A Microbial Cell Factory Combining an Artificial Metathase with a Fatty Acid Decarboxylase for New-to-Nature Synthesis of Cycloalkenes

doi:10.21203/rs.3.rs-3706252/v1

Download PDF

Article

A Microbial Cell Factory Combining an Artificial Metathase with a Fatty Acid Decarboxylase for New-to-Nature Synthesis of Cycloalkenes

https://doi.org/10.21203/rs.3.rs-3706252/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 27 Jun, 2024

Read the published version in Nature Synthesis →

Version 1

posted

You are reading this latest preprint version

Artificial metalloenzymes (ArMs) are powerful tools for catalyzing new-to-nature reactions. ArMs offer great potential to complement natural enzymes in concurrent cascade reactions. For in cellulo applications, however, the abundance of thiols in the cytosol of aerobic organisms limits the use of ArMs that rely on precious metal cofactors. To overcome this limitation, compartmentalization of ArMs either in the periplasm or on the surface of Escherichia coli has shown promise. We report herein on our efforts to combine a surface-displayed artificial metathase with UndB, an inner membrane decarboxylase. The resulting concurrent cascade converts medium-chain dicarboxylates to cyclic alkenes. To optimize the cascade, we evolved both enzymes independently and fine-tuned their relative expression levels. Compared to the wild-type whole-cell enzyme cascade, the engineered strain was improved by > 75-fold for the conversion of sebacic acid to cyclohexene. To the best of our knowledge, this represents the first example of a whole-cell enzyme cascade that combines a natural enzyme with an ArM.

Biological sciences/Biochemistry/Biocatalysis

Biological sciences/Systems biology/Synthetic biology

Physical sciences/Chemistry/Biosynthesis

Physical sciences/Chemistry/Coordination chemistry/Organometallic chemistry/Metathesis reactions

Artificial metalloenzymes (ArMs) result from the incorporation of an abiotic metal cofactor within a host protein. Over the years, such hybrid catalysts have been shown to combine attractive features of both homogeneous catalysts and enzymes^1–3. Thanks to the use of precious metal-containing cofactors, ArMs have been designed to catalyze various abiotic reactions, including cross-coupling⁴, C–H activation⁵, cyclopropanation⁶, ring-closing metathesis⁷, etc³. For optimization purposes, the power of directed evolution is unrivaled^8–13, a strategy that can be applied to ArMs. Thus far, however, most of the implementation effort with ArMs has focused on in vitro catalysis, relying on purified proteins. Catalyzing new-to-nature reactions within a cellular environment offers a wealth of potential applications in biotechnology^{6, 14–16}, therapy^{17, 18}, and biosensors¹⁹. For biotechnical applications, whole-cell enzymatic reactions alleviate purification procedures and are more cost-effective than cell-free systems^{20, 21}: More than half of all industrial enzymatic reactions rely on whole-cell biocatalysis²². Two challenges limit the implementation of ArMs in whole cells: i) the inactivation of precious metal cofactors by cellular nucleophiles, such as thiol derivatives and proteins^{23, 24}, and ii) the uptake and accumulation of the cofactor within the cytoplasm, assuming that substrates and products passively diffuse through non-specific pores^{25, 26}. To circumvent these challenges, several groups have relied on secreting the host protein scaffolds to the periplasm of E. coli^{7, 27}, or by displaying them on its cell surface^{22, 28}. For the latter strategy, monomeric host proteins are preferred, as the quaternary structure of a surface-displayed protein may not reflect its solution-phase structure²⁹.

Thanks to its versatility and bioorthogonality^{30, 31}, olefin metathesis –an abiotic reaction par excellence– is one of the most widely used reactions in synthetic chemistry. In this context, (derivatives of) the ruthenium-based Hoveyda-Grubbs catalyst (Ru-HG) has widely been used both in aqueous solutions and in cellular environments^32–34. Capitalizing on these findings, ArMs equipped with a Ru-HG cofactor have been extensively used to explore and validate various concepts in the field of ArMs^{17, 35, 36}. Capitalizing on a biotin-streptavidin or an arylsulfonamide-human carbonic anhydrase II couple, the authors have developed whole-cell platforms for new-to-nature biosynthesis, either in the periplasm or as E. coli surface-displayed ArMs^{7, 27, 28}. To the best of our knowledge, these have never been integrated into concurrent whole-cell biocatalytic cascades combining natural enzymes and ArMs.

