Results

Introduction

This project aims to systematically discover and engineer marine-derived ester hydrolases to address the persistent retention of aliphatic-aromatic polyester plastics (e.g., PET/PLA/PBAT) in the ocean. Combining marine enzyme discovery, computational protein redesign, inorganic immobilization, and cell-based enrichment/degradation modules, we discovered and biochemically characterized a marine polyester hydrolase (PEH-2013). Using ProteinMPNN and AI-driven mutagenesis, we redesigned it to produce a soluble, active variant, polygonase. To address the vulnerability of free enzymes to inactivation during industrial processing and environmental exposure, we developed an enzyme immobilization formulation based on calcium carbonate-silicon dioxide (CaCO₃-SiO₂) composite nanocarriers, balancing process compatibility with environmentally triggered release. Furthermore, we constructed a system for co-displaying an adhesion module and an engineered hydrolase on the surface of Escherichia coli (adhesion + SpyTag/SpyCatcher-tethered hydrolase) to enhance the accumulation and in situ degradation of polyester microplastics in water. To quantitatively assess the degradation process, a chemiluminescent detection method for the key intermediate terephthalic acid (TPA) was established. In the future, efforts will be made to design and develop an intelligent whole-cell biosensor by integrating degradation and detection capabilities using synthetic biology tools, moving toward the construction of an autonomous, controllable, and responsive bioremediation system for plastic pollution.

SectionⅠ-Enzyme mining and optimization

Enzyme mining

To identify promising marine-derived ester hydrolases, we used the highly efficient cutinase TfCut as a template and conducted homology mining using NCBI BLAST. We successfully identified two candidate enzymes originating from marine environments and named them PEH-AM and PEH-2013. For details, please check the page of MODEL-ENZYME MINING.

Based on the computer-assisted screening results, we synthesized genes through the International Gene Synthesis Consortium (IGSC) and cloned the selected sequences into the pET-28a(+) vector with an N-terminal His-tag. We heterologously expressed the two target proteins in Escherichia coli BL21 (DE3) and purified them by Ni-NTA chromatography. The results were then verified by SDS-PAGE gel electrophoresis (Fig. 1).

Theoretical molecular weight calculations indicate that the expected molecular weights of PEH-AM and PEH-2013 are 29.41 kDa and 32.71 kDa, respectively. In the SDS-PAGE analysis of the whole protein lysate, pellet, and supernatant samples in Fig. 1, a distinctly darker band, corresponding to the two target proteins, appears in the 25–35 kDa range, highly consistent with the expected size. This band is significantly enriched in both the soluble fraction and Ni-NTA elution/enrichment samples, providing preliminary evidence that PEH-AM and PEH-2013 were successfully expressed in E. coli BL21(DE3) and recovered in a soluble form.

SDS-PAGE analysis revealed that all eluted protein supernatants contained only a small fraction of the target enzyme, along with significant amounts of protein impurities and nonspecific contaminants. This suggests that the soluble expression levels of both PEH-AM and PEH-2013 in the E. coli system were generally low. In particular, even after scaling up PEH-AM to 10 L, its yield (A280 = 0.097) was still significantly lower than that of PEH-2013 (A280 = 1.256). Therefore, we discarded PEH-AM and focused our further analysis on PEH-2013.

After confirming the expression ability of the enzyme PEH-2013, we further evaluated its catalytic performance on actual polyester substrates. We first evaluated the degradation performance of the newly discovered enzyme PEH-2013 on PET substrates with varying degrees of crystallinity. We measured the depolymerization activity of PEH-2013 on PET with varying degrees of crystallinity (LC-PET and HC-PET) under standard reaction conditions (50°C, 24 hours, 500 nM enzyme loading) (Fig. 2).

The results showed that PEH-2013 exhibited significant hydrolytic activity against PET, effectively degrading it to form terephthalic acid (TPA). Its efficiency was significantly affected by the crystallinity of the PET. The concentration of TPA in the final product, LC-PET, was significantly higher (Fig. 2), indicating that the amorphous regions were more accessible to the enzyme. This result confirms that PEH-2013 is an active PET-degrading enzyme, and its catalytic efficiency is largely dependent on the crystallinity of the substrate.

