When biocatalysis meets flow chemistry

7

Introduction
The fast evolution of enzyme engineering during the last decade has dramatically impacted the number of industrial processes that involve one or more biocatalytic steps for the preparation of fine chemicals including pharmaceuticals, cosmetics, agrochemicals and bioactive compounds (1). Among the most impressive examples in the pharmaceutical sector, the enzymatic total synthesis of Islatravir, a nucleoside reverse transcriptase translocation inhibitor (NRTTI) for HIV infection, has been achieved by Merck in 2019 (2). Nine enzymes, most of which had to be carefully engineered, were required for the accomplishment of such a synthesis, with four of those biocatalysts serving exclusively as ancillary systems for cofactor recycling or equilibrium shifting. Another inspiring example regards the heavily engineered transaminase ATA 117 employed as a key catalyst to replace a rhodium-mediated asymmetric enamine hydrogenation step for the preparation of the blockbuster drug Sitagliptin (3). Additionally, starting from tropical berries a particular protein has been prepared by Amai Proteins company. Due to its sweetening power (3000 times sweeter than sugar), it has the potential to replace the commonly used sweeteners without health hazards and off-flavors, dramatically impacting the food sector (4).

 

What is clear is that protein preparation is no longer a problem, and the concern about biocatalysis as an expensive niche technology, incompatible with the majority of industrial processes has been dismantled. Especially, with the progress of enzyme engineering, which included Prof. Frances Arnold among the Nobel laureates, precise design of enzymes became possible, transforming biocatalysis in a smart strategy to be employed for industrial needs. In a parallel way, a special attention has been paid to highly-performing enzyme-mediated reactions, shifting the technical set-up from batch mode to alternative solutions. Among them, the combination of catalyst stabilization and reutilization via enzyme immobilization and flow chemistry facilities, represents a leap forward in increasing the sustainability, safety and productivity of biotransformations, eventually impacting on process efficiency and relative costs (5-7).

 

Very recently, multi-step enzymatic flow reactions merging the advantages of continuous processing (i.e., better parameter control, high mass and heat transfer, modularity), with the mild reaction conditions and the selectivity of biocatalysts (i.e., water media, room temperature, chemo-, regio-, and stereoselectivity) were even able to efficiently mimic whole cell metabolic pathways (8, 9). In our research group we are keen in developing new approaches to obtain robust and durable biocatalysts to be specifically used for continuous biotransformations, achieving in many cases highly-performing reactions (i.e., high substrate loading, reduced reaction time, increased overall yield, easy scale-up, fully automated processes).

 

Enzyme immobilization technology
While both free and immobilized biocatalysts have been reported for flow facilities, those where the catalyst (enzyme) is immobilized are more commonly used (7,10,11). In these conditions, a heterogeneous system is generated where the enzyme retains sufficient flexibility for the reaction to take place. Various strategies have been described for the immobilization of whole-cell biocatalysts and purified enzymes on different carriers. Both the systems present advantages and disadvantages. As our research group favors working with cell-free enzymes (Figure 1), thus avoiding possible issues of permeability and stability of the cell wall, as well as the competing cellular metabolism which may reduce the overall final yields of the target biotransformation(s), we decided to focus on this topic suggesting the reading of the review by Pinto et al. for the most recent advances on whole cell immobilization (12).

Unlike whole cell systems, pure enzyme-based approaches allow for the use of variable amounts of a catalyst (i.e., catalyst loading) so that in a multi-enzyme cascade, each individual step can be fine-tuned, finally obtaining a fully optimized process (8,13). It has been already demonstrated that enzyme immobilization impacts on the catalyst stability; this is particularly true for immobilization strategies where covalent bonds are involved (14,15).

 

In our group we have developed a variety of different catalyst immobilization protocols involving covalent bonds between the enzyme and the support, to finally obtain robust and durable biocatalysts to be used in packed-bed reactors for continuous biotransformations. Among the most successful strategies the one where the poly-His-tag, which is commonly fused to proteins for ease of purification, selectively interacts with an epoxy-resin following metal derivatization, is noteworthy (16). One of our most representative examples is the immobilization of the transaminase from the haloadapted bacterium Halomonas elongata (HeWT) (17). Figure 2a shows schematically the chemistry involved to covalently bind an enzyme. The convenience of this method stands in the fact that the matrices are commercially available (e.g., different pore size, linkers etc.), while the tunability is outstanding as the metal involved in the coordination with the protein His-tag (Figure 2a) can be chosen, thus avoiding/reducing catalyst toxicity (if any) and its consequent deactivation. Moreover, the enzyme-matrix contact time can be varied before the addition of the capping agent, so that the number of covalent bonds can be reduced or increased with an effect on the final catalyst stability (e.g., rigidification/distortion of the protein quaternary structure). Furthermore, as the presence of the poly-His-tag is routinely used for protein purification through metal affinity chromatography, cell crude extracts can be employed as starting material for the immobilization, merging in “one-shot” protein purification and its immobilization (18). Although a variety of hydrophilic and hydrophobic carriers with different characteristics such as bead diameters, pore size and linkers are available, in our lab we prefer non-swelling matrices since better suited for flow processing, especially when segmented streams are involved (e.g., aqueous/organic solvents or gas/liquid phases). In this way back pressure is minimized and no change in the volume is observed.

