Introduction

Plastics are an integral part of modern life because of their durability and versatility. Polyethylene terephthalate (PET), a type of thermoplastic, is used in a wide range of products, including bottles, food packaging, and textiles. While PET offers many benefits in its usage, its recycling presents challenges. The PET depolymerization process involves treatment with harmful chemicals at temperatures as high as several hundred degrees Celsius1, a substantial environmental burden. Recently, we identified an enzyme named PETase (EC 3.1.1.101) in a PET-assimilating bacterium, Piscinibacter sakaiensis (formerly, Ideonella sakaiensis)2,3,4, isolated from a PET recycling plant in Japan. This enzyme can hydrolyze PET at low temperatures, ranging from 20 to 45 °C. This discovery has led to the development of environmentally friendly PET recycling methods using engineered PETases5,6,7. On the other hand, PET bio-recycling faces acids produced by PET hydrolysis that deactivate PETase, which is most active in alkaline environments4.

Recycling plastics is a challenge to the carbon cycle, which is similar to the decomposition of terrestrial plants on ancient Earth. Leaves and woody components of plants are difficult to decompose, but certain microorganisms overcame this difficulty and gained an advantageous position in acquiring carbon sources8. Certain insects have evolved unique feeding behaviors and digestive systems to participate in this cycle9. Some insects are known to obtain new abilities through symbiosis with microorganisms; however, few, e.g., termites, secrete the microbial enzymes from their own tissues, such as salivary glands10,11. Most insects lack the enzymes needed to break down the polymers in plant cell walls, such as lignin and cellulose, possibly because of their limited ability to evolve enzymes from their endogenous genes9. In contrast, microorganisms are much more diverse and have evolved a wide variety of digestive enzymes, including cellulases12. Additionally, microorganisms frequently engage in horizontal gene transfer13 and their short generation times result in a faster rate of molecular evolution, allowing them to develop enzymes more efficiently14. However, microorganisms capable of degrading plastics, a new organic compound on the planet, are few4.

A recent report showed that homogenates of the wax worm Galleria mellonella exhibited polyethylene (PE)-degrading activity, dramatically reducing the weight of a PE bag within a day15. A PE-degrading enzyme was subsequently identified from their saliva16, however, its activity could not be reproduced by another group, which suggested that the Fourier transform infrared spectroscopy signals used may have included artifact signals generated by contaminants17. These observations imply that other enzyme(s) may contribute to PE degradation in G. mellonella. The reported PE-degrading activity of G. mellonella15,16 motivated us to explore the use of insects as a platform for plastic degradation. Furthermore, the fact that a part of the insect’s gastrointestinal tract is alkaline18,19, providing an optimal environment for PET hydrolysis by PETase4. In this study, for degrading plastics, we generated a model insect D. melanogaster, which expresses PETase from P. sakaiensis in the intestinal tract and salivary glands. This approach was designed to investigate whether PETase can be heterologously expressed and functionally secreted in a multicellular insect host as a proof-of-concept. We then studied PET degradation in the recombinant flies and their environment. In addition, the glycosylated PETase proteins secreted by the D. melanogaster cells were studied in vitro to characterize their PET hydrolytic activity and stability.

Results

Generation of D. melanogaster that secretes PETase proteins

We generated D. melanogaster that secretes PETase proteins in both the gut and the salivary glands. To observe the pH of the intestinal tract of the L3 larvae, bromothymol blue was incorporated into their food. This indicated that the posterior midgut (pm) is neutral to alkaline (Fig. 1a). To express PETase proteins in the pm of D. melanogaster, we referred to the database “Flygut” (https://flygut.epfl.ch/anatomy), which serves as an atlas for the D. melanogaster adult midgut and contains GAL4 lines that can drive gene expression in the fly gut at the pm region20.

Fig. 1: Generation of PETase-expressing D. melanogaster.
figure 1

a Gastrointestinal pH of D. melanogaster 3rd instar larvae. Bromothymol blue was added to Drosophila food to visualize the pH of the digestive tract of L3 larvae: anterior midgut (am), middle midgut (mm), posterior midgut (pm), hindgut (hg), salivary gland (sg). b, c Expression pattern of NP1502 > EGFP. We crossed the GAL4 line NP1502 with the UAS-EGFP line to see where NP1502 could drive the gene. d Expression of PETase by D. melanogaster. Western blotting of whole L3 larvae from NP1502>PETase and NP1502 > AcGFP flies was performed using anti-PETase and anti-GFP antibodies.

Full size image

We tested the EGFP expression in several GAL4 lines and decided to use one of the GAL4 enhancer trap lines, NP1502-Gal4 (#112663, KYOTO Drosophila Stock Center). D. melanogaster carrying the NP1502-Gal4 and the UAS-EGFP, NP1502 > EGFP, showed strong EGFP expression in the anterior midgut (am), posterior midgut (pm), and salivary glands (Fig. 1b, c). Using the NP1502-Gal4 line, we expressed PETase with a secretory signal peptide of the D. melanogaster BiP protein and a hexahistidine tag through the GAL4-UAS system (Supplementary Fig. 1a, b). We successfully observed the secretory expression of PETase in D. melanogaster (NP1502>PETase) as 30–41 kDa bands in western blotting, which were larger than the expected size of 30 kDa (Fig. 1d).

