The pET-21b was selected as the backbone vector, and the composition of respective DNA fragments were amplified by Flash Super-Fidelity DNA Polymerase and constructed through one-step cloning method. The recombinant plasmid was transformed into Escherichia coli BL21 (DE3) for expression.

The catalytic activity of the fusion protein was determined in the 1.5 mL pH 8.0 0.1 M phosphate buffer system at 60°C, using pNPB as the substrate. And our preliminary results have revealed that the enzyme activities were significantly different when the PET-binding peptides were connected to the -NH₂ terminus or the -COOH terminus of PETase. Therefore, we tried to construct 32 fusion proteins by assembled these 16 peptides at the -NH₂ and -COOH terminus of PETase, respectively. Finally, 19 fusion proteins were successfully expressed and 60-1-fusion protein and 70-2-fusion protein were not expressed at either the C-terminus or the N-terminus. Enzyme activity of expressible fusion protein is shown in Fig. 1a. The results showed that the enzyme activities obtained from the expression of PETase-G4S-50-1-C, PETase-G4S-50-5-C, and 60-2-N-G4S-PETase were relatively high.

Fig. 1   Enzyme activity and PET degradation products release of fusion proteins with G4S as linker
(a) . Enzyme activity of fusion proteins. (b). Release of PET degradation products after 24 h.

A total of 35 nmol enzyme solution was purified and reacted with 25 mg PET microplastics at 60°C in 8.75 mL of Gly-NaOH buffer (pH 9.0), at the end of which the products were analyzed using High Performance Liquid Chromatography (HPLC) and calculated the total amount of TPA and MHET released (Fig. 2). The result showed that 50-3-N-G4S-PETase, 50-4-N-G4S-PETase, PETase-G4S-50-5-C, 60-2-N-G4S-PETase, and 70-1-N-G4S-PETase showed improved degradation efficiency of PET microplastics compared to PETase, with products releases of 139.48%, 144.67%, 126.22%, 126.80%, and 155.62% of PETase, respectively (Fig. 1b).

Fig. 2   HPLC results of catalytic products

The comprehensive results of the previous experiments revealed that the PET degradation efficiency of 50-4-N-G4S-PETase was high, but its expression level was low. Therefore, we took it as an example to optimized type of fermentation medium, concentration of IPTG, fermentation temperature, and fermentation time at the shaking flask-level to enhance the protein expression.

TB, LB, and ZYM were selected as culture media to investigate the optimal medium type (Fig. 3a); IPTG concentrations of 0.025 mmol/L, 0.050 mmol/L, 0.075 mmol/L, and 0.100 mmol/L were selected to investigate the optimal IPTG concentration (Fig. 3b); fermentation was carried out at 16°C and 25°C to investigate the optimal fermentation temperature (Fig. 3c); 24 h and 48 h were selected to investigate the optimal fermentation time (Fig. 3d).

Fig. 3   50-4-N-G4S-PETase fermentation conditions optimization results

(a) . Fusion protein enzyme activity in different media. (b). Fusion protein enzyme induced by different concentrations of IPTG. (c). Fusion protein enzyme activity at different fermentation temperatures. (d). Fusion protein enzyme activity at different fermentation times.

The results showed that fermentation with TB as the medium, addition of IPTG at the concentration of 0.05 mM, and fermentation at 16°C for 48 h were the optimal fermentation conditions.

Taking PETase-G4S-70-1-C as an example, we replaced the flexible linker G4S with EAAAK, AG, Tr, 10A, and SLE, respectively. Then, we constructed recombinant plasmids PETase-EAAAK-70-1-C, PETase-AG-70-1-C, PETase-Tr-70-1-C, PETase-10A-70-1-C, PETase-SLE-70-1-C and transformed them into E. coli BL21 (DE3) to express fusion proteins, respectively (Fig. 4). The enzyme activity of the fusion proteins was determined (Fig. 5). The results showed that except for G4S, PETase-SLE-70-1-C had highest enzyme activities.

Fig. 4   Construction of 50-4-N-SLE-PETase fusion protein

(a). PCR validation results. (b). Plasmid map of pET-21b-50-4-N-SLE-PETase.

Fig. 5   Fusion protein enzyme activity with different linkers

We attempted to replace the linker of the 50-4-N fusion protein, followed by protein structure simulations using AlphaFold2 with (G4S)3, (EAAAK)3, (SLE)5, Tr, (AG)5, and 10A linkers, respectively (Fig. 6).

Fig. 6   AlphaFold2 structure simulation results

Molecular dynamics simulations showed that the G4S, Tr, and AG linkers are flexible in the fusion protein, while the SLE, EAK, and 10A linkers function as rigid linkers in the fusion protein (Fig. 7).

Fig. 7   Molecular dynamics simulations
(a). RMSF analog. (b). RMSD analog.