However, neither the carrier nor the chemistry is universally ideal, and for every biocatalyst a trial-and-error approach should be employed to identify the best immobilization strategy. For example, the acyl transferase from Mycobacerium smegmatis (MsAcT) which has been immobilized on a range of different cariers, retained its best activity and immobilization yield using the hydrophilic activated glyoxyl agarose beads via imine formation and reduction (Figure 2b) (15).

 

While more challenging, more than one enzyme can be immobilized on the same bead. Co-immobilization of several proteins on the same carrier demonstrated to improve the catalytic performance, minimizing reaction steps, reducing by-product formation, shifting the equilibrium toward the desired product, in situ recycling enzyme cofactors (when necessary), finally increasing bioprocess productivity and cost-efficiency (20,21). Pivotal in this direction has been the work performed by the group of Prof. Fernando Lopez-Gallego, where tailored chemical modification of the matrix surface allowed for sequential immobilization of different enzymes (22). While we have previously used mix-bed approaches, where an enzyme pair was immobilized on separate matrices which were then mixed in the correct proportions within the bioreactor (18), our best example regards the preparation of a multi-active biocatalyst (imm-RN-HOR) employing an α-rhamnosidase (RN) and a β-glycosidase (HOR) which have been simultaneously co-immobilized on glyoxyl-agarose beads. The resulting high-performing biocatalyst has been subsequently utilized for the “one-shot” obtainment of aglycones starting from natural rutinosides (i.e., glycosides containing rhamnose and glucose moieties) (Figure 3) (23).

A significant amount of research goes into optimization of immobilization methodologies, especially in the development of a searchable database to drive the selection of the best immobilization procedure for a specific biocatalyst avoiding the time-, energy- and cost-consuming trial-and-error approach (24). Another key topic regards the “ungreenness” of the procedures currently employed for support preparation/functionalization before enzyme immobilization, which typically rely on the use of toxic or hazardous reagents and drastic conditions (e.g., strong acids, oxidants, low or high temperature etc.). Our research group is involved in the design of biocatalyzed strategies for the carrier functionalization as well as the utilization of natural matrices (better if recovered from waste, residues or by-products) as support for enzyme immobilization. On one hand we would like to enhance the sustainability of these processes, on the other hand we aim at increasing the use of immobilized enzymes in those fields where the consumer preference for compounds derived from natural sources is rapidly expanding (e.g., food, cosmetic fields).

 

Biocatalysis in Continuous Flow
As mentioned above, one of the main advantages of immobilized biocatalysts is their employment under flow conditions. The compartmentalization of the immobilized enzymes into packed-bed reactors (PBRs) let a high amount of catalyst to be accumulated in a small space where the substrate flows in a controlled manner. Additionally, mass and heat transfer have been shown to be more efficient when compared to batch conditions, allowing for faster reactions which typically outperform one-pot set-ups (25). Moving away from analytical scale, which is the main limitation of enzyme-mediated processes, we developed a continuous multi-gram production of melatonin (to date exclusively prepared synthetically) via direct acetylation of the commercially available 5-metoxy-tryptamine mediated by MsAcT (15).

 

The small reactor filled with less than 2 milligrams of the immobilized enzyme could handle substrate at 0.5 M concentration (95 g/L) with a productivity of the desired compound of 36 g/day (5 min of residence time). The process was further integrated with an in-line extraction for the recovery and reuse of both the organic solvent and aqueous phase (Figure 4). The same system has been utilized for the scaled-up preparation of natural esters as flavor compounds with excellent yields and residence times, highlighting the versatility and sustainability of the developed methodology (26). Different PBRs can be also sequentially connected to perform multi-step biotransformations, thus increasing the complexity of the cascade (18). The beauty of this integrated technology is given by its flexibility and tunability: by changing the cartridges or the way to combine them, different products can be synthesized starting from the same substrates. In some cases, artificial metabolisms simulating the cell pathways have been fully developed, thus demonstrating the use of flow biocatalysis as an artificial cell-factory (8).

