Part:BBa_K4102000:Experience

This experience page is provided so that any user may enter their experience using this part.
Please enter how you used this part and how it worked out.

Production of LanM from pET-19b(+)::lanM

After molecular cloning yielded the sequencing-verified expression plasmid pET-19b(+)::lanM, this vector was transformed into E. coli BL21(DE3). The recombinant strain was inoculated for two 500 ml-scaled main cultivations, applying the complex medium LB. At an optical density at 600 nm (OD600) of 0.4, the production of recombinant LanM was started by adding the inductor substance isopropyl-β-D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM. Incubation was performed at 37 °C in an incubation shaker (250 rpm). In Figure 1, the growth of the cultures is shown. During the exponential phase (2 to 4 hours after incubation start), production was started by IPTG addition. The stationary phase was reached comparably fast at 6 hours after incubation start. Final optical densities (OD600) were at 3.46 ± 0.12.

Figure 1: Optical density of E. coli BL21(DE3) pET-19b(+)::lanM cultures. Incubation was performed in LB medium at 37 °C and constant shaking (250 rpm). Production was induced by addition of isopropyl-β-D-thiogalactopyranoside at 3:40 hrs. 100 ml culture were harvested by centrifugation (4000 g, 5 min).

Afterwards, the pellet was resuspended in 10 ml Buffer W (100 mM Tris-Cl, 150 mM NaCl, pH 8.0) and the suspension was applied to ultrasonication (10 min) for cell disruption. Insoluble cell debris remaining in the clear cell lysate was removed in a subsequent centrifugation (15,000 g, 15 min). The whole protein pattern of the supernatant (crude cell extract) was analyzed by SDS-PAGE. Figure 2 shows that recombinant LanM (13 kDa) was successfully produced and constituted a substantial fraction of the total protein composition. As described below, a novel purification approach was developed, applying crude cell extract acidification to pH 2.5 to remove all proteins except LanM. To determine recombinant LanM yields, purified LanM was employed for protein quantification, relying on the BCA assay. The presented strain was able to produce 238 mg LanM per liter culture.

Figure 2: SDS-PAGE gel of E. coli BL21(DE3) pET-19b(+)::lanM crude cell extract. Gel was stained with silver staining, the recombinant LanM had an expected size of 13 kDa. Protein ladder: PageRuler™ Prestained Protein Ladder (Thermo Fisher Scientific). kDa: kilo daltons.

Purification of LanM from crude cell extract

On an industrial level, affinity chromatography-free approaches for recombinant protein purification from cell lysates is attractive due to lowering of process costs. Thus, different approaches were tested with respect to their potential to yield highly pure LanM.

Remarkable performance was shown in an acidification-based method. First, crude cell extract was acidified with HCl to a pH of 2.5 to precipitate all microbial proteins, except LanM. The suspension was applied for centrifugation (4000 g, 5 min) to pellet denatured proteins. An aliquot from the supernatant was applied for SDS-PAGE to investigate the protein pattern after the described acidic treatment. As shown in Figure 3, no protein bands were observed besides the 13 kDa lane of recombinant LanM. Thus, a process was established to purify LanM in a simple, fast and cheap approach.

Figure 3: SDS-PAGE of crude cell extract of E. coli BL21(DE3) pET-19b(+)::lanM crude cell extract after acidification to pH 2.5 and removal of precipitated proteins (centrifugation). Gel was stained with silver staining, the recombinant LanM had an expected size of 13 kDa. Protein ladder: PageRuler™ Prestained Protein Ladder (Thermo Fisher Scientific). kDa: kilo daltons.

A second approach relied on the increase of temperature to selectively denature and precipitate proteins other than LanM. A first protein pattern determination was performed with E. coli BL21(DE3) pET-19b(+)::lanM crude cell extract prepared at room temperature (RT, 20 °C) (Figure 4). Agreeing to the above reported observation, recombinant LanM (13 kDa) was present in comparably high amounts. Afterwards, crude cell extract was heated to 50 °C, resulting in a first step of protein precipitation. However, no strong change in protein pattern composition was observed (Figure 4). In contrast, the treatment at higher temperatures (65 °C and 75 °C) reduced the protein diversity. The final step (95 °C) yielded almost pure recombinant LanM (Figure 4).