PET degradation by D. melanogaster

D. melanogaster feeds most actively during the larval stage, but because the larvae were unable to gnaw on solid PET film, we supplemented their diet with a water-soluble PET copolymer (wsPET, TK1606301-5) to examine whether PET could be degraded through ingestion. We raised NP1502>PETase flies on this modified diet and collected the L3 larvae to test whether PET degradation occurred by detecting terephthalic acid, a degradation product of wsPET (Fig. 2a). On average, 0.7 nmol of terephthalic acid was detected in a single NP1502>PETase larva, whereas no degradation products were detected in the NP1502 > AcGFP control group. Additionally, we analyzed the terephthalic acid content in the feeding medium (Fig. 2b), which increased during the rearing period, reaching 1.4 nmol/mg of medium on day 12. Meanwhile, no degradation products were detected in the control group. These results suggest that PETase-expressing D. melanogaster contributed to the hydrolysis of wsPET.

Fig. 2: Degradation of a water-soluble PET copolymer (wsPET) by D. melanogaster.
figure 2

a PET degradation products in L3 larvae. The wsPET was mixed into a feeding medium for either NP1502>PETase or NP1502 > AcGFP flies, and the amount of TPA in the resulting L3 larvae was measured. PETase (n = 7); AcGFP (n = 4). No PET degradation products were detected in the NP1502 > AcGFP group (“ND”); therefore, no statistical test was performed. b Temporal changes of PET degradation products in the feeding medium. TK1606301-5/ wsPET was mixed with the feeding medium that housed either NP1502 > AcGFP or NP1502>PETase flies. PET degradation products in the medium were measured every 4 days over a 12-day period (days 0, 4, 8, and 12; n = 3). No PET degradation products were detected when NP1502 > AcGFP was reared. Therefore, no statistical test was performed. Source data underlying the figures are available in Figshare48.

Full size image

The NP1502-Gal4 flies drive protein expression in the intestinal tract and salivary glands, allowing the saliva and excreta of NP1502>PETase flies to degrade PET. In order to test this hypothesis, we provided flies with a solid PET film inserted into the fly breeding medium and maintained the same PET film continuously across generations (Fig. 3a). The breeding medium contains acidic preservatives such as n-butyl p-hydroxybenzoate and propionic acid. In addition, PET hydrolysis products potentially include acids such as terephthalic acid and mono(2-hydroxyethyl) terephthalic acid. To neutralize these acids, we added calcium carbonate to establish a slightly alkaline environment, which would help facilitate efficient hydrolysis of PET by PETase (Fig. 3b). PET degradation by NP1502>PETase flies proceeded slowly but continuously, and was enhanced in media with higher calcium carbonate content (Fig. 3b). In contrast, no degradation was visually observed by eyes in the NP1502 > AcGFP flies, even when calcium carbonate was included, after 60 days of rearing (Fig. 3b). To further investigate the surface morphology of the PET films, we examined them by SEM from both the top-down view (0°) and at a tilt angle of 80°. These observations revealed that increasing the amount of calcium carbonate promoted further degradation of the PET film surface (Fig. 3c). Furthermore, the surface elemental composition of these PET films was analyzed by X-ray photoelectron spectroscopy (XPS), which provided the atomic percentages of oxygen (O1s) and carbon (C1s), along with the calculated O/C ratios (Supplementary Table 2). The control group treated with 2% calcium carbonate NP1502 > AcGFP flies (2%_Day60_AcGFP, n = 8) exhibited an O/C ratio of 0.329 ± 0.011. In comparison, NP1502>PETase samples (n = 8–10) showed elevated O/C ratios. A significant increase in the O1s atomic percentage was observed in the group treated with 10% calcium carbonate and NP1502>PETase flies (10%_Day58_PETase group) (26.28 ± 1.66%, p < 0.05), along with a significantly higher O/C ratio (0.359 ± 0.029, p < 0.05), compared with the AcGFP control. These findings indicate that PETase treatment promotes surface oxidation and/or hydrolysis of PET films in a calcium-carbonate–concentration-dependent manner. Taken together, these results demonstrate that the D. melanogaster NP1502>PETase line can degrade even solid PET films under the optimal pH conditions for PETase activity.

Fig. 3: Degradation of solid PET by D. melanogaster.
figure 3

a Breeding D. melanogaster on a medium containing an inserted PET film. b, c PET film after housed D. melanogaster. D. melanogaster was reared for 15–60 days on the feeding medium containing a PET film-insert and 0–10% calcium carbonate (b). Scanning electron microscopy (SEM) observation of the PET film surface from both the top-down view (0°) and at a tilt angle of 80° (c).

