Team:Calgary/Metal Separation Wetlab
A Selective Solution:
Biological methods of lanthanides separation and extraction have gained traction recently as a sustainable alternative to the chemical methods [1,2]. However, they lack the selectivity required to successfully separate lanthanides from highly impure sources such as electronic waste.
What sets our system apart from these previous attempts is lanmodulin (LanM), a special lanthanide-binding protein that has very recently been isolated from the methylotroph Methylobacterium extorquens [3]. Lanmodulin displays a marvelous ability to selectively bind lanthanides, with more than 100-million-fold affinity for them over other competing metals such as aluminum, silver, gold, copper, and iron.This makes it an ideal tool for purification of lanthanides from electronic waste [3]. Furthermore, by speaking with Dr. Cotruvo- the founder of lanmodulin- we learned that besides its great selectivity, lanmodulin portrays incredible robustness and flexibility. It can withstand temperatures as high as 95°C, and showcases strong lanthanide affinity down to a pH of 2.5. This makes lanmodulin a promising protein for use in industry, which often necessitates non-ideal, harsh physiochemical conditions. Finally, Dr. Cotruvo highlighted that lanmodulin has great reusability, with protein tolerating at least 50 cycles of metal-binding and release without showing any signs of degradation.
We wanted to design a high throughput system of lanthanide purification from electronic waste using lanmodulin. To do so, we decided to immobilize lanmodulin on a solid support which would allow for continuous reusability of the protein in multiple rounds of rare earth element recovery.
Dr. Cotruvo highlighted lanmodulin’s great reusability, with protein tolerating at least 50 cycles of metal-binding without showing signs of degradation. This makes it ideal for use in an immobilization system. We considered a variety of solid supports, but upon discussion with Dr. Hu- an expert in microcellulose engineering- he recommended using cellulose beads as a choice of support. The advantage of cellulose over other supports is that it is not only cheap and relatively inert, but it is also biodegradable and can withstand harsh conditions. To immobilize lanmodulin on cellulose, we decided to fuse lanmodulin to a cellulose-binding module (CBM). Traditional methods of immobilization such as a covalent attachment typically require extra steps to immobilize the protein after it has been purified, which adds cost and decreases the efficiency of the system [4]. However, using a CBM we can perform a one-step purification and immobilization of lanmodulin onto cellulose without requiring additional processing [4,5]. We considered a variety of CBMs, but further discussion with Dr. Hu helped us narrow down our choices to two possibilities. Dr. Hu mentioned that the production of CBMs in E. coli is often challenging due to the low-solubility nature of these proteins. Therefore, we decided to work with Cellulomonas fimi’s Cex CBM and Clostridium thermocellum’s CipA CBM. These CBMs have showcased relative ease of expression in E. coli and a high affinity for cellulose for stable immobilization [4,6].
Genetic Construct
We designed three genetic constructs: one for expression of wildtype lanmodulin, and the other two for expression of lanmodulin fused to the CBMs. We decided to fuse the CBMs to the C-terminus of lanmodulin because when we spoke with Dr. Cotruvo and Dr. Park- both lanmodulin experts who work very closely with the protein- they mentioned that based on their preliminary experiments, immobilization on C-terminus tends to help with protein stability and function. Upon conducting a literature review on various CBM fusion proteins, we decided to use a 20 amino acid long, threonine-proline rich linker which has been shown to allow for great flexibility and freedom for the fused proteins [5]. All constructs also contained a polyhistidine tag (His-Tag) for efficient protein purification and a TEV cleavage site for eventual cleaving of the His-Tag after purification.
Results:
Cloning and Protein Expression:
We initially began conducting our experiments by expressing our respective lanmodulin constructs, which were pre-ordered in a IDT vector, into E.coli BL21. However, after multiple experiments and trials we were unable to produce our desired protein. Therefore, we spoke to Dr. Schryvers, a biochemistry and molecular biology professor at the University of Calgary, who suggested that certain heterologous proteins are poorly suited for expression in E. coli as the natural protein expression pathway in E. coli might not be optimal for the protein folding. Using a solubility tag such as Glutathione S-transferase (GST) could help the exogenous protein fold better by acting as chaperone. Dr. Schryvers kindly provided us with the Xpress, an easy to use and quick E. coli expression vector. Xpress contains the GST solubility tag and by cloning with the proper restriction enzymes, a GST fusion protein could be obtained that should help with the expression of the heterologous protein. After receiving the Xpress vector we began cloning our respective lanmodulin inserts into the Xpress vector to obtain a GST-lanmodulin (GST-LanM) fusion protein.
