Plastic oxidation by peroxidases

Abstract

The present invention relates to plastic degradation via oxidation of a polymer in the plastic using a peroxidase. The peroxidase may comprise a hydrophobic loop. Also provided are recombinant microorganisms expressing the peroxidase and methods for producing the peroxidase.

Description

[0001] PLASTIC OXIDATION BY PEROXIDASES

[0002] CROSS-REFERENCE TO RELATED APPLICATION

[0003] This application claims priority to United States Provisional Application No. 63 / 566,584, filed March 18, 2024, and United States Provisional Application No. 63 / 756,874, filed February 11, 2025, the contents of each of which are incorporated herein by reference in their entireties for all purposes.

[0004] REFERENCE TO U.S. GOVERNMENT SUPPORT

[0005] This invention was made with government support under grant numbers DE- SC0021166 and DE-SC0022018 awarded by Department of Energy. The United States has certain rights in the invention.

[0006] REFERENCE TO SEQUENCE LISTING

[0007] The present application is being filed along with a Sequence Listing in electronic format. The Sequence Listing is provided as a file entitled 2101715- 001282_SequenceListing.xml, created March 18, 2025, which is 23 KB in size. The information in XML file format of the Sequence Listing is incorporated herein by reference in its entirety.

[0008] FIELD OF THE INVENTION

[0009] The invention relates to plastic degradation with peroxidases and recombinant microorganisms and methods for producing the peroxidase.

[0010] BACKGROUND OF THE INVENTION

[0011] More than 300 million tons of plastics are produced globally each year of which >90% are not effectively recycled: this plastic waste ends up in the environment, landfilled, or incinerated, generating negative environmental impacts and increasing greenhouse gas emissions. Mechanical recycling can result in recycled plastics with inferior properties compared to virgin plastics and thus reduced value after many cycles. Chemical recycling and upcycling strategies produce virgin-quality recycled plastics through deconstruction and repolymerization or upcycled products through deconstruction and new synthesis. These technologies are promising but often require high temperatures and pressures for operation and use expensive catalysts that limit industrial adoption. Biological plastics waste valorization offers alternative approaches to chemical recycling with lower operating temperatures and pressures and still provides the opportunity to upcycle plastics waste into higher value products. Recent rapid progress in identifying and engineering poly(ethylene terephthalate) (PET) degrading enzymes have resulted in impressive rates of PET deconstruction up to 61.3 gPETL^h-1and the first commercial-scale PET biological recycling plant. Contrary to PET, methods for the biological deconstruction of non- hydroly sable plastics - which make up approximately 70% of plastics waste - remain undeveloped.

[0012] The most abundant plastics, polyethylene (PE) and polypropylene (PP), are composed exclusively of C-C backbones that lack bonds susceptible to enzymatic hydrolysis by enzymes such as esterases or hydrolases, meaning that they are difficult to deconstruct biologically. Moreover, these C-C bonds are stronger than the labile ester or amide bonds found in hydrolysable plastics. Therefore, C-C bonds must be activated through functionalization such as oxidation to allow for their cleavage. While enzymes have been reported to act on non-hydrolysable plastics, these reports have been met with healthy skepticism and have been difficult to reproduce. Thus, there exists a need to discover enzymes capable of C-C bond activation and cleavage required for deconstruction of non- hydrolysable plastics wastes into monomers or into bioavailable deconstruction products for upcycling.

[0013] Insect larvae, namely Tenebrio molitor (also known as the yellow mealworm), efficiently consume and degrade PE, PP, poly(vinyl chloride), and polystyrene (PS) at rates on the order of weeks - much faster than soil or marine environments in which deconstruction typically takes tens to hundreds of years. Thus, the digestive systems of such insects and their gut microbiota may contain efficient enzy mes for deconstruction of non- hydrolysable plastics; however, it remains unclear how much of this deconstruction is due to the activities of their gut microbiota or the host species itself. A recent report of PE deconstruction by the saliva of the greater wax worm suggested that deconstruction is initiated by the host, but the validity of these claims has been challenged. Most of the literature, though, hypothesizes that gut microbiota drive polymer deconstruction in insect systems through unvalidated oxidative mechanisms similar to alkane metabolism.

[0014] There remains a need for safe and efficient methods for plastic degradation.

[0015] SUMMARY OF THE INVENTION The present invention relates to plastic oxidation with peroxidases. The inventors have surprisingly discovered enzy mes, for example, peroxidases, from yellow mealworm guts that participate in deconstruction of low-density PE (LDPE).

[0016] The present invention provides a recombinant microorganism. The recombinant microorganism comprises a heterologous gene encoding a peroxidase. The peroxidase comprises a hydrophobic loop. The hydrophobic loop may consist of at least 15 amino acids. At least 30% of the amino acids in the hydrophobic loop have a hydrophobic side chain. The hydrophobic loop may consist of an amino acid sequence at least 50% homologous to the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: 21), wherein X is an amino acid. The hydrophobic loop may consist of an amino acid sequence at least 80% identical to an amino acid sequence selected from the group consisting of SEQ ID NO: 11-20.

[0017] According to the recombinant microorganism of the present invention, the peroxidase may consist of an amino acid sequence at least 80% identical to an amino acid sequence selected from SEQ ID NOS: 1-10. In one embodiment, the peroxidase may consist of an amino acid sequence at least 80% identical to the amino acid sequence of SEQ ID NO: 10 and may comprise a tryptophan or tyrosine residue corresponding to W231 in SEQ ID NO: 10 and a try ptophan or tyrosine residue corresponding to W356 in SEQ ID NO: 10.

[0018] The recombinant microorganism may express the peroxidase. The recombinant microorganism may secrete the peroxidase. The recombinant microorganism may display the peroxidase on cell surface of the recombinant microorganism.

[0019] For each recombinant microorganism of the present invention , a composition is provided. The composition comprises the recombinant microorganism. The composition may further comprise a culture medium.

[0020] A method for producing a peroxidase is provided. The production method comprises growing the recombinant microorganism of the present invention in a culture medium; and expressing the peroxidase by the recombinant microorganism, whereby the peroxidase is produced. The production method may further comprise secreting the peroxidase by the recombinant microorganism into the culture medium. The production method may further comprise displaying the peroxidase on cell surface of the recombinant microorganism. The production method may further comprise purifying the peroxidase.

[0021] A method for oxidizing a polymer in plastic is provided. The oxidation method comprises exposing plastic to an effective amount of a peroxidase, and functionalizing a polymer in the plastic, whereby the poly mer is oxidized. The peroxidase comprises a hydrophobic loop. The hydrophobic loop may consist of at least 15 amino acids. At least 30% of the amino acids in the hydrophobic loop may have a hy drophobic side chain. The hydrophobic loop may consist of an amino acid sequence at least 50% homologous to the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: 21), wherein X is an amino acid. The hydrophobic loop may consist of an amino acid sequence at least 80% identical to an amino acid sequence selected from the group consisting of SEQ ID NOS: 11-20.

[0022] According to the oxidation method of the present invention, the peroxidase may consist of an amino acid sequence at least 80% identical to an amino acid sequence selected from SEQ ID NOS: 1-10. In one embodiment, the peroxidase may consist of an amino acid sequence at least 80% identical to the amino acid sequence of SEQ ID NO: 10 and may comprise a tryptophan or tyrosine residue corresponding to W231 in SEQ ID NO: 10 and a tryptophan or tyrosine residue corresponding to W356 in SEQ ID NO: 10.

[0023] The oxidation method may further comprise forming an alcohol, aldehyde, and / or ketone.

[0024] According to the oxidation method of the present invention, the polymer may be low- density polyethylene (LDPE). The polymer may be nonhydrolyzable.

[0025] According to the oxidation method of the present invention, the peroxidase is produced according to the production method of the present invention.

[0026] BRIEF DESCRIPTION OF THE DRAWINGS

[0027] FIG. 1 shows column calibration of the HT-GPC instrument. Calibration was performed from the RI detector using 12 narrow polystyrene standards (6x102-2x106 g / mol) at 140 °C in 1.2.4-tri chlorobenzene. 5th-order calibration curve (R2 > 0.99) calculated using MATLAB's curve fitting toolbox. Due to the molecular weight range chosen for calibration and the column pore size, the upper limit of this 5th-order polynomial calibration curve occurs at ~ 36 minutes, which corresponds to ~ 250-300 g / mol. Beyond this retention time, solvent and molecules elute with minimal separation, and thus molecular weights cannot be calculated reliably.

[0028] FIGS. 2A-2D show Yellow mealworm gut microbiota oxidize LDPE films. (A) Experimental procedure for microbial isolation, and materials characterization. Meahvorms are fed LDPE to enrich their gut microbiome before microbial isolations from the gut were done aerobically, anaerobically, and in the presence of antibiotic and antifungal agents on several media. 335 total isolates were cultivated from PE enriched mealworm guts. (B) Hypothesized routes for enzymatic PE deconstruction. Each number represents a corresponding enzyme family: 1. Dioxygenase, 2. Peroxidase, 3. Monooxygenase, 4. Cytochrome p450 monooxygenase. 5. Monooxygenase, 6. Cytochrome p450 monooxygenase, 7. Esterase, 8. Alcohol dehydrogenase. Peroxidase 9. Alcohol dehydrogenase, Peroxidase, 10. Aldehyde dehydrogenase, 11. Decarboxylase, 12. Baeyer- Villiger Monooxygenase, 13. Esterase. (C) FTIR spectra of LDPE films after five 24-hour doses of mealworm gut isolates. (D) XPS spectral data detailing the C Is bond distribution (main y-axis) and the extent of oxygenation of the PE films by the five isolates (secondary y-axis). Data represents the average of four technical replicates, taken at different locations on the same treated film. Error bars are the standard error of those four independent measurements.

[0029] FIGS. 3A-3C show Characterization of yellow mealworm gut microbiota isolates via growth and degradation of PE substrates (A) Grow th data of bacterial isolates in a mineral medium (MM), MM + LDPE powder, or MM + ultra-high molecular weight PE (UHMWPE) powder. OD600 values were recorded after seven days of growth. (B) Zoom in of the carbonyl region in Fig. 1C detailing ranges where aldehydes, ketones, and esters reside in the carbonyl region. (C) SEC chromatograms detailing no deconstruction by select microbial isolates. * Pseudomonas aeruginosa and Pseudomonas putida were used as control strains and were not isolated from the gut of the yellow- mealworm. Full names for each strain can be found in Supplementary Table 1.

[0030] FIGS. 4A-4D show CvDyP oxidizes, but does not degrade, PE films. (A) Workflow' for target enzyme identification in from mealworm gut isolates. Meahvorm gut isolates were whole genome sequenced. Gene counts from each genome w ere compared with those of closely related taxa to determine families of enzymes from the classes outlined in Fig. IB that may be more abundant in the mealworm gut isolates. Target enzymes were heterologously expressed into E. coli BL21 and tested in vivo on LDPE films. (B) FTIR data detailing oxidation of LDPE films by pure CvDyP after washing with w ater and 70% ethanol. (C) SEC traces indicate that CvDyP does not cleave C-C backbone bonds. (D) Prevalence of DyPs (pfam04261) in mealworm gut microbiomes fed LDPE and HDPE for 30 days relative to naive (as they arrive in our laboratory), Day 0 (starved for 2 days), and oats (fed oats for 30 days) conditions. Protein families present in this analysis were selected as (1) putative enzymes that perform oxygenating chemistries (peroxidases, monooxygenases, dioxygenases) and that are (2) secreted.

