investigated the enzymatic synthesis of glutathione (GSH) and its application in the bioleaching of metals from electronic waste. We developed a potentially effective GSH biosynthesis system employing the GshF/PPK enzyme cascade after conducting a series of rigorous testing. The E. coli SHuffle T7 competent expression cell strain was the best option among others for the heterologous overproduction of the targeted enzymes, highlighting the importance of selecting optimal E. coli strains for protein expression. Overall, our study highlighted the possibility of employing GSH as a chelating agent in the environmental cleanup and stressed the significance of optimizing conditions for better metal recovery. These conditions will be clarified after the chemical engineering experiments.