To further characterize the enzymatic properties and environmental adaptability of PEH-2013, we systematically determined its optimal reaction temperature, salt tolerance, and thermal stability.

Using nano-PET as the substrate, the effect of temperature gradient on the enzymatic activity of the enzyme PEH-2013 was investigated. Reactions were carried out at 50 °C for 3 hours with 6 mg/L nano-PET and an enzyme concentration of 500 nM. Results showed that PEH-2013 exhibited catalytic activity across the temperature range of 40°C to 60°C, reaching a peak at 50°C, confirming its optimal reaction temperature (Fig. 3a). This temperature is consistent with the properties of many industrial enzymes derived from mesophilic microorganisms. Given that PEH-2013 originates from a marine environment, we evaluated the effects of various NaCl concentrations on its activity. Comparisons with common PET hydrolases, LCC and ICCG, revealed that PEH-2013 exhibited significant salt tolerance compared to common PET hydrolases (Fig. 3b). To further investigate its thermal stability, we determined its melting temperature using a fluorescent dye-based protein thermostability assay. The Tm value was found to be approximately 42°C, indicating that PEH-2013 maintains its structural integrity at temperatures between 40 and 50°C, suggesting potential applications at slightly higher temperatures (Fig. 3c).

After defining the basic enzymatic properties of PEH-2013, we evaluated its hydrolysis performance against various polyester substrates under gradient temperature conditions to examine its broad-spectrum degradation potential. The experimental conditions were as follows: 500 nM of the purified enzyme was added to a reaction system containing 5 mg·mL⁻¹ of polyester powder (PET, PBAT, PLA). The reaction was incubated for 12 hours at a range of temperatures. After the incubation period, the reaction supernatant was analyzed by high-performance liquid chromatography (HPLC) to quantify the hydrolysis products (Fig 4).

The graph (Fig. 4) shows that PEH-2013 produced detectable hydrolysis products for PET, PBAT, and PLA across the tested temperature range, demonstrating its broad-spectrum polyester hydrolysis activity. Significant differences in product amounts were observed between substrates, suggesting that PEH-2013's catalytic efficiency for different polyesters is substrate-dependent. Overall, PEH-2013 demonstrates potential as a candidate enzyme for treating multiple polyester components in complex plastic waste streams.

In summary, the PEH-2013 enzyme exhibits several excellent properties suitable for real-world environments, including salt tolerance, thermal stability, and broad substrate spectrum. However, its low soluble expression level in the E. coli system has become a major bottleneck in the current project's in-depth characterization, high-throughput mutation screening, and large-scale application verification.

Solubility improvement strategy based on proteinMPNN

To improve enzyme expression activity and stability, we used ProteinMPNN for sequence prediction optimization using enzyme PEH-2013 as a template. For details, please click MODEL-ENZYME OPTIMIZATION - Solubility Optimization: ProteinMPNN - Driven Sequence Redesign and Multi-Stage Screening.

During the wet lab validation phase, we performed small-scale protein expression and purification on 13 candidate sequences from the initial computer-assisted screening. Of these, only three mutants, SEQ4, SEQ11, and SEQ59, successfully achieved soluble expression and were purified to high purity via nickel affinity chromatography. SDS-PAGE results showed that, under the same purification volume (1 L of bacterial culture), the protein band of SEQ4 was significantly darker than that of the other mutants, indicating higher expression levels and solubility (Fig. 5).

Compared to the wild-type enzyme PEH-2013, the proteinMPNN-predicted enzyme expression levels were significantly increased (Fig. 6). Therefore, SEQ4 was selected as the template sequence for subsequent stability engineering.

Based on the successful acquisition of the highly expressed soluble mutant SEQ4, we further evaluated its enzymatic properties to confirm whether it has improved activity and stability while maintaining the excellent properties of the wild type.

