Search

Deconstruction of polyethylene terephthalate (PET) plastic waste generates opportunities for valorization to alternative products. We recently designed an enzymatic cascade that could produce terephthalaldehyde (TPAL) from terephthalic acid. Here, we showed that the addition of TPAL to growing cultures ofEscherichia coliwild-type strain MG1655 and an engineered strain for reduced aromaticaldehyderection (RARE) strain resulted in substantial reduction. We then investigated if we could mitigate this reduction using multiplex automatable genome engineering (MAGE) to create anE. colistrain with 10 additional knockouts in RARE. Encouragingly, we found this newly engineered strain enabled a 2.5-fold higher retention of TPAL over RARE after 24h. We applied this new strain for the production ofpara-xylylenediamine (pXYL) and observed a 6.8-fold increase in pXYL titer compared to RARE. Overall, our study demonstrates the potential of TPAL as a versatile intermediate in microbial biosynthesis of chemicals that derived from waste PET.

Aldehydes are attractive chemical targets given applications as end products in the flavors and fragrances industry and as intermediates due to their propensity for C-C bond formation. While biosynthetic routes to diverse aldehydes have been designed, a common challenge is the stability of these aldehydes in the presence of microbial hosts of engineered pathways. Here, we identify and address unexpected oxidation of a model collection of aromatic aldehydes, including many that originate from biomass degradation, in the presence ofEscherichia colistrains that were engineered to minimize aldehyde reduction. Of heightened interest to us were resting cell conditions as they offer numerous advantages for the bioconversion of toxic metabolites. Surprisingly, when diverse aldehydes are supplemented toE. coliRARE cells grown under aerobic conditions, they remain stabilized on the timescale of days, whereas when these same aldehydes are supplemented to resting cell preparations ofE. coliRARE that had been grown under the same conditions, we observe substantial oxidation. By performing combinatorial inactivation of six candidate aldehyde dehydrogenase genes in theE. coligenome using multiplexed automatable genome engineering (MAGE), we demonstrate that this oxidation can be substantially slowed, with greater than 50% retention of 6 out of 8 aldehydes when assayed 4 hours after their addition. Given that our newly engineered strain exhibitsReducedOxidationAndReduction of aromatic aldehydes, we dubbed it theE. coliROAR strain. Seeking to apply this new strain to resting cell biocatalysis, we compared the capability to synthesis the aldehyde furfural from 2-furoic acid via the carboxylic acid reductase enzyme fromNocardia iowensis. Here, we found that use of ROAR resting cells achieved 2-fold enhancement in furfural titer after 4 h and nearly 9-fold enhancement after 20 h as compared to resting cells of the RARE strain. Moving forward, the use of this strain to generate resting cells should allow aldehyde product isolation, further enzymatic conversion, or chemical reactivity under cellular contexts that better accommodate aldehyde toxicity.HighlightsWhen genes that encode aldehyde reductases are knocked out inEscherichia colistrains, supplemented aldehydes can experience oxidation instead of reduction, which is catalyzed by a different set of endogenous enzymes.Interestingly, we show for a collection of aromatic aldehydes that this oxidation is far more substantial when using resting cell preparations than during aerobic fermentation.We investigate the identity of the responsible genes by performing combinatorial gene inactivation using multiplex automatable genome engineering.The strain that we engineer exhibits Reduced Oxidation And Reduction (theE. coliROAR strain) and thereby enables design of more efficient aldehyde bioconversion processes under diverse formats.

To better incentivize the collection of plastic wastes, new chemical transformations must be developed that add value to plastic deconstruction products. Polyethylene terephthalate (PET) is a common plastic whose deconstruction through chemical or biological means has received much attention. However, a limited number of alternative products have been formed from PET deconstruction, and only a small share could serve as building blocks for alternative materials or therapeutics. Here, we demonstrate the production of useful mono-amine and diamine building blocks from known PET deconstruction products. We achieve this by designing one-pot biocatalytic transformations that are informed by the substrate specificity of an ω-transaminase and diverse carboxylic acid reductases (CAR) towards PET deconstruction products. We first establish that an ω-transaminase fromChromobacterium violaceum(cvTA) can efficiently catalyze amine transfer to potential PET-derived aldehydes to form the mono-aminepara-(aminomethyl)benzoic acid (pAMBA) or the diaminepara-xylylenediamine (pXYL). We then identified CAR orthologs that could perform the bifunctional reduction of TPA to terephthalaldehyde (TPAL) or the reduction ofmono-(2-hydroxyethyl) terephthalic acid (MHET) to its corresponding aldehyde. After characterizing 17 CARsin vitro, we show that the CAR fromSegniliparus rotundus(srCAR) had the highest observed activity on TPA. Given these newly elucidated substrate specificity results, we designed modular enzyme cascades based on coupling srCAR and cvTA in one-pot with enzymatic co-factor regeneration. When we supply TPA, we achieve a 69 ± 1% yield of pXYL, which is useful as a building block for materials. When we instead supply MHET and subsequently perform base-catalyzed ester hydrolysis, we achieve 70 ± 8% yield of pAMBA, which is useful for therapeutic applications and as a pharmaceutical building block. This work expands the breadth of products derived from PET deconstruction and lays the groundwork for eventual valorization of waste PET to higher-value chemicals and materials.GRAPHICAL ABSTRACT

