Stereoselective Bio‐Organocatalytic Cascade to Chiral Amides as Active Pharmaceutical Ingredient Intermediates Using ω‐Transaminase and Choline Chloride Under Microwave Irradiation

TL;DR

This paper presents a green and efficient method to create chiral amides, important for pharmaceuticals, using enzymes and microwave conditions.

Contribution

A solvent-free, microwave-assisted bio-organocatalytic cascade for chiral amide synthesis using ω-transaminases and choline chloride.

Findings

The cascade achieved full conversions and >99% enantiomeric excess for key pharmaceutical intermediates.

Amidation using choline chloride yielded amides in 60%-86% without traditional activating agents.

Solvent removal after transamination improved amidation efficiency by eliminating byproducts.

Abstract

Amide bond formation is a key transformation in organic synthesis, especially for the preparation of active pharmaceutical ingredients (APIs). In this work, we report the development of a bio‐organocatalytic cascade, combining stereoselective transamination catalyzed by ω‐transaminases (ω‐TAs) in neat organic solvent and choline chloride (ChCl)‐mediated direct amidation. This strategy enables the synthesis of chiral amides from prochiral carbonyl compounds and carboxylic acids under solvent‐free microwave conditions. After optimizing the biocatalytic transamination in MTBE, we applied the method to the synthesis of key intermediates of Racecadotril and AVR‐48, achieving full conversions and enantiomeric excess above 99%. The amidation step, promoted by ChCl without traditional activating agents, proved highly efficient for a wide range of aliphatic and aromatic carboxylic acids, affording the target amides in 60%–86% yields. Solvent evaporation after the transamination step was essential to remove interfering byproducts such as acetone, thus improving amidation yields. Overall, this integrated methodology provides a green, efficient, and scalable route to access amide‐based building blocks in high optical purity, opening new avenues for sustainable pharmaceutical manufacturing.

Amide bond formation is central to active pharmaceutical ingredient (API) synthesis. A bio‐organocatalytic cascade is reported, merging stereoselective ω‐transaminase (ω‐TA) transamination in neat organic solvent with choline chloride‐mediated amidation under solvent‐free microwave conditions. The method delivers chiral amides in high yields and >99% ee, enabling efficient, green access to pharmaceutical intermediates such as those of Racecadotril and AVR‐48.© 2026 WILEY‐VCH GmbH

Figures

Chemical structures of racecadotril and AVR‐48.

Retrosynthetic analysis for racecadotril (A) and AVR‐48 (B).

Comparison of transamination conversions by substrate, solvent and enzyme. All data are reported as the mean for three single replicates. Reaction conditions: ω‐TA (20 mg CFE at a w = 0.6), organic solvent (1 mL, saturated with 0.8% in volume of H2O), carbonyl compound as substrate (50 mM), 2‐propylamine (150 mM), 30°C. See experimental part for details.

General scheme for transamination reaction.

General approach for the bio‐organocatalytic amidation reaction.

Bio‐organocatalytic amidation of compounds 3 and 4. Enantiomeric excess determined as reported in Supporting Information pg. S aGeneral reaction conditions in MTBE with evaporation of solvent: MW 80°C, six cycles of 3 min each. bReaction conditions in MTBE without evaporation of solvent: MW 80°C, six cycles of 3 min each.

Full text

Introduction

1

Amide bond formation is a fundamental transformation in chemistry, particularly for the synthesis of active pharmaceutical ingredients (APIs); for instance, about 122 of the Top 200 Small Molecules Drugs by Retail Sales in 2024 contain at least an amide functional group [1]. However, conventional methods for amide synthesis often require stoichiometric amounts of activating reagents, harsh reaction conditions, and toxic solvents, leading to significant environmental concerns [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14]. As a result, the pursuit of sustainable and environmentally friendly amidation strategies remains a prominent and ongoing challenge in organic chemistry.