[0031] FIG. 5 show's enrichment of alkane metabolism protein families in yellow meahvorm gut microbiome isolates. Families that are statistically enriched relative to the taxonomic Order of an isolate, as measured by a Fisher’s exact test, are depicted with their enrichment (log2(fold change)). All pfams used in this analysis can be found in Table 3. Any enzyme in that table that does not appear in this figure was not statistically enriched, p<0.05.

[0032] FIG. 6 shows difference in CvDyP-treated LDPE FTIR spectra pre and postwashing. FTIR spectra of stripped LDPE treated w ith pure CvDyP before (black) and after (red) film washing with w ater and 70% ethanol. Protein contaminant peaks at 3400 cm’1, 1650 cm1, and 1550 cm’1are removed as a result of washing and carbonyl peaks between 1700 and 1740 cm’1persist.

[0033] FIG. 7 show-s FTIR carbonyl index baseline selection. When taking the baseline of FTIR spectra, we observed differences that lead to inconsistencies in carbonyl index calculations. In order to use appropriate baselines to calculate the carbonyl index, each spectrum was looked at individually to determine where the true baseline is. There are five types of common biological oxidation spectra that are observed in the carbonyl region (1) standard peak, (2) baseline jump, (3) baseline jump + standard peak (4) standard peak w ith protein / biomass contamination, and (5) standard peak with protein contamination and a baseline jump. In each case, a different, appropriate baseline value needs taken. This figure gives best practices for defining the appropriate baseline to use when normalizing spectral values.

[0034] FIG. 8 shows CvDyP activity dependence on H2O2. H2O2 concentration was varied in 300 pL reactions in a 96 well plate at room temperature in pH4 potassium phosphate buffer. Error bars represent the standard error across two biological replicates. The fit curve follows Equation 1 below- where Vmax = 1.0062, Ki = 250, and m = 7.8

[0035] Eq1uation

[0036] FIG. 9 shows change in carbonyl index on in-house stripped LDPE films on lysates from A. coll BL21 (DE3) over-expression cultures of each tested DyP. Carbonyl indices were calculated according to FIG. 6. BL21 : E. coli BL21 control, remainder of the bars in the bar chart correspond to the nomenclature codes provided in Table 5for each Type I DyP that was tested.

[0037] FIGS. 10A-10B show purified type 13 DyPs oxidize PE films. (A) CI from FTIR spectra (calculated as the ratio of the maximum peak height between 1700-1745 cm’1and of the maximum peak height betw-een 1400-1500 cm’1) and % oxygen on stripped LDPE films after treatment with purified DyP peroxidase samples and after washing with water and 70% ethanol. (B) YRB hydrophobicity maps of select inactive (B / DyP) and active (CvDyP) detailing hydrophobic atoms in yellow, positively charged atoms in blue, and negatively charged atoms in red.

[0038] FIGS. 1 1 A-l I B show materials characterization of stripped and un-stripped LDPE. (A) FTIR spectra showing stripped (red) vs. unstripped (black) LDPE. Unstripped plastic contains additive peaks around 3400, 3200, and 1600 cm'1. (B) SEC chromatogram showing molecular weight distribution of stripped (red) vs. unstripped (black) LDPE. In the additive stripping process, LDPE is stripped of lower molecular weight chains, leading to an increase in the average molecular weight.

[0039] FIGS. 12A-12B show that identified LDPE oxidases are active peroxidases. (A) Pyrogallol peroxidase assay endpoint demonstrating activity of purified class i3 DyPs at room temperature after 24 hours at pH4. Error bars represent the standard error of biological triplicate measurements. (B) Activity of LDPE-active (CvDyP, CnDyP, CsDyP) and LDPE- inactive (B12DyP) enzy mes on commercial PE films, measured by carbonyl index. CI were calculated according to FIG. 7.

[0040] FIG. 13 shows Structural alignment of LDPE-active (red) vs LDPE-inactive (grey) DyPs. Grey DyPs are inactive BLIDyP and BoDyP in each image. Each label represents the red protein overlaid atop the two inactive proteins in each case. Hydrophobic loop indicated by black circle. Images generated using Pymol version 2.5.2. Enzy me codes are defined in Table 5.

[0041] FIG. 14 shows sequence alignment of hydrophobic loop region across all tested DyPs. Sequences were aligned using ClustlW. Arrow indicates relative activity level per FIG. 9. A 50% minimum consensus sequence is provided in the top row.

[0042] FIG. 15 shows YRB maps of all tested Dy Ps. Images generated using Pymol version 2.5.2. Yellow represents hydrophobic atoms, blue are positively charged atoms, and red are negatively charged atoms. Enzyme codes are defined in Table 5.

[0043] FIG. 16 shows that the hydrophobic loop region in DyPs can be used to modulate PE-oxidase activity. Changes in carbonyl index (DCI) observed between stripped LDPE films before and after enzymatic treatment with purified enzyme, detailing LDPE-oxidase activity by loop-engineered DyPs relative to native DyPs. Films w ere washed with water and 70% ethanol prior to FTIR spectrum collection. xxDyP- nomenclature refers to the base (xx) enzy me hydrophobic loop being removed for the smaller, less hydrophobic loop of CpDyP. Cp+xxHL nomenclature refers to the addition of the hydrophobic loop from xxDyP to CpDyP in place of the native hydrophobic loop. Error bars represent standard error across four independent biological replicates. * Statistically significant increase in ACI relative to no enzyme control via one-tailed t-test, p<0.05. ** Statistically significant increase in ACI relative to Q>DyP via one-tailed t-test, p<0.05.

[0044] FIGS. 17A-17B show trends between loop hydrophobicity and LDPE oxidase activity. (A) Correlation between hydrophobic loop region hydrophobicity and enzyme activity of DyPs in (A). (B) YRB hydrophobicity maps of mutant and native Corynebacterium DyPs. Images generated using Pymol version 2.5.2. Yellow represents hydrophobic atoms, blue are positively charged atoms, and red are negatively charged atoms. Enzyme codes are defined in Table 5.

[0045] FIGS. 18A-18C show that surface exposed aromatic residues permit LDPE oxidation. (A) Pymol rendering of Alphafold predicted CvDyP structure with active site (DI 86, H280, and R297) and surface exposed aromatics (W231 and W356) displayed as sticks. Distances between LDPE-oxi dizing W231 and bulky aromatic substrate reactive blue 2 (green) and model alkane dotriacontane (pink) are displayed. (B) Relative enzyme activities of CvDyP mutants on model substrate pyrogallol. (C) Change in carbonyl index of CvDyP and mutants after treatment on stripped LDPE. * Statistically significant increase in ACI relative to no enzyme control via one-tailed t-test, p<0.05. ** Statistically significant difference in ACI relative to WT via one-tailed t-test, p<0.05.

[0046] FIG. 19 shows GC FID chromatograms from Cp-CNHL DyP reaction with liquid alkanes. Control chromatograms were gathered for solvent hexane, internal standard 2- decanone, the vacuum dried C10-C40 alkane mixture, and Cp-CNHL reactions with the dried alkane mixture. Each chromatogram is of a single measurement. Each Cp-CNHL chromatogram represents a single reaction at room temperature, pH 4. after 16 hours of reaction time.

[0047] FIG. 20 shows molecular docking of Reactive Blue 2 to aromatic surface residues and dotriacontane to the hydrophobic loop. The top pose from Autodock molecular docking simulations of dotriacontane (magenta) with Cp-CNHL and Reactive Blue 2 (RB2, green)) are overlaid. Simulations purposely forced the binding region to the hydrophobic loop region and to key aromatic residues W300 and Y424 (cyan) for dotriacontane and RB2, respectively. This particular binding was done in order to simulate probable locations for proposed binding of each substrate. A non-conservative minimum number of 44 carbons in chain length for LDPE oxidation were calculated by approximating that the dotriacontane molecule was ~14 angstroms from the active site but needed to be 4.5 angstroms away to be oxidized. This remaining distance of ~9.5 angstroms was then used in conjunction with an approximated 5 angstroms per 6 carbons on the dotriacontane chain, estimated by distance measurements in Pymol. Note that this distance represents the absolute minimum number of carbons if the chain is bound in the optimum position on the Cp-CNHLDyP loop as shown.

[0048] FIGS. 21A-21B show enzyme activity of Cp-CNHL mutants with stabilizing variants. (A) Activity on pyrogallol. (B) Activity on RB5.

[0049] FIGS. 22A-22B show enzyme activity of Cp-CNHL mutants with flexibilityinducing variants. (A) Activity on pyrogallol. (B) Activity on RB5.

[0050] DETAILED DESCRIPTION OF THE INVENTION

[0051] The present invention relates to degradation of plastics by oxidizing polymers in the plastics by peroxidases. The invention is based on the inventors’ surprisingly discovery7of enzymes from yellow mealworm guts that participate in deconstruction of low-density PE (LDPE). Also provided by the present invention are recombinant microorganisms and method for producing the peroxidases.

[0052] The invention selected LDPE due to favorable material properties relative to other PEs. such as lower crystallinity and lower rigidity. Using a lower crystallinity substrate increases the chances of finding biological deconstruction agents, as amorphous domains are more accessible to enzymes and, thus, more readily bio-deconstructable. Previous studies have shown that yellow mealworms deconstruct LDPE, reducing weight-average molecular w eight by approximately 61% and number-average molecular w eight by approximately 40%. The inventors have identified yellow mealworm gut microbes and microbial enzymes that oxidize LDPE in isolation from their native environment by using Fourier transform infrared spectroscopy (FTIR) and x-ray photoelectron spectroscopy (XPS) analyses. More importantly, these studies revealed a sub-class of type I dyedecolorizing peroxidases (DyPs) capable of oxidizing LDPE films. Upregulation in gene count of DyPs in the gut microbiota of mealworms with plastic fed diets underscores the importance of these enzymes to initiating biodeconstruction. This enzyme sub-class contains a distinguishing hydrophobic loop insertion proximal to the active site that modulates the extent of oxidation and may participate in plastics binding. Surface residues proximal to this hydrophobic loop were found to be necessary7for LDPE oxidation, suggesting a non-canonical route of substrate oxidation. The inventors have identified a novel subclass of enzymes that play a pivotal role in initiating deconstruction of LDPE. opening new avenues for research into enzymatic polyolefin deconstruction and upcycling efforts.

[0053] The term “plastic"’ as used herein refers to a material made of synthetic or semisynthetic organic polymers that can be molded into a solid object having a desirable shape.

[0054] The term “plastic degradation” as used herein refers to breakdow n of a plastic into small particles or chemical components through a physical, chemical and / or biological process.

[0055] The term “homolog” as used herein refers to a protein that is a counterpart to a naturally occurring protein. The homolog may consist of an amino acid sequence similar to that of its corresponding naturally occurring protein. The similarity or homology may be at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40- 60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%. 40-100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%. 50-99%, 50-100%. 60-70%, 60-80%, 60-90%, 60-95%. 60-99%, 60- 100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80-90%, 80-95%, 80-99%, 80-100%, 90-95%, 90-99%, 90-100%, 95-99%, 95-100%, or 99-100%.