Another important topic regards the cofactor necessity, which several enzymes have (e.g., redox systems transaminases, transferases etc.), thus adding a further complication to the flow reaction design as cofactors are typically expensive molecules, while recycling systems (e.g., via addition of co-substrates or ancillary enzymes) need to be carefully optimized. Moreover, also when everything works perfectly and just a catalytic cofactor concentration is added to the flow biotransformation, they are lost in the downstream waste. Following the work of the group of Prof. Lopez-Gallego (27), we applied the same co-immobilization strategy to an ω-transaminase (HeWT) and its cofactor (i.e., pyridoxal phosphate –PLP–). which has led to the assembly of fully self-sufficient systems which performed very well in continuous flow (28).
More in detail, whereas HeWT was covalently immobilized to a previously functionalized epoxy-resin (Fig. 2a), PLP was bound through ionic bridges which made this molecule free to shuttle between the enzyme active site without leaving the pore microenvironment of the carrier (Fig.5a)(29). This strategy was subsequently applied to the flow synthesis of biogenic aldehydes from the corresponding amines obtaining high yields (90-99%) and fast residence times (15 min)(29).

As our knowledge in the field progressed, we increased the complexity of the cascade, adding a second module in the assembly (Fig. 5b). Here, the amine-to-carbonyl catalysis was followed by a second bioreactor to perform the reduction to alcohol. This work integrating in situ cofactor regeneration with in-line work-up procedures, allowed for the collection of the waste water containing the cofactors and their recirculation in the flow system for other reaction cycles, giving rise to closed-loop, ultra-efficient processes (Fig. 5b)(13).

 

Outlooks
The progress in our understanding of enzymes and their potential as catalysts together with the worldwide goal to increase sustainability, reducing waste and emissions has defined a novel status quo where biocatalysis represents a strong ally for scientists in the field of chemistry and engineering, with a variety of alternatives now available to be used also at industrial level. In this context, flow biocatalysis is considered an attractive technology for both academics and industrials due to the reduced costs of chemical transformations, the small equipment footprint, the highest reaction selectivity and mild operational conditions. Several projects are underway in our laboratory where different enzymes are combined to prepare the desired molecules on large scale. But our interest is not limited to pure biocatalytic approaches, in fact working towards the development of multi-step syntheses integrating chemical procedures and enzyme-mediated reactions, (i.e., chemo-enzymatic reactions) as well as novel green techniques (e.g., photochemistry) will only strengthen and accelerate the uptake of biocatalysis in industry. In particular, photo-biocatalysis represents a novel chapter and just few examples have been described in the literature so far. These reports mainly regard light-driven enzymes, light-activated cofactor regeneration and the utilization of light-dependent organisms. In this context, the photochemical recycling of cofactors plays the major role, as it represents an effective approach to combine photo- and biocatalysis. However, up to now light-driven cofactor regeneration has no become a standard procedure yet, mainly due to the poor TTN (i.e., total turnover number) of the photo- or biocatalyst. Although this limitation can be overcome, long procedures are needed for the preparation of novel photosensitizers with a more efficient electron transfer or the modification of enzymes via protein engineering to simplify the electron acceptance (30). As an alternative to these artificial systems a new branch of photo-biocatalysis has emerged based on photo-autotrophic organisms (i.e., cyanobacteria or algae). While the photosystem converts energy deriving from light into redox equivalents, the cell presents highly specialized electron transport chains and mechanisms for the control of reactive species (31).

A challenge in the field of photo-biocatalysis regards its employment on a preparative scale. In fact most of the discussed reactions are proposed just as a proof-of-concept in an analytical scale. A big step would be the demonstration of the feasibility of these combined techniques when applied to larger volumes and concentrations. However, in contrast with classical biotransformations which typically allow for straightforward upscaling, parameters such as light intensity or light penetration depth cannot be easily scaled (32,33). An option to address this issue may be the availability of novel photoreactors for both batch and continuous processing together with the possibility to screen several reactions conditions at time (32,34). Taking all these considerations into account, to transform photo-biocatalysis in a robust technology widely employable it is crucial to develop efficient photoreactors for large scale production as well as standardize as much as possible the reactions at small scale. Basic concepts have been up to now set up but they need to be extended in a way that novel reactions and concepts could come up.

 

Figure 1. Most used cell free enzyme immobilization techniques, adapted from Pinto et al (12).