Additionally, many bioactive molecules and APIs feature amide groups alongside stereogenic centers, which critically influence their biological activity. As the FDA increasingly prioritizes the approval of single‐enantiomer drugs over racemic mixtures, the demand for stereoselective synthesis strategies has intensified. In this context, biocatalysis has emerged as a promising approach [15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25]. Among biocatalysts, ω‐transaminases (ω‐TAs) have garnered significant attention due to their ability to enable the asymmetric synthesis of amines from prochiral ketones, as well as the kinetic resolution of racemic amines [26, 27, 28, 29, 30, 31, 32, 33, 34]. Notably, these enzymes can retain catalytic activity in neat organic solvents at controlled water activity, albeit with reduced reaction rates compared to aqueous buffer systems [35, 36, 37]. The use of inexpensive amine donors, such as 2‐propylamine, in neat organic solvents or in an aqueous buffer in the presence of a high volume of an organic cosolvent further underscores the industrial relevance of ω‐TAs in organic synthesis [35, 36, 38, 39, 40, 41, 42, 43, 44]. Organocatalysis incorporating synergistic, multicomponent, or cascade reactions along with emerging activation strategies such as electrocatalysis and photocatalysis also represents a compelling alternative [45, 46]. Despite these developments, bio‐organocatalytic strategies for direct amide bond formation from carbonyl compounds remain largely underexplored, highlighting an urgent need for further innovation.

In a recent study, we developed an amide bond formation approach using choline chloride (ChCl) as a reaction mediator, eliminating the need for traditional activating reagents and hazardous solvents [47]. ChCl, in combination with hydrogen bond donor (HBD) molecules, forms complex hydrogen bond interactions, which facilitate direct amidation of amines and carboxylic acids under mild conditions, providing a green alternative for pharmaceutical synthesis [48].

In recent years, racecadotril has emerged as an important therapeutic agent for the treatment of acute diarrhea, owing to its ability to selectively inhibit the enzyme enkephalinase. Its clinical relevance is supported by its pharmacological profile as well as its distinct chemical structure, which features a methoxy‐substituted aromatic ring, a thioether linkage, and a defined amide moiety [49]. A structurally related compound, AVR‐48, currently under clinical investigation for the treatment of respiratory conditions such as bronchopulmonary dysplasia, shares a similar molecular scaffold but introduces stereoselectivity as a key design element [50].

Although racecadotril and AVR‐48 differ in therapeutic indication, both compounds represent valuable molecular targets due to their shared amide functionality and stereochemically rich frameworks (Figure 1). As an example, racecadotril is traditionally synthesized through multistep procedures that rely on the direct thioacetylation of benzylacrylic acid, followed by amidation with benzyl glycinate, as described in patent WO2011116490A1 [51]. A graphical overview of the industrial method, including its advantages, limitations, and environmental impact, is provided in Scheme S1. While this industrially adopted strategy provides high yields (up to 80%–85%), it requires hazardous solvents such as dichloromethane and coupling agents such as DCC and HOBt, which generate significant waste streams and raise safety concerns. Alternative synthetic routes have also been reported, yet these approaches generally afford low yields and involve the use of environmentally unfriendly reagents [52].

Chemical structures of racecadotril and AVR‐48.

From a synthetic perspective, the core structures of racecadotril and AVR‐48 could be efficiently accessed through a biocatalytic approach involving enzymatic transamination, followed by a chemical or enzymatic amidation step, offering a greener alternative that avoids stoichiometric activating agents and hazardous solvents.

In this work, we explore the potential of integrating biocatalysis and organocatalysis to establish a powerful strategy for the stereoselective synthesis of amide‐based building blocks of racecadotril and its analog AVR‐48. By leveraging the synergistic potential of ω‐TAs and ChCl‐mediated activation, this approach aims to achieve high yields and enantiomeric purity while adhering to the principles of green chemistry. This study seeks to expand the synthetic toolbox for sustainable and selective amide bond formation, paving the way for more environmentally benign methodologies in pharmaceutical development.

Materials and Methods

2

General

2.1

All chemicals (carbonyl compounds: benzaldehyde, acetophenone, benzyloxyacetaldehyde, 1‐(benzyloxy)propan‐2‐one, 1‐methoxypropan‐2‐one, and 1‐(benzyloxy)propan‐2‐one (1–4, 11, and 12); carboxylic acids: butyric acid, pentanoic acid, pimelic acid, 4‐bromo benzoic acid, phenylacetic acid, hydrocinnamic acid, malonic acid, and succinic acid (a–h)) were obtained from Merck and used as received.