Using nano-PET as the substrate, the temperature dependence of the enzymatic activity of SEQ4 was investigated under identical experimental conditions as those used for enzyme PEH-2013. Reactions were carried out at 50 °C for 3 hours, with 6 mg/L nano-PET and an enzyme concentration of 500 nM. The results indicated that SEQ4 exhibited measurable catalytic activity across the temperature range of 40 °C to 60 °C, with maximum activity observed at 50 °C (Fig. 7a). To assess its salt tolerance, SEQ4 was tested under varying NaCl concentrations and compared directly with PEH-2013. The results showed no significant difference between the two enzymes, confirming that SEQ4 also possesses strong salt tolerance and can maintain catalytic performance under marine-like salinity conditions (Fig. 7b). To further investigate its thermostability, we determined the melting temperature (Tm) of SEQ4 using a fluorescent dye-based protein thermostability assay. The Tm of SEQ4 was found to be approximately 58 °C, substantially higher than that of PEH-2013 (42 °C) (Fig. 7c).

This increase in thermal stability suggests that SEQ4 maintains its structural integrity and enzymatic function over a broader temperature range, making it a more robust candidate for industrial polyester degradation and marine environmental applications.

Stability Enhancement and Activity Improvement of SEQ4 through Consensus Mutation Screening

To further enhance the stability and catalytic efficiency of SEQ4 for industrial application in polyester processing, we performed Multi-Tool Consensus Mutation Screening to identify beneficial amino acid substitutions. For details, please check the page of MODEL – Enzyme Optimization – Stability Enhancement: Multi-Tool Consensus Mutation Screening.

Based on the computational simulation results, several potential stabilizing mutations were proposed and subsequently evaluated experimentally. The enzymatic activities of each mutant were quantified using a p-nitrophenyl butyrate (pNPB) hydrolysis assay, monitored by UV spectrophotometry under identical reaction conditions. Among the tested variants, the S210Y mutation exhibited the highest relative activity, showing approximately 1.75-fold improvement compared to the wild-type SEQ4 enzyme (Fig. 8) This mutation likely enhances local hydrophobic packing near the catalytic region, contributing to both increased stability and improved catalytic efficiency.

These results demonstrate that consensus-based computational design effectively guides rational enzyme optimization, enabling SEQ4 to maintain superior performance under the elevated temperatures and processing conditions relevant to industrial polyester production. We renamed SEQ4 (S210Y) to polygonase, and employed it as the key enzyme for the following researches and applications.

SectionⅡ - Immobilization

Thermal stability test of immobilized enzyme

To evaluate the thermal stability of the engineered enzyme polygonase immobilized on a calcium carbonate–silicon dioxide (CaCO₃–SiO₂) nanocomposite support via a sol-gel method, we subjected the immobilized preparation to high-temperature incubation and activity testing For details, please check the page of Experiments - Calcium Carbonate-Silicon Dioxide Composite Immobilization - Thermal Stability Testing . As shown in Fig. 9, after a brief high-temperature treatment, the immobilized enzyme retained approximately 40% of its relative catalytic activity at temperatures up to 200°C. No residual activity was detected after calcination at 240 °C for only 3 minutes, demonstrating that immobilization significantly improved the enzyme's thermal stability. This thermal stability enables the enzyme preparation to maintain partial functionality within the typical industrial thermoforming temperature range of PBAT and PBAT/PLA blends (approximately 160–200°C), providing a promising approach for integrating bioactive components into plastic matrices to create process-compatible "functional plastics." Direct adaptation to PET-related high-temperature processes still requires further evaluation and process optimization.

Distribution of Immobilized Enzymes in Polymer Films

To verify whether the lamination process can achieve uniform distribution of immobilized enzymes within the film, we first focused on PBAT films, the most common and strongest type of film. In this project, fluorescein isothiocyanate (FITC) was used as a tracer surrogate. FITC solution and enzyme solution were coupled and mixed at a ratio of 50 µl:1 ml. Sample preparation and immobilization and lamination were performed using the same method. The surface, edge, and cross-section of the fluorescently labeled enzyme-containing PBAT plastic film were observed under a confocal microscope. As shown in the figure, confocal laser scanning microscopy reveals that FITC-stained polygonase fluoresces green. The distribution of the fluorescent green patches on the film surface reflects the distribution of the polygonase inorganic nano-calcium carbonate-silica complex enzyme within the film. The enzyme exhibits uniform distribution across the surface, edge, and cross-section of the film, with no apparent aggregation or distribution loss (Fig. 10). These results indicate that the adopted sol-gel immobilization and lamination process can uniformly integrate the immobilized enzyme with the CaCO₃ - SiO₂ composite carrier into the polyester matrix, providing an experimental basis for the preparation of spatially uniform and repeatable enzyme-plastic composite films.

