Results

Introduction：

We aim to construct an bioremediation system that utilizes E. coli to secrete PETase and MHETase for microplastic PET degradation and to perform primary enrichment of microplastic degradation product TPA, followed by secondary enrichment by Caenorhabditis elegans (C. elegans), while relying on C. elegans to evaluate the physiological toxicity of each component of the degradation process. For this purpose, we designed five plasmids, pSec-PETase, pSec-MHETase, pTransporter, and pBiosensor, which were transformed into E. coli, and pTransporter-elegans, which was transformed into C. elegans.

We transformed these plasmids into E. coli and C. elegans for expression of our target proteins. Then we will verify that our proteins are properly expressed and perform their specific functions through a series of experiments.

Part 1：Plasmid construction

The plasmid containing the target gene was synthesized by company, and primers with homology arms were designed to amplify the target gene from the corresponding plasmid by PCR. Next, we combined the parts and linear plasmid vectors through homologous recombination to form a circular plasmid, and then the constructed plasmid was transformed into E. coli Trans-T1 strain. The transformed bacterial were selected by antibiotic and we designed sequencing primers for sequencing the plasmids to confirm that we obtained expected plasmids with correct sequences, and the successfully constructed plasmids will be used for subsequent characterization experiments.

1.1 Construction of pSec-PETase and pSec-MHETase

We constructed plasmids that can secrete PETase and MHETase into the extracellular space of E. coli and verified that the plasmids were successfully constructed and transformed . pSec-PETase contains TurboPETase, signaling peptides (OmpA, PleB, MalE, LamB) that allow TurboPETase to be secreted extracellularly, and an enhancer that enhances the effect of the signaling peptides. pSec-MHETase contains MHETase and signaling peptides that allow MHETase to be secreted extracellularly (PelB, Amcut).

Analysis:The PCR amplification bands of LamB (approximately 110bp), PelB (approximately 100bp), MalE (approximately 110bp), OmpA (approximately 100bp), pET-28a vector (approximately 5000bp), and TurboPETase (approximately 750bp) align with the theoretical lengths estimated based on the designed primer positions, demonstrating that these target genes have been successfully obtained.

Analysis:The PCR amplification bands of MHETase (approximately 500bp), Amcut (approximately 150bp), PelB (approximately 100bp), pET-21a vector (approximately 5000bp) align with the theoretical lengths estimated based on the designed primer positions, demonstrating that these target genes have been successfully obtained.

Analysis: From the result of plate coating of LB solid medium with antibiotic Amp/Kan, the medium successfully grew dispersed single colonies, which indicated that the bacterium was successfully screened and transformed.

Analysis:The sequencing result can show that the target element and the plasmid vector were correctly connected; there were no heterogeneous peaks in the sequencing result (sequencing sequences were more accurate in the range of 200-800bp), and the sequence of the actual plasmid is identical with the designed plasmid.

1.2 Construction of pTransporter coli and pBiosensor

We constructed two plasmids pTransporter and pBiosensor, then we transformed them into one E. coli BL21 strain to apply the detection function. pTransporter can transport TPA into the cell, and pBiosensor, which detects TPA in the environment and fluoresces. pTransporter contains TpiA and TpiB, which encode two TPA transporter proteins, as well as TphC, which is a binding protein that facilitates the transmembrane transport of TPA. pBiosensor contains PtphC, an inducible promoter that responds to TPA. TphR can encode a regulatory protein of the PtphC promoter, and eGFP for indirect characterization of the intracellular concentration of TPA.

Analysis: The PCR amplification bands of TphR (approximately 500bp), PtphR (approximately 100bp), eGFP (approximately 500bp), terminator (approximately 100bp), TpiA (approximately 2000bp), TpiB (approximately 500bp), TphC (approximately 750bp), pET21a vector (approximately 5000bp), pET28a vector (approximately 5000bp) align with the theoretical lengths estimated based on the designed primer positions, demonstrating that these target genes have been successfully obtained. However, the plasmid vector pET21a/28a are amplificated by PCR. But they have two bright lines, but at the designed position it’s bright and clear, so we think the fragments here is our target elements. PtphC and Terminator fragments both have two bright lines, but at the designed position it’s bright and clear, so we think the fragments here is our target elements.