Full size image

PET degradation characteristics by glycosylated PETase

PETase secreted by D. melanogaster was larger in size than predicted (Fig. 1d), suggesting that it undergoes glycosylation due to the eukaryotic nature of the organism. Although glycosylation is rarely observed in bacterial proteins21, the size of native PETase proteins (27.6 kDa) obtained from P. sakaiensis culture and recombinant hexahistidine-tagged PETase proteins produced in E. coli (EcPETase, 28.6 kDa) were in agreement with the predictions (Fig. 4a), indicating a lack of glycosylation. We used D. melanogaster-derived S2 cells and human-derived 293 cells to secrete PETase proteins. The resulting proteins, DmPETase and HsPETase, showed bands of 30–41 kDa and 41–53 kDa (Fig. 4b), respectively, which are larger than the predicted molecular weight (28.6 kDa) of the histidine-tagged PETase protein (Fig. 4a). To confirm their glycosylation, we treated the PETase proteins with Protein Deglycosylation Mix II from New England Biolabs that includes O-glycosidase, PNGase F, α2-3,6,8,9 neuraminidase A, β1-4 galactosidase S, and β-N-acetylhexosaminidasef, which reduced the size of DmPETase and HsPETase (Fig. 4c), indicating that the PETases secreted by eukaryotes are glycosylated. Next, we searched possible N-glycosylation sites on PETase using NetBGlyc-1.022, a server for N-linked glycosylation sites in human proteins, and identified three asparagine (Asn, N) residues at positions 114, 138, and 264 as the most likely sites (Supplementary Fig. 2a). These Asn residues were substituted individually with alanine and expressed them in human 293 cells, resulting in a substantial size reduction for the mutants N114A and N138A, while N264A was comparable to the wild type and N30A as well as N37A (Supplementary Fig. 2b). These results indicate that N114 and N138 are the N-glycosylation sites of PETase in human cells. To further assess glycosylation sites in different expression systems, we digested HsPETase, deglycosylated HsPETase [Hs(deG)PETase], DmPETase, and deglycosylated DmPETase [Dm(deG)PETase] with trypsin, followed by LC-MS analysis (Supplementary Fig. 3). This analysis revealed N-glycosylation of N138 and N264, with N138 being predominantly glycosylated and N264 partially glycosylated in both expression systems. Peptides including N114 were not detected in the trypsin-digested samples; however, additional digestion with chymotrypsin enabled its minimal detection in the Dm(deG)PETase sample, although the glycosylation status of N114 could not be conclusively established from this experiment.

Fig. 4: Effects of glycosylation on PETase.
figure 4

PETases synthesized by prokaryotes P. sakaiensis (Ps) and E. coli (Ec) (a), eukarya, D. melanogaster (Dm) and H. sapiens (Hs) (b). c Deglycosylation (deG) of PETases produced by eukaryotic cells. PETases were detected by western blotting with an anti-PETase antibody. PET hydrolysis by EcPETase (d), DmPETase and Dm(deG)PETase (e), and HsPETases and Hs(deG)PETase (f). Deglycosylation treatment for DmPETase and HsPETase was followed by incubation with PET film at either 30 °C or 37 °C for 24 h. *P < 0.05 and **P < 0.01. Significance was calculated by two-way ANOVA followed by Šídák’s multiple comparison test. Data indicate mean with SE (n = 3). TPA, terephthalic acid; MHET, mono(2-hydroxyethyl) terephthalic acid; BHET, bis(2-hydroxyethyl)terephthalic acid. g, h Long-term PET degradation by PETases. Each PETase was incubated with PET film at 30 °C (g) and 25 °C (h). *P < 0.05 and **P < 0.01. Significance was calculated by two-way ANOVA followed by a two-stage linear step-up procedure of Benjamini, Krieger, and. Yekutieli. Data indicate mean with standard error (SE) (n ≥ 3). i SEM observation of PET film surfaces digested by EcPETase (Ec), DmPETase (Dm), and deglycosylated DmPETase (Dm(deG)). All samples were incubated at 25 °C. PET films treated with EcPETase were collected after 44 days, whereas those treated with DmPETase and Dm(deG)PETase were collected after 71 days. Scanning electron microscopy (SEM) images were acquired from both a top-down view (0°) and at a tilt angle of 80° to visualize surface morphology after enzymatic degradation. Source data underlying the figures are available in Figshare48.

