logo

Our mission is to ensure the generation of accurate and precise findings.

Contact Us No. 111, Ren'ai Road, Dushu Lake Science and Education Innovation Zone, Suzhou Industrial Park iGEM@xjtlu.edu.cn
Weizhuo Chen Oct 10, 2022

Results

Part A: Metallothionein(MT) for metals adsorption

Background:

Metallothionein(MT) family proteins are heterogeneous proteins with low molecular weight(<10 kDa) that bind to metals[1]. In our project, we chose ShMT3 and SmtA as our metal binding proteins. According to previous research, shMT3 was designed from the freshwater crab Sinopotamon henanense (ShMT) using sequence-based multiple sequence alignment (MSA) and structure-based molecular docking simulation (MDS). ShMT3 and SmtA have the ability to bind and sequester multiple heavy metal ions, such as Cu2+ and Cd2+ , due to their Cys-X-Cys or Cys-X-X-Cys structures[1].

1. CsgA-MT Constructions

Design:

One of the most important parts of the biofilm matrix of E.coli is a series of highly functional amyloids called curli fibers. To express metallothionein on the surface of E.coli MG1655, we linked the metallothionein with CsgA, which is the major subunit protein of curli fibers, via GS-linker(GGGS). As a result, metallothionein would express along with curli fibers on the surface of E.coli MG1655 to adsorb metals. Furthermore, aiming at enhancing the expression of MTs, we added the SUMO coding sequence before CsgA-MT coding sequence[2]. As a result, there would be four different MT constructs, which are SUMO-csgA-shMT3, csgA-shMT3, SUMO-csgA-SmtA and csgA-SmtA respectively. Since pET-28a-SUMO plasmid was selected as the backbone, IPTG was then added to bind with LacI protein and turn on the lacI promoter, so that (SUMO)csgA-MT would express.

Figure 1: The pathway of (SUMO)csgA-MT. We fused csgA with (SUMO)MT via GS(GGGS) linker. After being induced by IPTG, the lacI promoter would be turned on and the (SUMO)csgA-GGGS-MT would express.
Build:

After obtaining the plasmids that contain the targeted sequences from the company(GENEWIZ), four different plasmids(pET28a-SUMO-csgA-shMT3, pET28a-csgA-shMT3, pET28a-SUMO-csgA-SmtA and pET28a-csgA-SmtA) were transformed into E.coli MG1655. In the meanwhile, pET28a-SUMO was transformed into E.coli MG1655 as the control group. After being induced by 1 mM IPTG, metallothionein(MT) would express along with CsgA.

Test:

RT-qPCR:Constructions

To verify the expression level of metallothionein(MT), we conducted RT-qPCR(Real Time Quantitative PCR) to test the expression of csgA since MT would be linked with csgA if the plasmids were successfuly transformed into E.coli MG1655. We induced the production of csgA with 1 mM IPTG and extracted the RNA of cells harboring pET28a-SUMO, pET28a-SUMO-csgA-shMT3, pET28a-csgA-shMT3, pET28a-SUMO-csgA-SmtA and pET28a-csgA-SmtA. Then we obtained the cDNA of the five kinds of RNA using reverse transcription PCR and tested the expression level of csgA to reflect the expression level of csgA using RT-qPCR.

Figure 2: The result of RT-qPCR that demonstrates the CsgA expression level of E.coli MG1655 that containing pET28a-SUMO, pET28a-SUMO-csgA-shMT3, pET28a-csgA-shMT3, pET28a-SUMO-csgA-SmtA and pET28a-csgA-SmtA. Data represent mean ± SD of three independent experiments (n = 3). Statistical significances were analyzed using t-test compared with the control (cells harboring pET28a-SUMO),*p/#p˂0.05, **p/##p ˂0.01, ***p/###p ˂0.001.
Learn:

Figure 2 demonstrates the difference of the expression level of CsgA between the modified bacteria harboring MT constructs and the control group. As a result, the csgA genes in the inserted sequences: SUMO-csgA-shMT3, SUMO-csgA-SmtA and csgA-SmtA have overexpressed comparing with the control group(pET28a-SUMO).