Effect of immobilized enzyme addition amount on the mechanical properties of enzyme-containing PBAT plastic films

This project tested the mechanical properties of enzyme-containing PBAT plastic films with different addition amounts of polygonase calcium carbonate-silicon dioxide complex enzyme preparation. The Young's modulus, elongation at break, and tensile strength values obtained are shown in Table 1.

Table 1 Material mechanical parameters of enzyme-containing PBAT plastic films prepared with different addition amounts of polygonase calcium carbonate-silica complex enzyme preparation

Enzyme addition percentage (%)	Young's modulus（MPa）	Elongation at break（%）	Tensile strength（MPa）
0	1.62	615.71	22.84
0.5	1.92	425.57	20.16
1.0	1.58	307.80	14.80
2.0	1.80	213.80	15.70
5.0	1.99	44.86	11.60

Mechanical analysis revealed that increasing the loading of CaCO₃–SiO₂–immobilized polygonase in PBAT films introduced interfacial microgaps that reduced tensile strength and elongation while the Young’s modulus remained largely unchanged, leaving stiffness largely unaffected, indicating that with enzyme concentrations between 0.5% and 2% providing the optimal balance between mechanical integrity and biodegradability.

Degradation Behavior of Films Containing Immobilized Enzymes in Real Seawater

PBAT and PBAT/PLA blend films embedded with a 1% enzyme preparation were cut into 3 cm × 3 cm samples and placed in an 80 cm × 50 cm × 50 cm aquarium. Real seawater collected approximately 10 km from the coast was added, and an oscillator was used to simulate seawater flow. Degradation experiments were conducted at room temperature for 60 days.

During the experiment, the degradation of the film samples was macroscopically observed, and structural damage and surface erosion processes were systematically evaluated. The results showed that in the simulated marine environment, the enzyme-containing films exhibited progressively deeper cracking, pore formation, and surface roughening, which were significantly different from the control group without enzyme (Figure 11).

Scanning electron microscopy (SEM) was also performed on the samples to analyze their surface microstructure (Fig. 12). Comparison of plastic films that degraded over 30 days. High-resolution SEM imaging further revealed distinct microstructural differences between the enzyme-embedded membranes and the control membranes. While the surfaces of the control PBAT and PBAT/PLA (1:1) membranes remained smooth and intact, the membranes containing immobilized polygonase exhibited distinct morphological degradation, with the formation of irregular cavities visible upon further magnification. Together, these SEM observations confirm that immobilized polygonase effectively triggers localized polymer hydrolysis, thereby accelerating surface fragmentation under marine-like conditions.

Together, these results demonstrate the successful development of an innovative strategy for uniformly embedding heat-stable immobilized esterases into a degradable polyester film matrix. This type of "active film" exhibits significant and progressive degradation ability under conditions close to real ocean conditions, providing a strong experimental basis and methodological foundation for the development of a new generation of degradable polyester materials with controllable life cycles.

SectionⅢ - Lab stage degradation platform

Adhesion Protein Display on the Escherichia coli Cell Surface

To obtain a functional module with ideal adhesion to polyester microplastics, we selected the mussel foot protein mfp-3, reported in the literature for its excellent overall performance and strong environmental adaptability, as our experimental adhesion protein. Using a truncated form of the ice nucleation protein INPNC as the membrane anchoring domain, mfp-3 was fused to the cell surface of E. coli BL21(DE3). To directly visualize and verify the system, we constructed a reporter plasmid, pETlac-Plpp1.2-INPNC-mCherry, containing the red fluorescent protein mCherry. Observations under fluorescence microscopy showed that the cell surface of the engineered bacteria expressing the induced protein showed a clear red fluorescent signal, indicating that the anchoring system was effectively displayed on the cell surface (Fig. 13A).