Analysis: From the result of plate coating of LB solid medium with antibiotic Amp/Kan, the medium successfully grew dispersed single colonies, which indicated that the bacterium was successfully screened and transformed.

Analysis: The sequencing result show the target element and the plasmid vector of pTransporter were correctly connected. The sequence of the actual plasmid is identical with the designed plasmid. But for the pBiosensor sequencing, it failed or appeared bad signal for first several times. We discuss that it might be the copy numbers of plasmid vector pET28a is low, and it’s not suitable for the biosensor function, so we redesigned a pBiosensor using pUC57 plasmid vector. But later new sequencing results show that our old pBiosensor also succeed in construction.

Part 2: Degradation ability of PETase in gradient temperature and pH

In the treatment of domestic wastewater, we consider that we may have to deal with different physical and chemical environments of the treatment object, such as changes in pH in different water samples and seasonally influenced changes in ambient temperature, leading to changes in treatment efficiency. Therefore, we wished to explore the optimal pH and optimal temperature of our working enzyme, TurboPETase (hereafter all PETases are referred to as TurboPETase unless otherwise specified), and to quantify the changes in degradation capacity beyond the optimal pH and temperature.

We transformed the pBAD-PETase-6His plasmid into E. coli BL21 strain, for expressing the working enzyme by large-scale fermentation. We use high pressure homogenizer to break bacteria as unpurified PETase samples. And next we purified and concentrated to obtain PETase purified samples.

2.1 Protein purification and SDS-PAGE gel verification:

Analysis: #1- 4 PETase showed obvious bands near 25kDa, which is consistent with the theoretical value of PETase, but the width of the bands indicates that the protein concentration is not high, i.e., E. coli expresses PETase less efficiently; and the number of stray bands is less indicating that the protein is more pure. While the eluate showed obvious band missing near 25kDa, which indicated that PETase was purified and the affinity chromatography method was effective.

2.2 Degradation ability of gradient pH enzymatic samples

(1) The color reaction was performed using p-nitrophenol as a substrate

Analysis: The optimum pH of PETase as a lipase is about 5, but the activity of PETase as a lipase is not necessarily equal to its ability to degrade plastics, and many experiments have shown that this method as an indicator of the activity of PETase has a large error and the results are not accurate, so we chose to use the method of high-performance liquid chromatography (HPLC) with Bis (2-hydroxyethyl terephthalate) (BHET) as the substrate for the subsequent experiments. hydroxyethyl terephthalate (BHET) as the substrate in our subsequent experiments.

(2) Detection of BHET degradation ability in gradient pH
BHET substrate solution of 0.4 g/10 mL was configured, and 0.9 mL of BHET substrate solution plus 0.1 mL of unpurified/purified enzyme reaction system was configured, and the reaction was carried out at 65°C for 60 min with a gradient pH setting. After the reaction was completed, the reaction was terminated by the addition of 1 mL of methanol solution, and then the reaction was allowed to stand at 4°C for one night, and the supernatant was centrifuged for high-performance liquid chromatography (HPLC) to analyze the difference in peak areas of BHET characteristic peaks. Chromatography (HPLC) was used to analyze the difference in peak area of BHET characteristic peaks.

Analysis: The optimum pH for BHET degradation by PETase was measured to be around 8 when using unpurified reaction enzyme. Although the highest average substrate production was observed at pH=10 when protein purified protein was used as the reaction enzyme, there may be a certain error in this data due to the large standard deviation.

Conclude: BHET is a dimer of PET monomer, and it takes shorter time to degrade equal amount of BHET compared with multimeric PET, and the polymerization number of BHET is determined, so we chose BHET as the substrate to better quantify the ability of PETase to degrade microplastics.

2.3 For gradient temperature enzyme sample degradation BHET ability detection

Configure 0.9 mL of BHET substrate solution and 0.1 mL of purified enzyme reaction system, and react for 30 min at a gradient temperature setting of pH=8. At the end of the reaction, terminate the reaction by adding 1 mL of methanol solution, let it stand overnight at 4℃, and then centrifuge the supernatant for HPLC liquid phase analysis.