2. Assessment of E. coli metal tolerance

Design:

According to previous research, both ShMT3 and SmtA have the ability to increase the tolerance of E.coli against heavy metals, which allows E.coli that express MT could survive under high concentration of CuSO4_4 and CdCl2_2[3]. To testify if the modified bacteria in our project could survive high concentrations of heavy metals and be applied to actual industrial sewage treatment, we tested the growth of the modified bacteria under different concentrations of CuSO4_4(1200 μM, 1400 μM, 1600 μM, 2000 μM) and CdCl2_2(150 μM, 300 μM, 600 μM, 900 μM). The concentrations were determined based on the previous research[3]. In the meanwhile, we have learned from an interview of one local factory that the actual concentrations of Cu2+ and Cd2+ in industrial sewage were lower than the concentrations we selected.

Build:

Four different plasmids were transformed into the bacteria respectively: pET28a-SUMO-csgA-shMT3, pET28a-csgA-shMT3, pET28a-SUMO-csgA-SmtA and pET28a-csgA-SmtA. After obtaining the modified bacteria, we induced the expression of the inserted gene by adding 1mM IPTG.

Test:

We cultured the modified bacteria under different concentrations of CuSO4_4 and CdCl2_2 and then measured the growth condition of each construct in 8 hour using microplate reader by reading its OD600 value per hour.

Figure 3: The results of growth of the E. coli cells expressing SUMO-csgA-shMT3, csgA-shMT3, SUMO-csgA-SmtA, csgA-SmtA and control in LB medium with CuSO4_4(1200 μM, 1400 μM, 1600 μM, 2000 μM). Data represent mean ± SD of three independent experiments (n = 3)
Figure 4: The results of growth of the E. coli cells expressing SUMO-csgA-shMT3, csgA-shMT3, SUMO-csgA-SmtA, csgA-SmtA and control in LB medium with CdCl2_2(150 μM, 300 μM, 600 μM, 900 μM). Data represent mean ± SD of three independent experiments (n = 3)

Learn:

As above, all E. coli cells grew well under lower concentration of metals(1200 μM Cu2+ and 150 μM Cd2+), whereas the control group was inhibited in higher concentrations of Cd2+ and Cu2+(1600 μM, 2000 μM Cu2+ and 300 μM, 600 μM, 900 μM Cd2+). However, there may be some possible bias that had influenced the results. Firstly, although the growth of the control group showed the tendency to be inhibited by 1600 μM and 2000 μM CuSO4_4, it was not significantly obvious. Also, in 900 μM CdCl2_2, there was a dramatic decline of the OD value of the control group. We inferred that it may due to the evaporation of the fluid in the well of the microplate. In conclusion, the modified bacteria could survive under high concentrations of CuSO4_4 and CdCl2_2 and have the ability to tolerate heavy metals comparing with the control group.

3. Assessment of metal adsorption

Design:

To test the ability of MT constructs to adsorb metals, we measured the amount of metals binding with MT on bacterial biofilm using ICP-MS. We chose to verify the adsorbing ability under 600 μM CuSO4, 1600 μM CuSO4_4 and 300μM CdCl2_2 respectively according to the concentrations given in the previous research[1] and our preliminary experiments.

Build:

Five different plasmids were transformed into the bacteria respectively:pET28a-SUMO, pET28a-SUMO-csgA-shMT3, pET28a-csgA-shMT3, pET28a-SUMO-csgA-SmtA and pET28a-csgA-SmtA. After harboring the four MT constructs and one control, we induced the expression of csgA along with MTs by 1 mM IPTG.

Test:

Briefly, we cultured four kinds of modified E.coli and the control group harboring pET28a and incubated them along with a certain concentration of heavy metal ion. Cells were then cultured for 8 h at 37◦C, spun down for 10 min at 3000 rpm, washed three times with PBS, after which cells were then digested in HNO3_3 and H2_2O2_2 with an automatic microwave digestion instrument. After removing residual HNO3_3, samples were then added to a 50 mL tube containing 2% HNO3_3, and metal ions therein were assessed via ICP-MS.

(1)Adsorption of CuSO4_4

Figure 5: The results of metal adsorption of the E. coli cells harboring SUMO-csgA-shMT3, csgA-shMT3, SUMO-csgA-SmtA, csgA-SmtA and control in the presence of 600 μM CuSO4_4 for 8 h at 37 ℃. Data represent mean ± SD of three independent experiments (n = 3). Statistical significances were analyzed using t-test compared with the control (cells harboring pET28a-SUMO),*p/#p˂0.05, **p/##p ˂0.01.
Figure 6: The results of metal adsorption of the modified bacteria harboring SUMO-csgA-shMT3, csgA-shMT3, SUMO-csgA-SmtA, csgA-SmtA and control in the presence of 1800 μM CuSO4_4 for 8 h at 37 ℃. Data represent mean ± SD of three independent experiments (n = 3). Statistical significances were analyzed using t-test compared with the control (cells harboring pET28a-SUMO),*p/#p˂0.05, **p/##p ˂0.01.
(2)Adsorption of CdCl2_2
Figure 7: The results of metal adsorption of the E. coli cells harboring SUMO-csgA-shMT3, csgA-shMT3, SUMO-csgA-SmtA, csgA-SmtA and control in the presence of 300 μM CdCl2_2 for 8 h at 37 ℃. Data represent mean ± SD of three independent experiments (n = 3). Statistical significances were analyzed using t-test compared with the control (cells harboring pET28a-SUMO),*p/#p˂0.05, **p/##p ˂0.01.
Learn:

After testing the adsorption of 600 μM Cu2+, 1800 μM Cu2+ and 300 μM Cd2+, it has been demonstrated that the ability of the modified bacteria harboring MT constructs(SUMO-csgA-shMT3, csgA-shMT3, SUMO-csgA-SmtA and csgA-SmtA) to adsorb Cu2+ and Cd2+ was significantly higher than the control group. Also, modified bacteria harboring MT constructs with SUMO tag has showed a significantly higher adsorbing rate of Cu2+ and Cd2+ than those without SUMO tag, which indicated that the SUMO tag enhances the ability of ShMT and SmtA to bind metals.

However, the absorption rates of MT constructs remain lower than expected. One possible reason may be the spoilage of cells during the process of preparing samples for ICP-MS when centrifuging and washing cells with PBS. To verify this assumption, we centrifuged the supernatant that were stored before and it turned out that there were still amounts of cells in the supernatant. As a result, we assumed that the actual adsorbing rate would be higher than that in our experimental results, but due to a limit of time, we could not conduct new experiments to give a more precise adsorbing rate.

Part B: Metal binding peptide (MBP) for precious metals adsorption

Background

Metal binding peptide (MBP3) is a kind of silver-binding peptide identified from phage surface display[4]. MBP3 has chair-like three-dimensional structure. After capturing silver ions with four amino acids: Leucine(L), Phenylalanine (F), Arginine (R) and Tyrosine (Y), it has ability to reduce Ag+ to silver nanoparticles [5]. In our project, we expected to adsorb and reduce Ag+ with our engineered bacteria in order to further explore the potential of nanomaterial synthesis and precious metal recycling using a green approach in the future.

1. csgA-MBP3 construction

To express MBP3 on the surface of E.coli MG1655, we linked csgA sequence with MBP3 via GS-linker (GGGS). For plasmid construction, pGEX-4T-1 is a bacterial plasmid expressing target protein through tac promoter. CsgA-GS linker (GGGS)-MBP3 was inserted in the downstream of tac promoter. The tac promoter, a hybrid of the trptophan (trp) and lacUV5 promoter is induced by Isopropyl β-D-thiogalactoside (IPTG) and other galactosides[6].

Figure 8: Left: CsgA protein and MBP short peptide are represented by dark lines and blue balls respectively. Right: The pathway of csgA-MBP3. csgA is fused with MBP3 via GS(GGGS) linker. We transformed pGEX-4T-1-CsgA-MBP3 into E.coli MG1655. After induced by IPTG, CsgA-MBP3 would be expressed.
Colony PCR to verify plasmid successful transformation
Design

pGEX-4T-1-csgA-MBP3 (synthesized by GENWIZ) was transformed into E.coli MG1655 competent cells and selected by Lysogeny broth (LB) agar dish containing ampicillin.

Test

Colony PCR and agarose gel electrophoresis were conducted to verify plasmid transformed successfully with csgA-MBP3 insert.

Figure 9: Results of electrophoresis for pGEX-4T-1-CsgA-MBP3 plasmid construction. Samples were loaded as following orders: 100 bp DNA ladder,1000 bp DNA ladder, 673 bp positive control(+), three pGEX-4T-1-CsgA-MBP3 colonies(1,2,3), negative control(-)
Learn

According to PCR primers we synthesize (synthesized by GENEWIZ), theoretical PCR product would contain 673 base pairs. In electrophoresis results, no band was found in negative control (-). Bands in experiment group (1,2,3) and positive control (+) were at same location, between 600 and 700 bp of DNA ladder. Therefore, We could conclude that pGEX-4T-1-CsgA-MBP3 was successfully transformed into E.coli MG1655 strain.