Base on this, we further evaluated the binding ability of the engineered INPNC–mfp-3 bacteria to polyester substrates through adhesion experiments. The engineered bacteria were incubated with 1 cm × 1 cm polyester plastic sheets at room temperature for 12 hours. Quantitative analysis of bacterial cell density revealed that the mfp-3 surface-displaying strain exhibited significantly increased adhesion to the plastic sheet compared to the empty vector control strain (Fig. 14).

In summary, these results validated the successful display of the INPNC–mfp-3 fusion protein on the cell surface and clearly demonstrated its ability to significantly enhance cell adhesion to polyester substrates, laying a solid experimental foundation for the subsequent development of dual-anchor systems with both adhesion and degradation capabilities.

E. coli cells display protein hydrolysis by polygonase using a SpyCatcher/SpyTag coupling.

We constructed a polyester hydrolase degradation system for assembly on the surface of E. coli using the SpyC (SpyCather)/SpyT (SpyTag) proteins. Using a lipoprotein signal peptide (Lpp) outer membrane protein A (OmpA) fusion system as a vector, we constructed the pET28a (+)-Lpp-OmpA-SpyC plasmid for presentation of the SpyC protein on the surface of E. coli BL21 (DE3) cells. Furthermore, we constructed a SpyT-sfGFP fusion protein plasmid using super folded green fluorescent protein (sfGFP) as a visual reporter gene. Recombinant SpyT-sfGFP was overexpressed in E. coli and purified. SDS-PAGE analysis confirmed the purity and integrity of the isolated protein, revealing a single band at the expected molecular weight (Both are approximately 30 kDa) (Fig. 15). To test covalent binding, we incubated purified SpyT-sfGFP with E. coli cells expressing SpyC at room temperature for 1 hour. After the reaction, unbound proteins were completely removed by multiple centrifugation washes and observed using fluorescence microscopy. Observations under fluorescence microscopy analysis revealed a strong green fluorescence signal surrounding the bacterial cells, indicating that SpyT-sfGFP successfully and specifically bound to the cell surface (Fig. 13B). These results demonstrate that SpyC is functionally expressed on the E. coli surface and can effectively form a covalent bond with the SpyT-sfGFP fusion protein, confirming that the generation of the fluorescence signal depends on the specific interaction between SpyC and SpyT, successfully validating the effectiveness of our anchoring system.

Synergistic Verification of Nano-PET Degradation Efficiency by the Dual-Anchor System

To verify the synergistic effect of the adhesion protein and the ester hydrolase polygonase, we constructed a dual-anchor fluorescent strain. First, the pET28a (+)-Lpp-OmpA-SpyC plasmid was introduced into Escherichia coli BL21 (DE3) and cultured. Then the engineered bacteria were prepared into electroporation competent cells and electroporated with the pETlac-Plpp1.2-INPNC-mCherry plasmid. The bacteria were then cultured on ampicillin-kanamycin dual-resistance plates. Single colonies were selected for induction. After induction, purified SpyT - sfGFP was incubated with the dual-anchor engineered E. coli cells at room temperature for 1 hour. After the reaction, the cells were washed several times by centrifugation to completely remove unbound proteins, and the cells were observed by fluorescence microscopy. Observations under fluorescence microscopy showed that strong green and red fluorescence signals were observed around the same bacterial cell perspective (Fig. 13C). This result indicates that the E. coli dual surface display system was successfully and simultaneously localized on the surface of the engineered bacteria.

We replaced the fluorescent protein SpyT-sfGFP with SpyT- polygonase and used INPNC carrying mfp-3 to evaluate the degradation efficiency of this dual-anchor system using nano-PET as a substrate and compared it with the single-anchor strain and the free enzyme. After 6 hours of degradation at room temperature, the dual-anchor strain produced the highest concentration of TPA (Fig. 16).