We thus set out to address this challenge and selected a cascade consisting of an UndB-catalyzed decarboxylation to afford an olefin, combined with a ring-closing olefin metathesis (RCM) step, catalyzed by an artificial metathase. As medium-chain dicarboxylates are readily accessible from either glucose or oleic acid in concurrent enzyme cascades, we set out to produce cycloalkenes from the corresponding bio-derived dicarboxylates. Cycloalkenes, including cyclopentene (Cy5), cyclohexene (Cy6), and cycloheptene (Cy7), are traditionally derived from petroleum feedstocks and are used as bulk chemicals in a broad range of industrial applications³⁷. Using glucose as a carbon source and relying on the fatty acid-acyl carrier protein pathway (fatty acid-ACP), fatty diacids can be produced in E. coli (Fig. 1a, right)^38–40. We previously reported an alternative whole-cell enzyme cascade that converts olive oil to either azelaic acid (hereafter C9 diacid) and sebacic acid (hereafter C10 diacid)⁴¹, Fig. 1a, left. Unfortunately, our efforts to assemble a whole-cell enzyme cascade that converts these dicarboxylates to cyclopentene Cy5 and cyclohexene Cy6, respectively, by combining UndB with an artificial metathase failed. Instead, we had to resort to a biphasic mixture, combining UndB with a homogeneous Hoveyda-Grubbs metathesis catalyst Ru-HG. We present below a strategy to assemble a whole-cell microbial factory that converts fatty diacids into the corresponding cycloalkenes, Fig. 1b.

Selection of the Ru-HG catalyst and the protein scaffold for assembling the E. coli surface-displayed artificial metathase

While streptavidin (Sav) has proven highly versatile as a host protein for various ArMs, we selected human carbonic anhydrase II (hCAII) to assemble the concurrent whole-cell cascade for this study. Indeed, we have shown that the quaternary structure of homotetrameric Sav was compromised when displayed on E. coli’s outer membrane²⁹. Additionally, the artificial metathase based on Sav requires acidic conditions (typically pH < 6)^{7, 35}, incompatible with UndB’s activity, Figure S3a⁴². In contrast, we have shown that the artificial metathase based on hCAII maintains its activity up to pH 7.0 ²⁷. Accordingly, we evaluated the catalytic performance of a Ru-HG cofactor equipped with an arylsulfonamide anchor (Ru1, displayed in Fig. 1b) at pH 7.4. As a model substrate for ring-closing olefin metathesis (RCM) relying on purified hCAII samples, we selected 1,7-octadiene as substrate at pH 7.4. Ru1·hCAII wild-type (WT) yielded higher turnover nubmers(TON) at all tested cofactor concentrations than the corresponding free cofactor Ru1, Fig. 2a. Using [Ru1] = 30 µM, 1,7-octadiene was converted to cyclohexene with 25 and 42 TON for Ru1 and Ru1·hCAII WT, respectively. Accordingly, Ru1·hCAII was selected for assembling the microbial cell factory combined with UndB⁴².

For the compartmentalization of hCAII on E. coli cell surface display, two E. coli surface display constructs were evaluated: Lpp-OmpA-hCAII WT²⁸ and PelB-INPN-hCAII WT. Overall, the PelB-INPN-hCAII (hereafter hCAII^CSD) led to the highest hCAII concentrations, as revealed by whole-cell sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE, Figure S1a) and carboxyfluorescein-staining of the cells, Figure S2. The hCAII^CSD construct was thus selected for the assembly of Ru1·hCAII on the cell surface of E. coli. Similar constructs were also selected by Yang and Jia for the use of surface-displayed hCAII for CO₂ capture purposes^{43, 44}.

Optimization of the activity of the whole-cell based artificial metathase

In order to identify the optimal pH for engineering the microbial cell factory, we evaluated the pH-dependent activity of the UndB WT and cell-surface displayed Ru1·hCAII^CSD WT in distinct strains, Figure S3. As can be appreciated, the trends of both enzymes follow opposite trends. A good compromise was identified at physiological pH. Accordingly, the pH = 7.4 was selected for the remainder of the studies. We further investigated the effect of the concentration of Ru1 in the assembly of Ru1·hCAII^CSD WT. At [Ru1] = 10 µM, a concentration of [Cy6] = 10.0 mg·L^− 1 was reached within 20 hours, Figure S4a. The empty vector control produced only [Cy6] = 0.7 mg·L^− 1. Reducing [Ru1] = 2 µM, resulted in a significant reduction in Cy6 yield (i.e. 0.25 mg·L^− 1). Increasing the concentration to [Ru1] = 20 µM yielded a comparable yield of Cy6 as with [Ru1] = 10 µM (after subtracting the background activity obtained with the empty vector control). The temperature and incubation time for the whole-cell catalyzed RCM were also optimized: the highest Cy6 yield was achieved after 20 hours at 20°C ([Cy6] = 13.7 mg·L^− 1, Figure S4b). To further improve the TON, we optimized the whole-cell RCM activity of Ru1·hCAII^CSD by directed evolution.