Using Xpress’s quick screening system, we were able to easily identify and sequence verify positive clones with the proper GST-LanM insert. Protein production was then induced using an autoinduction media and after cell lysis, the soluble portion of lysate was subject to His-Tag purification and analysis on SDS-PAGE gel (Fig 3).
Figure 1. SDS analysis of his-tagged purified lanmodulin
The dark band in lane 9 (purple box) was observed around the 39 kDa range, which corresponds to the expected molecular weight of GST-LanM protein (Fig 3). The positive control GST was also observed in lane 5 (green box) (Fig 3). Overall, these results show successful production and purification of GST-LanM, suggesting that GST has helped to stabilize the protein, making it more suitable for expression in E. coli.
The dark band in lane 9 (purple box) was observed around the 39 kDa range, which corresponds to the expected molecular weight of GST-LanM protein (Fig 3). The positive control GST was also observed in lane 5 (green box) (Fig 3). Overall, these results show successful production and purification of GST-LanM, suggesting that GST has helped to stabilize the protein, making it more suitable for expression in E. coli.
With the permission of Dr. Schryvers, we have added the Xpress vector (BBa_K3945014) to the iGEM registry as an easy-to-use and reliable vector for expression of difficult heterologous proteins in E. coli.
Lanmodulin Characterization
After successfully producing our protein, we conducted a metal recovery assay to test the REE binding capacity of the lanmodulin protein. In this assay, lanmodulin was incubated with a known concentration of neodymium then filtered using centrifugal filtration.. The concentration of neodymium before and after the filtration step was measured using Arsenazo III assay. The protein and neodymiums bound to it would be stuck on the filter while unbound metals wash through and end up in the filtrate. Therefore, the concentration of neodymium is expected to decrease after incubation with lanmodulin.
Figure 2. Arsenazo III assay of samples before and after centrifugation. A starting neodymium sample of 25uM was treated with either water of 5uM lanmodulin. After centrifugal filtration, concentration of filtrate were assessed using Arsenazo III assay.
Arsenazo III is a colorimetric assay where a dark blue color represents high neodymium concentration and a light pink color represents little to no neodymium. Treatment with lanmodulin turned the solution from a dark blue color to an almost completely pink color similar to that of a zero neodymium sample. This indicates that lanmodulin was able to grab onto most of the neodymium present in the solution Whereas when samples were treated with just water and no protein, the color stayed dark, suggesting little to no neodymium was recovered in the absence of lanmodulin (Fig 4).
Figure 3. Concentration of neodymium in the initial sample, the filtrate and after treatment with HCl. The samples were treated with either 5 uM of lanmodulin, 5 uM of GST or water.
We measured the absorbance at 650 nm to quantify our Arsenazo Assay results. It was shown that after treatment of an initial 146 uM neodymium solution with LanM, the concentration dropped to 28.4 uM, indicating that LanM successfully recovered the majority of the neodymium with an 81% recovery success. Whereas, GST which is non-selective for neodymium performed relatively poorly with more than half the neodymium left in the sample. The negative control samples that were treated with water showed little to no neodymium recovery (Fig 5&6). Furthermore, upon treatment with a low acidic solution, lanmodulin released most of its bound neodymium, suggesting that acid treatment could be an efficient method of collecting the metals from the protein after recovery (Fig 5).
Figure 4. Percent neodymium recovery efficiency of 5 uM lanmodulin, 5 uM GST and water from a 146 uM initial sample of neodymium.
The above results show successful characterization of the lanmodulin protein for its ability to efficiently recover lanthanides and release them upon treatment with an acidic solution.
Future Experimental Plan:
Characterization of the Cellulose Binding Module
Unfortunately due to limited access as a result of the COVID-19 pandemic we were unable to express and purify the LanM-CBM fusion proteins. After successful cloning of the fusion proteins, the next step would be to conduct a cellulose binding assay and determine which CBM will serve as a greater immobilization medium. This will be determined by testing the binding efficiency of each of the CBM to cellulose. Thus, the CBMs will be incubated with a constant amount of bacterial cellulose. Following incubation, the cellulose, and any proteins bound to the cellulose, will be centrifuged and pelleted. The concentration of protein in the pellet will be compared to the initial protein concentration, to determine and compare the binding efficiency of the two separate CBMs to cellulose.
