Team:XMU-China/Results
Verified that EGFP could be displayed on the surface of the cell by the function of BrkA.
Verified that EGFP could be displayed on the surface of the cell by the function of AIDA.
Verified that EGFP may be displayed on the surface of the cell by the function of INPNC.
Further certification of the expression and function of INPNC-EGFP.
Our detection system focuses on NADPH, which is widely found and involved in many biochemical reactions. In order to eliminate the interference of endogenous NADPH, we wanted to display our enzymes employed in the detection system on the surface of E. coli, more details are shown on the Design page. A comprehensive analysis of the characteristics of different kinds of anchor proteins was applied and we ultimately determine INPNC, BrkA, and AIDA for experimental implementation.
For the necessity of determining the anchoring ability of these three proteins, we fused anchor protein with EGFP and constructed three genes: inpnc-egfp (BBa_K3332061), egfp-aidA (BBa_K3332063 ) and egfp-brkA (BBa_K3332064 ). The gene circuits were subsequently constructed to express these three genes of fusion proteins (Fig. 1a, Fig. 2a and Fig. 3a), carried by the pSB1C3 vector.
Fig. 1 Gene circuit involved in protein expression and SDS-PAGE analysis of anchored EGFP by silver staining. a, Gene circuit was constructed by BBa_K3332061 to express INPNC-EGFP. b, Target bands of INPNC-EGFP (white arrow, 70.4 kDa).
Fig. 2 Gene circuit involved in protein expression and SDS-PAGE analysis of anchored EGFP by silver staining. a, Gene circuit was constructed by BBa_K3332064 to express EGFP-BrkA. b, Target bands of EGFP-BrkA (black arrow, 76.6 kDa).
Fig. 3 Gene circuit involved in protein expression and SDS-PAGE analysis of anchored EGFP by silver staining. a, Gene circuit was constructed by BBa_K3332063 to express EGFP-AIDA. b, Target bands of EGFP-AIDA (white arrow, 80 kDa).
Target protein from the extracted membrane protein was observed on the SDS-PAGE gel, showing that EGFP-BrkA, INPNC-EGFP and EGFP-AIDA were successfully expressed and anchored into the membrane (Fig. 1b, Fig. 2b and Fig. 3b).
Positive colonies were cultivated to express the target protein. Then we harvested bacteria by centrifugation. The harvested bacteria, the lysate underwent centrifugation and its supernatant, were photographed, and then the fluorescence changes were quantitatively detected via microplate reader (Fig. 4 and Fig. 5).
Fig. 4 Quantitative analysis of fluorescence intensity of EGFP-AIDA and EGFP-BrkA. a, bacteria of E. coli with EGFP-AIDA#3 (left), EGFP (middle) and Blank (right) (J23100-B0034). b, bacteria of E. coli with EGFP-AIDA#4 (left), EGFP (middle) and Blank (right) (J23100-B0034). c, bacteria of E. coli with EGFP-BrkA, EGFP, and Blank (right) (J23100-B0034). d, the lysate underwent centrifugation and its supernatant of E. coli with Blank (J23100-B0034) (left), EGFP-AIDA#3 (middle) and EGFP (right), e, the lysate underwent centrifugation and its supernatant of E. coli with Blank (J23100-B0034) (left), EGFP-AIDA#4 (middle) and EGFP (right). f, the lysate underwent centrifugation and its supernatant of E. coli with Blank (J23100-B0034) (left), EGFP-BrkA (middle) and EGFP (right). g, Quantitative detection of fluorescence intensity of the resuspension and supernatant of broken samples.
Fig. 5 Quantitative analysis of fluorescence intensity of INPNC-EGFP. a, bacteria of E. coli with Blank (J23100-B0034) (left), INPNC-EGFP (middle) and EGFP (right). b, the lysate underwent centrifugation and its supernatant of E. coli with Blank(J23100-B0034) (left), INPNC-EGFP (middle) and EGFP (right). c, Quantitative detection of fluorescence intensity of the resuspension and supernatant of broken samples.
By comparing the fluorescence intensity of samples shown on the photos, we could find that both the bacteria, the lysate underwent centrifugation and its supernatant of E. coli with J23100 had no fluorescence signal. While both the bacteria, the lysate underwent centrifugation and its supernatant of E. coli with EGFP had a strong fluorescence signal. Notably, after ultrasonication, the fluorescence signal was dispersed to both the supernatant and the precipitation, which meant the EGFP protein was normally expressed on the intracellular. However, samples of EGFP-fusion proteins had fluorescence signals most on the precipitation, which suggests that our anchor proteins successfully display EGFP on the surface of E. coli.
We further quantitatively detected fluorescence of the resuspension and supernatant of broken samples. And we found ratios of fluorescence intensity of the resuspension and supernatant had an obvious difference between experimental groups.
We used a confocal laser scanning microscope (CLSM) to further confirm the cell-surface display system successfully worked and the results are shown in Fig. 6.
Fig. 6 Confocal laser scanning microscope (CLSM) imaging. J23100-B0034 was a negative control and EGFP was a positive control. Scale bars, 5 μm.