Directed evolution of the artificial metathase for enhanced RCM activity

In the past decade, directed evolution strategies have been applied to the genetic optimization of artificial metalloenzymes^{3, 45–50}. Inspection of the docked structure of Ru1·hCAII WT revealed six residues located in the proximity of the Ru1 cofactor, Figs. 2b and S7a. Based on previous directed evolution efforts on artificial metathases^{7, 35, 51}, we focused on introducing hydrophobic residues at the following positions: R58, N62, H64, N67, E69, and D72. These polar or charged amino acid residues were mutated by site-directed mutagenesis to Ala, Gly, Phe, Ile, Leu, Pro, and Val, Table S2. A library comprising forty-two variants was expressed and screened using an optimized protocol, Figure S5. Among them, seven variants (R58A, E69A, E69F, E69I, E69L, E69V, and D72F) displayed ≥ 1.4-fold improved activity compared to Ru1·hCAII^CSD WT, Figure S7b. The variant hCAII E69A presented the highest activity (1.8-fold) and was selected as the parent for further recombination. In the first round of recombination, twelve mutations (R58A/F/G/I/L/V and D72A/F/G/I/L/V) were recombined with hCAII E69A, Table S2. The resulting variants R58A/E69A and E69A/D72L exhibited up to a 2.2-fold improvement in the production of Cy6 (vs. Ru1·hCAII^CSD WT). In the second round of recombination, six mutations at position 72 (D72A/F/G/I/L/V) were recombined with R58A/E69A. The resulting variant hCAII R58A/E69A/D72L (hereafter hCAII^CSD R3) was identified with a 2.8-fold improved activity vs. hCAII^CSD WT, Fig. 3b and S7b.

To further enhance the activity of Ru1·hCAII^CSD R3, site-saturation mutagenesis libraries (SSM) at positions Q92, L198, and T199 were generated. These positions were selected as they lie within 4 Å of the sulfonamide anchor of the Ru1 cofactor, Figure S7c. Upon screening the hCAII^CSD R3 92_SSM, 198_SSM, and 199_SSM libraries, three improved mutations were identified (hCAII^CSD R3 Q92P, hCAII^CSD R3 T199H, and hCAII^CSD R3 T199P), resulting in up to a 3.3-fold improvement in Cy6 production compared to Ru1·hCAII^CSD WT, Figure S6 and S7d. Recombination of the beneficial mutations resulted in a penta-variant hCAII^CSD R58A/E69A/D72L/Q92P/T199P (hereafter hCAII^CSD R5) which further improved the yield of Cy6 (i.e. 50.3 mg·L^− 1 vs. 13.7 mg·L^− 1 for Ru1·hCAII^CSD R5 and Ru1·hCAII^CSD WT, respectively, Fig. 3b and S7d).

By truncating the PelB-INPN sequence, the variant hCAII R5 was expressed in the cytoplasm and purified by affinity chromatography, Figure S8b⁵². The binding affinity of Ru1 for hCAII WT and hCAII R5 was determined. The evolved hCAII R5 had a 5.8-fold higher binding affinity towards Ru1 compared to hCAII WT, Fig. 3c and Figure S16 (SI). The purified hCAII R5 variant was used to assemble the evolved artificial metathase Ru1·hCAII R5, and its in vitro activity towards three medium chain terminal dienes to afford Cy5, Cy6, and Cy7 was determined. In the presence of Ru1·hCAII R5, all three dienes led to improved yields for the production of Cy5, Cy6, and Cy7, albeit to different extents. The improvement in RCM performance for Ru1·hCAII R5 vs. Ru1 is most pronounced for Cy5: 7.1- fold for Cy5 vs. 3.7 fold for Cy6, Fig. 3c and Table S6.

