Amide Bond Formation Using 4-Coumarate: CoA Ligase from Arabidopsis thaliana

Abstract

Amide bond formation is one of the most fundamental reactions in organic chemistry, and amide bonds constitute the key functional groups in natural products, peptides, and pharmaceuticals. Here we demonstrate the chemoenzymatic syntheses of 4-coumaroyl- and hexanoyl-amino acids, using 4-coumarate: CoA ligase from the model plant Arabidopsis thaliana (At4CL2). At4CL2 accepts 4-coumaric acid and hexanoic acid as the carboxylate substrates to generate acyl adenylates, which are captured by the amino group of amino acids to afford a series of N-acyl amides. This study shows the potential of 4CL for application as a biocatalyst to generate a series of biologically active amide compounds.

Introduction

Amide functional groups are widely distributed in nature and among the most important motifs in pharmaceuticals, agrochemicals, and other valuable products.^1,2) The condensation reactions between carboxylates and amines generally require toxic reagents and solvents, although they are synthetically relatively simple reactions. Furthermore, the reactive functional group must be protected to prevent the generation of undesired side products, which increases the reaction steps required to introduce a single amide bond.³⁾ In contrast, the recently reported chemoenzymatic synthesis of an amide bond using adenylate-forming enzymes, including fatty acyl-CoA synthetase, the adenylation domains of the modular non-ribosomal peptide synthetase, and firefly luciferase (the ANL superfamily enzymes), is expected to be an environmentally friendly, gentle and efficient method.^4–13) The amide bond formation by the ANL superfamily enzymes is thought to proceed through the initial activation of the carboxylate substrate with ATP to form the acyl–AMP intermediate as a common mechanism, which is followed by a nucleophilic attack from various amino compounds to generate the amide bond.

4-Coumarate: CoA ligase (4CL, EC 6.2.1.12) is a member of the ANL superfamily enzymes, and plays a pivotal role in phenylpropanoid metabolism by providing precursors for a variety of plant secondary metabolites such as flavonoids and lignin.^14–16) The enzyme catalyzes 4-coumaroyl-CoA formation by the initial activation of 4-coumaric acid with ATP and MgCl₂ to form an acyl–AMP intermediate, which is followed by thioester bond formation with CoASH and the concomitant release of AMP (Fig. 1a). The mechanistic similarity with other members of the ANL superfamily enzymes prompted us to test 4CL as a biocatalyst for generating biologically active coumaroyl amide compounds (Fig. 1b). Notably, many 4-coumaroyl amides are inhibitors of microbial growth against various bacteria, including Bacillus subtilis and Staphylococcus aureus,¹⁷⁾ and N-(4-coumaroyl)-L-homoserine lactone (HSL) is regarded as the quorum-sensing signal in the photosynthetic bacterium Rhodopseudomonas palustris.¹⁸⁾ Here we report that the 4CL from Arabidopsis thaliana (At4CL2)^{14–16,19–21)} shows remarkable substrate promiscuity and catalytic versatility, and catalyzes amide bond formation to produce a series of 4-coumaroyl- and hexanoyl-amino acids, including N-(4-coumaroyl)-HSL and hexanoyl-HSL, acyl-HSL quorum-sensing signals.²²⁾

Fig. 1. The Enzyme Reaction of 4CL

(a) 4CL catalyzes the adenylation of 4-coumaric acid and the following thioesterification reaction to produce 4-coumaroyl-CoA. (b) The chemoenzymatic reaction using 4CL to generate the 4-coumaroyl-amides described in this study.

Results and Discussion

The recombinant At4CL2, with an additional hexa-histidine tag at the C-terminus, was heterologously expressed in Escherichia coli and purified by Ni-chelate chromatography. We tested the At4CL2 enzyme reactions in which an L-amino acid was used as a substrate instead of CoASH, along with 4-coumaric acid, ATP, and Mg²⁺. When L-alanine was employed as the amino substrate for At4CL2, the LC-MS analysis of the enzyme reaction products revealed a new compound with m/z = 236 [M + H]⁺, which is consistent with the molecular mass of N-(4-coumaroyl)-L-alanine (Fig. 2). The structure of the product was confirmed by direct comparison with the chemically synthesized authentic compound (Figs. 2 and S1). We also tested the enzyme reactions of At4CL2 with the other 19 proteogenic L-amino acids. As a result, the assumed N-(4-coumaroyl)-amino acids were detected in the LC-MS analysis under all conditions (Fig. S2). The MS/MS analysis revealed the generation of a characteristic coumaroyl 147 fragment of the amide bond cleavage, indicating the production of N-(4-coumaroyl)-L-amino acid conjugates (Fig. S3). Notably, the reaction with L-cysteine afforded two new peaks with m/z = 268, which suggested the formation of an S-(4-coumaroyl)-L-cysteine thioester, in addition to the N-(4-coumaroyl)-L-cysteine amide, by the nucleophilic attack of the cysteine thiol on the 4-coumaroyl–AMP intermediate (Fig. S3). Similarly, the reaction with L-lysine also afforded two new peaks with m/z = 293, suggesting the formation of N-α-(4-coumaroyl)-L-lysine and N-ε-(4-coumaroyl)-L-lysine (Fig. S3). The different efficiencies of the amide bond forming reactions among the various amino acids could be attributed to the differences in the size of the side chain or the pK_a of the amino group.