E. coli BL21 (DE3) carrying BBa_E0040 (EGFP) was rod-shaped and the fluorescence was equably distributed in the bacteria. The fluorescence of E. coli BL21 (DE3) carrying BBa_K3332063 (EGFP-AIDA) or BBa_K3332064 (EGFP-BrkA) was observed to be dotted and dispersed on the surface of bacteria. The results proved that EGFP has been anchored to the surface of the E. coli by the function of AIDA or BrkA, which is very exciting.
However, the fluorescence of E. coli BL21 (DE3) carrying BBa_K3332061 (INPNC-EGFP) was observed to be equally distributed in the bacteria and rarely dispersed on the surface of E. coli. The results proved that EGFP might not be anchored to the surface of the E. coli by the function of INPNC. More certifications are needed for this part.
Assayed the enzymatic activities of the glyphosate oxidase GOX.
Verified that GOX could be displayed on the surface of the cell by the function of INPNC.
Verified that GOX could be displayed on the surface of the cell by the function of BrkA.
Verified that GOX could be displayed on the surface of the cell by the function of AIDA.
Assayed the enzymatic activities of the three fusion protein of GOX and anchor protein and found that INPNC-GOX has the best activity.
The kinetics of GOX.
In order to detect glyphosate, we selected three genes from different sources: gox (BBa_K3332006 ), grhpr (BBa_K3332011 ), and inap (BBa_K3332001 ). GOX and GRHPR can degrade glyphosate into glycolic acid while consuming NADPH as a cofactor, and iNap can bind NADPH and show apparent fluorescence changes.
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies were selected by chloramphenicol preliminarily and then verified by colony PCR. The colonies, confirmed by sequencing finally, were cultivated to express target protein. And the total protein was gained from the supernatant after ultrasonication and centrifugation. GE AKTA Prime Plus FPLC System was used to harvest purified protein from the broken supernatant. An apparent protein peak in the AKTA FPLC System demonstrated the correct purified protein. Purified protein was verified by sodium dodecyl sulfate (SDS)-12% (wt/vol) polyacrylamide gel electrophoresis, and Coomassie blue staining (Fig. 7).
Fig 7. Gene circuit involved in protein expression and SDS-PAGE analysis of GOX by Coomassie blue staining. a, Gene circuit was constructed by BBa_K880005 and BBa_K3332006 to express GOX. b, Target bands of GOX (pink arrow, 34.6 kDa).
As shown in SDS-PAGE of GOX, the target protein could be observed at 34.6 kDa on the purified protein lanes and broken supernatant lanes, but not in the control groups. These results demonstrated that the GOX protein with His-Tag was successfully expressed in E. coli BL21 (DE3) and purified using GE AKTA Prime Plus FPLC System.
After transforming glyphosate into glyoxylic acid through GOX, we employed GRHPR, a glyoxylate reductase from the human liver, to reduce glyoxylic acid with NADPH consumption. NADPH was easy to be detected by the signal of OD340.
GE AKTA Prime Plus FPLC System was used to purify GOX and GRHPR protein. Then, glyphosate solution, NADPH solution, and GOX & GRHPR purified protein dissolved in Tris-HCl (pH 7.5) were mixed to detect the changes of OD340 immediately using TECAN® Infinite M200 Pro to investigated the performance of this system. (Fig. 8)
Fig. 8 OD340-Time curve of GOX & GRHPR and negative control.
As shown in Fig. 8, OD340 decreased as time went on in the group of GOX & GRHPR, while no decrease was observed in the negative control group (using GRHPR only). These results proved that GOX catalyzed the degradation reaction of glyphosate.
In order to display our protein on the surface of E. coli, GOX, GRHPR, and iNap with three types of anchor protein were fused (Figure 9a). They were inserted into the expression vectors with J23100 and B0034 (BBa_K880005 ) to obtain composite parts. Then, these recombinant plasmids were transformed into E. coli DH5α, and the correct recombinant one was confirmed by chloramphenicol, PCR or restriction enzyme digestion, and sequencing (Fig. 9b-i).
PCR and enzyme-digested products were purified by agarose gel electrophoresis (Fig. 9b-d). Due to the limited time, we just successfully constructed eight fusion protein designed.
Fig. 9 Gene circuit involved in protein expression and agarose gel electrophoresis of PCR or enzyme-digested product. Lane M: Marker. a, Gene circuits of enzymes used in our detection system fused with anchor proteins. b, the target band of inpnc-gox. c, the target band of gox-aidA. d, the target band of gox-brkA. e, the target band of inpnc-grhpr. f, the target band of grhpr-aidA. g, the target band of grhpr-brkA. h, the target band of inpnc-inap. i, the target band of inap-aidA.
After centrifugation, the harvested cells were resuspended and treated by membrane protein extraction kit to extract membrane protein. Membrane protein was verified by SDS-PAGE and silver staining (Fig. 10).
Fig. 10 Gene circuit involved in protein expression and SDS-PAGE analysis of INPNC-GOX by silver staining. a, Gene circuit was constructed (BBa_K2922052) to express INPNC-GOX. b, Target bands of INPNC-GOX (black arrow, 68.8 kDa).
Target protein from the membrane protein was observed through SDS-PAGE, which indicated that INPNC-GOX was successfully expressed and anchored into the membrane. Due to the limited time, the proof of expression of GOX-AIDA and GOX-BrkA was not finished finally, which was a great regret for us indeed.