Directed evolution of the UndB decarboxylase for enhanced production of terminal alkenes

To date, three classes of enzymes have been reported for the decarboxylation of fatty acids to the corresponding terminal alkenes: OleT, UndA, and UndB^{53, 54}. Among these enzymes, UndB has been identified as the most efficient enzyme for in vivo decarboxylation^{41, 55, 56}. UndB is a membrane-associated (Figure S1b), oxygen-dependent, non-heme diiron enzyme that converts medium-chain fatty acids (C10–16) into the corresponding 1-alkenes through oxidative decarboxylation^{54, 57}. Given the potential of medium-chain 1-alkenes as biofuels, UndB holds promise for various biotechnological applications. Despite this versatility, the catalytic potential of UndB remains underexplored. We applied directed evolution to improve the activity of UndB for the bis-decarboxylation of fatty diacids, Figure S9. Guided by the predicted structure from AlphaFold (AF-A4YOK1-F1)⁵⁸, we subjected fifteen residues located within 5 Å of the active site to alanine scanning, Figure S10a. This effort aimed at identifying positions where potential improvements to UndB’s activity may be achieved. Two variants, UndB I89A and UndB M257A, exhibited higher activity for the bis-decarboxylation of C10 diacid to 1,7-octadiene, compared to the UndB WT, Figure S10. Subsequent screening of an SSM library at position 89 yielded a variant, UndB I89V, which exhibited two-fold improved activity vs. UndB WT, Figure S11a. Next, we generated SSM libraries at positions UndB 92_SSM, 126_SSM, 254_SSM, and 257_SSM using UndB I89V as the parent. Screening these four SSM libraries resulted in six beneficial variants, Figure S11b. Specifically, the variant UndB I89V/126F exhibited a 3.1-fold higher activity than UndB WT, Fig. 3a and Table S7. Further recombination of the beneficial substitutions of M92R, M257A, or M257V into UndB I89V/L126F did not lead to improved variants, Figure S11c.

To achieve higher yields of 1,7-octadiene, the conditions for bis-decarboxylation involving the UndB I89V/L126F variant were optimized. We previously demonstrated that supplementing resting UndB cells with glucose enhanced the activity of bis-decarboxylation,⁴¹ presumably due to improved regeneration of electron sources such as NADH or NADPH. By substituting glucose (1% (w/w)) with glycerol (0.5% (v/v)), the yield of 1,7-octadiene by UndB I89V/L126F variant significantly increased from 57.8 to 90.2 mg·L^− 1, highlighting the potential of glycerol as a more efficient electron-donor, Fig. 4a and Table S7.

We further optimized the bis-decarboxylation of UndB by varying the 1 mM ≤ [C10 diacid] ≤ 5 mM and the UndB cell density (12.5 ≤ OD₆₀₀ ≤ 100). The optimal production of 1,7-octadiene (i.e. 133 mg·L^− 1) was achieved with a 1.5 mM ≤ [C10 diacid] ≤ 2 mM. Beyond this range, the yield of 1.7-octadiene decreased. For instance, at [C10 diacid] = 5 mM, the yield of 1.7-octadiene dropped to 63.4 mg·L^− 1. These results suggest that high fatty diacid concentrations should be avoided to maximize the yield of 1.7-octadiene. Moreover, the maximum yield of 1,7-octadiene was achieved with OD₆₀₀ = 25. Below or above this threshold, the yield of 1.7-octadiene declined: at OD₆₀₀ = 100 resulted in a 93% decrease in the yield of 1.7-octadiene compared to an OD₆₀₀ = 25. We hypothesize that high cell densities may hamper the efficient extraction of 1.7-octadiene. Upon lowering the temperature and extending the reaction time (e.g., 20°C, 24 hours) the yield of 1,7-octadiene increased to 191.5 mg·L^− 1, corresponding to > 86% yield, Fig. 4c. Decarboxylation of lauric acid (mono fatty acid, C12) by the UndB I89V/L126F variant afforded 263.3 mg·L^− 1 undec-1-ene (> 85% yield, Fig. 4c).

UndB catalyzes the decarboxylation of a variety of medium-chain-length fatty diacids⁵⁴. We evaluated the activity of UndB I89V/L126F variant towards C9–C14 diacids. Under optimized conditions, the UndB I89V/L126F variant exhibited the highest yield for the bis-decarboxylation of C12–C13 diacids, with > 245 mg·L^-1 titers corresponding to >85% yield. With either shorter or longer diacids, the yields decreased. These results substantiate a previous in vivo study with fatty acids⁵⁴.