2.3.1 Construct BHET and TPA standard curves in HPLC:

In the experiments to analyze the effect of gradient pH on enzyme degradation ability, we only considered the difference in the peak area of the substrate BHET to characterize the enzyme degradation ability, but this led to a potentially incomplete analysis and excessive bias. Therefore, we constructed standard curves for BHET and TPA to characterize the enzyme degradation capacity by the difference in the consumption of substrate BHET and the production of product TPA.

Gradient concentrations of BHET and TPA standards were configured, and HPLC liquid phase analysis was performed to obtain standard curves of concentrations and characteristic peak areas.

2.3.2 Protein concentration determination of two batches of purified PETase samples

Because we purified two batches of purified enzyme samples, considering the difference in enzyme amount between the two samples led to the property test affected by this factor, we need to carry out the purified enzyme protein concentration measurement, and can therefore calibrate the enzyme activity.

A gradient concentration PBS protein standard was configured, two batches of purified protein samples were sampled, and the absorbance of A595 was measured using an enzyme marker. Sample protein concentrations were calculated from the standard curve using the Calculation Tool.

Standard fit curve: polynomial function fit curve y = 0.264x4 - 0.9697x3 + 0.9756x2 + 0.3684x + 0.5246 R1² = 0.9998; linear fit curve y = 0.5717x + 0.5407 R2² = 0.9799; since R1 > R2, a polynomial function fit was chosen to calculate the first batch enzyme protein concentration : 0.379 mg/ml and the second batch of enzyme protein concentration: 0.282 mg/ml.

2.3.3 HPLC analysis of degradation capacity by temperature gradient

The concentration of each component was analyzed by standard curves to calculate the consumption of BHET and the production of TPA and MHET at different temperatures.

Analysis: The amount of BHET degradation (concentration of the blank group - concentration of the experimental group) increased and then decreased with increasing temperature, but considering that the amount of naturally degraded BHET increased with increasing temperature in the blank group, which in turn affected the analysis of the enzyme degradation part. To further optimization of the experiment, we should shorten the reaction time to avoid the substrate being completely consumed; the amount of TPA production (concentration of the experimental group - concentration of the blank group) increased with increasing temperature and then The amount of TPA generated (concentration of experimental group - concentration of blank group) decreased with increasing temperature, which can be a stronger indication of the increased substrate generation capacity in this gradient temperature probe; the intermediate product MHET was not calibrated with a standard curve, and the difference in its peak area (concentration of experimental group - concentration of blank group) decreased with increasing temperature by a relatively small amount, therefore, we suspect that most of the substrate BHET is degraded only to the intermediate product MHET, rather than to TPA. Next we could work on the optimization of the degradation efficiency of MHET, which can further increase the working enzyme's degradation ability for PET.

Part 3: Validation of Signal Peptide Function

3.1 Reaction enzyme sample preparation:

The Enhancer-SP-PETase plasmid ( SP: OmpA, PelB, LamB, MalE ) was constructed and transferred into BL21 strain, while the original pBAD-PETase-6His was transferred as a blank strain. The working enzyme was induced to be expressed and secreted.

The bacterial fluid was centrifuged and the supernatant was concentrated and fixed to 4 mL to be prepared as supernatant enzyme sample. 4 mL of pH=8 Tris-HCl Buffer was added to the precipitate and sonicated to break the bacteria, and prepared as a sample of the intracellular enzyme.

3.2 Reaction system configuration：

The reaction system was configured with 0.8 mL of BHET substrate solution + 0.1 mL of PH=8 Tris-HCl Buffer + 0.1 mL of reaction enzyme samples, and the reaction was carried out for 30 min at a gradient temperature setting of pH=8. At the end of the reaction, the reaction was terminated by the addition of 1 mL of methanol solution, and then the reaction was allowed to stand at 4°C overnight, and the supernatant was centrifuged and analyzed by UPLC (Ultra- Performance Liquid Chromatography (UPLC).

3.3 UPLC Standard Curve re-builted

The analytical speed and resolution of the UPLC instrument, as well as the sensitivity, were superior to that of HPLC, and we attempted to use UPLC for the next analytical investigations. However, because of the change of the liquid phase instrument and the column, we need to reconstruct the standard curve of the reaction components.