After mixed the glyphosate solution, NADPH solution, purified GRHPR protein dissolved in Tris-HCl (pH 7.5), and bacteria solution carrying GOX fused with anchor protein, OD340 was immediately detected using TECAN® Infinite M200 Pro to investigate the performance of GOX fused with anchor protein (Fig. 11).
Fig. 11 OD340-Time curve of three fusion proteins of GOX and negative control.
As shown in Fig. 11, OD340 decreased as time went on, when using bacteria solution carrying three types of anchored GOX. The OD340 decreased rate of J23100-B0034 and GOX-His samples was less than that of the INPNC-GOX samples. The results proved that INPNC-GOX, GOX-AIDA, and GOX-BrkA have been displayed on the surface and convert glyphosate as expected.
Assayed the enzymatic activities of the glyoxylate reductase GRHPR.
Assayed the kinetics of GOX.
Verified that GRHPR could be displayed on the surface of the cell by the function of INPNC.
Verified that GRHPR could be displayed on the surface of the cell by the function of BrkA.
Verified that GRHPR could be displayed on the surface of the cell by the function of AIDA.
Assayed the enzymatic activities of the three fusion protein of GRHPR and anchor protein and found that INPNC-GRHPR has the best activity.
GRHPR can convert glyoxylic acid into glycolic acid while NADPH is consumed as a cofactor. So, we constructed BBa_K3332056 which consists of BBa_K880005 and BBa_K3332011 .
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies were selected by chloramphenicol preliminarily and verified by colony PCR. The colonies, confirmed by sequencing finally, were cultivated to express target protein. And the total protein was gained from the supernatant after ultrasonication. GE AKTA Prime Plus FPLC System was used to harvest purified protein from the broken supernatant. An apparent protein peak in the AKTA FPLC System demonstrated the correct purified protein. A purified protein was verified by the sodium dodecyl sulfate (SDS)-12% (wt/vol) polyacrylamide gel electrophoresis and Coomassie blue staining (Fig. 12).
Fig 12. Gene circuit involved in protein expression and SDS-PAGE analysis of GRHPR by Coomassie blue staining. a, Gene circuit was constructed by BBa_K880005 and BBa_K3332011 to express GRHPR. b, Target bands of GRHPR (blue arrow, 36.7 kDa).
As showed by SDS-PAGE of GRHPR, the target protein was observed at 36.7 kDa on the purified protein lanes and broken supernatant lanes, but not in the control groups. The results showed that the GRHPR protein with His-Tag was successfully expressed in E. coli BL21 (DE3) and was successfully purified using GE AKTA Prime Plus FPLC System.
GRHPR was a glyoxylate reductase from the human liver, which reduce glyoxylic acid with NADPH consuming. NADPH was easy to be detected by the signal of OD340.
GE AKTA Prime Plus FPLC System was used to get purified GRHPR protein. After mixed Glyoxylic acid solution, NADPH solution, Tris-HCl, and GRHPR purified protein dissolved in Tris-HCl (pH 7.5), OD340 was immediately detected using TECAN® Infinite M200 Pro to investigate the performance of GRHPR (Fig. 13).
Fig.13 OD340-Time curve of GRHPR and negative control.
As shown in Fig. 13, OD340 of samples adding GRHPR decreased very quickly while the OD340 of control stay almost the same. These results proved that GRHPR works very well.
In order to display our protein on the surface of E. coli, GOX, GRHPR, and iNap with three types of anchor protein were fused. Then recombinant plasmids were transformed into E. coli DH5α, and the correct recombinant one was confirmed by chloramphenicol, PCR or restriction enzyme digestion, and sequencing.
PCR and enzyme-digested products were purified by agarose gel electrophoresis (Fig. 9e-g). INPNC-GRHPR, GRHPR-AIDA, and GRHPR-BrkA were successfully constructed.
After centrifugation, the harvested cells were resuspended and treated by membrane protein extraction kit to extract membrane protein. Membrane protein was verified by SDS-PAGE and silver staining (Fig. 14).
Fig 14. Gene circuit involved in protein expression and SDS-PAGE analysis of GRHPR-BrkA by silver staining. a, Gene circuit was constructed (BBa_K2922060 ) to express GRHPR-BrkA. b, Target bands of GRHPR-BrkA (Black arrows, 85.4 kDa).
The target protein from the membrane protein was observed through the SDS-PAGE gel, which indicated that GRHPR-BrkA was successfully expressed and anchored into the membrane. Due to the limited time, the proof of expression of INPNC-GRHPR and GRHPR-AIDA was not finished finally, which was a great regret for us indeed.
After mixed glyoxylic acid solution, NADPH solution, and bacteria solution carrying GRHPR fused with anchor protein, OD340 changes were immediately measured. When NADPH is consumed, OD340 declines (Fig. 15).
Fig. 15 OD340-Time curves of GRHPR fused with 3 types of anchor protein.
We successfully got OD340-Time curves of GRHPR fused with 3 types of anchor protein. When using GRHPR fused with anchor protein, the decrease of OD340 was observed as the reaction went on. Compared with J23100-B0034 and GRHPR-His-Tag, great changes took place in the slope of the experimental group, especially for INPNC-GRHPR, which means that fusion protein is displayed on the surface and works very well.
Notably, the enzyme activity of INPNC-GRHPR was the best one of them.