RT-qPCR to confirm CsgA-MBP3 successful expression
Design
After confirming pGEX-4T-1-csgA-MBP3 successful transformation, we verified CsgA-MBP3 successful expression and IPTG optimal concentration for induction of previous engineered bacteria for CsgA-MBP3 expression meanwhile.

Test

According to modeling, csgA-MBP3 expression is maximized around 1 mM IPTG induction. When IPTG concentration is above 1 mM, it reaches the plateau phase (See modeling page for details). Therefore, We used 0.5 mM and 1 mM IPTG for tac promoter induction to verify modeling accuracy. RNA extraction, reverse transcription and RT-qPCR were conducted to measure CsgA-MBP3 expression level at different IPTG concentrations in control group(MG1655) and MG1655 harboring pGEX-4T-1-csgA-MBP3.

Figure 10: results of RT-qPCR for CsgA-MBP3 expression level. The Control group, 0.5 mM IPTG, and 1mM IPTG in x-axis represent the RT-qPCR results of MG1655 without IPTG, MG1655 harboring pGEX-4T-1-csgA-MBP3 induced by 0.5mM and 1mM IPTG, respectively. Data represent mean ± SD of three independent experiments (n = 3). Statistical significances were analyzed using t-test **** p/####p<0.0001
Learn

From RT-qPCR results, CsgA-MBP3 gene expression has extremely highly significant difference between control, engineered bacteria with 0.5 mM and 1 mM IPTG induction and CsgA-MBP3 is overexpressed through IPTG induction.

In addition, experiment results validated the accuracy of IPTG addition modeling. According to modeling, IPTG has no effect on CsgA-MBP3 expression at more than 1 mM. Therefore, we chose 1 mM IPTG as the concentration to induce the expression of tac CsgA-MBP3.

2.MG1655 pGEX-4T-1-CsgA-MBP3 for silver ion adsorption

After successful plasmid construction, we identified MG1655 harboring pGEX-4T-1-CsgA-MBP3 ability in silver ion adsorption. In order to investigate E.coli environment maximum tolerance under different silver ion concentrations, We firstly measured kinetic growth curves at different of silver ion , so as to identify how much silver ion in sewage our modified bacteria can survive. We then identified nanosilver synthesis through scanning electron microscope (SEM) and energy dispersive X-ray spectroscopy (EDS) analysis and detected engineered bacteria adsorption by inductively coupled plasma mass spectrometry (ICP-MS).

Growth curve of MG1655 pGEX-4T-1-CsgA-MBP3 at different silver ion concentration
Design

According to China Electroplating Pollutant Emission Standards, Ag+ discharged in sewage shall not exceed 0.1 mg/L (0.93 μM). We observed bacteria growth inhibition at different Ag+ concentration, guiding Ag+ addition in subsequent experiments.

Test

We conducted kinetic growth assays on E.coli MG1655, E.coli MG1655 harboring pGEX-4T-1 and pGEX-4T-1-CsgA-MBP3 in AgNO3_3 respectively. 5 μM to 80 μM AgNO3_3 were added in 96 microplate wells along with LB and E.coli. The culture were incubated in orbital shaker and absorbance of OD600 was measured by microplate reader every 2 hours for two days.

Figure 11: Results of E. coli MG1655, MG1655 harboring pGEX-4T-1 and pGEX-4T-1-CsgA-MBP3 growth curves from 0 h to 31 h under 5 μM to 80 μM AgNO3_3. Data represent mean ± SD of three replicates.
Learn

In this kinetic growth assays, bacteria growth was inhibited differently in the first few hours. It is noted that the higher concentration of Ag+ resulting in longer inhibition time. For example, the growth inhibition time of bacteria reached 19 h under 80 μM AgNO3_3. Additionally, with culture time increase, E.coli growth started off and finally reached stationary phase.

For three strains kinetic growth curves, E.coli MG1655 harboring pGEX-4T-1-csgA-MBP3 had the longest growth inhibition time, while MG1655 and MG1655 pGEX-4T-1 inhibition time were approximately the same. This excludes the effect of plasmid transformation on bacterial growth and difference was mainly caused by CsgA-MBP3 fragment insert.

Moreover, with the purpose of reducing inhibition time so as to improve adsorption efficiency, we chose 0 μM to 20 μM as final AgNO3_3 concentration in subsequent experiments.

SEM imaging for nanosilver synthesis
Design

To confirm MBP3 ability of binding with Ag+ , we observed silver nanoparticles on E.coli biofilm through SEM and EDS analysis.