[0056] The term “peroxidase” as used herein refers to a naturally occurring enzy me that catalyzes oxidation of a hydrogen donor substance using peroxides or a homolog thereof. A primary or secondary alcohol may be formed. As an alcohol group has a lower activation energy7for subsequent oxidation than non-oxidized carbons, the primary or secondary alcohol group may be further oxidized into an aldehyde or ketone, respectively. The peroxidase may be expressed by a microorganism, which lives in a gut microbiome. The microorganism may be a yellow mealworm. The peroxidase may be expressed naturally by Brevibacterium linens, Brevibacterium ocianii, Corynebacterium glutamicum, Corynebacterium neomassiliense, Coryneb acterium provencense. Corynebacterium sp., Corynebacterium variabile. Rothia halotolerans, and Unclassified Brevibacterium. Table 6 shows amino acid sequences of exemplary^ peroxidases.

[0057] The term “hydrophobic loop" as used herein refers to a region in a protein structure rich in amino acids having a hydrophobic side chain. Exemplary amino acids having a hydrophobic side chain include alanine, valine, isoleucine, leucine, methionine, phenylalanine, tyrosine, tryptophan. Table 7 shows amino acid sequence of exemplary hydrophobic loops of peroxidases. The terms “oxidizing” and “oxidation” are used herein interchangeably and refer to a process in which an atom or a group of atoms lose electrons.

[0058] The term “functionalizing” or “functionalization” are used herein interchangeably and refer to introducing or modifying a functional group on a molecule or surface to impart specific properties or reactivity.

[0059] The term “nonliydrolyzable” as used herein refers to a chemical structure that contains no bonds susceptible to hydrolysis such as ester or amide bonds.

[0060] The term “effective amount” as used herein refers to an amount of the peroxidase sufficient to achieve a goal, for example, oxidation of a polymer in plastic.

[0061] The present invention provides a recombinant microorganism. The recombinant microorganism comprises a heterologous gene encoding a peroxidase.

[0062] The recombinant microorganisms may be Yarrowia, E. coli, Pichia pastoris, or S. cerevisiae. The Yarrowia may be Yarrowia lipolytica. The recombinant microorganism may be selected from the group consisting of Brevibacterium linens, Brevibacterium ocianii. ('orynebacterium glutamicum, Corynebacterium neomassiliense, Corynebacterium provencense, Corynebacterium sp.. Corynebacterium variabile, Rothia halotolerans, and Unclassified Brevibacterium.

[0063] According to the recombinant microorganism of the present invention, the peroxidase may comprise a hydrophobic loop. The hydrophobic loop may have at least about 10, 11, 12. 13. 14, 15, 17. 18, 19, 20, 25. 30, 35, 40, 45. or 50 amino acids, or about 10-50, 15-35 or 16-27 amino acids.

[0064] According to the recombinant microorganism of the present invention, at least about 20%. 30%. 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 20-50%, 20-60%, 20- 70%, 20-80%, 20-90%, 20-95%, 20-99%, 20-100%, 30-50%, 30-60%, 30-70%, 30-80%, 30-90%, 30-95%, 30-99%, 30-100%. 40-50%, 40-60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%, 40-100%, 50-60%, 50-70%. 50-80%, 50-90%, 50-95%, 50-99%. 50-100%, 60- 70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80-90%, 80-95%, 80-99%, 80-100%, 90-95%, 90-99%, 90-100%, 95-99%, 95- 100%. or 99-100%, of the amino acids in the hydrophobic loop may have a hydrophobic side chain.

[0065] According to the recombinant microorganism of the present invention, the hydrophobic loop may consist of an amino acid sequence at least about 40%. 50%. 60%. 70%. 80%. 90%. 95%. 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%, 40- 100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%, 80-100%, 90-95%, 90-99%, 90-100%, 95-99%, 95-100%, or 99- 100%. identical to the amino acid sequence of a hydrophobic loop in a naturally occurring peroxidase. The hydrophobic loop may consist of the amino acid sequence of a hydrophobic loop in a naturally occurring peroxidase.

[0066] According to the recombinant microorganism of the present invention, the hydrophobic loop may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%, 40- 100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%. 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%. 80-100%, 90-95%. 90-99%, 90-100%. 95-99%. 95-100%, or 99- 100%, identical to the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: 21), wherein each X may be any amino acid. The hydrophobic loop may consist of the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: 21), wherein each X may be any amino acid. Amino acid residues YNY and TG in SEQ ID NO: may be conserved on either side of the hydrophobic loop, and the hydrophobic loop length may vary depending on an insertion in the region between the amino acid residues YNY and TG.

[0067] According to the recombinant microorganism of the present invention, the hydrophobic loop may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40-60%, 40-70%. 40-80%, 40-90%, 40-95%, 40-99%, 40- 100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%, 80-100%, 90-95%, 90-99%, 90-100%, 95-99%, 95-100%, or 99- 100%. identical to an amino acid sequence selected from SEQ ID NOS: 11-20. The hydrophobic loop may consist of an ammo acid sequence selected from SEQ ID NOS: 11-20.

[0068] According to the recombinant microorganism of the present invention, the peroxidase may consist of an amino acid sequence at least about 40%. 50%. 60%. 70%. 80%. 90%. 95%. 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%, 40- 100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%, 80-100%, 90-95%. 90-99%, 90-100%, 95-99%, 95-100%, or 99- 100%, identical to the amino acid sequence of a naturally occurring peroxidase. The peroxidase may consist of the amino acid sequence of a naturally occurring peroxidase.

[0069] According to the recombinant microorganism of the present invention, the peroxidase may consist of an amino acid sequence at least about 40%. 50%. 60%. 70%. 80%. 90%. 95%. 99% or 100%, or about 40-50%. 40-60%, 40-70%, 40-80%. 40-90%. 40-95%, 40-99%, 40- 100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%, 80-100%, 90-95%. 90-99%, 90-100%, 95-99%, 95-100%, or 99- 100%. identical to an amino acid sequence selected from the group consisting of SEQ ID NOS: 1-10 (Table 6). The peroxidase may consist of an amino acid sequence selected from the group consisting of SEQ ID NOS: 1-10.

[0070] According to the recombinant microorganism of the present invention, the peroxidase may be a homolog of a naturally occurring peroxidase and may comprise one, two or more amino acids, for example, tryptophan or typrosine, corresponding to a tryptophan or tv prosin residue on the surface of the naturally occurring peroxidase. The peroxidase may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%. or about 40- 50%, 40-60%, 40-70%. 40-80%, 40-90%, 40-95%. 40-99%. 40-100%, 50-60%. 50-70%. 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80-90%, 80-95%, 80-99%, 80- 100%. 90-95%, 90-99%, 90-100%, 95-99%, 95-100%. or 99-100%, identical to the amino acid sequence of CvDyP (SEQ ID NO: 10), and may comprise a tryptophan or tyrosine residue corresponding to W231 in CvDyP (SEQ ID NO: 10) and / or a tryptophan or tyrosine residue corresponding to W356 in CvDyP (SEQ ID NO: 10). The peroxidase may consist of the amino acid sequence of CvDyP (SEQ ID NO: 10), and may comprise a try ptophan or ty rosine residue corresponding to W231 in CvDyP (SEQ ID NO: 10) and / or a try ptophan or tyrosine residue corresponding to W356 in CvDyP (SEQ ID NO: 10).

[0071] According to the recombinant microorganism of the present invention, the recombinant microorganism may express the peroxidase. The recombinant microorganism may secrete the peroxidase. The recombinant microorganism may display the peroxidase on cell surface of the recombinant microorganism.

[0072] For each recombinant microorganism of the present invention, a composition is provided. The composition comprises the recombinant microorganism. The composition may further comprise culture medium. The culture medium may be suitable for the recombinant microorganism to grow and / or express the peroxidase.

[0073] For each recombinant microorganism of the present invention, a method for producing a peroxidase is provided. The production method may comprise growing the recombinant microorganism in a culture medium; and expressing the peroxidase by the recombinant microorganism such that the peroxidase is produced. The production method may further comprise secreting the peroxidase by the recombinant microorganism into the culture medium. The production method may further comprise displaying the peroxidase on cell surface of the recombinant microorganism. The production method may further comprise purifying the peroxidase. The peroxidase may be purified from the recombinant microorganism by, for example, using nickel affinity chromatography in case an affinity tag His is used.

[0074] The present invention also provides a method for oxidizing a polymer in plastic. The oxidation method comprises exposing plastic to an effective amount of a peroxidase; and functionalizing a polymer in the plastic, whereby the polymer is oxidized.

[0075] According to the oxidation method of the present invention, the peroxidase may comprise a hydrophobic loop. The hydrophobic loop may have at least about 10, 11, 12. 13. 14, 15, 17. 18. 19, 20, 25, 30, 35, 40, 45, or 50 amino acids, or about 10-50, 15-35 or 16-27 amino acids.

[0076] According to the oxidation method of the present invention, at least about 20%. 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 20-50%, 20-60%, 20-70%, 20- 80%, 20-90%, 20-95%, 20-99%, 20-100%, 30-50%, 30-60%, 30-70%, 30-80%, 30-90%, 30-95%, 30-99%. 30-100%, 40-50%. 40-60%, 40-70%, 40-80%, 40-90%. 40-95%, 40-99%, 40-100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60- 80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80-90%, 80-95%, 80-99%, 80-100%, 90-95%, 90-99%, 90-100%, 95-99%, 95-100%, or 99- 100%. of the amino acids in the hydrophobic loop may have a hydrophobic side chain.

[0077] According to the oxidation method of the present invention, the hydrophobic loop may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%, 40- 100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%. 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%. 80-100%, 90-95%. 90-99%, 90-100%. 95-99%. 95-100%, or 99- 100%, identical to the amino acid sequence of a hydrophobic loop in a naturally occurring peroxidase. The hydrophobic loop may consist of the amino acid sequence of a hydrophobic loop in a naturally occurring peroxidase. According to the oxidation method of the present invention, the hydrophobic loop may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%. 40-90%, 40-95%, 40-99%, 40- 100%. 50-60%. 50-70%, 50-80%, 50-90%. 50-95%, 50-99%, 50-100%. 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%, 80-100%, 90-95%, 90-99%, 90-100%, 95-99%, 95-100%, or 99- 100%. identical to the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: 21), wherein each X may be any amino acid. The hydrophobic loop may consist of the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: 21), wherein each X may be any amino acid. Amino acid residues YNY and TG in SEQ ID NO: 21 may be conserved on either side of the hydrophobic loop, and the hydrophobic loop length may vary7depending on an insertion in the region between the amino acid residues YNY and TG.

[0078] According to the oxidation method of the present invention, the hydrophobic loop may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%, 40- 100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80- 90%, 80-95%, 80-99%. 80-100%, 90-95%. 90-99%, 90-100%, 95-99%. 95-100%, or 99- 100%, identical to an amino acid sequence selected from SEQ ID NOS: 11-20. The hydrophobic loop may consist of an amino acid sequence selected from SEQ ID NOS: 11-20.

[0079] According to the oxidation method of the present invention, the peroxidase may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%, 40-90%, 40-95%, 40-99%, 40-100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%. 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%. 70-95%, 70-99%, 70-100%. 80-90%, 80-95%, 80- 99%, 80-100%, 90-95%, 90-99%, 90-100%, 95-99%, 95-100%, or 99-100%, identical to the amino acid sequence of a naturally occurring peroxidase. The peroxidase may consist of the amino acid sequence of a naturally occurring peroxidase.