Assayed the enzymatic activities of the NADPH sensor iNap and got the working curve in the presence of NADPH.
Verified that iNap could be displayed on the surface of the cell by the function of INPNC.
Verified that iNap could be displayed on the surface of the cell by the function of AIDA.
Assayed the enzymatic activities of the two fusion protein of iNap and anchor protein and found that INPNC-iNap has better activity.
iNap-BrkA has not been got.
iNap is a chimera of circularly permuted YFP (cpYFP) and the NADP(H) binding domain of Rex from Thermus aquaticus (T-Rex). It has an excitation peak near 420 nm and an emission peak near 528 nm. After NADPH binding, iNap showed apparent fluorescence changes. So, we constructed BBa_K3332046 which consists of BBa_K880005 and BBa_K3332001.
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies were cultivated to express the target protein. And the total protein from the supernatant after ultrasonication and centrifugation. GE AKTA Prime Plus FPLC System was used to harvest purified protein from the broken supernatant. An apparent protein peak in the AKTA FPLC System demonstrated the correct purification of His-Tag-iNap protein. Purified protein was verified by sodium dodecyl sulfate (SDS)-12% (wt/vol) polyacrylamide gel electrophoresis and Coomassie blue staining (Fig. 16).
Fig.16 Gene circuit involved in protein expression and SDS-PAGE analysis of GRHPR by Coomassie blue staining. a, Gene circuit was constructed by BBa_K880005 and BBa_K3332001 to express iNap. b, Target bands of iNap (black arrow, 42.2 kDa).
As shown in SDS-PAGE of GRHPR, the target protein can be observed at 42.2 kDa on the purified protein lanes and broken supernatant lanes, but not in the control groups. The results showed that the iNap with His-Tag was successfully expressed in E. coli BL21 (DE3) and purified.
We mixed different concentrations of NADPH half-in-half with purified iNap protein dissolved in Tris-HCl (pH 7.5) and measure the fluorescence changes immediately. TECAN® Infinite M200 Pro was used to detect fluorescence intensity. The samples were excited at 420 nm, and the emission was measured at 528 nm (Fig. 17).
Fig. 17 iNap fluorescence intensity which was excited at 420 nm in the presence of different concentrations of NADPH. Emission was measured at 528 nm.
We successfully got a working curve of iNap in the presence of different concentrations of NADPH, which indicated that iNap is a sensitive detector of NADPH.
In order to display our protein on the surface of E. coli, GOX, GRHPR, and iNap with three types of anchor protein were fused. Then constructed plasmids were transformed into E. coli DH5α, and the correct recombinant one was confirmed by chloramphenicol, restriction enzyme digestion, and sequencing.
Enzyme-digested products were purified by agarose gel electrophoresis (Fig. 9h-i). INPNC-iNap and iNap-AIDA were successfully constructed.
After culturing the engineering bacteria to OD600 = 1.8~2.0, E. coli with INPNC-iNap or iNap-AIDA were obtained by centrifuging at 4000 rpm. Then, the cell precipitation was washed three times with PBS buffer (pH 8.0), and the final precipitation was resuspended in PBS buffer, which was equal to the volume of the original medium.
NADPH solutions were mixed with washed INPNC-iNap cell precipitation dissolved in PBS buffer half-in-half (pH 8.0), and fluorescence changes were measured in the presence of different concentrations of NADPH. TECAN® Infinite M200 Pro was used to detect fluorescence intensity (Fig. 18).
Fig. 18 a, Fluorescence Intensity-Time curve of INPNC-iNap and negative control in the presence of different NADPH concentrations. b, Fluorescence Intensity-Time curve of iNap-AIDA, and negative control in the presence of different NADPH concentrations.
When using engineered bacteria expressing INPNC-iNap or iNap AIDA, we successfully got fluorescence intensity gradient in a certain concentration range of NADPH. However, when using E. coli carrying BBa_K880005 (negative control in this case), no significant gradient could be observed through a vast range of NADPH’s concentration. In conclusion, our anchor proteins selected could work well and iNap could sense extracellular NADPH even displayed on the surface of bacteria.
Taken the experiments’ results above together, GOX, GRHPR, and iNap could work effectively and some of our enzymes in fusion protein-form with different anchor proteins were well characterized anyway. In view of all of this, we finally completed the design and construction of the ultimate gene circuits and some necessary characterization works in advance of demonstration.
Verified that PhnE1E2 can obviously enhance the transport capability of chassis bacteria to glyphosate.
Verify the function of PhnE1 and PhnE2 separately.
We use the phnE1 and phnE2 gene from Sinorhizobium meliloti 1021 to enhance the ability of our chassis bacteria to transport glyphosate. We expressed the two genes in a polycistron way on the pSB1C3 vector (Fig. 19).
Fig. 19 Gene circuit of phnE1 and phnE2.
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies that were selected by chloramphenicol preliminarily and then by colony PCR, restriction enzyme digestion (Fig. 20) and sequencing were cultivated to express target protein. And the total protein was gained by ultrasonication. The supernatant of culture and cells was separated by centrifugation. The harvested cells were then resuspended, treated by membrane protein extraction kit and subsequently got membrane protein, followed by SDS-PAGE and silver staining (Fig. 21). Notably, the bands of PhnE1 and PhnE2 were a little bit obscure. We attributed this to the difficulty of membrane protein’s expression in the engineered bacteria because the exogenous expression of membrane protein might change the permeability of membrane, to some extent, which seemed to be slightly poisonous.