Fig. 2. The Enzyme Reaction of At4CL2 with 4-Coumaric Acid and L-Alanine

Negative control indicates the incubation of 4-coumaric acid with L-alanine, ATP, and Mg²⁺ without enzyme.

In addition to the proteogenic amino acids, we also performed the At4CL2 enzyme reactions with non-proteogenic amino acids, such as D-serine, D-cysteine, D-leucine, D-phenylalanine, and L-homoserine, (S)-α-amino-γ-butyrolactone (L-homoserine lactone) as substrates. As a result, the enzyme reactions also afforded N-(4-coumaroyl)-D-amino acids, N-(4-coumaroyl)-L-homoserine, and N-(4-coumaroyl)-L-homoserine lactone (Fig. 3). Thus, the enzymatic adenylation of 4-coumaric acid by At4CL and the subsequent attack by various amino nucleophiles provided a broad range of 4-coumaroyl-amino acids. However, the yields of the N-(4-coumaroyl)-D-amino acids were considerably lower than those of the N-(4-coumaroyl)-L-amino acids. For example, the yield of N-(4-coumaroyl)-D-serine was only 2% as compared to that of N-(4-coumaroyl)-L-serine (Fig. 3c). Interestingly, 4-coumaroyl (or caffeoyl/cinnamoyl)-amino acids have been isolated from several plants, including Robusta coffee (Coffea canephora)²³⁾ and cacao (Theobroma cacao),²⁴⁾ and they exhibit remarkable biological properties, such as hepatoprotective effects,²⁵⁾ anti-neuroinflammatory activities,²⁶⁾ and human immunodeficiency virus type 1 integrase inhibition.²⁷⁾ Although their biosynthetic pathways and physiological roles in plants have not been clearly identified, 4CL and its homologous enzymes could be involved in the formation of the 4-coumaroyl-amino acid conjugates.

Fig. 3. The Enzyme Reactions of At4CL2 with 4-Coumaric Acid and Non-proteogenic Amino Acids

(a) Enzyme reactions of At4CL2 with D-amino acids, including D-serine, D-cysteine, D-leucine, and D-phenylalanine. (b) The enzyme reaction of At4CL2 with L-homoserine. (c) Comparison of the enzyme reactions between L-serine and D-serine.

To further explore the catalytic potential, we also performed in vitro enzyme reactions using hexanoic acid as a starter substrate to generate various hexanoyl–amino acid conjugates. The LC-MS analyses of the At4CL2 enzyme reaction products with hexanoic acid, ATP, Mg²⁺, and L-alanine, L-phenylalanine or L-homoserine lactone, afforded two new peaks with m/z = 188, m/z = 264, and m/z = 200, which indicated the generation of N-hexanoyl-L-alanine, N-hexanoyl-L-phenylalanine, and N-hexanoyl-L-homoserine lactone, respectively (Figs. S4a and b). The structure of N-hexanoyl-L-alanine was confirmed by direct comparison with the chemically synthesized authentic compound (Fig. S4c).

As described above, the amide bond formation reaction by ANL superfamily enzymes is proposed to proceed through the enzymatic adenylation of the carboxylate and subsequent nucleophilic attack by the amino substrate.¹³⁾ Indeed, the LC-MS analysis of the At4CL2 enzyme reaction in the absence of either amino substrates or CoASH detected the 4-coumaroyl–AMP intermediate 1, which is thought to be released from the active site of the enzyme (Fig. 4). Furthermore, the non-enzymatic incubation of the 4-coumaroyl–AMP intermediate 1 with L-alanine or L-phenylalanine also produced N-(4-coumaroyl)-L-alanine and N-(4-coumaroyl)-L-phenylalanine, respectively (Fig. S5). These results suggest that the enzyme is only involved in the adenylation of the carboxylate substrates to generate the acyl–AMP intermediates, and the subsequent nucleophilic attack by the amino substrate is non-enzymatic (Fig. 1b). However, the significant differences in the product yields observed between the L- and D-amino acid substrates also suggest enzymatic control for the amide bond forming reactions. To fully understand the intimate structural details of the At4CL2 amide bond forming reactions, further structure-function analyses of the enzyme are required.

Fig. 4. Detection of 4-Coumaroyl–AMP Intermediate 1

(a) The enzyme reaction of At4CL2 without amine compounds or CoASH. (b) MS/MS chart of the peak at 22.4 min (m/z 494.3). (c) Proposed MS/MS fragmentation of 4-coumaroyl–AMP intermediate 1.