Analysis：According to the chromatograms of samples, we think the UPLC method we use is not suitable, as the peak of TPA and MHET is not fully separated and it would lead to deviation. So we chose BHET consumption as the evaluation standard to measure the degradation power of the processed enzyme samples function and reflecting the secretory ability of SPs.


Conclude:
Based on the data analysis, the best signaling peptide for TurboPETase is OmpA, while MalE signaling peptide seems to inactivate it completely, we speculate that there may be a shifting mutation in the plasmid containing MalE signaling peptide, or MalE signaling peptide itself leads to the inactivation of PETase. Surprisingly, the three others signaling peptide acted more strongly than expected, and the PETase in the extracellular fluid degraded LamB and OmpA much more than the intracellular one, which was not consistent with our assumption that the activity of intracellular enzyme was much larger than that of extracellular enzyme. This is not consistent with our assumption that the activity of intracellular enzymes is much greater than that of extracellular enzymes. This may be due to several reasons: 1 the expression-secretion balance of intracellular enzyme levels under prolonged induction, the accumulation of extracellular enzymes due to secretion, 2 the reduction of enzyme activity by intracellular signal peptide sequences, and 3 the damage to enzymes during the process of ultrasonication. However, in view of the fact that the stronger the total enzyme activity of the extracellular fluid and the weaker the total enzyme activity of the broken fluid in the figure can be used as a supporting evidence for secretion, we believe that the strength of the signal peptide is objectively reasonable, and that a strong secretion ability of the signal peptide is more conducive to secretion applications in real environments.

Part 4: Sequence comparison of TTT proteins from E. coli and Comamonas sp. E6

We also found three proteins belonging to the tripartite tricarboxylate transporter (TTT) family in E. coli. We compared the three proteins of E. coli with those of Comamonas sp. E6 to confirm whether the TTT proteins in E. coli are similar to TpiA, TpiB, and TphC in structure and function, which will guide us to partially modify transmembrane proteins for heterologous expression in the following experiments.

We used BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi) to perform the sequence comparison of the proteins, and the results were as follows:

Analysis：From the results of blast, for the TTT of E. coli and Comamonas sp. E6, the similarity of proteins from TctA family or TctC family was high, but there is no significant identity of proteins from TctB family.This similarity in some degree provides a partial basis for the possibility of heterologous expression of membrane proteins in Escherichia coli, and the low similarity of TctB provides support for further optimization of heterologous expression of membrane proteins in the future.

Part 5: Validation of Biosensor and Transporter plasmid

In environmental wastewater treatment, there are difficulties in the application of LC (Liquid Chromatography, including HPLC and UPLC), which is currently the most commonly used means of detecting PET degradation, for large-scale continuous monitoring. We designed and constructed a transporter plasmid that can transport the PET degradation product TPA, and a detection plasmid with the TPA-specific promoter Ptpa. Transforming them into the BL21 strain to verify the ability of the transporter as well as the ability to detect the fluorescence of TPA expression in the samples.

Strains transformed with pBiosensor, pTransporter, and both plasmids were cultured for fluorescence detection at 12h, 18h, and 24h under TPA induction.

Analysis：Fluorescence intensity per unit OD600 value increases from 12h to 18h, but it decrease from 18h-24h, we guess it’s because the nutrition of the LB medium went out, and the bacterial would consume the heterologous protein eGFP in starvation, which leads the decreasing. And in the Figure 5.1, the Biosensor+Transporter+ inducer TPA is slightly higher than the two others group. It might meets our first expectation—E. coli BL21 strain don’t have the orginal ability to transport the TPA, and specific inducible promoters PtphC reacts to the TPA transported inside bacterial. But the ANOVA analysis in Figure 5.2 tells us the difference among the groups at 12h, 24h is not significant enough to support our expectation.

Conclude：
In fact, we repeated this experiment twice. In the first experiment, there were also some differences among the three groups, but because there might be some problems in the operation during our first experiment, which led to large differences among the data within the groups and the data were not credible, so we repeated the fluorescence detection experiment once.