Fig. 20 The agarose gel electrophoresis of enzyme-digested product (EcoR I & Pst I).
Fig. 21 SDS-PAGE analysis of PhnE1 and PhnE2 by silver staining.
After verifying that the two proteins can be expressed, we used HPLC to verify the effect of PhnE1 and PhnE2 on the transport ability of glyphosate by chassis bacteria. After being cultured in LB medium containing glyphosate for a certain time, the content of glyphosate in the medium was determined by HPLC. The amount of glyphosate was represented by peak area. We found that compared with the blank control, the transformed bacteria could absorb more glyphosate (Fig. 22).
Fig. 22 Relationship between the peak area of glyphosate and culture time.
Verified the relative inhibit effect to target genes of phnJ 0.94, phnJ 0.69, phnF 0.97 and phnF 0.75.
Verified the relative inhibit effect to target genes of phnF 0.97-phnJ 0.94 and phnF 0.97-phnJ 0.75.
Verify the relative inhibit effect to target genes of phnF 0.97-phnJ 0.94 and phnF 0.97-phnJ 0.75.
Verify the relative inhibit effect to target genes of each RNAi sequence with different promoters.
Verify the degradation activity of phnJ mut45.
We designed siRNA sequences through siRCon which is developed by Team Bielefeld-CeBiTec and used them to hinder endogenous phnF and phnJ gene from expression. OmpA 5’ UTR and Hfq binding sequence are added to the flanks of siRNA sequences to enhance the binding its ability with target sequence as well as increase the half-life of siRNA. Promoter BBa_J23101 and terminator BBa_J61048 were used to transcribe the siRNA sequences on the pSB1C3 vector (Fig. 23).
Fig. 23 Gene circuit of RNAi sequences.
Due to sequencing errors of some sequences, we finally chose phnF 0.97, phnJ 0.94 and phnJ 0.75 for subsequent experiments. We assembled the RNAi sequence of phnF 0.97 with phnJ 0.94 and phnJ 0.75 separately, each of them was transcribed by promoter BBa_J23101 and terminator BBa_J61048 on the pSB1C3 vector (Fig. 24).
Fig. 24 Gene circuit of final RNAi sequences we used.
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies that were selected by chloramphenicol preliminarily and then by colony PCR, restriction enzyme digestion (Fig. 25) and sequencing were cultivated to extract total RNA and then verified the hindering rate of RNAi sequences to target genes by reverse transcription and qPCR (Fig. 26).
Fig. 25 The agarose gel electrophoresis of enzyme-digested product (EcoR I & Pst I).
Fig. 26 The transcription rate of phnJ and phnF verified by reverse transcription and qPCR.
It can be seen that the RNAi sequences are able to downregulate the expression of target genes, and the relative hindering rate was consistent with the prediction given by the siRCon.
The crack of C-P bond is a crucial step during the degradation process of glyphosate, where PhnJ protein is an essential subunit from C-P lyase that can crack C-P bond. Therefore, the phnJ gene from Enterobacterales was used to enhance the capability of our chassis bacteria to degrade glyphosate. We use promoter BBa_J23100 , RBS BBa_B0034 and terminator BBa_B0015 to express the exogenous phnJ on the pSB1C3 vector (Fig. 27).
Fig. 27 Gene circuit of phnJ.
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies that were selected by chloramphenicol preliminarily and then by colony PCR, restriction enzyme digestion (Fig. 28) and sequencing were cultivated to express target protein. However, there are too many intracellular proteins with similar molecular weight to PhnJ that we couldn’t distinguish the target band on the polyacrylamide gel after staining. We also tried to purify the protein PhnJ-His-Tag with GE AKTA Prime Plus FPLC System, but due to the special structure of PhnJ, the His-Tag will be wrapped inside the protein so the protein couldn’t be ‘captured’ by the column. Finally, we did not get the SDS-PAGE results of PhnJ.
Fig. 28 The agarose gel electrophoresis of enzyme-digested product (EcoR I & Pst I).
Based on our modeling results, we designed three mutated phnJ sequences to further improve the glyphosate degradation ability of chassis bacteria (Fig. 29). In the first sequence, the 21st arginine was mutated to methionine; in the second sequence, the 16th threonine was mutated to serine, and the 40th arginine was mutated tyrosine; in the third sequence, the 45th proline was mutated to glutamine.
Fig. 29 Gene circuits of three mutated phnJ sequences.
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies that were selected by chloramphenicol preliminarily and then by sequencing were cultivated to express target protein. For the same reason, we did not get the SDS-PAGE results of PhnJ mutated sequences. We didn’t perform colony PCR and restriction enzyme digestion since these methods could not verify whether the mutation was successful or not.
Verify that PhnO can obviously enhance the degradation capability of chassis bacteria to AMPA.
None.
We use the phnO gene from Salmonella enterica LT2 to enhance the ability of our chassis bacteria to degrade aminomethylphosphoric acid (AMPA), which is the main metabolites of glyphosate. We expressed the proteins with promoter BBa_J23100 and RBS BBa_B0034 on the pSB1C3 plasmid backbone, BBa_B0015 was used as a terminator (Fig. 30).