Full size image

Next, we investigated the effect of glycosylation on PET degradation by comparing PETases with and without glycosylation. EcPETase was incubated with PET film at 30 °C and 37 °C for 24 and 48 h, respectively, and the PET degradation products were examined (Fig. 4d). PET degradation products for 24 h were detected similarly at 30 °C and 37 °C. However, for 48 h, the concentration of PET degradation products was lower (0.6-fold) at 37 °C compared to 30 °C (Fig. 4d), suggesting reduced enzyme activity of EcPETase at 37 °C. We also investigated PETase produced by D. melanogaster S2 cells (DmPETase) and Dm(deG)PETase on the PET film hydrolysis activity for 24 h at 30 °C and 37 °C (Fig. 4e). DmPETase showed similar activities at 30 °C and 37 °C, whereas Dm(deG)PETase activity at 37 °C was significantly lower (0.7-fold) compared to 30 °C. The same trend was observed for Hs(deG)PETase, where its activity at 37 °C was lower (0.6-fold) compared to 30 °C (Fig. 4f). Interestingly, HsPETase activity for PET film was dramatically reduced, where no detectable activity at 30 °C but slight activity at 37 °C (Fig. 4f). In both DmPETase and HsPETase, their deglycosylated form showed higher PET hydrolytic activity after 24 h of reaction (Fig. 4e, f). Taken together with the result that PETase-expressing flies degraded solid film PET for a longer time (Fig. 3b), this suggests that the glycosylated PETase is less active but more durable. We, therefore, further examined the PET degradation by these PETase proteins over a longer period (22 days) at 30 °C. The deglycosylation treatment was also performed on EcPETase (Fig. 4g). Both EcPETase and Ec(deG)PETase showed similar degradation trends over 22 days, confirming that deglycosylation treatment itself did not affect their PET hydrolytic activity. DmPETase and Dm(deG)PETase showed lower activity compared to EcPETases. For the initial 3 days, their activities were relatively similar, but thereafter Dm(deG)PETase activity was much lower than DmPETase. Notably, PET degradation products produced by DmPETase after 22 days reached the same level as those produced by EcPETase. Furthermore, PET degradation was examined at 25 °C, the temperature at which the insects were reared, over an extended period (Fig. 4h). EcPETase exhibited the highest initial hydrolytic activity, but degradation nearly plateaued after 28 days. In contrast, DmPETase continued to degrade PET over the long term, and by day 64, the concentration of degradation products (~3.6 mM) had reached a level comparable to that of EcPETase. Interestingly, Dm(deG)PETase initially produced higher amounts of degradation products than DmPETase up to day 8, but its activity gradually declined thereafter, with little further increase observed beyond day 47. Moreover, SEM observations revealed that DmPETase induced punctate, pit-like localized degradation on PET films, and this degradation progressed in a localized manner. This behavior was in clear contrast to the more uniform and widespread surface erosion observed with Dm(deG)PETase and EcPETase (Fig. 4i), indicating that glycosylation alters the mode of action of PETase. Together, these results demonstrate that glycosylation plays a critical role in the stability and sustained activity of DmPETase during long-term PET degradation.

PETase localization in PET degrading system

Deglycosylation of DmPETase and HsPETase dramatically enhanced their PET degradation activities (Fig. 4e, f), indicating that the glycans hinder the interaction between the catalytic site of PETase and the PET film surface. To explore this possibility, we aimed to determine the location of PETase proteins in the reaction mixture. PETase and its variants were incubated with PET film for 24 h (Fig. 5a), after which their localization—whether in the supernatant and/or on the PET film surface—was determined. The localization of Dm(deG)PETase and Hs(deG)PETase was more frequently observed in the PET film fraction compared to that of DmPETase and HsPETase (Fig. 5b). In contrast, DmPETase and HsPETase were more frequently localized in the aqueous phase than Dm(deG)PETase and Hs(deG)PETase, respectively (Fig. 5b), suggesting that glycans limit the access of PETase to PET films. Interestingly, a band corresponding to the EcPETase dimer was detected at approximately 60 kDa (Fig. 5b), whereas no band corresponding to the dimer was detected for either glycosylated or deglycosylated DmPETase and HsPETase (Fig. 5b). Thus, PETase that has an absence of any glycosylation is more likely to form dimers when it is present in a raw state.

Fig. 5: Localization of PETase proteins.
figure 5

a Schematic diagram of PET film degradation reaction system by PETase. EcPETase (Ec), DmPETase (Dm), Dm(deG)PETase [Dm(deG]), HsPETase (Hs), and Hs(deG)PETase [Hs(deG)] were incubated with PET film (pH 7.0, 30 °C) for 24 h. b Then, the samples, as indicated in (a), from the aqueous phase (1), its supernatant after centrifuge (5000 × g, 10 min) (2), and solubilized PETase proteins on PET film (3) were detected by western blot with an anti-PETase antibody. Asterisk points a band corresponding to the EcPETase dimer.

