Microbial Upcycling of Waste PET to Adipic Acid

Microorganisms can be genetically engineered to transform abundant waste feedstocks into value-added small molecules that would otherwise be manufactured from diminishing fossil resources. Herein, we report the first one-pot bio-upcycling of PET plastic waste into the prolific platform petrochemical and nylon precursor adipic acid in the bacterium Escherichia coli. Optimizing heterologous gene expression and enzyme activity enabled increased flux through the de novo pathway, and immobilization of whole cells in alginate hydrogels increased the stability of the rate-limiting enoate reductase BcER. The pathway enzymes were also interfaced with hydrogen gas generated by engineered E. coli DD-2 in combination with a biocompatible Pd catalyst to enable adipic acid synthesis from metabolic cis,cis-muconic acid. Together, these optimizations resulted in a one-pot conversion to adipic acid from terephthalic acid, including terephthalate samples isolated from industrial PET waste and a post-consumer plastic bottle.


S1 General Materials and Methods
Unless otherwise stated, starting materials and reagents were obtained from commercial suppliers and were used without further purification.All water used experimentally was purified with a Suez Select purification system (18 MΩ.cm, 0.2 µM filter).The following analyte abbreviations are used throughout: adipic acid (AA), 2-hexenedioic acid (2HDA), cis,cis-muconic acid (ccMA), catechol (Cat), protocatechuic acid (PCA), terephthalic acid (TA), polyethylene terephthalate (PET).Disodium terephthalate was used for all experiments using TA, except when waste PET was used.2HDA was synthesized as described previously 1 .Hot stamping foils were donated from API Foilmakers Ltd. in Livingston, UK.PET bottle samples were collected from domestic rubbish in Edinburgh, UK.
NMR: Proton nuclear magnetic resonance spectra ( 1 H NMR) were recorded using a Bruker AVA600 NMR spectrometer at the specified frequency at 298 K. Proton chemical shifts are expressed in parts per million (ppm, δ scale) and are referenced to residual protium in the NMR solvent.NMR solvents were used as purchased from commercial suppliers.
HPLC: High performance liquid chromatography (HPLC) analysis was carried out using a Thermo Fisher Scientific Dionex UltiMate 3000 Series UHPLC instrument equipped with a HyperSil Gold C18 column (150x3 mm x 3 µm).Analytes were detected at 206 nm and quantified by comparison to a caffeine internal standard, added to 50 µM final concentration.All HPLC solvents were purchased from commercial suppliers.Samples were analyzed using the following method: Solvent A: Water+0.1% v/v trifluoroacetic acid (TFA).

Media Recipes and Microbiology
Lysogeny Broth (LB) Medium: Bacto-tryptone (10 g/L), yeast extract (5 g/L) and NaCl (10 g/L) were dissolved in Milli-Q H2O.LB was autoclaved at 121 ˚C for 20 min, cooled and stored at room temperature.Solid media was made using the same recipe but with the addition of agar (15 g/L).
Stock solutions of 20% w/v glucose and 20% w/v glycerol were prepared and autoclaved at 121 ˚C for 20 min, cooled and stored at room temperature.Stock solutions of MgSO4 (1 M), CaCl2 (50 mM), thiamine hydrochloride (10 mg/mL) were prepared and filter sterilized.
Unless stated otherwise, E. coli cells were cultured at 37 ˚C with shaking at 220 rpm in an incubator shaker with a 5.1 cm orbit throw.Optical densities of E. coli cultures were determined using a DeNovix DS-11 UV/Vis spectrophotometer by measuring absorbance at 600 nm.For bacterial growth and protein induction, antibiotics required by each strain were added to media at the following concentrations: 100 µg/mL ampicillin, 30 µg/mL chloramphenicol, 50 µg/mL kanamycin, 25 µg/mL spectinomycin.