[0080] According to the oxidation method of the present invention, the peroxidase may consist of an amino acid sequence at least about 40%, 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, or about 40-50%, 40-60%, 40-70%, 40-80%. 40-90%, 40-95%, 40-99%, 40-100%, 50-60%, 50-70%, 50-80%, 50-90%, 50-95%, 50-99%, 50-100%, 60-70%, 60-80%, 60-90%, 60-95%, 60-99%, 60-100%, 70-80%, 70-90%, 70-95%, 70-99%, 70-100%, 80-90%, 80-95%, 80- 99%, 80-100%. 90-95%, 90-99%, 90-100%, 95-99%, 95-100%, or 99-100%, identical to an amino acid sequence selected from the group consisting of SEQ ID NOS: 1-10 (Table 6). The peroxidase may consist of an amino acid sequence selected from the group consisting of SEQ ID NOS: 1-10.

[0081] According to the oxidation method of the present invention, a homolog of a naturally occurring peroxidase may further comprise one, two or more amino acids, for example, tryptophan or typrosine, corresponding to a try ptophan or typrosin residue on the surface of the naturally occurring peroxidase. For example, the homolog of a naturally occurring peroxidase may comprise a tryptophan or tyrosine residue corresponding to W231 in CvDyP (SEQ ID NO: 10) and / or a tryptophan or tyrosine residue corresponding to W356 in CvDyP (SEQ ID NO: 10).

[0082] According to the oxidation method of the present invention, the recombinant microorganism may express the peroxidase. The recombinant microorganism may sccrctc the peroxidase. The recombinant microorganism may display the peroxidase on cell surface of the recombinant microorganism.

[0083] The oxidation method may further comprise forming an alcohol, aldehyde, and / or ketone. According to the oxidation method of the present invention, the polymer may comprise polypropylene (PP), polyethylene (PE), polystyrene (PS), or polyethylene terephthalate (PET). The PE may be ultra-high molecular weight polyethylene (UHMWPE), high-density polyethylene (HDPE), or low-density polyethylene (LDPE). In one embodiment, the polymer may be LDPE.

[0084] According to the oxidation method of the present invention, the polymer may be nonhydrolyzable.

[0085] According to the oxidation method of the present invention, the peroxidase may be produced according to the production method of the present invention.

[0086] The term “about” as used herein when referring to a measurable value such as an amount, a percentage, and the like, is meant to encompass variations of ±20% or ±10%. more preferably ±5%, even more preferably ±1%, and still more preferably ±0.1% from the specified value, as such variations are appropriate.

[0087] Example 1. Biological polyethylene deconstruction initiated by oxidation from DyP peroxidases

[0088] Polyethylene (PE) is the most commonly used plastic on Earth due to its favorable material properties such as high ductility7, mechanical strength, and bond homogeneity7that make the material resistant to deconstruction. However, the lack of robust recycling infrastructure for PE end-of-life management is leading to an estimated 4 million tons of environmental accumulation annually, with implications for human and environmental health. Biological deconstruction and upcycling could potentially aid in PE waste management by allowing for high-yield conversion of waste plastics to high value products, although such processes are not yet possible. In this work, we mined the gut of low-density PE (LDPE) fed mealworms that can reduce LDPE molecular weight by >40% and discovered dye decolorizing peroxidases (DyPs) that oxidized LDPE, initiating biological deconstruction. A plastic-active DyP is characterized by a hydrophobic loop near its active site that helps mediate binding and tunes activity . LDPE oxidation is driven by surface exposed residues proximal to the active site enabling activity on polymeric substrates. Our work provides robust evidence for enzymatic LDPE deconstruction and identifies molecular targets for further development to realize scalable biological LDPE upcycling.

[0089] Materials and Methods:

[0090] Mealworm cultivation: For each gut cultivation experiment, 100 yellow mealworms (Medium yellow mealworms were purchased from Rainbow Mealworms. Compton, CA, USA) were starved for 48 hours upon arrival and then placed in a Tupperware box with approximately 0.5 g of LDPE in a static incubator at 25 °C, 70% humidity7. Control worms fed oats were given 1.0 g of oats every 3 days to supplement feeding. In order to avoid beetles and mealworm cannibalism, dead and pupated larvae were removed every 2 days. Fourier transform infrared spectroscopy FTIR: Commercial (0.035 mm thick, biaxially oriented LDPE films, purchased from Goodfellow Cambridge Limited, Huntingdon, England. Catalog number LS580568) LDPE films and in-house stripped LDPE films were analyzed for changes to the inherent functional groups using FTIR (Thermo Scientific Nicolet iS5 FTIR Spectrometer, Pittsburgh, PA). Spectra for each powdered sample were recorded on a diamond crystal, attenuated total reflectance (ATR) cell. Spectra were recorded in the range of wavelengths 4000-500 cm1with a minimum of 32 scans and a spectral resolution of 0.482 cm Films were rinsed with 500 pL of water followed by 500 pL of 70% ethanol by vortexing for 1 minute at 200 rpm. If FTIR spectra indicated microbial or protein contamination, 70% ethanol on a cotton swab was lightly wiped over the surface to clean residual contamination. Size Exclusion Chromatography (SEC): Molecular weight characterization through SEC was conducted on a TOSOH HLC-8312GPC / H with two TSKgeiGMHHR-H(20)HT columns and one TSKgeiG2000HHR in series, coupled with refractive index (RI) and viscosity detectors. HPLC-grade 1 ,2,4-trichlorobenzene (TCB) stabilized with 500 ppm butylated hydroxytoluene (BHT) was used as the mobile phase. LDPE samples (3-10 mg) were dissolved in mobile phase (3-6 mL) at 140 °C for a minimum of two hours to generate nominal concentrations of 2 mg / mL. 300 pL of samples were injected and eluted for 80 minutes at a flow rate of 0.8 mL min'1at 140 °C.

[0091] MWDs were determined using the RI responses, and SEC samples were calibrated against linear polystyrene standards (12 runs between 6xl02-2xl06g / mol'1; FIG. 1). Molar masses of LDPE obtained were corrected using the Mark-Houwink relationship.

[0092] Metagenomic DNA extraction and sequencing: Following mealworm cultivation, guts from LDPE-fed, HDPE-fed, and oats fed guts, along with naive (as they arrive from Rainbow Mealworms) and starved (starved for 2 days after arrival from Rainbow Mealworms) were extracted from 10 mealworms, pooled together, and vortexed for 5 minutes at 2000 rpm to extract gut contents. A quick ~5 second centrifuge step was then used to pull remaining gut tissue to the bottom of a microcentrifuge tube without disturbing microbial biomass. The supernatant containing microbial content was removed and DNA was extracted using Monarch Genomic DNA extraction kit and sent to the Joint Genome Institute (JGI) for sequencing. An input of 50 ng of genomic DNA was sheared to 6 kb - 10 kb using the Megaruptor 3 (Diagenode). The sheared DNA was treated with exonuclease to remove single-stranded ends, DNA damage repair enzyme mix, end-repair / A-tailing mix and ligated with amplification adapters using SMRTbell Express Template Prep Kit 2.0 (PacBio) and purified with ProNex Size-Selective Purification System (Promega). The purified ligation product was split into t o reactions and enriched using 10-18 cycles of PCR using SMRTbell gDNA Sample Amplification Kit (PacBio). The amplified product was combined and treated with DNA damage repair enzyme mix, end-repair / A-tailing mix and ligated with barcoded overhang adapters. Up to sixteen libraries w ere pooled in equimolar concentrations and the pooled libraries were size-selected using the 0.75% agarose gel cassettes with Marker SI and High Pass protocol on the BluePippin (Sage Science). PacBio Sequencing primer w as then annealed to the SMRTbell template library and sequencing polymerase w as bound to them using Sequel II Binding kit 2.0. The prepared SMRTbell template libraries were then sequenced on a Pacific Biosystems' Sequel lie sequencer using SMRT Link 10.2, tbd-sample dependent sequencing primer, 8M vl SMRT cells, and Version 2.0 sequencing chemistry with 1x1800 sequencing movie run times.

[0093] Metagenomes sequenced by the JGI are publicly available on Integrated Microbial Genomes & Microbiomes (IMG / MER) under the following IMG Genome portal numbers:

[0094] Naive mealworm gut metagenome: 3300066518

[0095] Day 0 (starved) mealworm gut metagenome: 3300066519 30-day HDPE fed mealworm gut metagenome: 3300056814 30-day LDPE fed mealworm gut metagenome: 3300056790 30-day oats fed mealworm gut metagenome: 3300066518 Metagenomic data analysis to identify protein families of interest: All genes corresponding to pfams containing function peroxidase, monooxygenase, dioxygenase, and laccase were identified using the IMG Genome Function Profile tool across the metagenomes. These Genes obtained from JGI sequenced metagenomes were screened then with signalP 6.0 Fast version to determine secretion tag presence. Signal P settings were organism-other, and format none. All genes that SignalP tagged as having any output besides "other" were considered to be secreted. Any gene without a secretion tag was eliminated from a .gff annotation file for pfams generated by the JGI standard pipeline, and then the gene count of each pfam was used for plotting.

[0096] Microbial isolation from mealworm guts: Mealworms were fed diets of HDPE, LDPE, PP, and PS with or without oats, according to the mealworm cultivation method listed above. Mealworms were enriched on plastic for 20 days, with gut extractions and isolations occurring on days 5, 10, 15 and 20. Mealworm guts were extracted according to protocol in mealworm cultivation section. One mL of sterile phosphate buffered saline (PBS) was added to each sample and samples were vortexed for 5 minutes at 2000 rpm. Mealworms gut solutions were plated onto the following media agar: LB (Sigma- Aldrich Catalog No. L3147-1KG), YPD (Sigma- Aldrich, Catalog No. Y1500-250G), PDA (Millipore Sigma. Catalog No. 110130), Sheep's Blood (Rockland Chemical Company. Item No. R111-0050), and MacConkey agar plates (Becton, Dickinson and Company, SK.U: 211387). Mealworm guts were also plated onto each type of agar including 50 mg / rnL of antibiotics penicillin and streptomycin or including 50 mg / mL of antimycotic amphotericin. Individual colonies from morphologically unique isolates were picked, grow n in the corresponding liquid medium, and stocked into -80 °C storage in 20% glycerol. Starting with mealworm gut extractions, this process was repeated anaerobically to cultivate any anaerobic microbes from the mealworm gut.

[0097] Growth assay screening for PE deconstruction: All cultivated microorganisms from mealw orm gut isolations were screened for PE deconstruction ability through a grow th assay. Microbes were grown in a mineral medium (MM) (1 g NaEEPC . 0.25 g MgSO-i * 7H2O, 0. 1 g KH2PO4, and 0.5 g of yeast extract per 500 mL water) as a control, and then with 0.3% w. / v.LDPE powder (LDPE powder was purchased from Goodfellow^ Cambridge Limited, Huntingdon, England; catalog number LS563303). Microbes that show ed improved growth after 7 days on MM + LDPE relative to MM via OD 600 measurement were selected for further study. The mineral medium was bootstrapped with a low concentration of yeast extract to allow microorganisms to adapt and produce the enzymes needed to grow on PE as a primary carbon source. Microbial isolates that showed an increase in growth on MM + LDPE relative to MM were selected as potential plastic degraders.