Fig. 30 Gene circuit of phnO.
The constructed plasmid was transformed into E. coli BL21 (DE3). Positive colonies that were selected by chloramphenicol preliminarily and then by colony PCR, restriction enzyme digestion (Fig. 31) and sequencing were cultivated to express target protein. And the total protein was gained by ultrasonication. We used GE AKTA Prime Plus FPLC System to get purified protein from broken sup and successfully corrected purified protein. Purified protein was electrophoresed on a sodium dodecyl sulfate (SDS)-12% (wt/vol) polyacrylamide gel, followed by Coomassie blue staining (Fig. 32).
Fig. 31 The agarose gel electrophoresis of enzyme-digested product (EcoR I & Pst I).
Fig. 32 SDS-PAGE analysis of PhnO by Coomassie blue staining.
Velocities for the reaction of PhnO were determined by using 4,4-dithiodipyridine (DTDP) to continuously detect the formation of the product CoA at 324 nm (thiopyridone: ε = 19800 M-1·cm-1) at 25 °C. Enzyme activity was determined through the microplate reader.
Km and kcat of different substrates were determined with a concentration gradient (Fig. 33, Table 1).
Fig. 33 The relationship of 1/enzyme activity and 1/concentration of acetyl-CoA (A) and AMPA (B).
| Compound | Km (mM) | kcat (s-1) |
|---|---|---|
| acetyl-CoA | 0.1986 | 17.1394 |
| AMPA | 0.5637 | 3.2248 |
Table 1. Enzyme kinetic constants with two substrates.
In conclusion, the experimental result of phnE1E2, RNAi and phnO works well respectively. While phnJ was just verified in ultimate gene circuits (please seehttps://2020.igem.org/Team:XMU-China/Proof_Of_Concept ). These works ensured our following experiments of characterization of ultimate gene circuits of degradation system.
Fig.34 Gene circuit of kill switch in detection system (BBa_K3332081).
This kill switch is mainly composed of three parts: pBAD/araC (BBa_I0500), inverter (BBa_Q04510) and mazF (BBa_K1096002). When the E. coli is in our detection device where L-arabinose exists, the toxin gene mazF won't express and the E. coli won’t die. As long as the E. coli escape from our device to somewhere without L-arabinose, the mazF will express and the E. coli will be killed. More details could be seen in the Design page.
Verified that pBAD/araC can express downstream genes when induced by L-arabinose.
Verified that inverter has a reverse induction effect.
Verified that MazF have the function of killing E. coli.
Assayed the killing effect of kill switch in detection system (BBa_K3332081).
Given the use of L-arabinose inducible promoter involved in the kill switch (Fig. 34) and the purpose to characterize the function of inverter which is also employed in the kill switch in Fig. 34, we use pBAD/araC (BBa_I0500), inverter (BBa_Q04510) and B0034-eyfp-B0015 (BBa_E0430) to construct the following circuits:
Fig. 35 Construction of gene circuits involved in characterization of inverter. a, Comparing circuit (BBa_K3332082) of L-arabinose inducible promoter, pBAD/araC. b, Characterization circuit for inverter (BBa_K3332079). c, DNA gel electrophoresis of BBa_K3332082 (vector: pSB1C3, digested by Spe I and Pst I). d, DNA gel electrophoresis of BBa_K3332079 (vector: pSB1C3, digested by EcoR I and Pst I).
eyfp was designed in the downstream of pBAD/araC (Fig. 35a and 35c) to be set as a comparison of characterization circuit for inverter (Fig. 35b and 35d). By comparing the fluorescence intensity of induction group and non-induction group, the function of inverter can be characterized.
0.2% L-arabinose solution was added to induce for 6 hours, then the fluorescence intensity and corresponding OD600 of two groups was measured.
Fig. 36 Fluorescence intensity/OD600 for induction group and non-induction group (6 hours). Data are collected and analyzed according to iGEM standard data analysis form.
The fluorescence intensity indicates the expression level of eyfp. The result shows an extremely high fluorescence intensity of the characterization circuit without inducer added and a much lower fluorescence intensity after inducing. It demonstrated that the inverter did actually convert the input inducible signal of L-arabinose to the shut-down of the downstream gene’s (eyfp in this case) expression (Fig. 36). While in the control group, which has no inverter constructed in it, had an opposite behavior to the characterization circuit. Taken together, the function of inverter was characterized and this part could work well to support subsequent constructions and applications.
We use pBAD/araC (BBa_I0500), RBS (BBa_B0034 ), Terminator (BBa_B0015 ), mazF (BBa_K1096002) to construct the following circuit (Fig. 37), BBa_K3332083 , to verify the MazF's function to kill the bacteria carrying the mazF expression plasmid.
Fig.37. Construction of characterization circuit for MazF. a, Gene circuit involved in characterize the function of MazF (BBa_K3332083 ). b, DNA gel electrophoresis of BBa_K3332083 (vector: pSB1C3, colony PCR).
CFU assay was implemented to characterize the effect of toxin MazF (Fig. 38a). We discovered that the number of colonies decreased significantly (about 2-fold) after 5 hours, while there was no significant change in the non-induced group (Fig. 38b). This indicates that the mazF’s expression is successfully induced and toxin MazF could be lethal to the bacteria which has just toxin gene expressed but no antitoxin gene expressed/in them.