Assembly and engineering of a microbial cell factory for the biosynthesis of cycloalkenes Cy5 Cy6 and Cy7 from the corresponding diacids

Performing enzymatic cascade-reactions in whole-cell offers a cost-effective strategy for various applications. Integrating ArMs in whole-cell concurrent cascade reactions has great potential for new-to-nature biosynthesis^{34, 59}. Co-expressing UndB and hCAII^CSD in a single E. coli cell, enables a bis-decarboxylation and ring-closing metathesis concurrent cascade, Fig. 1b. The optimization of UndB and hCAII^CSD co-expression required integrating their genes into one or two plasmid vectors, which were subsequently transformed into E. coli, Table S8. The productivity of the co-expressed UndB and hCAII^CSD was determined using C10 diacid as the model substrate. All the co-expression strains resulted in a significantly reduced yield of 1,7-octadiene, Figure S13. Among them, the co-expression strain harboring pRSF-UndB I89V/L126F and pACYC-hCAII^CSD R5 (hereafter CoEx3*) gave the highest yield, with a titer of 34.2 mg·L^− 1 (i.e. 15.5% yield).

To enhance the yield of 1,7-octadiene and Cy6 within CoEx3*, we set out to optimize both the absolute and relative expression levels of UndB I89V/L126F and hCAII^CSD R5 simultaneously. Leveraging on the RedLibs algorithm and uASPIre/SAPIENs-based RBS predictions^{60, 61}, a set of degenerate RBSs were designed to fine-tune the expression levels of the two enzymes, Table S9. The RBS libraries of UndB I89V/L126F and hCAII^CSD R5 were generated, co-transformed, and screened in a 24-well plate, Figure S14 and S15. The ten strains with the highest yields in Cy6 were rescreened in triplicate, Table S10. Ultimately, the most productive strain RBS P8_15 (hereafter CoEx3*-S183) yielded 96.6 and 15.5 mg·L¹ of 1,7-octadiene and Cy6, respectively. Notably, the yield of Cy6 by CoEx3*-S183 was 6.0-fold higher than the CoEx3*, corresponding to a remarkable 77.6-fold increase compared to the initial CoEx3 strain (co-expressed with UndB WT and hCAII^CSD WT, without the RBS optimization step, Fig. 5b).

Next, the engineered CoEx3*-S183 strain was evaluated for the production of Cy5 and Cy7, starting from C9 diacid and undecanedioic acid (C11 diacid) as substrates. The yields of 1,6-heptadiene and 1,8-nonadiene resulting from bis-decarboxylation were comparable to that of 1,7-octadiene, Table S11. The yield of Cy5 and Cy7 was significantly lower: 3.1 mg·L^− 1 and 1.1 mg·L^− 1, for Cy5 and Cy7 respectively, Fig. 5c. This can be traced back to the inherently lower propensity of 1,6-heptadiene and 1,8-nonadiene toward ring-closing metathesis⁶², Fig. 3c.

Artificial metalloenzymes catalyze abiotic reactions that complement the reaction repertoire of natural enzymes. Their integration into whole cells offers fascinating perspectives for biotechnological applications. In this study, we engineered a microbial cell factory incorporating a surface-displayed artificial metathase that functions in concert with an evolved bis-decarboxylase, thus enabling the production of cycloalkenes from fatty diacids in a concurrent cascade reaction. Optimization of the yield of the cycloalkene was achieved by combining the directed evolution of both enzymes, combined with the engineering of the reaction flux by fine-tuning the expression levels of both enzymes. Compared to the unoptimized wild-type enzyme cascade, this strategy leads to up to a > 75-fold increase in yield of Cy6. Under optimized conditions, we synthesized up to 15.5 mg·L^− 1 Cy6 from a bio-renewable fatty diacid, which can be produced either from glucose or oleic acid^38–41. To the best of our knowledge, this is the first report that combines a whole-cell reaction cascade integrating an artificial metalloenzyme with a natural metalloenzyme.

Data availability

The original materials, methods, and data that support the findings of this study can be accessed within the paper and its Supplementary Information.

Code availability

The source code for RedLibs v1.1.0, algorithm is available at GitHub: https://github.com/dgerngross/RedLibs. The code used in uASPIre/SAPIENs-based RBS predictions are available under: https://github.com/JeschekLab/uASPIre and https://github.com/BorgwardtLab/SAPIENs.

Acknowledgements

T. R. W. thanks the NCCR Molecular Systems Engineering (Grant Number 200021_178760) and the SNSF (Grant number 200020_212088) for generous funding. S. W. thanks the Federal Commission for Scholarships for Foreign Students for an ESKAS Scholarship. R. T. thanks the Naito Foundation for financial support.