In the data of the second experiment, although the data still show the trends that there exists the difference between three groups. However, due to the non-significant difference between the experimental groups and the fact that there is still a certain error between the parallel groups (there is a non-normal data). So, we still lack strong evidence of the actual effect of our components (or the effect of our components still needs to be optimized).

When analyzing the reasons for the unsuccessful characterization experiments of pBiosensor and pTransporter, we considered that Transporter is a heterologous Transporter element, so we predicted the protein structure of the membrane transporter protein elements in Transporter, and two of the elements are heterologous multiple transmembrane proteins (the number of transmembrane transitions for tpiA is 11, tpiB is 5), we think that the multiple transmembrane crossing is likely to lead to unsuccessful construction or difficult to function properly in our E. coli engineering bacteria. Due to the tight experimental time, we continued the subsequent experiments after the initial characterization of pBiosensor and pTransporter. Afterwards, we will continue to explore the function of this transmembrane protein and enhance its transporter ability to achieve and optimize the enrichment of TPA in our engineered bacteria E. coli.

Part 6: Validation of Signal Peptide-MHETase Function

In order to optimize the MHET accumulation problem analyzed in Part2 section, we tried to construct pET28a-SP-PETase-SP-MHETase (SP: PelB, Amcut) plasmid. However, this dual secretion plasmid was unsuccessful in designing primers for homologous recombination construction twice. Considering the time constraints for enzyme-linked design, we decided to construct a single secretion plasmid using a new plasmid backbone with different resistance from the pSec-PETase (pET28a, Kan resistance) plasmid: pSec-MHETase (pET21a, Amp resistance). Subsequently, a double plasmid double secretion BL21 strain can be constructed by double resistance selection.

Transform the newly constructed single-secretory plasmid pSec-MHETase, induce expression and secretion of the working enzyme, centrifugation of the bacterial fluid to take the supernatant for concentration and volume to 4mL, to the precipitate add 4mL pH=8 Tris-HCl Buffer and ultrasonic broken bacteria, prepared for the treatment of MHETase enzyme samples.

A complete blank reaction system was configured with 0.8mL of BHET substrate solution + 0.1mL of purified PETase sample + 0.1mL of treated MHETase enzyme/0.1mL of pH=8 Tris-HCl Buffer sample and no PETase or MHETase reaction enzyme samples, and the reaction was carried out for 60min at pH=8 and 65°C. The reaction was completed by adding 1mL of methanol to the precipitate and sonicated to break the bacteria. At the end of the reaction, 1 mL of methanol solution was added to terminate the reaction, which was allowed to stand overnight at 4°C, and the supernatant was centrifuged for HPLC liquid phase analysis.

In this experiment, there is no different between PETase blank group and MHETase experiment group, at first we thought that it might be the function of MHETase has some problems. But later we realized that our reaction condition is 65℃ (The best working temperature for modified TurboPETase), and it’s might be fatal to MHETase. We made a really big mistake here. But the freezing time is approaching, we could only correct this reaction experiment for MHETase in the future…….

Pure nano PET and TPA toxicity detection on nematode growth medium (NGM)

Compared with control group, exposure to 50, 100, 200, 400, 600, 800, 1000 μg/mL of polyethylene terephthalate (PET), or 125, 250, 500, 750, 1000 μg/mL terephthalic acid (TPA) on NGM show no adverse effects on Caenorhabditis elegans (C. elegans) intestinal ROS production, indicating that pure nano-size PET (Figure 1) and pure TPA (Figure 2) of these concentrations has almost no toxicity to C. elegans on NGM.

Pure nano PET and TPA toxicity detection in liquid culture system

Compared with control group, exposure to 125, 250, 500, 750, 1000 μg/mL pure PET and 125, 250, 500, 750, 1000μg/mL TPA in liquid culture system show no adverse effects on C. elegans intestinal ROS production, indicating that pure nano-size PET (Figure 3) and TPA (Figure 4) of these concentrations has almost no toxicity to C. elegans in liquid culture system.

The survival of C. elegans was examined under stereomicroscope before sampling and all C. elegans we observe survived.

Conclusion:

Based on our result, both pure nano-size PET and TPA are not toxic to C. elegans, which indicate that it can function normally in our liquid co-culture system.