 

Figure 2. Covalent immobilization strategies A. Chemistry on epoxy-matrix. IDA = aminodiacetic acid; M2+= bivalent metal; EDTA = Ethylenediaminetetraacetic acid. B. General scheme for the immobilization on aldehydic supports. Figure adapted from Donzella et al (19).

 

Figure 3. Preparation of Hesperetin and Quercetin from the corresponding rutinosides.

 

Figure 4. MsAcT-mediated biotransformations under flow condition for process implementation.

 

Figure 5. Strategies for the reuse of cofactors under flow conditions. A. Enzyme/cofactor co-immobilization. B. Waste water recirculation.

 

References and notes

1. Wu S, Snajdrova R, Moore JC, et al (2021) Biocatalysis: Enzymatic Synthesis for Industrial Applications. Angewandte Chemie – International Edition 60:88–119. https://doi.org/10.1002/anie.202006648
2. Huffman MA, Fryszkowska A, Alvizo O, et al (2020) Design of an in vitro biocatalytic cascade for the manufacture of islatravir. Science (1979) 368:1255–1259. https://doi.org/10.1126/SCIENCE.ABC1954
3. Savile CK, Janey JM, Mundorff EC, et al (2010) Biocatalytic Asymmetric Synthesis of Sitagliptin Manufacture. Science (1979) 329:305–310. https://doi.org/10.1126/science.1188934
4. Melton L (2023) So sweet, no spike. Nat Biotechnol 41:883–885. https://doi.org/10.1038/s41587-023-01860-2
5. Benítez-Mateos AI, Contente ML, Roura Padrosa D, Paradisi F (2021) Flow biocatalysis 101: Design, development and applications. React Chem Eng 6:599–611. https://doi.org/10.1039/d0re00483a
6. Leemans Martin L, Peschke T, Venturoni F, Mostarda S (2020) Pharmaceutical industry perspectives on flow chemocatalysis and biocatalysis. Curr Opin Green Sustain Chem 25:100350. https://doi.org/10.1016/j.cogsc.2020.04.011
7. Basso A, Serban S (2019) Industrial applications of immobilized enzymes—A review. Molecular Catalysis 479:110607. https://doi.org/10.1016/j.mcat.2019.110607
8. Donzella S, Colacicco A, Nespoli L, Contente ML (2022) Mimicking Natural Metabolisms: Cell-Free Flow Preparation of Dopamine. ChemBioChem 23:. https://doi.org/10.1002/cbic.202200462
9. Crotti M, Robescu MS, Bolivar JM, et al (2023) What ’s new in flow biocatalysis ? A snapshot of 2020 – 2022. 1–10. https://doi.org/10.3389/fctls.2023.1154452
10. Bolivar JM, López-Gallego F (2020) Characterization and evaluation of immobilized enzymes for applications in flow reactors. Curr Opin Green Sustain Chem 25:100349. https://doi.org/10.1016/j.cogsc.2020.04.010
11. Romero-Fernández M, Paradisi F (2020) Protein immobilization technology for flow biocatalysis. Curr Opin Chem Biol 55:1–8. https://doi.org/10.1016/j.cbpa.2019.11.008
12. Pinto A, Contente ML, Tamborini L (2020) Advances on whole-cell biocatalysis in flow. Curr Opin Green Sustain Chem 25:100343. https://doi.org/10.1016/j.cogsc.2020.04.004
13. Contente ML, Paradisi F (2018) Self-sustaining closed-loop multienzyme- mediated conversion of amines into alcohols in continuous reactions. Nat Catal 1:452–459. https://doi.org/10.1038/s41929-018-0082-9
14. Romero-Fernandez M, Paradisi F (2020) General Overview on Immobilization Techniques of Enzymes for Biocatalysis. In: Sons JW& (ed) Catalyst immobilization Eds. M Benagilia, A. Puglisi,. pp 409–435
15. Contente ML, Farris S, Tamborini L, et al (2019) Flow-based enzymatic synthesis of melatonin and other high value tryptamine derivatives: A five-minute intensified process. Green Chemistry 21:3263–3266. https://doi.org/10.1039/c9gc01374a
16. Mateo C, Fernandez-Lorente G, Cortes E, et al (2001) One-Step Purification, Covalent Immobilization, and Additional Stabilization of Poly-His-Tagged Proteins Using Novel Heterofunctional Chelate-Epoxy Supports. Biotechnol Bioeng 73:253–258. https://doi.org/10.1002/bit.1058