Molecular Biology
All synthetic genes were codon-optimized for E. coli BL21(DE3) and synthesized using GeneArt  Ligation Independent Cloning (LIC) 4 , or AquaCloning 5 .A typical SLiCE assembly was prepared as follows: a 10 µL reaction was set containing 1 µL of SLiCE extract (prepared from E. coli JM109 according to the published protocol 6 ), linearised vector (50 ng) and insert(s) at a molar concentration ration of 3:1 (insert:vector).SLiCE cloning reactions were incubated at 37 ℃ for 1 h before transformation of E. coli DH5α.Linearized DNA used for assemblies was gel-extracted.
Modular cloning was done following using JUMP (Joint Universal Modular Plasmids) backbones and plasmids and following the protocols indicated in 7 .

pPCA1
The backbone of pGro7 (Takara Bio Inc.) and the expression cassette of the pVan1 plasmid 8 (coding for tphA1, tphA2, tphB2, dcddh) were PCR amplified (using primers in Table S3) and combined using SLiCE cloning.

pPCA2 and pAA5
The expression cassettes and backbones of the pAA4 and pPCA1 plasmids were exchanged to generate pPCA2 and pAA5.Insert and backbones were amplified via PCR using the primers shown in Table S3, and the purified PCR products were assembled via SLiCE cloning.These PCR reactions used Phusion polymerase High GC buffer instead of standard 5x Phusion HF buffer.
Cassette integration in E. coli BL21(DE3) was performed by co-transforming the locus-specific gRNA, donor plasmid and pX2-Cas9 (Addgene plasmid #85811).After recovery at 37 ℃ for 4 h, positive transformants were selected on LB agar with ampicillin.Integration was confirmed by colony PCR using flanking (A, B) and cassette-specific primers (C) indicated in Table S5, using Phusion polymerase with High GC buffer.The CRISPR plasmids were cured by growing cells at 42 ℃ overnight without antibiotic selection and confirmed by loss of antibiotic resistance.

S4.1 Protein expression in E. coli BL21(DE3)
A single colony of E. coli BL21(DE3) cells co-transformed with desired plasmids was used to inoculate 10 mL LB containing the appropriate antibiotics and cultures were incubated with orbital shaking at 37 °C overnight.Unless stated otherwise, protein expression cultures were prepared by inoculating 250 mL of LB media in a 500 mL baffled Erlenmeyer flask containing the appropriate antibiotics with 1% vol/vol overnight culture.The cultures were incubated with orbital shaking at 37 °C until OD600 = 0.45-0.55 and isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to a final concentration of 0.4 mM.
Cultures were incubated at 21 °C for 24 h before harvesting cells for biotransformation reactions.

S4.2 Whole cell biotransformation of terephthalic acid to adipic acid
Freshly prepared cells from expression cultures were harvested by centrifugation (4000 xg, 19 °C, 20 min) and the supernatant carefully removed.Cell pellets were resuspended in biotransformation buffer (M9-3% glucose containing the appropriate substrate) to OD600=60, unless otherwise noted.For biotransformation, a 3 mL aliquot of resuspended cells in reaction buffer was added to a 15 mL conical centrifuge tube per screening reaction.Reactions were incubated at 21 °C at 220 rpm for the required time and then analyzed by HPLC using the method outlined in Section S4.5.For increased aeration where necessary 8 holes were pierced into the lid of the reaction tube using a sterile needle.

S4.3 Fermentation reactions
Cultures of E. coli BL21(DE3) transformed with pPCA1 and pAA were grown to OD600 = 0.45-0.55 in either M9-glucose, M9-glycerol or LB and the appropriate antibiotics, and induced with IPTG (0.4 mM).After incubation at 21 ˚C and 220 rpm for 18 h, TA (5.0 mM) was added, and cultures were incubated for the required time before being analyzed by HPLC using the method outlined in Section S4.5.
Unless otherwise specified, 3.5 mL of biotransformation reaction filtered supernatant containing ccMA (under conditions described in Section S4.2) was mixed with 20 mol% of Pd catalyst (based on initial TA concentration) and transferred to hydrogen-producing cultures through rubber seals with an 18G, 1.5" needle.Cultures were incubated for a further 22h at 37 °C at 220 rpm before being analyzed by HPLC using the method outlined in Section S4.5.

S4.5 HPLC sample preparation from biotransformation reactions
A 200 µL sample was removed from the biotransformation reaction and quenched with 400 µL acetonitrile containing 0.15% v/v trifluoroacetic acid (TFA).Samples were vortexed for 10 sec and incubated at room temperature for 30 min before being vortexed for 10 sec and clarified by centrifugation (15000 xg, 10 min).A 300 µL aliquot of the supernatant was transferred to a fresh tube and the solvent allowed to evaporate in a fume cupboard overnight.300 µL of MilliQ-H2O containing 0.1% v/v TFA and 51 µM caffeine was added and the samples were clarified by centrifugation (15000 xg, 10 min) before analysis by HPLC using the method outlined in Section S1.