Glycerol stock of E. coli BL21(DE3) cells harboring the plasmid for expression of the ω‐TAs have been used in this study. These ω‐TAs were overexpressed in E. coli BL21(DE3)/pET21 cells using the OnePot automated system (Katakem, Catanzaro, Italy; for details see Supporting Information pg. S3). Electrophoresis gel imaging using Gel Doc EZ from Biorad allows high‐quality analysis of enzyme expression during bacterial growth.

The MW‐assisted reactions were performed in the high‐energy‐density laboratory CEM Discover MW oven equipped with an external IR sensor for temperature control; reactions were run in dynamic mode working with borosilicate glass vessels equipped with a silicon cap and septum.

Biocatalytic transaminations were performed in an Eppendorf Thermomixer compact 5350. The protein expression analyses were performed using a Gel Doc EZ from Biorad. The optical density (OD) calculation was performed by a WPA Bioware.

GC–MS analyses were performed by a Focus GC (Thermo Scientific) equipped with a Varian VF‐5 m capillary column (30 m × 0.25 mm × 0.25 μm) and coupled to a single quadrupole mass spectrometer (DSQII Thermo Scientific).

Conversions of amines were determined by GC using a 7890A GC system (Agilent Technologies), equipped with a FID detector, using H_2_ as carrier gas, and a DB1701 column from Agilent (30 m, 250 μm, 0.25 μm). The enantiomeric excess of the chiral compounds was measured using a ChiraSil DEX‐CB column from Agilent (25 m, 320 μm, 0.25 μm) after derivatization (see Supporting Information pg. S4).

Proton nuclear magnetic resonance (^1^H NMR) spectra and carbon nuclear magnetic resonance (^13^C NMR) spectra were recorded on a Bruker AV400.

Cell‐Free Extract Enzyme Preparation

2.2

The detailed procedure for preparation of cell‐free extract (CFE) enzyme is described in the Supporting information (see Supporting information pg. S3).

Amination in MTBE with 2‐Propylamine on Analytical Scale

2.3

The lyophilized CFE containing overexpressed ω‐TA (20 mg) was placed in an Eppendorf tube and methyl tert‐butyl ether (MTBE) was added (1 mL, saturated with 0.8% in volume of distilled water). The sample was shaken for about 30 min, at 25°C and 750 rpm in an Eppendorf thermomixer. Then, an additional amount of water (16 μL to reach finally approximately a_w_ of 0.6) was added and the sample was shaken for another 30 min to equilibrate the system. Carbonyl substrate (50 mM from a MTBE stock solution) and 2‐propylamine (150 mM) were added and the sample was shaken in an Eppendorf thermomixer at 30°C and 750 rpm. The enzyme was removed by filtration with a 0.2 μm filter and the conversions were determined by GC‐FID with the method reported in the Supporting Information (Supporting information pg. S4). The procedure for the determination of enantiomeric selectivity is described in the Supporting Information (Supporting information pg. S4).

General Procedure for MW‐Assisted Direct Amidation

2.4

In a typical reaction, the reagents were added into vessels equipped with a silicon cap and septum with a ratio of 1.0:1.0:1.0 eq. of ChCl (2 mmol), carboxylic acid (2 mmol), and amine (2 mmol). The mixture was heated in the MW reactor (CEM Discover Sp) at 80°C for six reaction cycles lasting 3 min in “dynamic mode” (P 80 W), and the temperature was read with an IR probe. Solid reagents were heated in an oil bath, up to 80°C, for 2 min before inserting them into the MW oven. The reaction was monitored by GC–MS. After phase separation, the organic phase was taken, 10 mL of a saturated solution of CuSO_4_ (×3) was added, and the organic phases collected were dried over Na_2_SO_4_, followed by evaporation under reduced pressure to give the corresponding products 9a–f, (R)‐10a–f, (S)‐10a–f. Products 9a–f were obtained in pure form after the workup, while compound (R)‐10a–f and (S)‐10a–f were purified by column chromatography on silica (DCM:MeOH 9:1). Yields were determined as isolated products. All unknown products were characterized with GC–MS and ^1^H and ^13^C NMR (See Supporting Information pg. S10–S27).