Nitroaromatic functional groups can impart valuable properties to chemicals and to biological macromolecules including polypeptides. Para-nitro-L-phenylalanine (pN-Phe) is a nitroaromatic amino acid with uses including immune stimulation and fluorescence quenching. As the chemical synthesis of pN-Phe does not follow green chemistry principles and impedes provision of pN-Phe to engineered bacterial cells in some contexts, we sought to design a de novo biosynthetic pathway for pN-Phe in Escherichia coli. To generate the nitro chemical functional group, we identified natural diiron monooxygenases with measurable in vitro and in vivo activity on envisioned amine-containing precursors of para-amino-L-phenylalanine (pA-Phe) and para-aminophenylpyruvate. By expressing one of these N-oxygenase genes together with previously characterized genes for the biosynthesis of pA-Phe, we achieved the synthesis of pN-Phe from glucose. Through further optimization of the chassis, plasmid constructs, and media conditions, we were able to improve the selectivity of pN-Phe biosynthesis, resulting in a maximum titer of 819 µM in rich defined media under shake-flask conditions. These results provide a foundation for the biosynthesis of related nitroaromatic chemicals and for downstream biological applications that could utilize pN-Phe as a building block.HighlightsPara-nitro-L-phenylalanine (pN-Phe) is a valuable small molecule for its applications in genetic code expansion.We establish de novo biosynthesis of pN-Phe from glucose in E. coli, which is also the first example of a de novo pathway design for an unnatural but commonly used non-standard amino acid.We show the first use of an N-oxygenase enzyme in the de novo synthesis of a nitroaromatic product.Screening of natural N-oxygenases and strain engineering resulted in final pN-Phe titers of 820 ± 130 µM in shake flask experiments with rich defined media.

Efforts are underway to construct several recoded genomes anticipated to exhibit multi-virus resistance, enhanced non-standard amino acid (NSAA) incorporation, and capability for synthetic biocontainment. Though we succeeded in pioneering the first genomically recoded organism (Escherichia colistrain C321.ΔA), its fitness is far lower than that of its non-recoded ancestor, particularly in defined media. This fitness deficit severely limits its utility for NSAA-linked applications requiring defined media such as live cell imaging, metabolic engineering, and industrial-scale protein production. Here, we report adaptive evolution of C321.ΔA for more than 1,000 generations in independent replicate populations grown in glucose minimal media. Evolved recoded populations significantly exceed the growth rates of both the ancestral C321.ΔA and non-recoded strains, permitting use of the recoded chassis in several new contexts. We use next-generation sequencing to identify genes mutated in multiple independent populations, and we reconstruct individual alleles in ancestral strains via multiplex automatable genome engineering (MAGE) to quantify their effects on fitness. Several selective mutations occur only in recoded evolved populations, some of which are associated with altering the translation apparatus in response to recoding, whereas others are not apparently associated with recoding, but instead correct for off-target mutations that occurred during initial genome engineering. This report demonstrates that laboratory evolution can be applied after engineering of recoded genomes to streamline fitness recovery compared to application of additional targeted engineering strategies that may introduce further unintended mutations. In doing so, we provide the most comprehensive insight to date into the physiology of the commonly used C321.ΔA strain.Significance StatementAfter demonstrating construction of an organism with an altered genetic code, we sought to evolve this organism for many generations to improve its fitness and learn what unique changes natural selection would bestow upon it. Although this organism initially had impaired fitness, we observed that adaptive laboratory evolution resulted in several selective mutations that corrected for insufficient translation termination and for unintended mutations that occurred when originally altering the genetic code. This work further bolsters our understanding of the pliability of the genetic code, it will help guide ongoing and future efforts seeking to recode genomes, and it results in a useful strain for non-standard amino acid incorporation in numerous contexts relevant for research and industry.

Progress in genetic code expansion requires accurate, selective, and high-throughput detection of non-standard amino acid (NSAA) incorporation into proteins. Here, we discover how the N-end rule pathway of protein degradation applies to commonly used NSAAs. We show that several NSAAs are N-end stabilizing and demonstrate that other NSAAs can be made stabilizing by rationally engineering the N-end rule adaptor protein ClpS. We use these insights to engineer a synthetic quality control method, termed “Post-Translational Proofreading” (PTP). By implementing PTP, false positive proteins resulting from misincorporation of structurally similar standard amino acids or undesired NSAAs rapidly degrade, enabling high-accuracy discrimination of desired NSAA incorporation. We illustrate the utility of PTP during evolution of the biphenylalanine orthogonal translation system used for synthetic biocontainment. Our new OTS is more selective and confers lower escape frequencies and greater fitness in all tested biocontained strains. Our approach presents a new paradigm for molecular recognition of amino acids in target proteins.