This result clearly demonstrates that co-localizing the adhesion module and the catalytic module on the surface of a single bacterial cell can produce significant synergistic effects. We analyzed that the strong adhesion mediated by mfp-3 tightly anchors the bacteria to the polyester microplastic surface, creating a localized high-concentration substrate environment for the polygonase hydrolase, thereby significantly improving catalytic efficiency, thereby significantly improving the catalytic efficiency by 2.7 times.

Quantitative Detection of the Degradation Product TPA Based on Hydroxyl Radical Oxidation

To accurately evaluate the degradation efficiency of polyester substrates by polygonase ester hydrolase, we developed a highly sensitive fluorescence detection method for TPA. Based on the specific hydroxylation of terephthalic acid (TPA) by hydroxyl radicals (·OH) to produce the highly fluorescent substance 2-hydroxyterephthalic acid (HTPA), we optimized the reaction conditions and established a standard curve between TPA concentration and fluorescence intensity within a 1-minute reaction time (Fig. 17). The curve showed a good linear relationship between TPA concentration and fluorescence intensity (R² > 0.99) within the range of 0.00 to 0.10 mol/L, making it suitable for the quantitative detection of TPA in aquatic environments such as marine environments.

After a 6-hour degradation experiment of nano-PET using the dual-anchor system, 1 ml of the reaction supernatant was added to an equal volume of Fenton's reagent (containing Fe(II) and H₂O₂). The reaction was incubated at room temperature for 1 minute, and the fluorescence intensity was immediately measured. By comparing with a standard curve, the amount of TPA generated in each experimental group was successfully quantified.

The laboratory-scale degradation platform successfully demonstrated the feasibility of constructing a dual-anchor whole-cell system that integrates both adhesion and catalytic functions on the E. coli surface. The INPNC–mfp-3 fusion enabled strong and specific adhesion of engineered cells to polyester substrates, while the SpyCatcher/SpyTag-mediated display of polygonase provided efficient enzymatic degradation capability. When combined in a single E. coli strain, the dual-anchor system exhibited a synergistic degradation effect on nano-PET, yielding a 2.7-fold increase in TPA production compared with single-anchor or free-enzyme systems. Furthermore, the fluorescence-based hydroxyl radical oxidation assay for TPA quantification achieved high linearity (R² = 0.998), enabling accurate and sensitive measurement of polyester degradation products.

These results systematically validate the effectiveness of our constructed dual-anchor whole-cell catalytic platform. By spatially colocalizing the adhesion protein module (mfp-3) and the hydrolase module (polygonase) on the same cell surface, this system significantly enhances the interfacial interaction between the cell and polyester microplastics, making the substrate more accessible and significantly improving the overall hydrolysis efficiency and reaction rate. Furthermore, the dual-anchor system achieves synergistic adhesion and catalysis while maintaining the integrity of the cell structure, demonstrating superior degradation capacity and system stability, providing a solid experimental foundation for the development of an efficient and sustainable biodegradation platform.

Prospective

Building on the successful establishment of a marine-derived polyester biodegradation system, our future work will focus on advancing this platform toward practical, intelligent, and sustainable applications across multiple dimensions.

At the molecular level, we plan to integrate biological sensing modules with our current dual-anchor system to develop a fully biological monitoring and degradation platform. By coupling the detection of degradation intermediates (such as terephthalic acid, TPA) to gene circuit responses, we aim to construct a cell-based biosensor capable of autonomously quantifying plastic degradation efficiency and dynamically regulating enzyme expression, achieving closed-loop control of microplastic degradation in aquatic environments.

At the systems level, we envision expanding this approach into microplastic purification and environmental restoration technologies. By combining immobilized hydrolase films with water filtration units, we plan to construct microplastic purification membranes that can continuously capture and degrade polyester contaminants in marine and freshwater systems. This concept can be integrated with real-time biosensing units for in situ detection of plastic degradation products, forming an intelligent, self-reporting remediation network.

In summary, our project establishes a robust scientific foundation for next-generation biodegradable polyesters and biological microplastic remediation systems. By integrating synthetic biology, material science, and environmental engineering, we aim to create a scalable, intelligent, and ecologically responsible biotechnological solution for global plastic pollution—advancing toward a sustainable circular bioeconomy.