Figure 4: SDS-PAGE of crude cell extract of E. coli BL21(DE3) pET-19b(+)::lanM crude cell extract after stepwise heat treatment and removal of precipitated proteins (centrifugation). After crude extract was exposed to the indicated temperature (see lanes), denatured proteins were removed by pelleting (13,000 g, 1 min) and a sample of the supernatant was applied for SDS-PAGE. Gel was stained with silver staining, the recombinant LanM had an expected size of 13 kDa. Protein ladder: PageRuler™ Prestained Protein Ladder (Thermo Fisher Scientific). kDa: kilo daltons. RT: room temperature (20 °C).

Purification of LanM from solutions containing LanM as the sole protein

Besides the purification of proteins from crude cell extracts, their isolation from other matrices is of relevance in bioprocess design as well. For example, the REE-LanM complex needs to be separated from the ore leachate during application of LanM for rare earth element (REE) isolation from ores. Thus, two approaches were tested to selectively precipitate LanM from solutions.

The first approach relied on the application of the cosmotrophic salt ammonium sulfate, (NH4)2SO4. It is commonly used for salting out enzymes and does not impair the functionality of the biomolecules (Matulis 2016). To investigate the suitability for salting out of LanM, a LanM-containing crude cell extract was prepared and the protein was purified following the above described acidification approach. The pH of the resulting acidic supernatant, containing LanM as sole protein, was risen to 7.0. Then, five aliquots were prepared, each containing about 2.4 mg of LanM. After increasing the volume to 1 ml with dH2O, (NH4)2SO4 was added. Relative to 100% saturation (750 mg (NH4)2SO4 to 1 ml solution), the following test concentrations were prepared: 0, 50, 75, 90, 100%. The salt was dissolved by incubation at 60 °C and stirring. After full dissolving, the volumes were incubated at room temperature for 1.0 hour and protein was pelleted (13,000 g, 1 min). Afterwards, the pellet was dissolved in dH2O and the protein content was determined via the BCA approach. As shown in Figure 5, the application of 50% ammonium sulfate only led to a slight precipitation compared to the 0% reference. In contrast, about 74% of applied LanM was isolated by precipitation in the presence of ammonium sulfate at a saturation of 75%. No further increase was obtained at higher concentrations of the cosmotrophic salt (Figure 5).

Figure 5: Amount of recombinant lanmodulin (LanM) precipitated from a solution by salting-out with ammonium sulfate (AS) or by adding ethanol (≥ 99.8%). 100% AS saturation was 750 mg to 1 ml solution. Amount of LanM added: 2.4 mg LanM (AS-based precipitation), 4.8 mg (ethanol-based precipitation). The weight of precipitated LanM was determined in a BCA assay with re-dissolved LanM.

A second method for isolation of LanM from solutions was based on increasing the solution hydrophobicity by addition of ethanol. 1 ml of purified LanM solution (4.8 mg) was mixed with 9 ml of ethanol (≥ 99.8%, pre-cooled to -20 °C). Immediately, the solution turned turbid. After incubation at -80 °C for 90 min, the precipitate was pelleted (13,000 g, 5 min) and the supernatant was discarded. To evaporate remaining solvent, the microcentrifuge tubes were incubated with opened lid at 60 °C. Finally, the dry pellet was dissolved in dH2O and the solution was applied for BCA assay-based protein determination. As shown in Figure 5, a similar precipitation yield was observed (71% of applied LanM) as for the maximum yields in ammonium sulfate-based precipitation.

To sum up, a first step of removing all proteins other than LanM (e.g. by acidification of the crude cell extract) can be followed by selectively isolating LanM from solutions to obtain a highly pure fraction of this biological chelating agent for downstream processes.

Activity assessment of recombinant LanM

LanM is known to be a potent rare earth element (REE) chelating agent. Previous works showed that the binding activity can be monitored in a competition assay (Cotruvo et al. 2018). In addition to LanM, a second metal ion chelating agent, xylenol orange (XO), is added to the assay. XO is initially present as a disodium salt. In the presence of other metal cations, XO can form complexes with these, resulting in a characteristic absorbance increase at 574 nm (A574). However, free REE-binding sites on LanM outcompete XO in REE acquisition. In consequence, LanM binding activity can be assessed by determining the amount of lanthanide cations which can be added without an increase in A574. After this assay was established into lab routine, purified recombinant LanM was characterized with respect to its lanthanum (La(III))-binding activity.