Results and Discussion

3

As illustrated in Scheme 1, a retrosynthetic analysis suggests that these molecules could arise from analogous carbonyl precursors—such as ketones or aldehydes—that are amenable to stereoselective transamination. This strategy not only provides access to enantiomerically enriched intermediates but also enables a greener and potentially more sustainable route to pharmaceutically relevant amide‐containing compounds.

Retrosynthetic analysis for racecadotril (A) and AVR‐48 (B).

Optimization of the Biotransamination Conditions in Organic Solvent

3.1

In this study, we selected two specific transaminases, Cv‐ωTA and AsR‐ωTA, due to their demonstrated high enantioselectivity for the production of S‐ and R‐configured amines from ketones, as reported in the literature [35, 53, 54, 55, 56]. At the beginning, we focused on optimizing the reaction conditions in organic solvent to allow for compatibility between the biocatalytic and the organocatalytic steps. For this initial screening, we employed the benchmark substrate acetophenone (2), as reported in a previous work [35], and the structurally analogous benzaldehyde (1) (Scheme 2). Although these model substrates had limited relevance to our final synthetic goals, they allowed us to establish the optimal reaction parameters. With a suitable protocol in hand, we turned our attention to the synthesis of two pharmacologically relevant API intermediates as targets (compounds 7, (R)‐8 and (S)‐8, see Scheme 2). Remarkably, these compounds were obtained in full conversion due to the favorable thermodynamic equilibrium, confirming the applicability and robustness of the developed method for the synthesis of the desired amine intermediates (Figure 2).

Comparison of transamination conversions by substrate, solvent and enzyme. All data are reported as the mean for three single replicates. Reaction conditions: ω‐TA (20 mg CFE at a w = 0.6), organic solvent (1 mL, saturated with 0.8% in volume of H2O), carbonyl compound as substrate (50 mM), 2‐propylamine (150 mM), 30°C. See experimental part for details.

General scheme for transamination reaction.

Various solvents, including MTBE, toluene, and hexane, were tested to optimize the reaction conditions for the transamination steps (Figure 2). Our data show that the solvent effect is highly substrate‐dependent, specifically, for the benchmark substrates benzaldehyde (1) and acetophenone (2). For those substrates, the best conversions were observed in toluene and hexane, respectively, indicating that these solvents provide more favorable environments for the conversion of smaller carbonyl substrates. Figure 2 also shows the conversion results of the reactions in MTBE with substrates 1 and 2. As observed, both enzymes provided good conversions, in agreement with the influence of the solvent and reagents' nature.

Reactions performed in hexane and toluene exhibit higher conversion with substrates 1 and 2, but as we can see from the values, they showed more variability than the reaction performed in MTBE, especially with substrates 3 and 4.

However, when the reaction was extended to the bulkier substrates 3 and 4, a significant drop in conversion was observed in both hexane and toluene, while reactions performed in MTBE consistently achieved conversions above 99% with enantiomeric excess (ee) between 97 and 99%. This behavior can be explained by the impact of solvents with different hydrophobicities (log p values) on enzyme flexibility and substrate accessibility [57, 58, 59]. It is known that in more hydrophobic solvents, such as hexane and toluene, enzymes may adopt more rigid conformations, which can hinder the access of larger substrates to the active site, thus explaining the poor conversion observed with 3 and 4 in these solvents. MTBE, on the other hand, appears to balance retention of enzyme activity and substrate solubility, while enabling sufficient enzyme flexibility for the access of larger substrate to the active site (Figure 2 and Table S1, Supporting Information pg. S5) [57, 58, 59]. Moreover, among ethers, MTBE is one of the best alternatives in terms of health and environmental score according to the CHEM21 solvent selection guide, and it is considered “usable” by the solvent selection guide of Pfizer, GSK and Sanofi [60]. The main concern is about safety, due to its volatility. Nevertheless, safer alternatives, such as CPME, were not suitable for the reaction because, as will be discussed in the next section, the organic solvent must be removed before amidation without losing volatile amines.