Full size image

Discussion

In this study, we developed an insect model capable of degrading PET by genetically engineering D. melanogaster to secrete PETase. The D. melanogaster was able to degrade a water-soluble PET copolymer, wsPET, in its diet by processing it through the intestinal tract. Additionally, solid PET films placed in the same environment were degraded. Recently, it has been reported that several insect larvae can degrade plastics. For example, larvae of the cereal-feeding pest Plodia interpunctella, with two bacterial strains isolated from their gut contents, degraded polyethylene film23. Similarly, larvae of the litter beetles, Tenebrio molitor, can degrade polystyrene24,25, polyvinyl chloride26, and polyethylene27, with the contribution of gut bacteria. Therefore, employing a symbiotic approach is an interesting concept. However, since P. sakaiensis is an aerobe, colonization of the insect gut is likely unfeasible. Other gut microorganisms, such as E. coli, may represent more suitable candidates. Nonetheless, selecting an E. coli strain capable of establishing symbiosis in the insect gut and secreting PETase would present significant and complex challenges. As an alternative strategy, we turned our attention to D. melanogaster, which has long been used as a model organism, in which many fundamental principles of developmental biology28,29, immunology30, and genetics31 have been established. Its short generation time and ease of genetic manipulation have further enabled functional analyses of disease-related genes and drug discovery screenings32. The extensive genetic engineering tools accumulated through decades of D. melanogaster research enabled us to engineer insects capable of degrading plastics by expressing exogenous plastic-degrading enzymes, without relying on symbiotic bacteria, as is often seen in the degradation of plastics and cellulose in other insects10,11,16.

No overt adverse effects were observed in PETase-expressing D. melanogaster under either standard culture conditions or PET co-culture experiments. This includes general appearance, survival, and behavior, as compared with control flies. Although a slight reduction in fertility was observed under high concentrations of CaCO₃, it was comparable to that observed in the control group under the same conditions. In the context of this study, it is worth noting that NP1502-Gal4 may also drive gene expression in glial cells33,34, a possibility that should be taken into account when interpreting the absence of observable phenotypes. The current level of PET degradation in our system did not appear to affect D. melanogaster survival, however, higher concentrations of PET degradation products could still impact the host, other organisms, or the environment, and this possibility will be carefully evaluated in future studies. With the establishment of this system, future studies can now evaluate the biological effects of feeding on wsPET-derived PET and its degradation products, carefully assess these effects at larger scales and over longer periods. Such evaluations should include not only general physiological impacts but also potential effects on lifespan, reproductive capacity, and overall physiological health.

In the context of enzymatic recycling of PET, technologies that enhance the life cycle of enzymes are essential. Although glycosylated PETase secreted from D. melanogaster and human cells exhibited reduced activity toward solid PET and far from practical application, its durability was improved. The deglycosylation of DmPETase and HsPETase significantly enhanced their binding to the PET film surface. This effect may be attributed to the hydrophilic nature of the sugar chains, which could inhibit the access of glycosylated PETase to the hydrophobic PET. A putative dimer on the PET film surface observed only in EcPETase and absent in both the glycosylated and deglycosylated PETases, suggests that glycosylation prevents inter-protein interactions, such as hydrophobic interactions, which would otherwise lead to PETase aggregation. This reduced aggregation may result in increased PETase activity. In a previous study, leaf and branch compost cutinase (LCC), an enzyme known for its PET hydrolysis activity at high temperatures, was heterologously expressed in the eukaryote Pichia pastoris35. In that study, glycosylation was found to reduce the aggregation of LCC at higher temperatures. These findings indicate that glycosylation plays a key role in preventing the aggregation of PET-degrading enzymes, particularly at the liquid-solid interface where the local density of hydrophobic proteins increases.

In our research, we analyzed the glycosylation sites of PETase secreted from eukaryotic cells using Western blotting and LC–MS. LC–MS analysis revealed that both D. melanogaster- and human-derived PETases showed a high likelihood of glycosylation at N138 and N288, with additional moderate possibilities at N264 and N277. In our system, since insects continuously secrete PETase, it is equally important to provide conditions at biologically permissible temperatures. Therefore, we think the development of PETase variants that can maintain enhanced catalytic activity and durability at relatively low temperatures, through targeted amino acid substitutions, is of critical importance. Moreover, using thermostable PETase variants as scaffolds could provide synergistic effects on stability35.

PETase has also been expressed in eukaryotic hosts such as microalgae and yeast36,37, where a larger version of PETase protein than the native form was detected. In our study, D. melanogaster degraded wsPET and solid PET in the presence of calcium carbonate. The solid PET was primarily degraded in areas where D. melanogaster could crawl, rather than the embedded part in the medium. Interestingly, a study using Phaeodactylum tricornutum observed that this microalga attaches near regions of PET decomposition36. Similarly, the study designed for PETase expression in P. pastoris requires that the fungus attach to PET for effective functioning37. Recently, PETase and class II hydrophobin I from Trichoderma reesei (HFBI) were co-displayed on the yeast cell surface38. In this study, HFBI, an amphipathic protein with both hydrophilic and hydrophobic patches, was utilized to enhance the interaction between yeast cells and PET, resulting in a dramatic increase of PET degradation in the co-display system compared to displaying PETase alone. These studies imply that it is important to have additional factors that effectively facilitate the physical access of glycosylated PETase to solid PET.