At4CL2 reportedly accepts 4-coumaric acid, caffeic acid, and medium-chain fatty acids as starter substrates to produce the corresponding CoA thioesters, but sinapic acid, benzoic acid, phenylacetic acid, and certain amino acids including phenylalanine are not activated.¹⁶⁾ Further engineering to expand the substrate scope of At4CL2 would enable its employment as a versatile biocatalyst to generate various useful amide compounds for future drug discovery.

Acknowledgments

This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (JSPS KAKENHI Grant Number JP16H06443, JP19K15703, JP20H00490, and JP20KK0173), the New Energy and Industrial Technology Development Organization (NEDO, Grant Number JPNP20011), the PRESTO program from Japan Science and Technology Agency, and Fuji Foundation for Protein Research.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

References

• 1) Dunetz J. R., Magano J., Weisenburger G. A., Org. Process Res. Dev., 20, 140–177 (2016).
• 2) Carey J. S., Laffan D., Thomson C., Williams M. T., Org. Biomol. Chem., 4, 2337–2347 (2006).
• 3) Abe T., Hashimoto Y., Hosaka H., Tomita-Yokotani K., Kobayashi M., J. Biol. Chem., 283, 11312–11321 (2008).
• 4) Arai T., Arimura Y., Ishikura S., Kino K., Appl. Environ. Microbiol., 79, 5023–5029 (2013).
• 5) Goswami A., Van Lanen S. G., Mol. Biosyst., 11, 338–353 (2015).
• 6) Miki Y., Okazaki S., Asano Y., J. Ind. Microbiol. Biotechnol., 44, 667–675 (2017).
• 7) Wood A. J. L., Weise N. J., Frampton J. D., Dunstan M. S., Hollas M. A., Derrington S. R., Lloyd R. C., Quaglia D., Parmeggiani F., Leys D., Turner N. J. S., Flitsch L., Angew. Chem. Int. Ed., 56, 14498–14501 (2017).
• 8) Hara R., Hirai K., Suzuki S., Kino K., Sci. Rep., 8, 2950 (2018).
• 9) Kano S., Suzuki S., Hara R., Kino K., Appl. Environ. Microbiol., 85, e00120-e19 (2019).
• 10) Marchetti P. M., Richardson S. M., Kariem N. M., Campopiano D. J., Medchemcomm, 10, 1192–1196 (2019).
• 11) Zhu M., Wang L., He J., ACS Chem. Biol., 14, 256–265 (2019).
• 12) Petchey M. R., Rowlinson B., Lloyd R. C., Fairlamb I. J. S., Grogan G., ACS Catal., 10, 4659–4663 (2020).
• 13) Winn M., Richardson S. M., Campopiano D. J., Micklefield J., Curr. Opin. Chem. Biol., 55, 77–85 (2020).
• 14) Stuible H. P., Kombrink E., J. Biol. Chem., 276, 26893–26897 (2001).
• 15) Schneider K., Hövel K., Witzel K., Hamberger B., Schomburg D., Kombrink E., Stuible H. P., Proc. Natl. Acad. Sci. U.S.A., 100, 8601–8606 (2003).
• 16) Schneider K., Kienow L., Schmelzer E., Colby T., Bartsch M., Miersch O., Wasternack C., Kombrink E., Stuible H. P., J. Biol. Chem., 280, 13962–13972 (2005).
• 17) Guzman J. D., Molecules, 19, 19292–19349 (2014).
• 18) Schaefer A. L., Greenberg E. P., Oliver C. M., Oda Y., Huang J. J., Bittan-Banin G., Peres C. M., Schmidt S., Juhaszova K., Sufrin J. R., Harwood C. S., Nature (London), 454, 595–599 (2008).
• 19) Ehlting J., Büttner D., Wang Q., Douglas C. J., Somssich I. E., Kombrink E., Plant J., 19, 9–20 (1999).
• 20) Hamberger B., Hahlbrock K., Proc. Natl. Acad. Sci. U.S.A., 101, 2209–2214 (2004).
• 21) Li Y., Kim J. I., Pysh L., Chapple C., Plant Physiol., 169, 2409–2421 (2015).
• 22) Fuqua C., Greenberg E. P., Curr. Opin. Microbiol., 1, 183–189 (1998).
• 23) Clifford M. N., Knight S., Food Chem., 87, 457–463 (2004).
• 24) Stark T., Justus H., Hofmann T. J., Agric. Food Chem., 54, 2859–2867 (2006).
• 25) Yuan X., Lin L., Zhang X., Deng S., Phytochem. Lett., 7, 137–142 (2014).
• 26) Hu X. L., Lin J., Lv X.-Y., Feng J.-H., Zhang X.-Q., Wang H., Ye W. C., Eur. J. Med. Chem., 151, 261–271 (2018).
• 27) Lee S. U., Shin C.-G., Lee C.-K., Lee Y. S., Eur. J. Med. Chem., 42, 1309–1315 (2007).