Optimization of the Bio‐Organocatalytic Amidation Reaction of Carbonyl Compounds

3.2

Once the reaction conditions for the biocatalytic transamination were established, the focus shifted to the bio‐organocatalytic cascade reaction (Scheme 3). As shown in Scheme 3, the objective is to synthesize the amide starting from the amine intermediate obtained after the first transamination step on the carbonyl compound, by simply adding ChCl and the carboxylic acid at the end of the transamination to initiate the second step. This method effectively bridges biocatalytic transamination of carbonyl compounds with organocatalytic amidation between carboxylic acids and amines.

General approach for the bio‐organocatalytic amidation reaction.

Therefore, we initiated a series of preliminary tests, including cascade reactions with and without solvent evaporation (product 9a, (S)‐10a and (R)‐10a, see Scheme 4). Specifically, we performed solvent evaporation using a dinitrogen stream after the transamination. Scheme 4 shows the results of the cascade reaction: starting from compound 3, amides 9a–f were produced, while starting from compound 4 led to the formation of chiral amides (R)‐10a–f and (S)‐10a–f.

Bio‐organocatalytic amidation of compounds 3 and 4. Enantiomeric excess determined as reported in Supporting Information pg. S aGeneral reaction conditions in MTBE with evaporation of solvent: MW 80°C, six cycles of 3 min each. bReaction conditions in MTBE without evaporation of solvent: MW 80°C, six cycles of 3 min each.

The results from the tests conducted without solvent evaporation revealed incompatibility under such condition, with the production of amides falling below 2% (procedure [b]) confirming the need for solvent evaporation.

As shown in Scheme 3, during the conversion of the carbonyl compound to the amine, one equivalent of 2‐propylamine is transformed into acetone. We speculate that this byproduct could interfere with the binding of ChCl, potentially reducing or even preventing its coordination during the amidation reaction under microwave conditions. By evaporating the solvent using a dinitrogen stream, it is possible to remove this byproduct as well, thereby greatly enhancing the yield of the ChCl‐organocatalyzed step. We noticed that the organocatalytic amidation of intermediate aldehyde 7 yielded the final products 9a–f with excellent purity (>99%) without the need for further work‐up, whereas the same reaction on ketone intermediate 8 resulted in detectable traces of byproducts, possibly due to the lability of the phenoxy moiety. The products (R)‐10a–f and (S)‐10a–f required purification by column chromatography, and in a few cases, the removal of byproducts proved difficult.

In all cases, yields ranged from 60% to 86% across the various substrates tested. Aliphatic acids afforded good yields with both amines (entries 9a, 9b, 9c, (R)‐10a, (R)‐10b, (S)‐10a, (S)‐10b, Scheme 4). As expected, the reactions with aromatic acids also proceeded with good yields (9d, 9e, 9f, (R)‐10d, (R)‐10e, (R)‐10f, (S)‐10d, (S)‐10e, (S)‐10f, Scheme 4). Moreover, the absolute configuration of the stereogenic center of the amine 8 (R or S) did not influence the outcome of the subsequent organocatalytic reactions. The only exception was observed in the reaction between substrate 8 (R or S) and acid c, where no conversion was obtained. In these cases, traces (<2%) of undesired side‐products were formed.

It is known from literature that pimelic acid (c) can form crystalline or semi‐crystalline aggregates with molecules containing amino or amide functionalities [61, 62], where the stability of these structures depends on the nature of the interaction partner and the presence of specific templates [63]. It is therefore conceivable that the amines derived from ketones, such as the 1‐(benzyloxy)propan‐2‐amine (8), could participate in the formation of such aggregates, thereby hindering the progress of the reaction—unlike amines derived from aldehydes such as 7, which may not exhibit this behavior. Indeed, the formation of a stable supramolecular complex, where ChCl enables the activation of the carboxylic acid and stabilization of the ammonium intermediate through hydrogen bonding, remains critical for the success of the amidation reaction, particularly under solvent‐free microwave conditions [47].