Furthermore, one of the reported strategies for enzymatic PET depolymerization is the moist-solid reaction mixture approach, in which PET can be degraded in the presence of only a minimal amount of buffer solution39. This approach has been shown to enhance enzyme stability and substrate accessibility under low-water conditions. In our system, several larvae and pupae were consistently present on the PET film surface, which remained moist but was not submerged in an aqueous solution. The observation that PET degradation proceeded under these conditions suggests a resemblance to the moist-solid reaction mixture environment. These findings indicate that insect-associated systems could serve as a potential realization of the moist-solid reaction concept in future studies.

Taken together, these findings provide a foundation for developing insect-based biocatalytic systems for PET degradation. In recent years, the potential of insects for bioremediation and waste management has been increasingly explored, accompanied by growing interest in genetically engineered insects. Examples include the use of black soldier flies for sustainable waste processing40 and studies employing D. melanogaster as a model platform41. In the current study, although proof-of-concept studies can be successfully conducted in D. melanogaster under tightly controlled laboratory conditions, extrapolating such approaches to other insect species requires considerable caution. The husbandry of genetically engineered insects entails multifaceted challenges, encompassing potential ecological risks, stringent regulatory and ethical constraints, as well as substantial practical and economic burdens. To address these concerns, incorporating a tet-off based lethal gene expression system could provide an additional safeguard, as insects that escape into the environment would not survive. Such tools have already been validated in D. melanogaster42,43, and in principle, could be adapted for other insect species as well. Nevertheless, rigorous biocontainment measures, comprehensive and transparent risk assessments, and the proactive establishment of broad societal consensus will be essential prerequisites for responsible future deployment.

Conclusions

In this study, by feeding D. melanogaster larvae expressing PETase on medium containing wsPET, we demonstrated the PET degradation capacity of the engineered insects. We also found the degradation of solid PET in the appropriate pH environment. Our results showed that the glycosylated PETase maintains its activity for a longer period both in vitro and in vivo, highlighting the advantages of glycosylated enzymes functioning on plastics in the external environment. Importantly, the study also revealed the challenge of effectively attaching PETase to PET surfaces for optimal degradation. Our study highlights the feasibility of employing engineered insects as platforms for plastic biodegradation.

Methods

Plasmid construction

Codon-optimized petase for D. melanogaster with Bip signal sequence fused to the N-terminus and hexahistidine tag fused to the C-terminus (Bipss-DmPETase-6xHis), humanized petase with IL2 signal sequence fused to the N-terminus and hexahistidine tag fused to the C-terminus (IL2ss-HsPETase-6xHis) were artificially synthesized (Eurofins Genomics, Tokyo, Japan). To establish transgenic flies, we subcloned Bipss-DmPETase-6xHis as well as Bipss fused AcGFP (Addgene #54705) into pENTR/D-TOPO (pENTR-Bipss-DmPETase-6xHis or pENTR-Bipss-AcGFP) and then further subcloned them into pUASg.attB (a kind gift from Dr. K. Basler, University of Zürich)44. The plasmid, pUAS-Bipss-DmPETase-6xHis and pUAS-Bipss-AcGFP were injected into fly embryos. For D. melanogaster S2 cellular expression, we subcloned Bipss-DmPETase-6xHis into pTuba-pCopia-Puro vector to obtain pTuba-Bipss-DmPETase-6xHis-pCopia-Puro vector. For human 293-F cellular expression, IL2ss-HsPETase-6xHis gene was subcloned into pCAG-SV40-Hygro cells to obtain pCAG-IL2ss-HsPETase-6His-SV40-Hygro. Mutated hsPETases were obtained using inverse PCR (Supplementary Table 1).

Flies

All fly stocks were maintained and crossed at 23 °C or 25 °C on standard cornmeal medium (KYOTO Stock Center recipe, https://kyotofly.kit.jp/stocks/documents/dgrcqandae.html#lc90bdb5)45 at 25 °C under a 12/12-h light/dark cycle. Transgenic fly lines, NP1502-GAL4 (y[*] w[*] P{w[+mW.hs]=GawB}rdgB[NP1502] / FM7c; stock #112663)46 and UAS-EGFP (w1118; P{w+mC = UAS-EGFP}5a.2; #108231) were obtained from the KYOTO Drosophila Stock Center. To generate UAS-PETase and UAS-AcGFP lines, we injected the plasmids into y[1] M{vas-int.Dm}ZH-2A w[*]; PBac{y[+]-attP-3B}VK00037 (KYOTO Drosophila Stock Center #130449), and then established two recombinant strains, y[1] w[*]; PBac{y[+mDint2] w[+mC]=UAS-hsp-Bipss.PETase.6His}VK00037 / CyO, P{ry[+t7.2]=sevRas1.V12}FK1 (UAS-PETase) and y[1] w[*]; PBac{y[+mDint2] w[+mC]=UAS-hsp-Bipss.AcGFP}VK00037 / CyO, P{ry[+t7.2]=sevRas1.V12}FK1 (UAS-AcGFP). After establishing the lines, we crossed NP1502-GAL4 flies to UAS-PETase or UAS-AcGFP flies to obtain double homozygous GAL4/UAS transgenic lines NP1502>PETase and NP1502 > AcGFP flies, respectively. The following D. melanogaster lines generated in this study have been deposited at the KYOTO Drosophila Stock Center: DGRC #119680 (y[1] w[]; PBac{y[+mDint2] w[+mC]=UAS-hsp-Bipss.AcGFP}VK00037/CyO, P{ry[+t7.2]=sevRas1.V12}FK1), DGRC #119681 (y[1] w[]; PBac{y[+mDint2] w[+mC]=UAS-hsp-Bipss.PETase.6His}VK00037 / CyO, P{ry[+t7.2]=sevRas1.V12}FK1), DGRC #119682 (y[] w[] P{w[+mW.hs]=GawB}rdgB[NP1502]; PBac{y[+mDint2] w[+mC]=UAS-hsp-Bipss.AcGFP}VK00037), and DGRC #119683 (y[] w[] P{w[+mW.hs]=GawB}rdgB[NP1502]; PBac{y[+mDint2] w[+mC]=UAS-hsp-Bipss.PETase.6His}VK00037).