Test

Bacteria induced by IPTG (1 mM) were incubated in LB medium for two days to expressed modified curli fibers. Subsequently, cells were centrifuged at 3000g and media was replaced with fresh LB broth. AgNO3_3 was added to medium and incubated in dark room for 36h.

E. coli harboring pGEX-4T-1-csgA-MBP3 with/without 6μM AgNO3_3 induction were prepared for SEM and EDS assays. Samples were centrifuged and washed by phosphate-buffered saline (PBS). After that, they were fixed, dehydrated and dried for SEM test (done by Yuantest Laboratory).

Learn

Bacteria aggregation and suspected biofilm structures could be identified in E. coli samples with/without Ag+ . Unfortunately, Peaks corresponding to signals of Ag element were not found in EDS analysis. Possible reasons could be less nano-silver synthesis on biofilm and loss of silver nanoparticles in SEM sample preparation, for example, dehydration operation and critical point drying. Due to a limit of time, we could not conduct new experiments to testify the synthesis of silver nanoparticles. We would demonstrate the adsorption of Ag+ in the subsequent ICP-MS experiments so as to validate the function of MBP3.

ICP-MS for silver ion adsorption
Design

We next measured Ag+ adsorption of engineered bacteria in AgNO3_3 solution. The adsorption capacity was reflected by the amount of silver element in supernatant or in bacteria and biofilm surface.

Test

Bacteria induced by IPTG (1mM) were incubated in LB medium for two days to expressed modified curli fibers. Subsequently, cells were centrifuged at 3000g and media was replaced with fresh LB broth. AgNO3_3 was added to the medium and incubated in the dark room for 36h.

Method 1: Measurement of E.coli silver ion adsorption

We firstly measured silver ion adsorption in E.coli. Procedures are as followed. Samples were centrifuged and washed by phosphate-buffered saline (PBS). Then organic matter of cells was removed by microwave digestion. Finally, silver concentration in bacteria and biofilm surface was measured by ICP-MS.

However, the results did not meet our expectations. PBS washing away bounded silver ions, negative charged membrane attracting large amount of silver ions and inadequate centrifugation were potential reasons for not reaching desired outcomes.

Method 2: Measurement of supernatant silver ion concentration

We then measured Ag+ concentration in supernatant and calculated adsorption rate along with different silver ion concentration.

Silver ion adsorption =Total silver ion addition - Supernatant silver ion concentration, this formula could be used to calculate silver ion adsorption and illustrates heavy metal ions' concentration after adsorption in the meanwhile.

Figure 12 : Results of supernatant Ag+ concentration and Ag+ adsorption rate of E.coli harboring pGEX-4T-1(control), pGEX-4T-1-CsgA-MBP3 in the presence of AgNO3_3 for 36h at 37℃: Data represent mean ± SD of two independent experiments (n = 2). Statistical significances were analyzed using t-test compared with the control (cells harboring pFEX-4T-1),*p/#p˂0.05, **p/##p ˂0.01, ***p/###p ˂0.001
Learn

In supernatant measurement, adsorption has significant difference between these two strains in the presence of AgNO3_3 addition (3 μM to 9 μM). Additionally, the csgA-MBP3 group had less silver ion in the supernatant and higher adsorption rate than the control group.

Compared with MG1655 harboring pGEX-4T-1, bacteria inserted CsgA-MBP3 fragments has higher adsorption rate (60%-70%) of silver ion in sewage.

Part C: Bacteria culture in MBBR carriers

In device design aspects, we coated cells in moving bed bioreactor (MBBR) carriers. MBBR carriers are microbial carriers used in MBBR process, providing suitable environment for microorganism growth due to its large specific surface[7].

We first observed the morphology of biofilm through SEM, then we coated bacteria on carriers and measured its biofilm quantity. Biofilm quantity was reflected by the mass increment of the carriers' dry weight and the amount of bacteria on carriers.

Suggestions of bacteria addition in real sewage were put forward based on experiment data so as to prove project concept that biofilm on carriers could reduce heavy metal ions in sewage.

SEM imaging for morphological observation of biofilm

Design

SEM microscope was used to observed biofilm microstructure.

Test

Bacteria (E.coli MG1655 harboring pGEX-4T-1-csgA-MBP3) were incubated in LB medium for two days by IPTG (1 mM) to express modified curli fibers. Subsequently, cells were centrifuged at 3000g and media was replaced with fresh LB broth. After that, bacteria were centrifuged and washed by phosphate-buffered saline (PBS). Samples were fixed, dehydrated and dried then tested by SEM (done by Yuantest Laboratory).