During preparation of the first assay, the applied LanM was purified in two steps: Removal of non-LanM proteins (acidification of crude cell extract) and ammonium sulfate-based precipitation of LanM from the supernatant. 80 µg of LanM were applied into the assay reaction, corresponding to 6.2 nmol. As shown in Figure 6, no considerable absorbance increase was observed during the addition of 14 nmol La(III). At 16 nmols, however, A574 was increased, indicating that the additional lanthanum cations were not acquired by LanM any longer but by XO, resulting in the characteristic assay absorbance shift. Thus, a binding ratio of ~2.3 lanthanide cations per LanM molecule was derived from the binding assay. This does not exactly agree to the binding of three lanthanide cations to strong EF-hand motifs and one lanthanide cations to a fourth, weaker EF-hand on LanM (Cook et al. 2018). However, the results indicate that the produced and purified protein was actively chelating lanthanides.

Figure 6: Uptake of added trivalent lanthanum cations (La3+) by xylenol orange, buffered by the lanthanide-chelating protein lanmodulin (LanM) in a competition assay. 80 µg (6.2 nmol) of LanM were added to the competition assay and La3+ was added as LaCl3. Formation of the La3+-XO complex was determined photometrically (A574nm).

An identical assay was run with LanM precipitated via the ethanol approach (see above) to investigate if this step impaired functionality. As shown in Figure 7, the applied 170 µg (13.1 nmol) LanM bound added trivalent lanthanum until 25 nmol of this REE were added. A continued addition led to the increase of A574, indicating that all LanM-located REE binding sites were saturated and XO acquired the freely available cations. Starting at 26 nmol, the amount of La(III) cations bound to XO exactly met the surplus of La(III) added after the proposed point of LanM saturation. The REE:LanM ratio was 1.9 at the LanM saturation point (25 nmol La(III) added). This ratio was lower than the value determined with ammonium sulfate-precipitated protein but still indicated that ethanol-based precipitation likely does not impair LanM functionality.

Figure 7: Uptake of added trivalent lanthanum cations (La3+) by xylenol orange, buffered by the lanthanide-chelating protein lanmodulin (LanM) in a competition assay. 170 µg (13.1 nmol) of LanM were added to the competition assay and La3+ was added as LaCl3. Formation of the La3+-XO complex was determined photometrically (A574nm).

Isolation of rare earth elements (REE) from ores using recombinant lanmodulin (LanM)

After progress in downstream process development of LanM-containing solutions, purified recombinant LanM was applied for isolation of REE from ores.

Preparation and characterization of apatite ore leachate

As feedstock, apatite ores from the Bafq district/Iran were applied. Based on previous composition analysis, comparably low fractions of the mineral weight are made up of REE cations (about 2 wt.-%, Stosch et al. 2010). Due to the extensive mining of ores with high REE content in the past decades and centuries, the mining of substrates with lower REE concentrations becomes increasingly attractive. Based on its high affinity for REEs down to the picomolar range, LanM is a promising chelating agent to extract lanthanides from ore leachates with comparably low REE concentrations (Cotruvo et al. 2018).

Apatite was manually ground to a pulverized state and acidified with hydrochloric acid (HCl) to pH 0.3. After several heat/stirring-cycles, all mineral particles were fully dissolved. A sample of the created leachate was analyzed via ICP-MS. The overall REE content was at 1.2 wt.-% of the applied apatite ore while the divalent calcium (Ca(II)) content was at 27.7 wt.-% (Figure 8). This agreed to the expected dominance of Ca(II) and to the REE content deduced from previous characterizations of this apatite ore (Stosch et al. 2010). In consequence, the calculation of the Ca(II)-to-REE ratio was a suitable indicator for detecting REE purification. The content of the particular REE cations is shown in Figure 9. As expected, trivalent cerium (Ce(III)) dominated the REE mixture (0.6 wt.-%), followed by La(III) and Nd(III) (both at ~0.2 wt.-%) and Y(III) (0.13 wt.-%) (Figure 9). All other REE cations were present in concentrations of <0.1 wt.-%. Thus, the applied apatite ore was a source of mainly light REEs.

Figure 8: Ca(II) and REE content of apatite, determined in an acidic ore leachate. REE: rare earth elements, Ca(II): divalent calcium. Wt.-%. Weight percentage.

Figure 9: REE content of apatite, determined in an acidic ore leachate. REE: rare earth elements, wt.-%. weight percentage.

First extraction process design

A first REE extraction process design (Figure 10) initiated with the combination of REE-containing ore leachate and purified LanM. At pH > 4.0, the LanM-REE complex should form. By salting-out with the cosmotrophic salt ammonium sulfate, (NH4)2SO4, the complex should be precipitated from the leachate. After removal of the supernatant and dissolving of the pellet (LanM-REE complex), the REE cations were released from the complex by acidification of the solution (pH < 1.5). Applying the acidic solution to an ultrafiltration unit (molecular weight cut-off value: 3 kDa), apo-LanM (retentate) was separated from highly pure REEs (flow-through). Apo-LanM was to be reused for further extraction cycles (Figure 10).