Then, we replaced all the linkers in the 14 heterologously expressible recombinant plasmids with SLE and transformed them into E. coli BL21 (DE3) for expression. We determined the enzyme activity (Fig. 8a) and PET degradation ability (Fig. 8b) of the fusion proteins.

Fig. 8   Enzyme activity and PET degradation products release of fusion proteins with SLE as linker
(a). Enzyme activity of fusion proteins. (b). Release of PET degradation products after 24 h.

The results showed that the enzyme activity of the fusion proteins PETase-SLE-50-5-C, 60-2-N-SLE-PETase, PETase-SLE-70-1-C was relatively high. And the fusion proteins 40-1-N-SLE-PETase, 50-1-N-SLE-PETase, 60-2-N-SLE-PETase, and 70-1-N-SLE-PETase showed improved degradation efficiency of PET microplastics compared to PETase, with the products releases 103.75%, 146.97%, 176.37%, and 135.45% of PETase, respectively.

Combining the results of the previous experiments, we found that the PET degradation effect and enzyme activity of the fusion proteins 50-4-N-G4S-PETase, PETase-G4S-50-5-C, and 60-2-N-SLE-PETase were both high at expression-level and degradation-level. Thus, in order to verify the binding effect of the above three short peptides with PET microplastics, we constructed recombinant plasmids by assembling these three peptides with eGFP with pET-20b as vector and G4S as linker, respectively, and transformed them into E. coli BL21 (DE3) for expression.

The eGFP solution and eGFP-fusion protein solution with equal fluorescence intensity were taken and incubated with PET film, protected from light at 37°C for 24 h, respectively. Then, free proteins were eluted, and the PET film was observed under a fluorescence microscope at 488 nm excitation wave (Fig. 9).

Fig. 9   Fluorescence characterization of 50-4-N, 50-5-C, 60-2-N linked eGFP fusion proteins.
(a) . eGFP of equal fluorescence intensity to 50-4-N-G4S-eGFP binding PET.(b). 50-4-N-G4S-eGFP binding PET. (c). eGFP of equal fluorescence intensity to 50-5-C-G4S-eGFP binding PET. (d). 50-5-C-G4S-eGFP binding PET. (e). eGFP of equal fluorescence intensity to 60-2-N-G4S-eGFP binding PET. (f). 60-2-N-G4S-eGFP-binding PET.

Comparing with the eGFP, 50-4-N-G4S-eGFP, 50-5-C-G4S-eGFP, and 60-2-N-G4S-eGFP showed different degrees of enhanced fluorescence signals under the microscope. The results demonstrated that all three peptides, 50-4-N, 50-5-C, and 60-2-N, could bind to PET efficiently, thus enhancing the degradation of PET of PETase-linker-peptide fusion proteins.

Based on the multiple mutation sites of 50-4-N, 50-5-C, and 60-2-N predicted by the dry lab, we successfully constructed 19 recombinant plasmids carrying respective mutants using MEGAWHOP cloning method. And these plasmids were transformed into E. coli BL21 (DE3) for expression, respectively.

The results showed that the enzyme activity of 50-4-N-Y33F-G4S-PETase was higher than that of its wild type (Fig. 10a), which was 175.25% of the wild type.

The enzyme activities of the mutants of PETase-G4S-50-5-C were all higher than that of the wild type (Fig. 10c), among which, the enzyme activities of PETase-G4S-50-5-C-R287V and PETase-G4S-50-5-C-C293V were much higher than that of the wild type, which were respectively 184.67%, and 181.39% of wild type.

The 60-2-N-SLE-PETase mutants all had lower enzyme activities than the wild type (Fig. 10e).

Fig. 10   Enzyme activity and PET degradation products release of single mutants of fusion proteins
(a) . Enzyme activity of PETase-G4S-50-4-N and its mutants. (b). PET degradation products release from PETase-G4S-50-4-N and mutants after 24 h. (c). Enzyme activity of PETase-G4S-50-5-C and its mutants. (d). PET degradation products release of PETase-G4S-50-5-C and its mutants after 24 h. (e). Enzyme activity of PETase-SLE-60-2-N and its mutants. (f). PET degradation products release from PETase-SLE-60-2-N and its mutants after 24 h.

The results showed that product release of 50-4-N-S14A-G4S-PETase, 50-4-N-V19I-G4S-PETase, 50-4-N-L23K-G4S-PETase, 50-4-N-Y33F-G4S-PETase, 50-4-N-S61A-G4S-PETase were all higher than the wild type (Fig. 10b), 103.59%, 133.47%, 117.53%, 115.94%, and 126.10% of the wild type, respectively.

The product release of all mutants of PETase-G4S-50-5-C were higher than that of the wild type, in which the PET degradation efficiency of the PETase-G4S-50-5-C-R287V was the most efficient (Fig. 10d), which was 164.38% of that of the wild type.

The product release of 60-2-N-N29D-SLE-PETase and 60-2-N-L60I-SLE-PETase was higher than that of the wild type (Fig. 10f), which was 102.94%, and 106.21% of the wild type, respectively.