[0098] Plastic oxidation assays by microbial isolates: Starter cultures of each isolate were grown in Tryptic Soy Broth (Sigma-Aldrich, Catalog No. 22092-500G) from freezer stocks. Optical densities were recorded of each overnight culture and the number of microbes for inoculation was normalized accordingly, assuming OD600 as a proxy for number of bacterial cells. 3 mL of OD 1.0 overnight culture was pelleted, washed twice using MM and then resuspended in MM for inoculation. MM was used to mimic conditions used in growth screening assays and to provide microorganisms with non-carbon-based essential nutrients. Inoculum was dosed onto an MM plate, covered with an LDPE film (Goodfellow Cambridge Limited, Huntingdon. England. Catalog number LS580568). and then a second inoculum was dosed atop the film. Following the same protocol, microbes were dosed onto films daily for 5 days. After treatment, films were washed via vortexing in 500 pL w ater for 2 minutes at 2000 rpm, follow ed by vortexing in 500 pL of 70% ethanol under the same conditions. Films were then subject to FTIR and X-Ray Photoelectron Spectroscopy (XPS) analysis. Films subject to XPS analysis were additionally washed in Sodium Dodecyl Phosphate (SDS) to remove additional adventitious carbon contaminants. Additionally, films were analyzed via SEC to determine changes in the MWD of the polymer to delineate which isolates can degrade LDPE.

[0099] XPS: XPS spectral scans. Cis, and Ols scans were collected on four locations on each film, using the following settings: Ci s spectra were analyzed assuming a C-C, C-H peak at 285.0 eV and then peak shifts of 1.5 eV, 3.0 eV, and 4.5 eV for C-O, C=O, and O- C=O peaks, respectively. All peak fitting analysis was performed using Casa XPS. In survey spectra that contained residual contamination in the form of nitrogen or sodium from protein or SDS from washing, residual oxygen was subtracted by subtracting 1% O for every 1% N and by subtracting 1% O for every 4% Na, respectively. All spectral analysis was assessed by averaging the results from survey scans on four randomly selected points each film.

[0100] Microbial isolate DNA extraction and genome sequencing: Microbial isolates Staphylococcus lentus, Enterococcus termitis, Corynebacterium Variabile, Brevibacterium Epidermidis, and Kocuria halotolerans were grown overnight in TSB and DNA was extracted using NEB Monarch Genomic DNA Purification Kit (Monarch Genomic DNA purification Kit #T3010L, New England BioLabs, Ipswich, MA. USA.), following the kit protocol. Genomic DNA quality was confirmed via qubit and DNA gel electrophoresis. Genomic DNA was sequenced by the Joint Genome Institute.

[0101] Isolate Minimal Draft Genome Sequencing and Assembly: The draft genome of each isolate Staphylococcus lentus, Enterococcus termitis, Corynebacterium Variabile, Brevibacterium Epidermidis, and Kocuria halotolerans was generated at the DOE Joint Genome Institute (JGI) using Illumina technology. An Illumina standard shotgun library was constructed and sequenced using the Illumina NovaSeq S4 platform which generated at least 6.7 million reads for each isolate, and completeness of 98.42 - 100%. Raw Illumina sequence was quality filtered using BBTools per SOP 1061. The following steps were then performed for assembly: (1 ) artifact filtered and normalized Illumina reads were assembled with SPAdes (version v3.15.3; -phred-offset 33 -cov-cutoff auto -t 16 -m 64 -careful -k 25,55,95); (2) contigs were discarded if the length was <lkb (BBTools reformat.sh: minlength=1000 ow=t).

[0102] Genomes sequenced by the JGI are publicly available on Integrated Microbial Genomes & Microbiomes (IMG / MER) under the following IMG Genome portal numbers:

[0103] Brevibacterium epidermidis'. 8012935351 Corynebacterium variabile'. 8081958687

[0104] Enterococcus termitis.' 8012939035 Staphylococcus lentus'. 8012942269 Genome annotation:

[0105] All genomes were annotated using annotation System IMG (IMGAP v5.1.13). Gene calling program was GeneMark.hmm-2 vl.25_lic; INFERNAL 1.1.3 (Nov 2019); Prodigal v2.6.3; tRNAscan-SE v.2.0. 12 (Nov 2022). The annotation algorithm was: lastal 1256; HMMER 3. Ib2; signalp 4.1; decodeanhmm 1.1g. The following databases were used to annotate each gene: Rfam 13.0; IMG-NR 20211118; SMART 01_06_2016; COG 2003; TIGRFAM vl5.0; SuperFamily vl.75; Pfam v34.0; Cath-Funfam v4.

[0106] Comparative genomic analysis for enzyme selection: Genomic sequences of each isolate were mined for protein families (pfam) of monooxygenases, dioxygenases, laccases, and peroxidases using the JGI Integrated Microbial Genomes (IMG) tool. The raw pfam gene counts in the microbial genomes were compared to the average gene count of the same pfam across the taxonomic order. The Joint Genome Institute (JGI) integrated microbial genomes (IMG) workspace was used to compile all genomes from the taxonomic order of each strain Staphylococcus lentus. Enterococcus termitis. Corynebacterium variabile, Brevibacterium epidermidis, and Kocuria halotolerans . The IMG Genome Function Profile tool was then used to generate a count of all genes corresponding to putative plastic degrading pfams (Supplementary Table 3) across the genome in the taxonomic order. These jobs were submitted to IMG and subsequently downloaded on July 22, 2022. The number of genes in each pfam w ere then averaged across all genomes from the order. Gene counts were compared statistically using a Fisher’s exact test to determine which enz me families have a higher gene count in the isolate genome than the average of >1000 genomes from the same bacterial Order. Those pfams having a gene count log base 2-fold change greater than one were selected for further study.

[0107] Gene synthesis, cloning, and enzyme expression: Protein sequences were codon optimized for E. coli and purchased from Twist Biosciences (South San Francisco, CA, USA) either in the pET-28a vector, or as gene fragments. Gene fragments were cloned into the pET-28a vector by using restriction-ligation cloning at the BamHI and Hindlll restriction enzyme sites. Upon selection of colonies using kanamycin as a resistance marker, plasmids were transformed into E. coli BL21(DE3) using standard heat shock transformation. Each construct had a 6x histidine tag on the N or C terminus for purification.

[0108] For protein expression, BL21 strains containing the plasmid of interest were inoculated at 37°C until they reached an optical density of approximately 0.7 OD600. 0.3 mM IPTG was added to each culture for induction of protein expression. Cultures were then grown at 30 °C for 6 hours to express protein, which were then visualized via SDS-PAGE. For proteins that did not express at 30 °C (CvDyP, and subsequently all DyPs for the remainder of the study), cultures were instead grown for 18-22 hours at 18 °C.

[0109] Protein purification: After protein expression is complete, cultures w ere pelleted, lysed using Solulyse bacterial protein extraction reagent (Genlantis Inc. San Diego, CA, USA. Catalog number L100500) per the manufacturer’s instructions. Soluble protein fraction in Solulyse solution was brought to a concentration of 10 mM imidazole and then purified , through nickel affinity chromatography via FPLC. Using a HisTrap HP 1 mL column (Cytiva, Marlborough, MA, USA), binding buffer was comprised of 20 mM NaHzPCty 500 mM NaCl, and 20 mM Imidazole and elution buffer was comprised of 20 mM NaH2PO4, 500 mM NaCl, and 500 mM Imidazole. A gradient from 0-50% elution buffer, followed by isocratic flow of 100% of elution buffer was used to purity the protein. All purifications w ere carried out at 4°C.

[0110] Production of stripped LDPE films:

[0111] LDPE (low-density polyethylene pellets [melt index:25 g / 10 min] were purchased from Sigma Aldrich Chemical Company, St. Louis, MO, USA; catalog number 428043 - 250 g) (7.2 g) was dissolved in xylenes (150 mL) under reflux (130 °C) with constant stirring for 3 hours. The solution was left to cool to 60 °C without stirring and ice-cold methanol (19.95 mL) was added dropwise before pouring the solution into room temperature methanol (129.2 mL). The solution was vacuum filtered, and the recovered polymer washed with fresh methanol 3 times before drying at room temperature for 48 hours. Recovered LDPE (7-8 g) was added to a cellulose extraction thimble and placed in a Soxhlet apparatus with chloroform (200 mL). The heating rate w as controlled to achieve Soxhlet extraction cycles of ~15 minutes duration over a total of 24 hours. The remaining polymer was left to dry at room temperature for 1 hour and dried under vacuum for 16 hours. The stripped PE w as then hot pressed between Kapton films at 11 MPa at 180 °C for 5 minutes to form thin films (< 1 mm). Enzyme assaying on plastic films: For activity screening, proteins were heterologously expressed in E. coli BL21(DE3) and crude lysates were dosed onto LDPE films (Goodfellow Cambridge Limited. Huntingdon. England. Catalog number LS580568) in three 90-minute doses and allowed to dry overnight prior to screening for oxygenation via FTIR. A more thorough analysis of enzyme activity on films started with purified enzymes that were stored at a concentration of 1.0 g / L. verified by a Bradford assay. To screen enzymes for activity, 10 pL of 1.0 g / L enzyme was dosed onto a film (Goodfellow Cambridge Limited, Huntingdon, England. Catalog number LS580568 or LDPE stripped of additives from (Sigma Aldrich Chemical Company, St. Louis, MO, USA; catalog number 428043 - 250 g) with 1 mM of the appropriate cofactor in a sodium phosphate buffer to a final volume of 30 pL. Three, two-hour doses of this nature were performed, with the last dose allowed to react and dry overnight for 16 hours. Enzyme re-dosing was deemed necessary as a single dose saturating the surface proved insufficient for deconstruction. Moreover, the film surface dried as liquid evaporates, requiring enzyme to be re-dosed to sustain the reaction. Each enzyme was initially tested at pH 7, for ease of assaying. Plastic films were then washed in water and ethanol, air dried, and analyzed via FTIR to monitor chemical changes. Enzy matic reactions of DyPs were carried out at 1 mM H2O2 (Sigma Aldrich Chemical Company. St. Louis, MO, USA Catalog number H1009 - 500ML ) and a pH of 4.0 as a result of enzy me optimization studies.

[0112] Pyrogallol peroxidase assay: Pyrogallol (Sigma Aldrich Chemical Company. St. Louis, MO, USA Catalog number P0381 - 25 g) was used as a standard substrate for measuring enzyme activity of peroxidases. The enzyme assay used followed the ‘Enzymatic Assay of Peroxidase (EC 1.11.1.7)' protocol from Millipore sigma. Briefly, 0.027% v / v hydrogen peroxide was added with 0.5% w / v pyrogallol and 0.75 units of peroxidase for the reaction. The generation of purpurgallin was measured using absorbance at 420 nm.

[0113] DyP-Peroxidase sequence similarity screening: From the metagenomic data in ‘Metagenomic data analysis to identify protein families of interest’, 6 enzymes were randomly selected from LDPE fed mealworm gut sample that fell within pfam 04261 and were thus classified as dye-decolorizing peroxidases. Additionally, a protein BLAST search on CvDyP was performed to find enzymes with similar sequences from the same genus. From this search, two hits within the top 10 BLAST results by7E-value were specifically selected to ensure that the test database encompassed the four subclasses of class I DyPs per classification in. This list was generated to encompass a variety of DyP subclasses and taxonomic origins, including several in the same 13 subclass that were from the same genus as CvDyP. In total, 9 enzymes across class I DyPs were tested for PE activity in addition to CvDyP (Supplementary Table 5). Species with DyPs included in this list were: Rothia halotolerans (RhDyP), unclassified Brevibacterium (UnBDyp), Brevibacterium oceani (BoDyP), Brevibacterium linens (BlDyP, Bl2DyP, BL3DyP), Corynebacterium provencense (CpDyP), Corynebacterium sp. (CsDyP) and Corynebacterium neomassiliense (CnDyPfi where CsDyP and CnDyP are those from the BLAST search and all others listed are from the LDPE-fed metagenomic data.