Lewis, J.C. Artificial Metalloenzymes and Metallopeptide Catalysts for Organic Synthesis. ACS Catal. 3, 2954-2975 (2013).
Bos, J. & Roelfes, G. Artificial metalloenzymes for enantioselective catalysis. Curr. Opin. Chem. Biol. 19, 135-143 (2014).
Schwizer, F. et al. Artificial Metalloenzymes: Reaction Scope and Optimization Strategies. Chem. Rev. 118, 142-231 (2018).
Li, J. et al. Ligand-Free Palladium-Mediated Site-Specific Protein Labeling Inside Gram-Negative Bacterial Pathogens. J. Am. Chem. Soc. 135, 7330-7338 (2013).
Yu, K. et al. Artificial Metalloenzyme-Catalyzed Enantioselective Amidation via Nitrene Insertion in Unactivated C(sp(3))-H Bonds. J. Am. Chem. Soc. 145, 16621-16629 (2023).
Huang, J. et al. Unnatural biosynthesis by an engineered microorganism with heterologously expressed natural enzymes and an artificial metalloenzyme. Nat. Chem. 13, 1186-1191 (2021).
Jeschek, M. et al. Directed evolution of artificial metalloenzymes for in vivo metathesis. Nature 537, 661-665 (2016).
Arnold, F.H. Directed Evolution: Bringing New Chemistry to Life. Angew. Chem. Int. Ed. 57, 4143-4148 (2018).
Reetz, M.T. & Carballeira, J.D. Iterative saturation mutagenesis (ISM) for rapid directed evolution of functional enzymes. Nat. Protoc. 2, 891-903 (2007).
Bornscheuer, U.T., Hauer, B., Jaeger, K.E. & Schwaneberg, U. Directed Evolution Empowered Redesign of Natural Proteins for the Sustainable Production of Chemicals and Pharmaceuticals. Angew. Chem. Int. Ed. 58, 36-40 (2019).
Wang, Y.J. et al. Directed Evolution: Methodologies and Applications. Chem. Rev. 121, 12384-12444 (2021).
Packer, M.S. & Liu, D.R. Methods for the directed evolution of proteins. Nat. Rev. Genet. 16, 379-394 (2015).
Bornscheuer, U.T. et al. Engineering the third wave of biocatalysis. Nature 485, 185-194 (2012).
Samanta, A., Sabatino, V., Ward, T.R. & Walther, A. Functional and morphological adaptation in DNA protocells via signal processing prompted by artificial metalloenzymes. Nat. Nanotechnol. 15, 914-921 (2020).
Okamoto, Y. et al. A cell-penetrating artificial metalloenzyme regulates a gene switch in a designer mammalian cell. Nat. Commun. 9, 1943 (2018).
Wittwer, M. et al. Engineering and emerging applications of artificial metalloenzymes with whole cells. Nat. Catal. 4, 814-827 (2021).
Eda, S. et al. Biocompatibility and therapeutic potential of glycosylated albumin artificial metalloenzymes. Nat. Catal. 2, 780-792 (2019).
Ghattas, W. et al. Receptor-Based Artificial Metalloenzymes on Living Human Cells. J. Am. Chem. Soc. 140, 8756-8762 (2018).
Vong, K. et al. An artificial metalloenzyme biosensor can detect ethylene gas in fruits and Arabidopsis leaves. Nat. Commun. 10, 5746 (2019).
Coelho, P.S. et al. A serine-substituted P450 catalyzes highly efficient carbene transfer to olefins in vivo. Nat. Chem. Biol. 9, 485-487 (2013).
Tufvesson, P., Lima-Ramos, J., Nordblad, M. & Woodley, J.M. Guidelines and Cost Analysis for Catalyst Production in Biocatalytic Processes. Org. Process. Res. Dev. 15, 266-274 (2011).
Grimm, A.R. et al. A Whole Cell E. coli Display Platform for Artificial Metalloenzymes: Poly(phenylacetylene) Production with a Rhodium-Nitrobindin Metalloprotein. ACS Catal. 8, 2611-2614 (2018).
Mallin, H., Hestericova, M., Reuter, R. & Ward, T.R. Library design and screening protocol for artificial metalloenzymes based on the biotin-streptavidin technology. Nat. Protoc. 11, 835-852 (2016).
Wilson, Y.M., Durrenberger, M., Nogueira, E.S. & Ward, T.R. Neutralizing the Detrimental Effect of Glutathione on Precious Metal Catalysts. J. Am. Chem. Soc. 136, 8928-8932 (2014).