S4.6 Cell immobilization in alginate hydrogels
Following the preparation of whole cell reactions as outlined in Section S4.

Figure S5 .
Figure S5.Response curves for pathway intermediates in comparison to a caffeine internal standard.Reaction samples were diluted to within the linear range prior to analysis by HPLC.Response curves generated using the mean values of triplicate runs for each analyte concentration.
2, cell samples were resuspended in 1.5% w/vol sodium alginate solution.The resulting solution was added dropwise to a 0.1 M aqueous solution of calcium chloride.The resulting beads were left to solidify for 10 min at room temperature.Alginate beads containing 1.5 mL cell samples were then added to 1.5 mL of M9 reaction buffer containing 2X TA before being incubated at 21 ˚C at 220 rpm for 24 h.Reaction supernatants were then analyzed by HPLC using the method outlined in Section S4.5.

Figure S11 .Figure S12 .
Figure S11.Production of ccMA from various E. coli BL21(DE3) strains, expressing TPADO from a chromosomal integration at different loci or from plasmid pPCA1.For each strain to cell concentrations were tested (n = 1).

Figure S13 .S26Figure S14 .
Figure S13.Bio-reduction of ccMA by engineered E. coli cells harboring pAA4 and pAA3.Reactions were performed at two different cell concentrations with n=3 for reactions at OD600=60 and n=2 for reactions at OD600=120.

Figure S20 .Figure S21 .
Figure S20.Control reactions showing increased BcER activity in alginate beads.A) reactions run with 2.5 mM ccMA for 24 h in aerobic conditions after 6 h of pre incubation without substrate.B) reactions run with 2.5 mM TA in various sizes of alginate beads.

Figure S22 .
Figure S22.Control reactions showing terephthalate dependent BcER inhibition.A) ccMA and AA production from 1 mM PCA, 1 mM TA (Low TA), 1 mM PCA or 5 mM TA (High TA).B) AA production from 2.5 mM ccMA after the addition of TA.

Figure S23 .
Figure S23.Effect of pH on whole pathway and PCA synthesis.A) Conversion of 5 mM ccMA by non-immobilized _pPCA1_pAA4 cells at various starting pH values.B) Effect of pH on conversion of 5 mM TA to PCA after 5 h.C) Synthesis of ccMA and AA from TA in immobilized E. coli_pPCA1_pAA4 cells at different pH values.

Figure S25 .
Figure S25.Whole-cell biotransformation experiments using commercial TA, and TA isolated from various PET sources, as outlined in Section S4.7.

1 HFigure S26. 1 H
Figure S26.1 H NMR spectra of TA, BHET, and ethylene glycol commercial standards, aligned with TA obtained by hydrolysis of either a plastic bottle or hot stamping foil.
Products from PCR were gel purified with a Zymoclean Gel DNA Recovery Kit (Zymo Research).All restriction enzymes and T4 DNA ligase were purchased from Thermo Fisher and used following manufacturer recommendations.Plasmid DNA was purified with a Miniprep Kit (Qiagen) from E. coli DH5α.All generated plasmids were confirmed by colony PCR and Sanger sequencing (Azenta).
TM(Thermo Scientific).Oligonucleotide primers were synthesized by Integrated DNA Technologies.OneTaq 2X (New England Biolabs, NEB) was used for colony PCRs, Phusion High-Fidelity DNA Polymerase (NEB) was used for all other PCR reactions, following manufacturer recommendations.Homology-mediated DNA assembly was done with SLiCE (Seamless Ligation Cloning Extract)

Table S1 .
List of plasmids used in this study.

Table S2 .
Nucleotide sequences of protein coding sequences used in this study

Table S3 .
Oligonucleotide primers used for plasmid construction.

Table S4 .
Oligonucleotide primers used for construction of plasmid donor DNA for genome modification.

Table S5 .
Oligonucleotide primers used for genotype confirmation of PCA-integration strains.