[0114] DyP-Peroxidase sequence secretion tag identification and structure prediction: All sequences were screened using signalP 6.0 using the fast model. The fasta file produced with secretion tags cleaved off was used to generate AlphaFold predictions of structure via Alphafold colab. Additionally, an outgroup protein of a similar peroxidase coming from the same genus as that of one of the screened DyP peroxidases was used. The outgroup protein used the AlphaFold prediction available on interpro and the protein has accession number A0A163AWC0, identified as a catalase peroxidase.

[0115] YRB map generation: YRB maps were generated according to Hagemans et. al. (Front Mol Biosci 2, 56 (2015)). The python script from the original manuscript was downloaded and used to generate YRB maps.

[0116] Cp-CNHL activity assay on C10-C40 alkanes: 750 pL C10-C40 alkane standards (Sigma Aldrich, catalog #68281) were run through a rotary evaporator at 150 mbar absolute pressure, 50°C jacket, -20°c condenser, 200 rpm rotation for 5 minutes and dried in a vacuum oven overnight at 150 mbar absolute pressure at 100°C, left uncapped overnight. pH4 buffer, Cp+CNHL and H2O2 were applied directly to dried alkane wax emulating “Enzyme assaying on plastic films / ’ Alkanes were resuspended into 150 pL hexane to redissolve alkanes for gas chromatography (GC) analysis. 2-decanone (Sigma Aldrich, 196207 catalog #) was added at 1 mg / mL as an internal standard. Chromatographic analyses were performed with a gas chromatography-flame ionization detector system (GC-FID) 7980A-5975C from Agilent Technologies. Separation of the metabolites was performed on a DB-5 Column coated with polyimide (30 m length, 0.25 mm inner diameter, and 0. 1 pm film thickness; Agilent Technologies, USA) for proper separation of substances, and Helium (He) was utilized as a carrier gas. The analysis w as performed using a split injector at 350 °C and an injection volume of 1 pl. The ion source temperature was 230 °C, mass spectral analysis was performed in scan mode, the quadrupole temperature of 150 °C, and a fragmentation voltage of 70 eV.

[0117] Cp-CNHL molecular docking analysis: Molecular docking simulations were prepared using Autodock 4 in Dockey for MacOS version Big Sur 11.7. A custom bounding box was drawn around the hydrophobic loop of the enzyme when docking dotriacontane and a custom bounding box was drawn around the aromatic residues W300 and Y424 when docking reactive blue. A custom bounding box was drawn around the active site when docking decane and dotriacontane to the active site of the enzyme. The Lamarckian GA property search algorithm, using Autodock 4 default parameters. Ligands were prepared using default parameters on prepare_ligand4 ligand preparation tool. Ligands were Z1NC6920423 (dotnacontane). CHEMBL134537 (decane), and CHEMBL5187239.

[0118] Results:

[0119] Yellow mealworm gut microbes oxidize low-density polyethylene

[0120] Yellow7mealworms deconstruct LDPE, reducing its molecular weight by >33-62% upon ingestion. In the case of another nonhydrolyzable plastic, expanded polystyrene (EPS) foam, deconstruction appears to be driven by the yellow' mealworm gut microbiota; plastic deconstruction is greatly reduced when mealworm guts are cleared of microbiota. Moreover, mealworms produce13C labeled CO2 from13C labeled PS only when microbes are present in their gut. As a direct result of these data and a growing number of reports of insect gut microbiota participating in LDPE deconstruction, we hypothesized that microbial enzy mes in mealworm gut contents were responsible for LDPE deconstruction. We thus isolated microbes from the guts of mealw orms fed diets of nonhydrolyzable plastics in the presence or absence of antimicrobial treatments (FIG. 2A). Using simple growth assays, we screened the LDPE deconstruction ability of over 300 isolates to reveal 21 taxonomically unique potential plastic degraders (Table 2). Five top performing isolates, Staphylococcus lentus, Enterococcus termitis, Corynebacterium variabile, Brevibacterium epidermidis, and Kocuria halotolerans showed at least a 25% increase in optical density' when grown in LDPE-containing mineral media compared to LDPE-free mineral media (FIG. 3A). These strains were selected for further evaluation of their plastics deconstruction abilities.

[0121] The deconstruction of LDPE by mealw orms and their gut microbiota is hypothesized to follow' microbial alkane metabolism pathw ays, leveraging promiscuous enzymes that are active on hydrocarbons (FIG. 2B). C-C and C-H bonds in LDPE fdms are proposed to be first enzymatically oxidized by monooxygenases, dioxygenases or peroxidases to generate primary or secondary alcohols. Further enzymatic oxidation of these primary and secondary alcohols produces aldehydes and ketones, respectively. In such a mechanism, the initial hydroxylation ‘activates’ the chain for deconstruction (the bond energy for a C-C bond is 607 kJ / mol compared to 314 kJ / mol in the CH3-CO bond) and is subsequently hydroxylated in a second reaction, forming a geminal diol. The geminal diol then rapidly forms the observed ketone or aldehyde due to spontaneous, irreversible dehydration of the unstable diol intermediate. Further enzymatic oxidation and hydrolysis or decarboxylation are expected to produce fatty acids for beta-oxidation and / or carbon dioxide (FIG. 2B).

[0122] We evaluated the propensity’ of top microbial isolates to deconstruct LDPE by evaluating the extent of oxidation on plastic substrates with diverse physicochemical properties (Table 1). Monocultures of the top five isolates were washed and then dosed daily on commercially available LDPE films over 5 days, adding fresh microbial inoculum to the film with each dose. The film was cleaned to remove bound proteins and characterized for chemical modification by X-Ray Photoelectron Spectroscopy (XPS) and Fourier Transform Infrared Spectroscopy (FTIR). All tested microbes increased the film oxygen content relative to untreated controls (one tail t-test, p<0.05), primarily in the form of carbonyl formation (C=O) (Fig. 1C-D). This carbony l formation is consistent with proposed mechanisms of LDPE bio-oxidation (FIG. 2B, FIG. 3B) and mimics chemical changes resulting from abiotic oxidation methods such as photo-oxidation or cold plasma oxidation. Corynebacterium variabile treatment led to the largest increase in oxygen content relative to an untreated control (6.9-fold from 1.6% to 11%), followed by Enterococcus termitis (5.1-fold), Staphylococcus lentus (3.7-fold), Brevibacterium epidermidis (3.4-fold), and Kocuria halotolerans (3. 1 -fold) (FIG. 2D). FTIR spectra suggest that most isolates introduced aldehydes into PE chains by terminal oxidation (FIG. 2C, Table 3). However, microbes had distinct preferences for terminal or subterminal oxidation routes as inferred by the predominant detected C=O species (FIGS. 2B-2D, Table 3). In addition to new' C=O bonds, treatment with each isolate resulted in minor increases in the C-0 content of the film, likely as a result of the initial hydroxylation (FIG. 2D).

[0123] Three isolates, Corynebacterium variabile, Staphylococcus lentus, and Kocuria halotolerans were chosen for SEC analysis due to different oxidation patterns on FTIR spectra (FIG. 3C, Table 3) that imply different enzymes responsible for oxidation events. Although isolated microbes induced chemical changes in the LDPE films, no change in the molecular weight distribution (MWD) of LDPE was detected via SEC (FIG. 3C). However, due to the overlap between low molar mass products and solvent peaks in SEC chromatograms and the column set used, SEC is unable to detect and resolve soluble deconstruction products below -250 g / mol (FIG. 1). Given the observed changes in LDPE film chemical modification (FIG. 2C) and LDPE-dependent growth of the microbial isolates (FIG. 3A), it is possible that some polymer chains are preferentially degraded into soluble products and metabolized upon initial oxidation, leaving high MW polymer chains for detection via SEC. An alternate interpretation, however, may be that efficient LDPE deconstruction is the result of synergistic action from a microbial consortium rather than isolates.

[0124] Microbial communities in the mealworm gut initiate PE deconstruction via a secreted PE-oxidizing enzyme

[0125] We assembled and mined genomes for our top-performing microbial isolates to identify putative PE-oxidizing enzymes responsible for the observed changes in FIG. 2C (FIG. 4A). All five isolate genomes were enriched relative to other species within their taxonomic Order in protein families proposed to oxidize alkanes such as dioxygenases, monooxygenases, and peroxidases (FIG. 2B, FIG. 5, Fisher’s exact test; p-val < 0.05). These strains were likely present in the mealworm gut for their ability to metabolize lignin and / or hydrocarbons such as cuticular waxes on leaves that form part of the mealworm diet. Thus, we hypothesized that these enzymes may have promiscuous activity on hydrocarbon- rich plastics.

[0126] The Corynebacterium variabile isolate encoded the highest diversify of putative PE- oxidizing enzyme families, which we evaluated for activity (FIG. 5.Table 4). Table 4 shows putative alkane metabolism pfams from comparative genomic analyses that yielded positive hits from JGI IMG. These enzyme families were present in genomes from the Order of each of the five tested isolates, Staphylococcus lentus, Enterococcus termitis. Corynebacterium Variabile, Brevibacterium Epidermidis. and Kocuria halotolerans . Ten protein families (pfams), consisting of four classes of monooxygenases, three classes of dioxygenases, and three classes of peroxidases were identified as enriched in C. variabile. We randomly selected one representative enzy me from each of the ten pfams identified for in vitro testing on LDPE films (Table 5). These enzymes were heterologously expressed in E. coli and lysates were used to treat LDPE films.

[0127] Among the tested enzymes, only the dye decolorizing peroxidase (DyP; pfam 04261), denoted CvDyP (IMG gene ID Ga0530663_0293_23766_25013), chemically modified LDPE films and acted as a LDPE oxidase (FIG. 4B). Three 90-minute treatments of LDPE films with purified CvDyp and cofactor H2O2 led to the formation of ketones and aldehydes, evidenced by FTIR spectral peaks at 1710 cm'1and 1740 cm1, respectively. Carbonyl peaks were confirmed to be a direct result of enzymatic activity, as the peak persists after washing with water and ethanol. Washing is critical for accurate identification of oxidation as it removes surface-bound protein that generates a confounding signal in the carbonyl region (FIG. 6). To quantify this activity, we calculated the carbonyl index (CI) or ratio of the maximum peak height between 1700-1745 cm'1and the maximum peak height between 1400-1500 cm'1(FIG. 7). Enzyme treatment increased film CI by more than 10- fold relative to a FLCh-free, inactive CvDyP control (FIG. 4B). Sequence analysis of the enzyme with SignalP confirmed that this DyP was secreted by its microbial host and peroxidase activity was confirmed using model substrate pyrogallol (FIG. 8). Despite the ability of CvDyP to oxidize PE films, it is unable to cleave C-C bonds in PE after 20x 90- minute treatments (FIG. 4C).

[0128] DyPs are absent in the y ellow mealworm genome but are highly abundant in the metagenomes of mealworm gut microbial communities fed on LDPE (41 gene counts) and high-density PE (HDPE) (37 counts) relative to oats fed (0 counts) mealworms (FIG. 4D). Since DyPs are present only in the gut microbiome and not the host, our work suggests that gut microbial communities may play an important role in LDPE deconstruction in yellow mealworms. DyP producing microbes are far more prevalent in plastic-containing guts, implying that this oxidative event is critical for the LDPE deconstruction process.

