Total Synthesis and Antibacterial Activity of 5-Geranyloxy-7-hydroxycoumarin and Murrayacoumarin A

Abstract

Coumarins are widely found in medicinal plants and exhibit diverse biological properties, including antibacterial activities. Herein, we report the total synthesis of 5-geranyloxy-7-hydroxycoumarin, 5-geranyloxy-7-methoxycoumarin, and Murrayacoumarin A. The asymmetric synthesis of (+)-Murrayacoumarin A was achieved via regioselective asymmetric dihydroxylation, allowing the determination of absolute configuration at the C7′-position. In addition, the antibacterial activities of the synthesized natural products and their derivatives were evaluated.

Introduction

The rising prevalence of drug-resistant bacteria has become a critical global public health issue, posing significant challenges in the clinical settings.¹⁾ Effective treatment options are limited, underscoring the need for novel therapeutic agents.^2,3) A diverse range of antibacterial compounds from natural sources has been discovered, greatly contributing to the advancement of medicinal chemistry research.⁴⁾ Considering this background, we performed growth inhibition assays against Methicillin-resistant Staphylococcus aureus (MRSA)—the most common antimicrobial-resistant organism identified in Japanese healthcare facilities⁵⁾—using our unique natural product library to identify novel hit compounds.⁶⁾ Among a library of 577 compounds, many of which are plant-derived, 81 of these compounds have the potential to inhibit the growth of MRSA.

In general, it is more challenging to identify antibacterial compounds active against Gram-negative bacteria than against Gram-positive bacteria, primarily because the outer membrane of Gram-negative bacteria serves as the first line of defense against toxic compounds.⁷⁾ It has been reported that the outer membrane protein TolC is essential for antibiotic susceptibility in Escherichia coli.^8,9) Using a TolC-deficient strain is an appropriate model for evaluating the antibacterial activity of compounds once they penetrate Gram-negative bacteria. Therefore, we investigated 81 compounds using a TolC-deficient strain of Salmonella (S.) Typhimurium to evaluate their effects on the growth of Gram-negative bacteria. This assay identified 5-geranyloxy-7-hydroxycoumarin (1), which was originally isolated from the stem bark of Clausena excavata by Ito in 1996,¹⁰⁾ as a promising candidate (Fig. 1).

Fig. 1. Target Natural Products

Coumarins are widely distributed in natural resources as secondary plant metabolites and exhibit diverse biological properties, including antibacterial activity, making them attractive targets in synthetic and medicinal chemistry.^11,12) However, the biological activity of compound 1 has been evaluated only in terms of its ability to inhibit the activation of the Epstein-Barr virus early antigen induced by 12-O-tetradecanoylphorbol-13-acetate,¹³⁾ and its total synthesis has not been reported. Additionally, structurally related natural products such as 5-geranyloxy-7-methoxycoumarin (2)¹⁴⁾ and (+)-Murrayacoumarin A (3)¹³⁾ have been isolated. Therefore, the synthesis of these compounds and their derivatives offers an opportunity to explore structure–activity relationships, potentially contributing to the development of novel antibacterial agents. Moreover, while compound 3 contains hydroxy groups at C7′- and C8′-positions on compound 1, the absolute configuration at the C7′-position has not been determined. Thus, we performed the asymmetric total synthesis of 3 to clarify its absolute configuration.

Here, we report the total syntheses of 5-geranyloxy-7-hydroxycoumarin (1), 5-geranyloxy-7-methoxycoumarin (2), and (+)-Murrayacoumarin A (3). We also evaluated the antibacterial activities of the synthesized natural products and their derivatives against MRSA and Salmonella TolC-deficient strains.

Results and Discussion

We synthesized 5-geranyloxy-7-hydroxycoumarin (1) from commercially available phloroglucinol (4) (Chart 1). Refluxing compound 4 with propiolic acid in water yielded 5,7-dihydroxycoumarin (5) in 58% yield.¹⁵⁾ Selective protection of the 7-hydroxy group with 1-nicotinoyl benzotriazole afforded compound 6 in 71% yield.¹⁶⁾ Geranylation of the 5-hydroxy group, followed by deprotection of the nicotinoyl group under K₂CO₃/MeOH conditions produced compound 1 in 69 and 97% yield, respectively. Additionally, the methylation of compound 1 with iodomethane and K₂CO₃ afforded 5-geranyloxy-7-methoxycoumarin (2) in 70% yield.