Result and discussion for fluorescence detection

We examined the intestinal fluorescence intensity of Caenorhabditis elegans (C. elegans) at different stages. For C. elegans N2 worms at different stages (L4 Larva/day 1 adult) fed with different strains of Escherichia coli (BL21/OP50) expressing the fluorescent protein tdTomato, we took 21 worms from each group, capturing fluorescence with MetaMorph (Molecular Devices Inc.) software on an IX73 invert microscope (Olympus), and analyzing fluorescence intensity with ImageJ software.

Then, we performed data analysis. For C. elegans of both L4 stage and day 1 stage, the data are not normally distributed, so we performed nonparametric test - Mann Whitney test, to detect whether there is significant difference of C. elegans intestinal fluorescence between different E. coli feeding strains.

The results are shown in Figure 5 and Figure 6 for L4 stage and day 1 adult stage, respectively, with representative images on the left and quantification graph on the right. N2 worms fed on normal OP50 without tdTomato expressing was set as negative control, showing no visible fluorescence as expected. For worms fed on E. coli expressing tdTomato both at the L4 (Figure 5) and day 1 adult stage (Figure 6), the BL21 group shows stronger fluorescence compared with OP50 group, suggesting that E. coli BL21 strain may be more effectively than OP50 strain colonized or retained within intestine of C. elegans.

These results above suggest that E. coli BL21 strain may be more effectively colonized or retained within intestine of C. elegans than OP50 strain.

C. elegans intestine lysis spreading assay

After detecting the fluorescence intensity inside the intestine of C. elegans, we wonder about the feasibility of engineered E. coli to colonize in the worm intestine. To address this issue, we performed the spread plate assay to evaluate the number of living bacteria in the worm intestine.

Since the plasmids carrying the tdTomato gene also carry the ampicillin resistance gene, by adding the antibiotic ampicillin to the solid LB medium, we can avoid the contamination of undesired microbe from the medium. Through a series of gradient dilutions, we obtained the following results shown in Table 1 and Figure 7.

Then we analyzed the data. Colony-forming unit (CFU) per 10 worms were normalized by subtracting the number of colonies observed in plates from the wash control.

For the OP50 group, we chose the number of colonies in group 0 as the reference for calculating CFU. For the BL21 group, we chose the number of colonies in group 4 as the reference for calculating CFU.

N (OP50) = N (group 0) *n (dilution ratio) -N (group supernatant) = 208*1-7 = 201 CFU/10 worms
N (BL21) = N (group 4) *n (dilution ratio) -N (group supernatant) = 115*10⁴-0 = 1.15*10⁶ CFU/10 worms

The results showed that the number of living bacteria inside the intestine of C. elegans in the BL21 group was much higher than that in the OP50 group, consistent with the previous fluorescence results.

Conclusion:

Based on the results of C. elegans fluorescence intensity detection and intestinal lysis spreading assay, we have reached the conclusion: The intake of E. coli BL21 strain by C. elegans is much higher than that of OP50 strain. As an uncommon food for C. elegans, E. coli BL21 strain is able to colonize in the intestinal tract of C. elegans and the ability is much better than that of OP50 strain. This proves that our modified BL21 strain with the TPA transportation ability is suitable to colonize in the C. elegans intestine for further accumulation.

Synthetic biology modification of C. elegans

We have designed a plasmid that enables the expression of the TPA transport system specifically within the intestinal tract of C. elegans. The detailed design of this plasmid is presented below.

Using homologous recombination, we designed homology arms on the primers and simultaneously recombined four fragments using the efficient C116 homologous recombinase.

The vector was linearized and the insertion fragments with homology arms were obtained by PCR amplification. Gel validation was followed by gel extraction, homologous recombination and transformation of competent DH5α cells. The resulting colonies were plated, picked, and cultured before plasmid extraction. PCR validation (Figure 9) confirmed the presence of the desired band, which was subsequently confirmed by sequencing.

Outlook

Although we have successfully constructed a new plasmid used for the enrichment of TPA inside the intestinal tract of C. elegans, we have not yet validated the function of it in vivo. Our future work is to microinject and integrate our constructed plasmid into C. elegans to study the actual effect of this plasmid on TPA enrichment.