Figure 10: Process design to extract rare earth elements (REE) from ores by application of the bacterial protein lanmodulin (LanM). In contrast to the refined process, this approach relied on removal of the LanM-REE complex by salting-out with the cosmotrophic salt ammonium sulfate. Figure created with BioRender.

However, it was observed that salting-out with ammonium sulfate required a high concentration of the salt. This led to the formation of precipitated both in the LanM-added and in a LanM-free assay. Thus, the selectivity of the process was not ensured and another option for isolation of the LanM-REE complex was to be identified.

Refined extraction process design

The refined, final design of the REE extraction process (Figure 11) replaced salting-out-based complex isolation with an ultrafiltration step, separating the solution into ore leachate (flow-through) and the LanM-REE complex (retentate).

Figure 11: Process design to extract rare earth elements (REE) from ores by application of the bacterial protein lanmodulin (LanM). In contrast to an initial approach, the LanM-REE complex was separated from the ore leachate in a first ultrafiltration step. Figure created with BioRender.

The extraction process performance was monitored by analysis of REE contents at different process steps and investigation of the presence of other ore-derived cations. In supplementation tests, it was shown that the XO assay did not respond to the presence of Ca(II) (main cation in the applied ore) and Na(I). Fe(II) cations were only able to affect an absorption increase if they were present in high concentrations. Thus, it was assumed that the XO assay was suited for approximated determination of REE contents with sufficient specificity. More dedicated analysis was performed by application of ICP-MS.

To test the refined process, 30 mg apatite (3.8 µmol REE) and 26.3 mg LanM (2.0 µmol) were combined. XO assay results indicated that, during the process, 32% of REEs were lost in the flow-through of the first ultrafiltration (UF1-FT, Figure 12). Although this was a considerable fraction, the major amount of REEs (68%) were probably kept within the retentate (UF1-Ret). Since UF1-FT (16.75 ml) had a manifold higher volume than UF1-Ret (0.5 ml) and the distribution of REE amounts within the fractions did not follow this difference, an enrichment of REEs within the retentate fraction was assumed. After acidification-based release of REEs from the complex and ultrafiltration 2 (UF2) to separate the REE from apo-LanM, the majority of initially applied REE (63%) was found in UF2-FT while 5% of the employed REEs were detected in UF2-Ret (Figure 12).

Figure 12: Share of REE applied into a protein-based REE extraction process present in different process solutions. Two ultrafiltration steps were performed to separate a protein-REE complex from ore leachate (UF1) and to isolate the REE after release from the protein (UF2). FT: flow-through, Ret: retentate. REE: rare earth elements.

Highly dedicated and global analysis of the cation composition at various process steps was performed via ICP-MS. This analysis widely agreed on the XO assay-based approximations. As shown in Figure 13, the initial ore leachate, created from 30 mg apatite, comprised about 8.32 mg Ca(II) cations and 0.38 mg REE cations. Accordingly, the leachate’s gravimetric Ca(II)/REE ratio was 22.0 (Figure 13). After the first ultrafiltration, most of the initial Ca(II) content was removed with the flow-through (UF1-FT, Figure 13) while a neglectable amount of REEs was detected in this fraction. Thus, the Ca(II)/REE ratio was comparably high in UF1-FT (~8.860). During analysis of the UF2-FT, almost all expectedly purified REE cations were detected. In this fraction, the Ca(II)/REE ratio was determined to be 1.6 (Figure 13) – strongly reduced compared to the starting Ca(II)/REE ratio (ore leachate: 22.0). Thus, the refined process yielded a final process solution with about 14-fold enrichment of the desired REEs compared to the initially dominating, unwanted cations (Ca(II)). If the volume remaining in the UF1 retentate (about 3% of the ore leachate) could be further reduced, the transfer of contaminating cations could be minimized even more. This might lead to a fully pure REE solution in the UF2-FT fraction.

Figure 13: Calcium ion (Ca(II)) and rare earth element (REE) content in different fractions during an REE purification process based on the bacterial protein LanM. The initial ore leachate was created by dissolving 30 mg of apatite ore. During the REE extraction process, illustrated in Figure 11, two ultrafiltration (UF) steps were employed for retaining the LanM-REE complex (UF1) and separating LanM from released REEs (UF2). Purified REEs were expected to be present in the UF2 flow-through (UF2-FT). The table gives an overview on the gravimetric ratio between Ca(II) and REE cations in the respective fractions.