Chart 1. Synthesis of 5-Geranyloxy-7-hydroxycoumarin and 5-Geranyloxy-7-methoxycoumarin

Coumarins with electron-donating groups at the C7-position exhibit red-shifted absorption owing to an intramolecular electron pushpull effect, making them promising candidates for fluorescent probes.¹⁷⁾ Compound 1, which possesses a hydroxyl group at the C7-position, can function as a fluorescent dye without additional chemical modifications. Therefore, we recorded the absorption and emission spectra of 1 and 2 in EtOH (Fig. 2). Compound 1 exhibited an absorption band of up to 430 nm and emitted at 452 nm upon excitation at 405 nm. By contrast, compound 2 absorbed only the UV region and showed no fluorescence upon excitation at 405 nm. These results indicate the importance of the free hydroxy group at the C7-position for the fluorescence properties of compound 1. The introduction of fluorescent probes into bioactive molecules often leads to a decrease in their bioactivity, posing a challenge for biological analyses. This unique property suggests that compound 1 can serve as both an antibacterial agent and a fluorescent probe.

Fig. 2. UV-Vis Spectra

Subsequently, Murrayacoumarin A (3) was synthesized (Chart 2A). Since the absolute configuration at the C7′ position was not determined, we carried out regio- and stereoselective dihydroxylation of the terminal olefin in compound 1. Racemic (±)-3 was synthesized from compound 1 using K₂OsO₄ as a catalyst and K₃Fe(CN)₆ as an oxidant at 0°C. Regioselective dihydroxylation of the terminal double bond resulted in the formation of (±)-3 in 39% yield (Chart 2A, entry 1). Applying Sharpless’s asymmetric dihydroxylation with (DHQD)₂PHAL as a chiral ligand produced (+)-3 in 54% yield with 98% enantiomeric excess (ee) (Chart 2A, entry 2).^18,19) Similarly, using (DHQ)₂PHAL yielded (–)-3 in 57% yield with –93% ee (Chart 2A, entry 3). To determine the absolute configuration at the C7′-position of (+)-3, we also synthesized 3 from compound 6 and known geranyl bromide derivative (R)-8¹⁹⁾ (Chart 2B). The allylation of the 5-hydroxyl group of compound 6 afforded compound 9 in 38% yield. Subsequently, deprotection of the acetal was carried out under TsOH·H₂O/MeOH conditions, followed by deprotection of the nicotinoyl group under K₂CO₃/MeOH conditions, and yielded compound 3 in 27% yield for two steps. HPLC analysis confirmed that ee of compound 3 remained at 94% and exhibited a peak identical to that of (+)-3. Based on these results, the absolute configuration at the C7′ position of (+)-3 was determined to be the R configuration.

Chart 2. Total Synthesis of Murrayacoumarin A

To investigate the structure–activity relationship of compound 1, we synthesized derivatives by replacing the geranyl group with various side chains (Chart 3). Reaction of 5-hydroxy-7-nicotinoyloxycoumarin (6) with prenyl bromide, neryl bromide, (–)-citronellyl iodide, and (E),(E)-farnesyl bromide produced the corresponding derivatives, 5-prenyloxy 10a, 5-neryloxy 10b, 5-(–)-citronellyloxy 10c, and 5-(E),(E)-farnesyloxy 10d, respectively. Subsequent deprotection of the nicotinoyl group under K₂CO₃/MeOH conditions yielded the final derivatives 11a–11d. Additionally, 4-methylcoumarin derivative 11e was synthesized from 5-hydroxy-4-methyl-7-nicotinoyloxycoumarin¹⁶⁾ using the same protocol.

Chart 3. Synthesis of 7-Hydroxy-5-geranyloxycoumarin Derivatives

a) Reaction was performed at 0 to 40°C. b) K₂CO₃ was used instead of Cs₂CO₃.

We evaluated the antibacterial activities of the synthesized compounds. The minimum inhibitory concentration (MIC) of synthesized 1 was 25 μM against an MRSA strain and 50 μM against a Salmonella TolC-deficient strain, which was comparable to that of extracted 1, indicating the potent growth-inhibitory effect. By contrast, compound 1 did not inhibit the growth of the wild-type strain of S. Typhimurium, suggesting that its activity was limited by the inability to effectively permeate the outer membrane of Gram-negative bacteria. To identify the key structural features contributing to antibacterial activity, we tested the MICs of eight compounds (Table 1). Compounds 1, 11b, and 11c inhibited growth at concentrations of 12.5–25 μM for the MRSA strain and 50 μM for the TolC-deficient strain. Compound 11a was effective against the TolC-deficient strain at 100 μM but did not inhibit MRSA growth. By contrast, compound 11d inhibited MRSA growth at 12.5 μM but showed no effect on the TolC-deficient strain, even at 100 μM. Other compounds, including 2, (+)-3, 5, and 11e, exhibited MICs exceeding 100 μM against both strains. These results identified key structural features that contribute to antibacterial activity. Notably, the free hydroxy group at the C7-position (compared with compound 2), the lipophilic side chain at the C5-position (compared with compounds 3 and 5), and the unsubstituted C4-position (compared with compound 11e). Moreover, side chain lengths similar to that of the geranyl group effectively inhibited the growth of the TolC-deficient strains (compounds 11b and 11c), whereas shorter or longer side chains resulted in higher MICs (compounds 11a and 11d).