Figure 13: These four images show E. coli morphology photographed by SEM at 20 kV with gold spray on the sample surface. Scale bars are 10 μm,2 μm,1 μm and 1 μm respectively.
Learn

We observed suspected biofilm-like structures and curli fibers. Despite the lack of evidence, clusters of E. coli can be observed, demonstrating E. coli presence in the form of populations.

The inconspicuous biofilm may be related to the pre-sampling of SEM, for example, the liquid carbon dioxide washed off the biofilm, and biofilm evaporated with water in critical point drying.

Biofilm quantity measurement in MBBR carriers

Measurement by bacteria amount coated on carriers
Design

In order to determine biofilm quantity in MBBR carriers, we eluted biofilm in carriers by PBS solution and then measured absorbance in OD600 to identify bacteria amount in carriers.

Test

E. coli MG1655 harboring pGEX-4T-1-csgA-MBP3 grew overnight was induced by IPTG and incubated in shaking incubator at 37℃ 171 rpm. Cell cultures, thirty MBBR carriers and AgNO3_3 (6μM) were added into each bottle. After that, all bottles were into 37℃ incubator.

For each bottle, Five carriers were taken out each day and washed three times with PBS. We centrifuged solution, collected bacteria pellet and resuspended in 3mL of PBS, measured the absorbance in OD600. The absorbance reflects the number of bacteria on the carriers.

Figure 14: Amount of Bacteria on MBBR carriers measured for four days. Data represent mean ± SD of two independent experiments (n = 2)
Learn

In the control group without E.coli addition, the solution was clarifying, indicating that the experiment was not contaminated with miscellaneous bacteria. Bacteria numbers in carriers increased on the first two days, and decreased on the third day, illustrating that the optimal days number for biofilm culture was two days. The absorbance measured in the second day was 0.4455, representing there are 2.14×108 cells on a carriers.

However, we initially conducted six replicated experiments, and absorbance values of each replicates were highly different. The results are less convincing since two replicate experiments data were analyzed as the final result. Handling of biofilm washing and biofilm fragility could be potential causes of large experimental errors. Therefore, we remeasured the biofilm quantity on carriers using dry weight method.

Measurement by dry matter determination
Design

We then measured biofilm quantity through dry weight determination. We weighed carriers mass and mass difference was regarded as biofilm quantity growing on carriers.

Test

The total mass of thirty carriers was measured in advance. Cells culture, thirty MBBR carriers, AgNO3_3 (6μM) were added into each bottle and incubated in 37℃.

For each two days, dried carriers were weighed several times after cooling to room temperature, which ignores temperature effect on measurement. The difference in mass represents biofilm amount, thus biofilm quantity on carriers could be measured accurately.

Figure 15: Amount of Bacteria on MBBR carriers measured each two days. Data represent mean ± SD of three independent experiments (n = 3)
Figure 16: Biofilm of E. coli was incubated for four days under 6 μM AgNO3_3
Learn

The quantity of biofilm reached its maximum on the fourth day. After four days, biofilm quantity decreased and possible reasons might be nutrients shortage and cell growth inhibition by metabolite byproducts.

On the fourth day, we observed biofilm coated on carriers. Based on the results of dry weight determination, four days of biofilm incubation would be most suitable, and incubated biofilm could be used in subsequent experiments for heavy metal ions adsorption.


  1. 1.  X. Li, Z. Ren, M. J. C. Crabbe, L. Wang and W. Ma, Ecotoxicology and Environmental Safety, 2021, 222, 112512↩︎

  2. 2.  T. R. Butt, S. C. Edavettal, J. P. Hall and M. R. Mattern, Protein Expression and Purification, 2005, 43, 1–9. ↩︎

  3. 3.  W. Ma, X. Li, Q. Wang, Z. Ren, M. J. C. Crabbe and L. Wang, Ecotoxicology and Environmental Safety, 2019, 181, 301–307. ↩︎

  4. 4.  U. O. S. Seker and H. V. Demir, Molecules, 2011, 16, 1426–1451. ↩︎

  5. 5.  E. Lee, D.-H. Kim, Y. Woo, H.-G. Hur and Y. Lim, Biochemical and Biophysical Research Communications, 2008, 4. ↩︎

  6. 6.  R. S. Donovan, C. W. Robinson and B. R. Glick, Journal of Industrial Microbiology, 1996, 16, 145–154. ↩︎

  7. 7.  A. Barwal and R. Chaudhary, Rev Environ Sci Biotechnol, 2014, 13, 285–299 ↩︎