D. melanogaster S2 cell culture

The D. melanogaster S2 cell lines were purchased from Gibco (#R69007) and cultured at 26 °C in Schneider’s Drosophila medium (Gibco, #21720001) supplemented with 10% fetal bovine serum. The plasmid vector pTuba-Bipss-DmPETase-6xHis-pCopia-Puro was transfected into S2 cells using calcium phosphate transfection. Puromycin (InvivoGen, final concentration, 10 mg L-1) was used to select cells stably expressing DmPETase. After selection, fetal bovine serum in the culture medium was gradually removed, and the cells were maintained and passaged in a medium containing 5 mg L-1 puromycin and secreted DmPETase. We collected the supernatant.

Human 293-F cell culture

Expi 293-F (derived from human 293 cells) cell line was purchased from Gibco (#A14527) and cultured in FreeStyle™ 293 Expression Medium (Gibco) at 37 °C in a humidified 95% air and 5–8% CO2 environment on a shaking platform at 120 rpm. The plasmid vector pCAG-IL2ss-HsPETase-6His-SV40-Hygro was transfected into 293-F cells using a NEPA21 electroporator (NEPA GENE). Hygromycin B (FUJIFILM Wako; final concentration, 125 mg L−1) was used to select the cells stably expressing HsPETase.

Protein purification

The medium collected after cell culture was subjected to ultrafiltration to eliminate histidine and loaded onto a TALON resin (Takara Bio) pre-equilibrated with binding buffer (50 mM Tris-HCl pH 7.5, 300 mM NaCl, 5 mM imidazole). After washing with the binding buffer, His-tagged PETase proteins were eluted with elution buffer (50 mM Tris-HCl, pH 7.5, 300 mM NaCl, and 150 mM imidazole). The eluent was desalted using a PD-10 column (Cytiva) into 50 mM Na2HPO4-HCl (pH 7.0). The concentration of purified proteins was determined by measuring the absorbance at 280 nm, where 1 Abs = 1 mg mL−1.

Deglycosylation of PETase

Purified proteins were treated with Protein Deglycosylation Mix II, according to the manufacturer instructions. Briefly, 25 μg of proteins were mixed with Deglycosylation Mix Buffer 1 (non-denaturing reaction) then incubated with Protein Deglycosylation Mix II for reaction at 25 °C for 30 min followed by 30 °C for 16 h.

Site-specific N-glycosylation profiling

DmPETase and HsPETase proteins (1.65 μg) treated with or without Protein Deglycosylation Mix II were digested with trypsin (1 μg) overnight at 37 °C. In addition, DmPETase (0.66 μg) treated with or without Protein Deglycosylation Mix II were digested with a mixture of trypsin (0.5 μg) and chymotrypsin (0.5 μg) overnight at 25 °C. Peptides were analyzed by LC-MS/MS using a Vanquish Neo nanoLC system (Thermo Fisher Scientific) coupled to an Orbitrap Exploris 240 mass spectrometer (Thermo Fisher Scientific). LC separation was performed on an ADVANCE UHPLC system (Michrom BioResources) with a PepMap Neo C18 trap column (5 µm, 300 µm × 5 mm; Thermo Fisher Scientific) and an Aurora Elute C18 analytical column (1.7 µm, 75 µm × 150 mm; IonOpticks), operated at a flow rate of 300 nL min-1. MS data were searched using Mascot v2.8.1 (Matrix Science) to identify deglycosylation events in which glycosylated Asn residues are converted to Asp, resulting in a + 0.984 Da mass shift. Deamidated Asn residues within the Asn-X-Ser/Thr motif were considered N-glycosylation sites.