[0129] PE-oxidizing DyPs are ubiquitous

[0130] As DyPs are ubiquitous in bacteria and fungi, we investigated whether there were unique sequence and structural features that dictate if a DyP can function as an LDPE- oxidase. DyPs are divided into three major classes based on size: i) P or primitive DyPs (<300 amino acids); ii) I or intermediate DyPs (300-400 amino acids); and iii) V or advanced DyPs (>400 amino acids). CvDyP is a class I (subclass 13) DyP. Thus, we heterologously expressed and tested 11 enzymes across all four sub-classes of class I DyPs, both native and external to the yellow mealworm gut microbiome to determine if LDPE oxidation is a general property of class I DyPs, or the 13 subclass of DyPs (Table 6). DyPs across all tested sub-classes oxidize LDPE as measured by an increase in CI; only 5 / DyP and 7?oDyP did not show activity over the .co / z-noDyP control (FIG. 9). Thus, PE- oxidizing DyPs are not restricted to the guts of yellow mealworms and may be found in a number of environments including well water, cheese rind metagenomes, and stool samples.

[0131] To ensure we identified enzymes that act on PE chains rather than potential additives, we tested purified class 13 DyP (CvDyP, C / 7 DyP. and CsDyP) activity on LDPE films that were stripped of additives in-house via Soxhlet extraction (FIG. 10A, Table 1, FIG. 11); a list of stripped additives identified via GC-MS can be found in Supplementary Data File 2. CM D P and CsDyP were selected for further study as the nearest phylogenetic neighbors of CvDyP. Again, each of these enzymes was confirmed as a peroxidase by demonstrating activity on model substrate pyrogallol (FIG. 12A). All three enzy mes were active on stripped LDPE, meaning that oxidation occurs on the polymer chain, not on any additives (FIG. 10A). However, quantitative performance is substrate specific. Relative enzyme activity on stripped LDPE (~83 kg / mol, FIG. 10A) differs from commercial LDPE (-133 kg / mol, FIG. 12B)., underscoring the importance of polymer characteristics for enzyme performance Nonetheless, our results indicate that the 13 class of DyPs both native and not native to the mealworm gut are able to oxidize PE chains in LDPE films.

[0132] A prominent hydrophobic loop near the active site modulates DyP PE-oxidase activity

[0133] We hypothesized that conserved structures across LDPE-active DyPs may be important for LDPE oxidation. Thus, we predicted the structure of each tested Dy P via AlphaFold 2.0 (all with pLDDT confidence scores >88.8) and used them in a multiple structural alignment. Secretion signal peptides were identified via SignalP version 6.0 and removed prior to structural simulations to aid alignment. LDPE-active DyPs contain a unique loop region extending outward from the body of the protein and proximal to the predicted active site (FIG. 10B, FIG. 13) that deviates from non-PE oxidizing DyPs. Importantly, the loop is hydrophobic, ranging from approximately 33% (5 / 2DyP) to 54% (CwDyP) of residues with hydrophobic side chains. This loop in CvDyP is encoded by: YNYDLPVTPSSADALVDADPVALSDT (SEQ ID NO: J, with YNY residues at the start of the sequence and GL residues at the end of the sequence being conserved across all tested ty pe I DyPs (FIG. 14). LDPE oxidase activity7directly correlates with hydrophobic loop length and hydrophobicity7(FIGS. 9 and 14); the enzymes with the longest and most hydrophobic loops are the top performing LDPE oxidases. Non-PE active enzymes contain a much smaller 12 amino acid loop that is more hydrophilic (FIG. 1 OB, FIGS. 13-15). The extension of the hydrophobic loop outward from the center of the protein and its proximity7to the predicted active site suggests that the hydrophobic loop may be important for productive binding of highly hydrophobic PE chains in orientations favorable for catalysis.

[0134] Mutant DyPs were generated to validate the role of the hydrophobic loop for LDPE- oxidase activity. The four tested DyPs from Corynebacterium strains ( ?Dy P. CvDyP, CsD P. and CnDy P) were selected as engineering targets due to their high sequence homology but differing hydrophobic loop structures and lengths. The smaller, less hydrophobic loop region from C / ?DyP was swapped with the hydrophobic loops in CvDyP, CwDyP. and CvDyP to create mutant hydrophobic loop reduction and extension chimeras. Reducing the hydrophobic loop size and hydrophobicity reduced LDPE-oxidase activity relative to the wild type (FIG. 16). Similarly, inserting the longer, more hydrophobic loops from CvDyP, CnDyP. and CvDyP into C DyP enhanced LDPE-oxidase activity over the wild type QoDyP (FIG. 16). Quantify ing the hydrophobicity of these mutants around the active site and loops via a sequence hydrophobicity index revealed a positive correlation with observed LDPE activity7(R2= 0.43; FIG. 17). These results confirm that this divergent hydrophobic loop region has a significant role in DyP LDPE activity7.

[0135] FIGS. 18A-18C show that surface exposed aromatic residues permit LDPE oxidation. (A) Pymol rendering of Alphafold predicted CvDyP structure with active site (DI 86, H280, and R297) and surface exposed aromatics (W231 and W356) displayed as sticks. Distances between LDPE-oxi dizing W231 and bulky aromatic substrate reactive blue 2 (green) and model alkane dotriacontane (pink) are displayed. (B) Relative enzyme activities of CvDyP mutants on model substrate pyrogallol. (C) Change in carbonyl index of CvDyP and mutants after treatment on stripped LDPE. * Statistically significant increase in ACI relative to no enzyme control via one-tailed t-test, p<0.05. ** Statistically significant difference in ACI relative to WT via one-tailed t-test, p<0.05.

[0136] FIG. 19 shows GC FID chromatograms from Cp-CNHL DyP reaction with liquid alkanes. Control chromatograms were gathered for solvent hexane, internal standard 2-decanone, the vacuum dried C10-C40 alkane mixture, and Cp-CNHL reactions with the dried alkane mixture. Each chromatogram is of a single measurement. Each Cp-CNHL chromatogram represents a single reaction at room temperature, pH 4. after 16 hours of reaction time.

[0137] FIG. 20 shows molecular docking of Reactive Blue 2 to aromatic surface residues and dotriacontane to the hydrophobic loop. The top pose from Autodock molecular docking simulations of dotriacontane (magenta) with Cp-CNHL and Reactive Blue 2 (RB2, green)) are overlaid. Simulations purposely forced the binding region to the hydrophobic loop region and to key aromatic residues W300 and Y424 (cyan) for dotriacontane and RB2, respectively. This particular binding was done in order to simulate probable locations for proposed binding of each substrate. A non-conservative minimum number of 44 carbons in chain length for LDPE oxidation were calculated by approximating that the dotriacontane molecule was ~14 angstroms from the active site but needed to be 4.5 angstroms away to be oxidized. This remaining distance of ~9.5 angstroms was then used in conjunction with an approximated 5 angstroms per 6 carbons on the dotriacontane chain, estimated by distance measurements in Pymol. Note that this distance represents the absolute minimum number of carbons if the chain is bound in the optimum position on the Cp-CNHL )yP loop as shown.

[0138] Surface-exposed tryptophans cataly ze non-canonical oxidation of LDPE chains external to the protein

[0139] We next tried to determine how DyPs non- terminally oxide PE substrates as they are unable to fit within the canonical active site. However, studies with reactive blue (RBI 9), a bulky dye that cannot fit into the DyP active site, demonstrate a non-canonical oxidation mechanism mediated by surface-exposed aromatic residues that can harbor radicals. These radicals allow for oxidation of the otherwise sterically hindered RBI 9 by facilitating long chain electron transport from the substrate to the heme cofactor and are conserved across type I DyPsiusing TcDyP as a basis, W263 is conserved and W356 is conserved as a try ptophan or a ty rosine subclass-wide. Therefore, we hypothesized that LDPE oxidation occurs via the same mechanism and identified W231 and W356 in CvDyP as analogous residues to W263 and W376 that are necessary for RB19 oxidation by TcDyP (FIG. 18A). To test this hypothesis, we mutated these two try ptophans conservatively to nonradical forming phenylalanine or serine known to preserve the environment around the heme cofactor. Control studies with model substrate pyrogallol, which binds the canonical active site, demonstrate that all mutants (W231 F, W356F, W231 F / W356F, and W231S / W356S) retained canonical peroxidase activity (FIG. 18B). However, there was a 40-70% decrease in activity relative to wild type, perhaps due to minor conformational shifts in the protein structure and / or loss of non-canonical surface activity. All W231 mutants completely lost LDPE oxidase activity relative to wild type control, demonstrating that it is necessary' for DyP activity on plastic substrates (FIG. 18C). W356 is involved in plastics deconstruction although it is not essential for activity; W356 mutants lost 48.5% of activity relative to wildtype on PE substrates (FIG. 18C). These results are consistent with our hypothesis and suggest that LDPE oxidation is catalyzed by surface-exposed ammo acid (W231 and W3566) residues on DyPs.

[0140] Our observations regarding the surface catalysis of PE substrates (FIG. 18) and the importance of a hydrophobic loop for activity (FIG. 16) suggest a model for LDPE catalysis via DyPs. LDPE oxidation proceeds via the surface-exposed radical mechanism only, as it cannot bind the relatively small canonical active site, and the hydrophobic loop aids with substrate binding and positioning against these residues for activity. Removing this loop reduces PE interaction with these surface residues and thus reduces activity (FIG. 1 ). However, this loop-mediated mechanism is only possible for substrates of a minimum size, as the substrate must be large enough to bind to the hydrophobic loop and reach the surface oxidation site. Model alkanes (C10-C40) that failed to dock to the canonical active site, as simulated by Autodock, were not oxidized by top performing DyP mutant, Cp+CNHL DyP (FIG. 19). Moreover, C32 alkane dotriacontane, the largest substrate we could simulate with off-the-shelf tools, is too far from W300 and Y424, analogous to CvDyP W231 and W356, to be oxidized in the same manner bulky dyes are oxidized when its docking was forced to the Cp+CNHLDyP hydrophobic loop using Autodock (FIG. 20). That is, surface-mediated activity is only possible when the substrate is large enough for the hydrophobic loop to correctly position it, such as PE chains and their deconstruction products. Under optimum conditions, dotriacontane (C32) binding was calculated to be ~9.5 A short from interacting with the surface-exposed radicals for catalysis (FIG. 20). By approximating a C-C bond length of 5 A per six carbons estimated by Pymol, the C32 alkane was found to be short ~12 carbons. In other words, a bulky substrate must be at least 44 carbons long to be correctly positioned by the hydrophobic loop to access the surface residues. This nonconservative estimate of the minimum substrate size is suggested by Autodock simulations as the specific position at which a PE chain binds to the hydrophobic loop. The orientation and branching density7of bound PE chains and the flexibility7of the hydrophobic loop region all impact actual binding.