Reynolds, E.W., Schwochert, T.D., McHenry, M.W., Watters, J.W. & Brustad, E.M. Orthogonal Expression of an Artificial Metalloenzyme for Abiotic Catalysis. ChemBioChem 18, 2380-2384 (2017).
Liu, Z.N. et al. Assembly and Evolution of Artificial Metalloenzymes within E. coli Nissle 1917 for Enantioselective and Site-Selective Functionalization of C-H and C=C Bonds. J. Am. Chem. Soc. 144, 883-890 (2022).
Zhao, J.M. et al. Genetic Engineering of an Artificial Metalloenzyme for Transfer Hydrogenation of a Self-Immolative Substrate in Escherichia coil's Periplasm. J. Am. Chem. Soc. 140, 13171-13175 (2018).
Rebelein, J.G., Cotelle, Y., Garabedian, B. & Ward, T.R. Chemical Optimization of Whole-Cell Transfer Hydrogenation Using Carbonic Anhydrase as Host Protein. ACS Catal. 9, 4173-4178 (2019).
Baiyoumy, A. et al. Directed Evolution of a Surface-Displayed Artificial Allylic Deallylase Relying on a GFP Reporter Protein. ACS Catal. 11, 10705-10712 (2021).
Furstner, A. Olefin metathesis and beyond. Angew. Chem. Int. Ed. 39, 3012-3043 (2000).
Grubbs, R.H. & Chang, S. Recent advances in olefin metathesis and its application in organic synthesis. Tetrahedron 54, 4413-4450 (1998).
Kingsbury, J.S., Harrity, J.P.A., Bonitatebus, P.J. & Hoveyda, A.H. A recyclable Ru-based metathesis catalyst. J. Am. Chem. Soc. 121, 791-799 (1999).
Garber, S.B., Kingsbury, J.S., Gray, B.L. & Hoveyda, A.H. Efficient and recyclable monomeric and dendritic Ru-based metathesis catalysts. J. Am. Chem. Soc. 122, 8168-8179 (2000).
Schunck, N.S. & Mecking, S. In Vivo Olefin Metathesis in Microalgae Upgrades Lipids to Building Blocks for Polymers and Chemicals. Angew. Chem. Int. Ed. 61, e202211285 (2022).
Vornholt, T. et al. Systematic engineering of artificial metalloenzymes for new-to-nature reactions. Sci. Adv. 7, eabe4208 (2021).
Mayer, C., Gillingham, D.G., Ward, T.R. & Hilvert, D. An artificial metalloenzyme for olefin metathesis. Chem. Comm. 47, 12068-12070 (2011).
Carretero-Gonzalez, J., Lopez, J.M., Barbero, M.A., Ramos, I.R. & Ruiz, A.G. Development of nanostructured catalytic membranes for partial benzene hydrogenation to cyclohexene. J. Nanosci. Nanotechnol. 7, 4391-4401 (2007).
Feng, Y. et al. Tuning of acyl-ACP thioesterase activity directed for tailored fatty acid synthesis. Appl. Microbiol. Biotechnol. 102, 3173-3182 (2018).
Seo, E.J. et al. Enzyme/whole-cell biotransformation of plant oils, yeast derived oils, and microalgae fatty acid methyl esters into n-nonanoic acid, 9-hydroxynonanoic acid, and 1,9-nonanedioic acid. Bioresour. Technol. 251, 288-294 (2018).
Malca, S.H. et al. Bacterial CYP153A monooxygenases for the synthesis of omega-hydroxylated fatty acids. Chem. Comm 48, 5115-5117 (2012).
Wu, S.K., Zhou, Y., Gerngross, D., Jeschek, M. & Ward, T.R. Chemo-enzymatic cascades to produce cycloalkenes from bio-based resources. Nat. Commun. 10, 5060 (2019).
Rui, Z. et al. Microbial biosynthesis of medium-chain 1-alkenes by a nonheme iron oxidase. Proc. Natl. Acad. Sci. USA. 111, 18237-18242 (2014).
Zhu, Y.Z., Liu, Y.R., Ai, M.M. & Jia, X.Q. Surface display of carbonic anhydrase on Escherichia coli for CO2 capture and mineralization. Synth. Syst. Biotechnol. 7, 460-473 (2022).
Fan, L.H., Liu, N., Yu, M.R., Yang, S.T. & Chen, H.L. Cell surface display of carbonic anhydrase on Escherichia coli using ice nucleation protein for CO(2) sequestration. Biotechnol. Bioeng. 108, 2853-2864 (2011).
Markel, U., Sauer, D.F., Schiffels, J., Okuda, J. & Schwaneberg, U. Towards the Evolution of Artificial Metalloenzymes-A Protein Engineer's Perspective. Angew. Chem. Int. Ed. 58, 4454-4464 (2019).