To better investigate this aspect, we evaluated the reactivity of pimelic (c), malonic (g), and succinic (h) acids with amines structurally analogous to (R) and (S)‐8, namely (R) and (S)‐1‐methoxypropan‐2‐amine ((R) and (S)‐13), and (R) and (S)‐1‐(benzyloxy)propan‐2‐amine ((R) and (S)‐14), as reported in Scheme S2. As expected, no conversion was observed for the reactions involving pimelic (c), as observed in the case of using amines (R) and (S)‐8 (see Table S2). In contrast, reactions with succinic acid (h) afforded measurable amounts of the desired products (R) and (S)‐15, along with unidentified side‐products. Similarly, reactions with malonic acid (g) yielded measurable amounts of the desired products (R) and (S)‐16, again accompanied by unidentified side‐products. The corresponding analytical data and GC–MS chromatograms are provided in the Supporting Information (pg. S6‐S9).

Concerning the stereoselectivity, the carbonyl substrates used in the cascade reaction were proven to be converted into enantiomerically pure amines, essential intermediates for APIs [64].

Conclusions

4

In this study, we successfully developed and optimized a bio‐organocatalytic cascade for chiral amide synthesis using prochiral carbonyl compounds and carboxylic acids as reagents. We also demonstrated its significant potential for applications in drug synthesis by synthesizing structural analogs of racecadotril and AVR‐48 building blocks. By establishing optimal conditions for the two steps of the cascade, we achieved high yields and excellent enantiomeric purity. A key finding was the essential role of solvent evaporation during the cascade reaction, after the transamination, in enhancing the reaction yields to the final amide product. By exploring a diverse set of acids, we expanded the reaction scope. Notably, microwave irradiation at 80°C for 18 min provided the most efficient conditions, yielding high‐purity amides with superior yields. In summary, this work contributes significantly to synthetic chemistry by offering robust and sustainable methods for amide synthesis. The techniques developed hold promise for advancing pharmaceutical research and production, facilitating the transition from laboratory‐scale experiments to industrial‐scale applications while respecting sustainability requirements. This work not only enhances the toolkit of synthetic chemists but also paves the way for future innovations in the green synthesis of chiral amides.

Looking ahead, several challenges must be addressed to further improve the environmental profile of the cascade. Increasing substrate concentration and achieving full conversion will be essential for process intensification and may be enabled by ω‐TA immobilization, enzyme stabilization strategies, or variants engineered for improved solvent tolerance. The sustainability of the organocatalytic step could be further enhanced by minimizing byproduct formation under microwave conditions and by exploring alternative hydrogen‐bond donors or ChCl analogs derived from renewable feedstocks. In addition, both MTBE and ChCl are amenable to closed‐loop recycling, a key requirement for future large‐scale applications that will be systematically evaluated in follow‐up studies. Overall, these improvements are critical to unlocking the full potential of the cascade for sustainable pharmaceutical manufacturing.

Supporting Information

Additional supporting information can be found online in the Supporting Information section. The Supporting Information contains general procedures for enzyme expression, lyophilized cell‐free extract preparation, analytical methods, and characterization of products by GC and NMR. Supporting Scheme S1: Comparison of the industrial method used for the synthesis of racecadotril and the biocatalytic method developed in this work. Supporting Scheme S2: General approach for the bio‐organocatalytic amidation reaction using different diacids (c, g, h) and (S) and (R) configured amines 13 and 14. Supporting Fig. S1: GC analysis for enantiomeric determination of product using AsR transaminase. Product (R)‐8 retention time: 29.387 min. Supporting Fig. S2: GC analysis for enantiomeric determination of product using AsR transaminase. Product (S)‐8 retention time: 29.124 min. Product (R) retention time: 29.275 min*.* Supporting Table S1: Biocatalytic transamination in analytical scale (total reaction volume of 1 mL). Supporting Table S2: Conversion values for products and side‐products obtained from reactions of chiral amines with dicarboxylic acids.

Author Contributions

This study was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Funding

This study was supported by Nederlandse Organisatie voor Wetenschappelijk Onderzoek (Grant OCENW.XS24.3.183), Horizon Europe Marie Skłodowska‐Curie Actions (Grants 101153173 and 101208425), and Fonds de la Recherche Scientifique (Grant 40029634).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material