Design of non-cytotoxic 6,7-dihydroxycoumarin-5-carboxylates with antibiofilm activity against Staphylococcus aureus and Candida albicans

Abstract

The 6,7-dihydroxycoumarin-5-carboxylates DHCou and 4-Me-DHCou have been synthesized via five-step route including a propargyl-Claisen rearrangement as key step. The compounds show antibiofilm activity against Stapylococcus aureus and Candida albicans but lack the cytotoxic activity of parent 6,7-dihydroxycoumarines such as esculetin and 4-methylesculetin.

---

The occurrence of resistance of microbial pathogens against market antimicrobial drugs has been continuously rising since 1970s and nowadays we see ourselves confronted with multidrug-resistant microbial pathogens causing a strong need for new drugs and concepts to counter act this threat for public health.^1–7 In addition, microbial biofilms provide challenges for the development of antimicrobials as bacterial^8–13 and fungal pathogens^14–16 tend to hide in biofilms preventing drug penetration and leading to recurring and persistent infections.¹⁵ Beyond planktonic cells, biofilms are a principal form of microbial growth on surfaces in which microbes embed themselves in sugar, peptide and lipid containing hydrogels. Biofilms are often critical to development of clinical infections in human host¹⁴ and can be found for many microbial pathogens such as Methicillin-resistant Staphylococcus aureus (MRSA),^10,17Pseudomonas aeruginosa^8,18–21 and Candida albicans^15,22,23 causing severe infections.

In biofilms these pathogens can even co-occur during infection.²⁴ Compounds which are able to disrupt biofilms or inhibit their formation are of vast importance to make these pathogens susceptible again against antimicrobial drugs.²⁵ The inhibition of biofilm formation can be achieved via different modes of action of compounds: while targeting of efflux pumps by efflux pump inhibitors disables the secretion of building blocks for the biofilms,²⁶ the inhibition of cell–cell communication by quorum sensing inhibitors is another successful strategy to avoid biofilm formation and impair virulence of the pathogens.²⁵

For several coumarins antibiofilm properties have been demonstrated,^27–30 such as coumarin (1), esculetin (2), dephnetin (3), umbelliferone (4), 4-hydroxycoumarine (5), and scopoletin (6) (Fig. 1).^31–36 The biofilm inhibitory effect of dihydroxycoumarins such as umbelliferone (4) and esculetin (2) has been reported to be mediated by both, efflux pump inhibition as well as impairment of quorum sensing,^31,33 making them interesting lead structures for the development of novel biofilm inhibitors. However, both compounds show distinct antiproliferative activities against human cells limiting their potential application.^37–40

Fig. 1 Structures of selected coumarin derivatives showing antibiofilm activities.

In this work, we aimed to design novel 6,7-dihydroxycoumarin-5-carboxylate derivatives with reduced cytotoxic activity bearing a molecular handle for attachment and hybridization with further antimicrobial drugs or siderophores moieties to enable future transfer of the antibiofilm activity of 6,7-hydroxycoumarins onto these entities as outlined in Scheme 1. We hypothesized that the 5-position of the coumarin core might be suitable for the incorporation of a carboxylate handle allowing for the conjugation of such compounds to artificial siderophores such as the recently published enterobactin derivative Ent_KL ^41–43 or additional antimicrobial drug moieties³ while retaining the antibiofilm properties. Furthermore, a main metabolic pathway of coumarins mediated by cytochrome P450 monooxygenases.^44,45 Leading to cytotoxic intermediates and additionally to depletion of cellular glutathione levels is the 3,4-epoxidation of the core. We expected substituents in 5-position to impair this oxidative metabolization leading to a reduced cytotoxicity of the respective 6,7-dihydroxycoumarines.

Scheme 1 Concept of 6,7-dihydroxcoumarin-5-carboxylates as potential antibiofilm compounds bearing a handle for conjugation to siderophores and antimicrobial drug moieties.

Therefore, we followed a semi-synthetic strategy to generate the 6,7-dihydroxycoumarin-5-carboxylate (DHCou) starting from the natural product esculetin (2) in 5 steps. First, the selective alkylation of the 7-hydroxy function⁴⁶ with benzyl bromide at −15 °C gave access to the 7-benzyloxy coumarin derivative 7 in 73% yield, which could be further O-alkylated at the 6-hydroxy function in presence of propargyl bromide obtaining the O,O-dialkylated coumarin 8 in 91% yield. Compound 8 was submitted to a cascade-reaction consisting of a thermal [3.3]sigmatropic propargyl-Claisen rearrangement and subsequent CsF-mediated nucleophilic 5-exo-dig cyclization of the intermediate allenylphenolate forming the 2-methylbenzo[d]furan 9 in 71% yield upon heating to 216 °C in the microwave over 3 h. This methodology had been invented by Ishii and co-workers⁴⁷ and was used earlier for the conversion of O-alkyl derivatives of scopoletin (6) (Scheme 2).⁴⁸ In order, to enable oxidative cleavage of the furan ring forming the 5-formyl coumarin derivative 10 we explored different oxidative reactions conditions. Most 2-step procedures forming first the corresponding 3,4-epoxide and subsequently furnishing the oxidative cleavage in the presence of sodium periodate or lead tetraacetate failed to give access to the product.

Scheme 2 Synthesis of DHCou from esculetin (2).