In Figure 14, an overview is given on the REE abundance pattern between the initial ore leachate and the final fraction with purified REEs (UF2-FT). As shown, the distribution pattern did not change remarkably. This indicated that LanM can be applied to effectively extract a broad range of REEs independent on the particular element.

Figure 14: Relative amount of rare earth element (REE) cations in different fractions of a protein-based REE purification process. REE amounts are given in relation to the total amount of REEs found in the respective fraction. UF2-FT: ultrafiltration 2 flow-through fraction.

A protein-free control run was performed to identify if the observed elevated presence of REE in the UF2-FT was attributable to the chelating activity of LanM. As shown in Figure 15, the protein-free control run (w/o LanM) yielded only a neglectable amount of REEs in the UF2-FT fraction (2.7 nmol) while the LanM-containing process run (w/ LanM) resulted in the presence of 502 nmol of REEs in the desired fraction (UF2-FT). Thus, it can be solidly stated that LanM contributed to the enrichment of REEs in the final solution.

Figure 15: Presence of rare earth elements (REEs) in the final solution of a protein-based (w/ LanM) and protein-free control (w/o LanM) process for REE extraction from ores. UF2-FT: ultrafiltration 2 flow-through.

Retainment of LanM for reuse in further REE extraction cycles

The loss of LanM during ultrafiltration was assessed by application of flow-through (FT) and retentate (Ret) samples to BCA-based protein determination. Of the 26.3 mg LanM applied for the process, 1.3 mg were lost in the UF1-FT (Figure 16) while 25 mg were kept in the retentate. Of this retained amount, almost all LanM remained in the retentate during the second ultrafiltration for LanM/REE separation (24.8 mg) while only a neglectable fraction (0.2 mg) passed the membrane and was detected in the flow-through (Figure 16). Thus, it can be assumed that the chelating agent can be quantitatively retained in the end of the process.

Figure 16: Amount of the bacterial protein lanmodulin (LanM) present in different fractions during a LanM-based rare earth element extraction process. LanM amounts were determined via the BCA assay. UF: ultrafiltration, FT: flow-through, Ret: retentate.

LanM, which was already used in two REE-capture-and-release cycles, was applied to the above described (Figure 11) REE purification process. Initial results indicated that LanM can be applied in several extraction cycles due to a conservation of binding activity (data not shown). Further experiments need to unambiguously determine the effect of many extraction cycles on LanM chelating performance. In previous studies, LanM reusability was principally verified (Deblonde et al. 2020).

Summary

To sum up, a newly designed, biometallurgical process yielded a 14-fold enrichment of rare earth elements (REEs) from ores. The approach was based on the bacterial protein lanmodulin (LanM) which served as a biological REE chelating agent. New LanM affinity chromatography-free purification processes were developed, enabling for rapid and cheap access to this biomolecule. In addition to successful enrichment of REEs by application of LanM, the protein was reconstituted to its apo-state finally and was rendered applicable for further extraction runs.

Literature

Cook, E.C.; Featherston, E.R.; Showalter, S.A.; Cotruvo, J.A. (2018): Structural Basis for Rare Earth Element Recognition by Methylobacterium extorquens Lanmodulin. In Biochemistry 59:120-125.

Cotruvo, J.A.; Featherston, E.R.; Mattocks, J.A.; Ho, J.V.; Laremore, T.N. (2018): Lanmodulin: A Highly Selective Lanthanide-Binding Protein from a Lanthanide-Utilizing Bacterium. In J. Am. Chem. Soc. 140: 15056-15061.

Deblonde, G.J.-P.; Mattocks, J.A.; Park, D.W.; Reed, D.W.; Cotruvo, J.A.; Jiao, Y. (2020): Selective and Efficient Biomacromolecular Extraction of Rare-Earth Elements using Lanmodulin. In Inorganic Chemistry 59: 11855-11867.

Matulis, D. (2016): Selective Precipitation of Proteins. In Curr. Prot. Prot. Sc.:4.5.1-4.5.37.

Stosch, H.-G.; Romer, R.L.; Daliran, F.; Rhede, D. (2010): Uranium–lead ages of apatite from iron oxide ores of the Bafq District, East-Central Iran. In Miner Deposita.

User Reviews

UNIQ79330386511d3e2a-partinfo-00000000-QINU UNIQ79330386511d3e2a-partinfo-00000001-QINU