17. Planchestainer M, Contente ML, Cassidy J, et al (2017) Continuous fl ow biocatalysis : production and in-line puri fi cation of amines by immobilised transaminase from Halomonas elongata †. Green Chemistry 19:372–375. https://doi.org/10.1039/c6gc01780k
18. Contente ML, Paradisi F (2018) Self-sustaining closed-loop multienzyme- mediated conversion of amines into alcohols in continuous reactions. Nat Catal 1:452–459. https://doi.org/10.1038/s41929-018-0082-9
19. Donzella D, Contente ML (2024). The joint effort of enzyme technology and flow chemistry to bring biocatalytic processes to the next level of sustainability, efficiency and productivity J Flow Chem 14:85-96. https://doi.org/10.1007/s41981-023-00286-w
20. Wheeldon I, Minteer SD, Banta S, et al (2016) Substrate channelling as an approach to cascade reactions. Nat Chem 8:299–309. https://doi.org/10.1038/nchem.2459
21. Schoffelen S, Van Hest JCM (2013) Chemical approaches for the construction of multi-enzyme reaction systems. Curr Opin Struct Biol 23:613–621. https://doi.org/10.1016/j.sbi.2013.06.010
22. Santiago-Arcos J,, Velasco-Lozano S., López-Gallego F. (2023) Multienzyme Coimmobilization on Triheterofunctional Supports, Biomacromolecules 24: 929 – 94213. https://doi.org/10.1021/acs.biomac.2c01364
23. Colacicco A, Catinella G, Pinna C, et al (2023) Flow bioprocessing of citrus glycosides for high-value aglycone preparation. Catal Sci Technol 4348–4352. https://doi.org/10.1039/d3cy00603d
24. Roura Padrosa D, Marchini V, Paradisi F (2021) CapiPy: Python-based GUI-application to assist in protein immobilization. Bioinformatics 37:2761–2762. https://doi.org/10.1093/bioinformatics/btab030
25. Britton J, Raston CL (2017) Multi-step continuous-flow synthesis. Chem Soc Rev 46:1250–1271. https://doi.org/10.1039/c6cs00830e
26. Contente ML, Tamborini L, Molinari F, Paradisi F (2020) Aromas flow: eco-friendly, continuous, and scalable preparation of flavour esters. J Flow Chem 10:235–240. https://doi.org/10.1007/s41981-019-00063-8
27. Velasco-Lozano S, Benítez-Mateos AI, López-Gallego F (2017) Co-immobilized Phosphorylated Cofactors and Enzymes as Self-Sufficient Heterogeneous Biocatalysts for Chemical Processes. Angew Chem 56:771-775. https://doi.org/10.1002/anie.201609758
28. Benítez-Mateos I., Contente M. L., Velasco-Lozano S., Paradisi F., Lopez-Gallego F. Self-sufficient flow-biocatalysis by co-immobilisation of pyridoxal 5´-phosphate and ω-transaminases onto porous carriers. ACS Sustain Chem Eng (2018), 6:13151-13159. https://doi.org/10.1021/acssuschemeng.8b02672
29. Contente M. L., Paradisi F. Transaminase‐catalyzed continuous synthesis of biogenic aldehydes. ChemBioChem (2019), 20, 2830-2833. https://doi.org/10.1002/cbic.201900356
30. Schmermund L, Jurkaš V, Özgen FF, et al (2019) Photo-Biocatalysis: Biotransformations in the Presence of Light. ACS Catal 9:4115–4144. https://doi.org/10.1021/acscatal.9b00656
31. Köninger K, Gómez Baraibar Á, Mügge C, et al (2016) Recombinant Cyanobacteria for the Asymmetric Reduction of C=C Bonds Fueled by the Biocatalytic Oxidation of Water. Angewandte Chemie – International Edition 55:5582–5585. https://doi.org/10.1002/anie.201601200
32. Heining M, Buchholz R (2015) Photobioreactors with internal illumination – A survey and comparison. Biotechnol J 10:1131–1137. https://doi.org/10.1002/biot.201400572
33. Glemser M, Heining M, Schmidt J, et al (2016) Application of light-emitting diodes (LEDs) in cultivation of phototrophic microalgae: current state and perspectives. Appl Microbiol Biotechnol 100:1077–1088. https://doi.org/10.1007/s00253-015-7144-6
34. Slattery A, Wen Z, Tenblad P, Pintossi D, Sanjose-Orduna J, den Hartog T, et al. An all-in-one multipurpose robotic platform for the self-optimization, intensification and scale-up of photocatalysis in flow. ChemRxiv. 2023; doi:10.26434/chemrxiv-2023-r0drq This content is a preprint and has not been peer-reviewed.