Characterization of the Cellulose-Immobilized Lanmodulin
Next step in our metal recovery experimental plan consists of conducting a metal recovery assay similar to the one performed with immobilized lanmodulin but this time the protein will be immobilized on bacterial cellulose beads. This will help determine the binding efficiency of lanmodulin immobilized cellulose which will be compared to that on immobilized lanmodulin to determine the impact of immobilization on the protein’s binding capabilities.
Proof of Concept
After fully characterizing both the lanmodulin and the CBMs, the final step would be to demonstrate the system's capacity to recover neodymium from a mixed metal mixture. For initial rounds of experiments, the recovery efficiency of our immobilized system will be tested in a synthetic metal solution representing the composition of metal found in e-waste. The system will be assessed for the percent REE recovered and the percent of non REE metals recovered. This will help us demonstrate that our system not only has a high recovery rate but also is able to remain quite selective in presence of high concentration of competing metals. Then, finally the system will be tested in a dissolved e-waste solution to create a proof of concept of selective recovery of REEs from e-waste.
Conclusion and Future Direction:
We have been able to successfully express and purify the novel lanmodulin protein using the Xpress vector. After purification we demonstrated lanmodulin’s great REE binding capacity by comparing it to the non-selective GST protein. With more than 80% REE recovery rate, we have proved that as advertised lanmodulin is the optimal solution to the REE scarcity. Once we have fully characterized the immobilized lanmodulin system, our next step is to move our experiments towards our adsorber column and ensure lanmoduliin has a similar binding capacity and functionality in the adsorber column, compared to our proof of concept experiments in the lab. Moreover, since our end goal is to produce a large-scale rare-earth recovery process we will be working on optimizing the conditions for the large-scale production and storage of lanmodulin. We hope that one day we can bring the full potential of this novel protein to life
References:
-
Park DM, Brewer A, Reed DW, Lammers LN, Jiao Y. Recovery of Rare Earth Elements from Low-Grade Feedstock Leachates Using Engineered Bacteria. Environmental Science and Technology. 2017 [accessed 2021 Sep 17];51(22):13471–13480. https://pubs-acs-org.ezproxy.lib.ucalgary.ca/doi/full/10.1021/acs.est.7b02414. doi:10.1021/ACS.EST.7B02414
-
GJ D, JA M, DM P, DW R, JA C, Y J. Selective and Efficient Biomacromolecular Extraction of Rare-Earth Elements using Lanmodulin. Inorganic chemistry. 2020 [accessed 2021 Sep 17];59(17):11855–11867. https://pubmed.ncbi.nlm.nih.gov/32686425/. doi:10.1021/ACS.INORGCHEM.0C01303
-
JA C, ER F, JA M, JV H, TN L. Lanmodulin: A Highly Selective Lanthanide-Binding Protein from a Lanthanide-Utilizing Bacterium. Journal of the American Chemical Society. 2018 [accessed 2021 Sep 17];140(44):15056–15061. https://pubmed.ncbi.nlm.nih.gov/30351021/. doi:10.1021/JACS.8B09842
-
Improved immobilization of fusion proteins via cellulose‐binding domains - Linder - 1998 - Biotechnology and Bioengineering - Wiley Online Library. [accessed 2021 Oct 14]. https://onlinelibrary.wiley.com/doi/pdf/10.1002/(SICI)1097-0290(19981205)60:5%3C642::AID-BIT15%3E3.0.CO;2-8
-
Myung S, Zhang X-Z, Zhang Y-HP. Ultra-stable phosphoglucose isomerase through immobilization of cellulose-binding module-tagged thermophilic enzyme on low-cost high-capacity cellulosic adsorbent. Biotechnology Progress. 2011 [accessed 2021 Oct 14];27(4):969–975. https://onlinelibrary.wiley.com/doi/full/10.1002/btpr.606. doi:10.1002/BTPR.606
- Ong E, Gilkes NR, Miller RC, Warren RAJ, Kilburn DG. The cellulose‐binding domain (CBDCex) of an exoglucanase from Cellulomonas fimi: Production in Escherichia coli and characterization of the polypeptide. Biotechnology and Bioengineering. 1993 [accessed 2021 Oct 14];42(4):401–409. doi:10.1002/BIT.260420402