Western blotting

Samples containing either D. melanogaster larvae expressing PETase or recombinant PETase proteins were mixed with EzApply (ATTO, Japan), denatured, and separated by e-PAGEL HR HER-T7.5 L (ATTO) using an EzRun MOPS (ATTO). After separation, the proteins were transferred onto an Immobilon-E Membrane (Millipore) by electroblotting using EzFastBlot (ATTO). Alternatively, samples were mixed with 2× SDS sample buffer supplemented with 0.2 mM dithiothreitol, denatured, and then separated on a 12.5% Super Sep Ace polyacrylamide gel (FUJIFILM Wako). After separation, the samples were transferred onto a PVDF membrane (ATTO) using blotting buffer (25 mM Tris-HCl, 192 mM glycine, and 20% MeOH, pH 8.3). We used immunoglobulins purified from rabbit serum that had been immunized with PETase protein as an antibody against PETase (1:5000, outsourced to Eurofins Genomics for production). The secondary antibodies used were donkey anti-rabbit IgG-horseradish peroxidase (1:5000; Jackson ImmunoResearch) or goat anti-rabbit IgG-horseradish peroxidase (1:5000; Life Technologies).

Water-soluble PET copolymer (wsPET) degradation by D. melanogaster

The PET copolymer TK1606301-5 (wsPET) was synthesized to contain acids (terephthalic acid/isophthalic acid/5-sulfoisophthalic acid monosodium salt = 60/30/10 mol%) and glycol (ethylene glycol/di-ethylene glycol = 80/20 mol%/acids). The intrinsic viscosity (IV)　of this polymer was 0.36, and the glass transition temperature was 60 °C. We crossed 10 male and 15 female D. melanogaster flies in media containing TK1606301-5 at a final concentration of 5 mg mL-1. We collected third-instar larvae, washed them with PBS, and stored at –80 °C before the analysis of PET degradation products. Moreover, we collected D. melanogaster breeding media on days 0, 4, 8, and 12 and stored at –80 °C. The freeze-stored larvae were sonicated with 10 µL of stopping buffer, 80% 20 mM NaH2PO4, pH 2.5/20% DMSO (v/v), per larva. The breeding media (150 mg) was vigorously suspended in 250 µL stopping buffer. These samples were heated at 80 °C for 10 min, centrifuged at 15,000 × g for 10 min, and the supernatant was filtered using a CosmoNice Filter-S with 0.45-µm pore size (Nacalai Tesque, Kyoto, Japan) to remove insoluble materials. The filtrates were subjected to reverse-phase HPLC to measure the TK1606301-5 hydrolysis products, as previously reported4,47.

PET film degradation by D. melanogaster

Calcium carbonate (CaCO₃) was added to the medium and mixed thoroughly before use. To assess PET film degradation by D. melanogaster, an amorphous PET film (200 µm thickness, 2.4% crystallinity) was inserted upright into the medium and gently embedded so that it remained standing (Fig. 3a), and 15 male and 20 female adult flies were introduced for crossing at that site. The flies were maintained at 25 °C and transferred every 3 weeks to fresh medium. Larvae or pupae attached to the PET film were not removed but were transferred together with it for subculture. Consequently, adult flies and their developing progeny (larvae and pupae) were continuously present on the same PET film throughout the experiment. After each culture period, the PET film was recovered, washed with 1% SDS, rinsed with purified water, and examined by scanning electron microscopy (SEM; SU6600, Hitachi, Japan). The PET film was further prepared for XPS analysis by immersion in isopropanol, ultrasonic washing, immersion in ethanol, and subsequent air-drying. Eight to ten areas of each film surface were analyzed by XPS using a PHI 5000 VersaProbe II photoelectron spectrometer (ULVAC-PHI, Japan) with a monochromatic Al Kα X-ray source (1486.6 eV). The charge shift was calibrated against the C1s peak of surface adventitious carbon at 284.8 eV.

PET film degradation by PETase

The PET film (6 mm diameter, 200 µm thickness, 2.4% crystallinity) was incubated with PETase enzyme at a final concentration of 100 nM in a reaction buffer containing 50 mM Bicine-NaOH (pH 9.0) at 25, 30, or 37 °C. After the reaction, a 5 µL sample of the reaction enzyme was mixed with 195 µL of the stopping buffer. The protein was then denatured by heating at 80 °C for 10 min, followed by centrifugation at 15,000 × g for 10 min. The supernatant was filtered through a Cosmonice Filter-S and subjected to reverse-phase HPLC.

Detection of PET film interacting PETase

One hundred nM of PETase was incubated with the PET film (6 mm diameter, 200 µm thickness, 2.4% crystallinity) at 30 °C for 24 h in 300 µL of 50 mM Na2HPO4-HCl (pH7.0). Subsequently, 20 µL of the supernatant of the reaction solution was mixed with 20 µL of 2× SDS-PAGE sample buffer. The PET film was washed with 50 mM Na2HPO4-HCl buffer (pH 7.0). After removal of residual water by brief spin-down, the film was incubated with 60 µL of 2× SDS-PAGE sample buffer for 10 min with vortexing every 2–3 min to solubilize proteins, followed by mixing with 60 µL of 50 mM Na2HPO4-HCl buffer (pH 7.0) and heating at 90 °C for 10 min. The samples were analyzed by western blotting using anti-PETase antibodies.