Fig. 38 CFU assay for characterizing the function of MazF. a, Solid medium plates of CFU assay, where plates marked “A” belonged to the non-induction group while those marked “B” belonged to the induction group. b, CFU (per mL liquid culture) calculated of different groups are plotted against time (h).
We use pBAD/araC (BBa_I0500), inverter (BBa_Q04510), RBS (BBa_B0034 ), Terminator (BBa_B0015 ), mazF (BBa_K1096002) to construct the following circuit:
Fig. 39 Construction of gene circuit of kill switch in detection system. a, Gene circuit of kill switch in detection system. b, DNA gel electrophoresis of BBa_K3332081 (vector: pSB1C3, colony PCR).
CFU assay was also carried out to characterize the effect of kill switch (Fig. 40a). We discovered that the number of colonies of E. coli carrying BBa_K3332081 (without induction) and pBAD/araC-mazF (induction) decreased significantly after 8 hours, while there was no significant change in pBAD/araC-inverter-mazF (induction) and pBAD/araC-mazF (without induction) groups (Fig. 40b), which could verify that BBa_K3332081 has the killing function without L-arabinose. In conclusion, the kill switch (BBa_K3332081) in detection system did work to prevent engineered E. coli escaping and could achieve our purpose of bio-containment.
Fig. 40 CFU assay for characterizing the killing effect of kill switch in detection system. a, Solid medium plates of CFU assay, where plates marked “C” belonged to the non-induction group while those marked “D” belonged to the induction group. b, CFU (per mL liquid culture) calculated of different groups are plotted against time (h).
In addition to the above design, we also came up with other kill switch schemes during the brainstorming, which were not well characterized due to the limited time.
“Deadman” kill Switch:
Fig. 41 “Deadman” kill switch.
In this circuit, LacI can repress pTrc-2 promoter and pTrc-2 derivative promoter while TetR can repress pLtetO-1 promoter. When ATc (anhydrotetracyclin) exits, the pLtetO-1 promoter can’t be repressed due to the combination of ATc and TetR repressor. Then LacI which is expressed by pLtetO-1 can repress pTrc-2 promoter and pTrc-2 derivative promoter. As a result, mf-lon and mazF can’t be expressed. As a kind of E. coli toxin, MazF can cause the E. coli death. So there comes the conclusion that as long as the engineered E. coli are cultured in the environment with ATc, it won’t be killed by the function of MazF, but when the E. coli escape from our detection device, the effect can be reversed, that is to say, the E. coli will be killed by the MazF. In the same way, we can conclude that at the presence of IPTG, mazF can be expressed to cause E. coli death.
Verified that pLtetO-1 promoter could be repressed by TetR protein, which can be inhibited by ATc.
Verified that pTrc-2 promoter and pTrc-2 derivative promoter could be repressed by LacI protein, which can be inhibited by IPTG.
Verify that mf-lon hydrolases has the function of recognizing and degrading the genes with a corresponding degron.
Assay the killing effect of Deadman kill switch.
We use pLtetO-1 (BBa_K3332034), B0034-ecfp-B0015 (BBa_E0420), BBa_J23106 and BBa_P0140 to construct the following circuits in Fig. 42.
Fig. 42 Construction of gene circuits for verifying the function of pLtetO-1 promoter and TetR repressor. a, Gene circuit (BBa_K3332084) for characterizing pLtetO-1 promoter. b, Gene circuit (BBa_K3332085) for verifying the function of TetR expressed by tetR gene. c, DNA gel electrophoresis of BBa_K3332084 (vector: pSB1C3 or pUC57, digested by EcoR I and Pst I). d, DNA gel electrophoresis of BBa_K3332085 (vector: pSB1C3, digested by Spe I and Pst I).
100 ng/mL ATc solution was added to induce for 6 hours, then measure the fluorescence intensity and corresponding OD600 of two groups of induction and non-induction (Fig. 43). There is no significant difference of fluorescence intensity/OD600 between the induction group and non-induction group of BBa_K3332084, indicating the constitutive expression of pLtetO-1 promoter when no TetR exists in the system. And the fluorescence intensity/OD600 of the non-induction group of BBa_K3332085 is rather low, compared to that of the induction group of this circuit, revealing that the gene tetR and its product TetR repressor could function well under the circumstance of the absence of ATc. Besides, the repressed pLtetO-1 promoter shows a dramatic change in expression level after induction, which presents an excellent inducibility for further utilization.
Fig. 43 Fluorescence intensity/OD600 for induction and non-induction group (6 hours). Data are collected and analyzed according to iGEM standard data analysis form.
We use pTrc-2 promoter (BBa_K3332038), pTrc-2 derivative promoter (BBa_K3332040), B0034-ecfp-B0015 (BBa_E0420), pLtetO-1 (BBa_K3332034), RBS1 (BBa_K3332036 ) and LacI-ssrAtag (mf-lon) (BBa_K3332037 ), Terminator (BBa_B0015 ) to construct the following circuits (Fig. 44):
Fig. 44 Construction of gene circuits for characterizing pTrc-2 promoter and pTrc-2 derivative promoter. Gene circuits: a, BBa_K3332087; b, BBa_K3332088; c, BBa_K3332089. DNA gel electrophoresis: d, BBa_K3332087 (vector: pUC57, digested by EcoR I and Pst I); e, BBa_K3332088 (vector: pSB1C3, digested by Pst I); f, BBa_K3332089 (vector: pUC57, digested by and Pst I).