Table 1. Minimum Inhibitory Concentrations (MICs) of the MRSA Strain and the TolC-Deficient Strain (μM)

Compound	1	2	(+)-3	5	11a	11b	11c	11d	11e	CPFXa)
MRSA	25	>100	>100	>100	>100	25	12.5	12.5	>100	37.7
∆TolC	50	>100	>100	>100	100	50	50	>100	>100	<0.78

a) Ciprofloxacin (CPFX) was used as an indicator of antibacterial activity.

Conclusion

5-Geranyloxy-7-hydroxycoumarin (1), 5-geranyloxy-7-methoxycoumarin (2), and (+)-Murrayacoumarin A (3) were synthesized from commercially available phloroglucinol. The absolute configuration at the C7′-position of (+)-3 was determined to be R by synthesizing both stereoisomers using Sharpless’s asymmetric dihydroxylation protocol. Antibacterial assays revealed structure–activity relationships, indicating that the free 7-hydroxy group, optimal side chain length at the C5-position, and the unsubstituted C4-position in compound 1 were essential for its antibacterial activity. Moreover, the dual antibacterial and fluorescent properties of compound 1 enable unique biological analyses. Ongoing studies by our group aim to elucidate the mechanism underlying its antibacterial activity.

Acknowledgments

This work was supported by AMED-CREST Grant No.: 23gm1610004 and 24gm1610004, and JSPS KAKENHI Grant No.: 23K18175.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1)  Saha M., Sarkar A., J. Xenobiot., 11, 197–214 (2021).
• 2)  Lewis K., Biochem. Pharmacol., 134, 87–98 (2017).
• 3)  Muteeb G., Rehman M. T., Shahwan M., Aatif M., Pharmaceuticals (Basel), 16, 1615 (2023).
• 4)  von Nussbaum F., Brands M., Hinzen B., Weigand S., Häbich D., Angew. Chem. Int. Ed., 45, 5072–5129 (2006).
• 5)  Uematsu H., Yamashita K., Kunisawa S., Fushimi K., Imanaka Y., PLOS ONE, 12, e0179767 (2017).
• 6)  Yamaguchi J., Manome T., Hara Y., Yamazaki Y., Nakamura Y., Ishibashi M., Takaya A., Front. Pharmacol., 15, 1365815 (2024).
• 7)  Nikaido H., Microbiol. Mol. Biol. Rev., 67, 593–656 (2003).
• 8)  Brown D. G., May-Dracka T. L., Gagnon M. M., Tommasi R., J. Med. Chem., 57, 10144–10161 (2014).
• 9)  Zhu S., Alexander M. K., Paiva T. O., Rachwalski K., Miu A., Xu Y., Verma V., Reichelt M., Dufrêne Y. F., Brown E. D., Cox G., iScience, 27, 109592 (2024).
• 10)  Ito C., Ohta H., Tan H. T.-W., Furukawa H., Chem. Pharm. Bull., 44, 2231–2235 (1996).
• 11)  Sharifi-Rad J., Cruz-Martins N., López-Jornet P., Lopez E. P.-F., Harun N., Yeskaliyeva B., Beyatli A., Sytar O., Shaheen S., Sharopov F., Taheri Y., Docea A. O., Calina D., Cho W. C., Oxid. Med. Cell. Longev., 2021, 6492346 (2021).
• 12)  Ranjan Sahoo C., Sahoo J., Mahapatra M., Lenka D., Kumar Sahu P., Dehury B., Nath Padhy R., Kumar Paidesetty S., Arab. J. Chem., 14, 102922 (2021).
• 13)  Ito C., Itoigawa M., Onoda S., Hosokawa A., Ruangrungsi N., Okuda T., Tokuda H., Nishino H., Furukawa H., Phytochemistry, 66, 567–572 (2005).
• 14)  Caldwell A. G., Jones E. R. H., J. Chem. Soc., 1945, 540–543 (1945).
• 15)  Cao J.-L., Shen S.-L., Yang P., Qu J., Org. Lett., 15, 3856–3859 (2013).
• 16)  Fatykhov R. F., Khalymbadzha I. A., Chupakhin O. N., Charushin V. N., Inyutina A. K., Slepukhin P. A., Kartsev V. G., Synthesis (Stuttg.), 51, 3617–3624 (2019).
• 17)  Sun X. Y., Liu T., Sun J., Wang X. J., RSC Adv., 10, 10826–10847 (2020).
• 18)  Xu D., Park C. Y., Sharpless K. B., Tetrahedron Lett., 35, 2495–2498 (1994).
• 19)  Marsico G., Pignataro B. A., Masi M., Evidente A., Casella F., Zonno M. C., Tak J.-H., Bloomquist J. R., Superchi S., Scafato P., Tetrahedron, 74, 3912–3923 (2018).