Markel, U. et al. Chemogenetic Evolution of a Peroxidase-like Artificial Metalloenzyme. ACS Catal. 11, 5079-5087 (2021).
Kato, S., Onoda, A., Schwaneberg, U. & Hayashi, T. Evolutionary Engineering of a Cp*Rh(III) Complex-Linked Artificial Metalloenzyme with a Chimeric beta-Barrel Protein Scaffold. J. Am. Chem. Soc. 145, 8285-8290 (2023).
Leveson-Gower, R.B., Zhou, Z., Drienovská, I. & Roelfes, G. Unlocking Iminium Catalysis in Artificial Enzymes to Create a Friedel-Crafts Alkylase. ACS Catal. 11, 6763-6770 (2021).
Upp, D.M. et al. Engineering Dirhodium Artificial Metalloenzymes for Diazo Coupling Cascade Reactions. Angew. Chem. Int. Ed. 60, 23672-23677 (2021).
Gu, Y., Bloomer, B., Liu, Z.N., Clark, D. & Hartwig, J.F. Directed Evolution of Artificial Metalloenzymes in Whole Cells. Angew. Chem. Int. Ed. 61, e202110519 (2022).
Sauer, D.F. et al. A Highly Active Biohybrid Catalyst for Olefin Metathesis in Water: Impact of a Hydrophobic Cavity in a β-Barrel Protein. ACS Catal. 5, 7519-7522 (2015).
Zhao, J.M., Kajetanowicz, A. & Ward, T.R. Carbonic anhydrase II as host protein for the creation of a biocompatible artificial metathesase. Org. Biomol. Chem. 13, 5652-5655 (2015).
Chatterjee, A., Eliasson, S.H.H. & Jensen, V.R. Selective production of linear alpha-olefins via catalytic deoxygenation of fatty acids and derivatives. Catal. Sci. Technol. 8, 1487-1499 (2018).
Rui, Z., Harris, N.C., Zhu, X.J., Huang, W. & Zhang, W.J. Discovery of a Family of Desaturase-Like Enzymes for 1-Alkene Biosynthesis. ACS Catal. 5, 7091-7094 (2015).
Cai, P. et al. Microbial synthesis of long-chain alpha-alkenes from methanol by engineering Pichia pastoris. Bioresour. Bioprocess. 9, 58 (2022).
Zhou, Y.J.J., Hu, Y.T., Zhu, Z.W., Siewers, V. & Nielsen, J. Engineering 1-Alkene Biosynthesis and Secretion by Dynamic Regulation in Yeast. ACS Synth. Biol. 7, 584-590 (2018).
Iqbal, T.M., S; K Rajendran, K; Sidhu, S. J; Das, D Unraveling the Conversion of Fatty Acids into Terminal Alkenes by an Integral Membrane Enzyme, UndB. ACS Catal. 13, 15516-15525 (2023).
Database, A.P.S. (2022).
Bloomer, B.J., Clark, D.S. & Hartwig, J.F. Progress, Challenges, and Opportunities with Artificial Metalloenzymes in Biosynthesis. Biochemistry 62, 221-228 (2022).
Jeschek, M., Gerngross, D. & Panke, S. Rationally reduced libraries for combinatorial pathway optimization minimizing experimental effort. Nat. Commun. 7, 11163 (2016).
Hollerer, S. et al. Large-scale DNA-based phenotypic recording and deep learning enable highly accurate sequence-function mapping. Nat. Commun. 11, 3551 (2020).
Ghosh, S., Ghosh, S. & Sarkar, N. Factors influencing ring closure through olefin metathesis - A perspective. J. Chem. Sci. 118, 223-235 (2006).

There is NO Competing Interest.

RobinSINatureSynthesisDec11.2023.docx
Supplementary Information

Download PDF

Journal Publication

published 27 Jun, 2024

Read the published version in Nature Synthesis →

Version 1

posted

You are reading this latest preprint version

A Microbial Cell Factory Combining an Artificial Metathase with a Fatty Acid Decarboxylase for New-to-Nature Synthesis of Cycloalkenes

Status:

Journal Publication

Version 1

Abstract

Figures

Main

Results and discussion

Optimization of the activity of the whole-cell based artificial metathase

Directed evolution of the artificial metathase for enhanced RCM activity

Directed evolution of the UndB decarboxylase for enhanced production of terminal alkenes

Conclusions

Declarations

Data availability

Code availability

Acknowledgements

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1