We added IPTG solution to induce for 6 hours, then measure the fluorescence intensity and corresponding OD600 of four groups. We can see, after adding IPTG to induce the two promoters, the fluorescence intensity/OD600 of BBa_K3332088 and BBa_K3332089 both increased. The change of fluorescence intensity/OD600 after induction of pLtetO-1-lacI-pTrc-2-E0420 (ECFP) group (BBa_K3332088) is larger than that of pLtetO-1-lacI-pTrc-2 derivative-E0420 (ECFP) group (BBa_K3332089) (Fig. 45), so it can confirm us that the LacI has a weak inhibitory effect on pTrc-2 promoter and a strong inhibitory effect on pTrc-2 derivative promoter.
Fig. 45 Characterization of pTrc-2 and pTrc-2 derivative promoters. a, photographs of liquid culture of the four groups. b, Fluorescence intensity/OD600 for induction and non-induction groups (6 hours). Data are collected and analyzed according to iGEM standard data analysis form.
Fig. 46 Gene circuit of kill switch in degradation system (BBa_K3332078).
Formaldehyde promoter (pHCHO) provides an inducible scheme for downstream genes. Inverter plays the role of gene inversion which can reverse induction effect when existing. mazF is designed to follow formaldehyde promoter and inverter, which makes sense of our kill switch system in degradation: Formaldehyde promoter can be activated after formaldehyde induction and expression of mazF is blocked by reversion effect of inverter. In contrast, without formaldehyde, mazF can express under pR promoter inside inverter and cause cell death by MazF’s protein toxicity.
Verified that formaldehyde promoter could be induced by formaldehyde.
Verified that inverter has a reverse induction effect.
Verified that MazF could kill E. coli.
Construct the total circuit and then verify its function.
We use formaldehyde promoter (BBa_K1334002), B0034_ecfp_B0015 (BBa_E0420) to construct the following circuit, BBa_K3332090 (Fig. 47a), which is used to characterize formaldehyde promoter.
Fig. 47 Construction of gene circuit for characterizing formaldehyde promoter. a,Gene circuitBBa_K3332090. b, DNA gel electrophoresis of BBa_K3332090 (vector: pSB1C3, digested by EcoR I and Pst I).
Because the formaldehyde (HCHO) is poisonous to E. coli, the growth curve of E. coli in different concentrations of formaldehyde was tested. Different concentration gradients of formaldehyde were added into the culture environment of E. coli and the growth curves were observed and recorded (Fig. 48).
Fig.48 BBa_K3332090 (ProFormaldehyde-ECFP) groups and BBa_K1334002 (ProFormaldehyde-pSB1C3) groups induced by different concentrations of formaldehyde.
After induction, BBa_K3332090 groups (ProFormaldehyde-ECFP) had a generally higher fluorescence intensity/OD600 than that of BBa_K1334002 (ProFormaldehyde-pSB1C3) groups (Fig. 48), which indicates that formaldehyde promoter could work well. However, according to the growth curves recorded, the toxicity of formaldehyde is not obvious.
We have designed different improvement versions of formaldehyde promoters, which were characterized together with the original formaldehyde promoter (BBa_K1334002). More details could be seen in the Improve page.
Refer to the above (kill switch in the detection system).
Refer to the above (kill switch in the degradation system).
Fig. 49 Alternative kill switch in degradation system.
PO promoter (pPO) and delta protein are used to realize the gradually increasing expression of toxin after ATc induction: when engineered E. coli are applied in soil, ATc can induce Tet promoter (pTet) to produce delta protein which can activate PO promoter. As a result, pPO could express delta protein and the toxin protein MazF, among which delta protein could continue activating pPO and generating more and more delta and toxin protein (forming a positive feedback). At the same time, the antitoxin MazE could be expressed by placing different constitutive promoters which have different expression levels in front of mazE. So the concentration of intracellular MazE is relatively constant, but as long as adding ATc, the kill switch will be kicked off, which means the concentration of MazF will increase gradually. When the amount of toxin exceeds the amount of antitoxin, the E. coli will be killed. By adding ATc before putting our engineered E. coli in the tea garden, the E. coli will be dead after a certain time which can be controlled by selecting different constitutive promoters for expressing antitoxin MazE.
Due to the limit of time, although we have these two mature kill switch designs for our degradation system, we choose the former as our main design. But the latter one inspired CSU and helped their modeling part.
In the sum of the experimental results above altogether, for detection system, the kill switch can effectively prevent engineered bacteria from escaping. Although the alternative has not been constructed completely, the three promoters: pTrc-2, pTrc2-derivative, pLtetO-1 of alternative were successfully characterized. For degradation system, the function of each part in the circuit of kill switch: formaldehyde promoter, MazF and inverter was verified but whether the total circuit could work or not hasn’t been testified yet. Meanwhile, the alternative of kill switch in degradation system still remains in the theory inquisition stage, which also inspired the modeling part of CSU.
Xiamen University, Fujian, China
No.422, Siming South Road, Fujian, P.R.China 361005