[0141] Discussion

[0142] In this work, we identified type I DyPs as a class of LDPE-oxi dizing enzymes and provide evidence for their role in LDPE deconstruction in yellow mealworms by showing an enrichment in gene count of DyPs in PE-fed mealworms relative to non-plastic-fed controls. DyPs initiate LDPE deconstruction by oxidizing PE chains to form aldehydes or ketones. Rigorous controls and characterization confirmed the validity of these observations and reduced false positives in activity7measurement. LDPE cleaning prior to materials characterization allowed for distinction between biological contaminants and true chemical modification. LDPE substrates stripped of additives were similarly oxidized, demonstrating enzymatic activity7on the polymer backbone, rather than polymer additives. Importantly, evidence of LDPE oxidation is absent on ELCh-free controls, confirming peroxidase activity. The reliability of these findings was demonstrated by independent quantitative spectroscopic techniques, FTIR and XPS. We identified a divergent loop characteristic of LDPE-active DyPs and demonstrated that its hydrophobicity can be tuned to enhance LDPE activity. Lastly, we provide support for a non-canonical mechanism of LDPE oxidation through solvent-exposed, radical-harboring amino acid residues by abolishing LDPE oxidation upon mutation of putative oxidative residue W231.

[0143] This work suggests that fully biological routes for LDPE deconstruction are feasible by confirming enzymatic oxidation of PE chains. Abiotic pre-treatments such as chemical treatment or thermal oxo-deconstruction were previously thought to be necessary to make plastics chemically available for biodeconstruction. Though such an oxidative step has been hypothesized as the first step in enzy matic LDPE deconstruction, this step had not been directly demonstrated until now. Activity of LDPE deconstructing enzymes such as laccases, manganese peroxidases, and alkane hydroxylases are limited to pre-oxidized or very7low molecular weight (Mw~4 kg / mol, Mn~1.7 kg / mol) polymers. Our data conclusively show that abiotic oxidation or pretreatment is not required for biological deconstruction and that biological activation chemistries do exist.

[0144] Phenol oxidases and multi-copper oxidases have been reported to oxidize and deconstruct non-pretreated LDPE. Both reports show carbonyl formation after enzyme treatment via FTIR or Raman spectroscopy, consistent with the oxidation performed by DyPs, but neither study leveraged inactive enzy me controls. This lack of controls, coupled with a lack of repeatability has led the phenol-oxidase study to be scrutinized, as the possibility7of non-specific enzyme binding to the plastic leading to the formation of new carbonyl peaks cannot be ruled out. We confirm that new carbonyl groups are a result of Dy P activity rather than non-specifically bound protein by including a DyP control in the absence of H2O2 and by including FTIR spectra before and after washing LDPE films. Additionally, any oxidative or deconstructive activity by phenol oxidases or multi-copper oxidases on LDPE could be the result of activity on additives found in the plastic. We confirmed here that DyP activity occurs directly on LDPE chains by stripping the plastic of additives prior to its use. Both of the aforementioned studies also report medium chain ketones and / or carboxylic acids as deconstruction products via gas chromatography, but these compounds may have been inherent to the plastic and leached out into the reaction mixture, a possibility that cannot be ruled out due, again, to the lack of inactive enzyme controls in this study. Moreover, the generation of medium chain deconstruction products implies that the single purified enzy me performs oxidation and cleavage of C-C bonds, which is highly unlikely for a single enzy me, as the chemistry for C-C oxidation is far different than that needed to cleave C-C bonds in the PE backbone. By confirming that oxidation occurs directly on PE chains, we definitively show that DyP peroxidases act as a first enzymatic step in that series by oxidizing the PE chains, priming them for deconstruction.

[0145] DyPs are knoyvn for their substrate promiscuity’ and propensity to oxidize bulky polymeric substrates such as lignin and have been shoyvn to oxidize and degrade UV- oxidized PS. Oxidation of substrates that cannot fit in the active site is made possible by catalytic radicals on surface exposed, aromatic residues, but the importance of these residues has only been demonstrated on the oxidation of bulky dyes such as RBI 9. We confirmed the importance of W231 in CvDyP, showing that it must be present to oxidize LDPE. This finding is consistent with the mechanism for RBI 9, as W231 is analogous to W263 of TcDyP, the residue responsible for RBI 9 oxidation. Therefore, we propose that surface residue W231 serves as a non-canonical oxidation site for LDPE chains. This non- canonical oxidation is enhanced by a hydrophobic loop region proximal to the solvent- exposed aromatics, where the loop region can act as an anchor and bind to the extremely hydrophobic LDPE, making LDPE chains sterically available for protein radicals that facilitate the transfer of electrons and LDPE oxidation. Thus, the hydrophobic loop and surface aromatic regions are promising targets for engineering to enhance LDPE oxidation.

[0146] DyPs provide a crucial first step in the LDPE biodeconstruction process through oxidation of PE chains. Due to the diverse nature of this enzyme class, there may be DyPs from sources other than mealworm guts that are more effective at PE oxidation than those reported in this study. Moreover, studying the oxidation mechanism of DyPs is essential to improve LDPE-oxidase activity. Such activities are informed by the identification of the hydrophobic loop region and residue W231 in this work. These findings can be further developed to create design rules for LDPE-oxidizing DyPs and to engineer LDPE-oxidases with improved activity or tolerance to industrial processing conditions. By providing the first step in biological LDPE deconstruction, this study provides a route to discover enzymes downstream of DyPs in the deconstruction pathway. Ultimately, this knowledge can enable the development of sustainable, biologically driven approaches for plastics deconstruction.

[0147] Ongoing / Future work

[0148] The results presented in this manuscript suggest that DyPs can be engineered for improved activity on various substrates, namely LDPE. Enzyme activity can be improved by making point mutations to improve enzyme stability, including but not limited to L226T, R252K, L244Q. D200N, C197T, N210E, K177E, F433Y, and E138T, supported byimproved activity on reactive black 5 (RB5) (FIG. 21). Additionally, point mutations that improve flexibility in the active site region that contains the two conserved residues corresponding to W231 and W263 in CvDyP can be used to improve activity on non- canonical substrates such as RB5 or LDPE that are too large to fit into the canonical enzy me active site. These mutations, including but not limited to Q308G, F245G, Q303G, R302G, E298G, N243G, and D392G. are made by mutating amino acid residues that participate in at least two hydrogen bonding events in the catalytic region to glycines to reduce hydrogen bonding and increase flexibility-. Again, mutations demonstrate improved activity on RB5 (FIG. 22). Lastly, we hypothesize that electron transport in the catalytic site can be improved by making mutations, for example. T393Y, Y424H, Y424W, D301Y, and D301W, that introduce additional tryptophan or tyrosine residues to the region where residues corresponding to CvDyP W263, W231, the hydrophobic loop, and heme cofactor residue. Importantly, combinations of successful point mutations can be used to further improve enzyme activity.

[0149] Table 1. Materials characteristics of plastic substrates used in this study

[0150] Mnrefers to the number average molecular weight K1-,rrefers to the weight average molecular weight Table 2. List of taxa isolated from mealworm guts under various feed conditions

[0151] Table 3. Location of maximum carbonyl peaks by each isolate on LDPE films

[0152] Table 4. Putative alkane metabolism pfams

[0153]

[0154] Table 5: Enzyme and sequence information from screened Corynebacterium variabile enzymes

[0155]

[0156] Table 6. Sequence information of class I DyPs tested for LDPE-oxidase activity

[0157] Table 7. Hydrophobic loop sequences

[0158] All documents, books, manuals, papers, patents, published patent applications, guides, abstracts, and / or other references cited herein are incorporated by reference in their entirety. Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the specification and examples be considered as exemplar)’ only, with the true scope and spirit of the invention being indicated by the following claims.

Claims

WHAT IS CLAIMED:

1. A method for oxidizing a polymer in plastic, comprising:(a) exposing plastic to an effective amount of a peroxidase, wherein the peroxidase comprises a hydrophobic loop; and(b) functionalizing a polymer in the plastic, whereby the polymer is oxidized.

2. The method of claim 1, wherein the hydrophobic loop consists of at least 15 amino acids.

3. The method of claim 1 or 2, wherein at least 30% of the amino acids in the hydrophobic loop have a hydrophobic side chain.

4. The method of any one of claims 1-3, wherein the hydrophobic loop consists of an amino acid sequence at least 50% homologous to the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: 21), wherein X is an amino acid.

5. The method of any one of claims 1-4, wherein the hydrophobic loop consists of an amino acid sequence at least 80% identical to an amino acid sequence selected from the group consisting of SEQ ID NOS: 11-20.

6. The method of any one of claims 1-5, wherein the peroxidase consists of an amino acid sequence at least 80% identical to an amino acid sequence selected from SEQ ID NOS: 1-10.

7. The method of any one of claims 1-6, wherein the peroxidase consists of an ammo acid sequence at least 80% identical to the amino acid sequence of SEQ ID NO: 10 and comprises a tryptophan or tyrosine residue corresponding to W231 in SEQ ID NOTO and a tryptophan or ty rosine residue corresponding to W356 in SEQ ID NO: 10.

8. The method of any one of claims 1-7, further comprising forming an alcohol, aldehyde, and / or ketone.

9. The method of any one of claims 1 -8, wherein the polymer is low-density polyethylene (LDPE).

10. The method of any one of claims 1-9, wherein the polymer is nonhydrolyzable.

11. A recombinant microorganism comprising a heterologous gene encoding a peroxidase, wherein the peroxidase comprises a hydrophobic loop.

12. The recombinant microorganism of claim 11, wherein the hydrophobic loop consists of at least 15 amino acids.

13. The recombinant microorganism of claim 11 or 12. wherein at least 30% of the amino acids in the hydrophobic loop have a hydrophobic side chain.

14. The recombinant microorganism of any one of claims 11-13, wherein the hydrophobic loop consists of an amino acid sequence at least 50% homologous to the amino acid sequence of XXRMLRRXYNYXXGXXXXXXLXTGLXFXSFQA (SEQ ID NO: _), wherein X is an amino acid.

15. The recombinant microorganism of any one of claims 11-14, wherein the hydrophobic loop consists of an amino acid sequence at least 80% identical to an amino acid sequence selected from the group consisting of SEQ ID NO: 11-20.

16. The recombinant microorganism of any one of claims 11-15. wherein the peroxidase consists of an amino acid sequence at least 80% identical to an amino acid sequence selected from SEQ ID NOS: 1-10.

17. The recombinant microorganism of any one of claims 11-16, wherein the peroxidase consists of an amino acid sequence at least 80% identical to the amino acid sequence of SEQ ID NO: 10 and comprises a tryptophan or tyrosine residue corresponding to W231 in SEQ ID NO: 10 and a try ptophan or tyrosine residue corresponding to W356 in SEQ ID NO: 10.

18. The recombinant microorganism of any one of claims 11-17. wherein the recombinant microorganism expresses the peroxidase.

19. The recombinant microorganism of any one of claims 11-18, wherein the recombinant microorganism secretes the peroxidase.

20. The recombinant microorganism of any one of claims 11-19, wherein the recombinant microorganism displays the peroxidase on cell surface of the recombinant microorganism.

21. A composition comprising the recombinant microorganism of any one of claims 11-20.

22. The composition of claim 21, further comprising a culture medium.

23. A method for producing a peroxidase, comprising:(a) growing the recombinant microorganism of any one of claims 11-17 in a culture medium; and(b) expressing the peroxidase by the recombinant microorganism, whereby the peroxidase is produced.

24. The method of claim 23, further comprising secreting the peroxidase by the recombinant microorganism into the culture medium.

25. The method of claim 23 or 24, further comprising displaying the peroxidase on cell surface of the recombinant microorganism.

26. The method of any one of claims 23-25, further comprising purifying the peroxidase.

27. The method of any one of claims 1-10, wherein the peroxidase is produced according to the method of any one of claims 23-26.

Images