While no conversion to the intermediate 3,4-epoxide was observed in the presence of mCPBA, decomposition occurred when DMDO was applied for epoxide formation. A first success was achieved forming the 3,4-epoxide in the presence of trifluoroacetic acid anhydride (TFAA) and hydrogen peroxide at 0 °C over 5 h and subsequently cleaving the epoxide with sodium periodate in a 1 : 1-mixture of MeOH and water. However, compound 10 was only obtained impure in roughly 26% yield as the reaction occurred with several side reactions. Similarly, 10 was obtained impure in roughly 25% yield, when 2-methylbenzofurane 9 was ozonolyzed at 78 °C in CH₂Cl₂ followed by reductive workup with dimethyl sulfide. Ozonolysis at 78 °C in a 1 : 1-mixture of CH₂Cl₂ : MeOH/1 : 1 and reductive workup with triphenylphosphine gave access to impure 5-formyl coumarin derivative 10 in roughly 61% yield. Subsequent Pinnick oxidation of the impure aldehyde 10 delivered the O-protected 6,7-dihydroxycoumarin-5-carboxylate 11b in roughly 45% and 27% over two steps from furane 9. We then found that a similar procedure for the direct conversion of compound 9 into compound 11a lacking the acetate was also possible in a one-pot fashion giving 48% yield. The hydrogenolytic cleavage of the O-benzyl moiety in the presence of palladium on charcoal and subsequent O-acetylation with acetic acid anhydride and pyridine gave access to the O,O-diacetate 12 in 64% yield. Furthermore, DHCou was obtained in 72% yield and an overall yield of 16% over 5 steps after hydrogenolytic cleavage of the O-benzyl moiety and subsequent acidic deacetylation in the presence of aqueous HCl.

When DHCou was evaluated by crystal violet staining assay for its effects against formation of S. aureus biofilms no inhibitory activity was observed at the highest concentration of 250 μg mL⁻¹, while esculetin (2) showed inhibition effects of 93% on the formation of S. aureus biofilms at the concentration of 250 μg mL⁻¹ and of 33% at 125 μg mL⁻¹ (Table 1). However, inhibitory effects of 62% could be observed against C. albicans biofilms when DHCou was applied at 250 μg mL⁻¹ in the early stage of biofilm formation (Table 1). Esculetin (2) showed a slightly higher biofilm inhibition against C. albicans of 77% at the same concentration but was not active against preformed biofilms of S. aureus.

Table 1 Inhibition of biofilm formation on S. aureus and C. albicans biofilms^49–51

Compound	Biofilm inhibition [% ± SD]
	S. aureus (DSM 1104)	C. albicans (DSM 11225)
(−): no activity, SD: standard deviation, references [%].a Microporenic acid A (MAA): 93 ± 0.3 (250 μg mL^−1), 93 ± 1 (62.5 μg mL^−1), 62 ± 6 (7.8 μg mL^−1).b MMA: 82 ± 6 (250 μg mL^−1), 81 ± 8 (62.5 μg mL^−1), 73 ± 17 (7.8 μg mL^−1).c Farnesol: 87 ± 3 (250 μg mL^−1), 79 ± 14 (31.3 μg mL^−1), 67 ± 11 (15.6 μg mL^−1).d Farnesol: 75 ± 6 (250 μg mL^−1), 58 ± 15 (31.3 μg mL^−1), 46 ± 14 (15.6 μg mL^−1).
Esculetin (2)	93 ± 2 (250 μg mL^−1)^a	77 ± 7 (250 μg mL^−1)^c
	33 ± 6 (125 μg mL^−1)^a	58 ± 17 (125 μg mL^−1)^c
4-Methylesculetin (14)	94 ± 1 (250 μg mL^−1)^a	76 ± 7 (250 μg mL^−1)^c
	48 ± 8 (125 μg mL^−1)^a	48 ± 14 (125 μg mL^−1)^c
DHCou	—^b	62 ± 9 (250 μg mL^−1)^d
4-MeDHCou	75 ± 5 (250 μg mL^−1)^b	60 ± 2 (250 μg mL^−1)^d
	43 ± 11 (125 μg mL^−1)^b
	31 ± 15 (62.5 μg mL^−1)^b

---

However, due to its cytotoxic effects, these values must be taken with care (Table 2) and a conclusion on the single standing influence of the C5 substitution is not possible. Although, a clear loss in antibiofilm activity against S. aureus was observed, a significant portion of the initial antibiofilm activity against C. albicans could be retained by C5 substitution. Beyond that, validating our initial hypothesis, DHCou showed neither cytotoxicity nor antiproliferative activity against the tested mammalian cell lines (L929 and KB3.1) at the highest concentration tested (1 mg mL⁻¹ = 4.5 mM). Nevertheless, to increase the overall biofilm inhibitory activity of the compounds, we planned to increase the lipophilicity by incorporation of a methyl substituent in the 4-position of the coumarin core. Therefore, we generated 4-methylesculetin (14) via a Pechmann condensation of 1,2,4-trihydroxybenzene (13) with ethylacetoacetate in TFA at 100 °C (Scheme 3). Following the earlier strategy, selective O-benzylation of the 7-hydoxy position and subsequent O-propargylation of the 6-hydroxy position gave access to the O,O-dialkylated precursor for the cascade rearrangement.

Scheme 3 Synthesis of 4-MeDHCou from 4-methylesculetin (2).

Table 2 Cytotoxic activity on different mammalian cell lines^a 52,53

Cell line	Cytotoxicity IC50 [μM]
	DHCou	4-MeDHCou	Esculetin (2)	4-Methylesculatin (14)
a For control references epothilon B see Table S4 in the ESI;† (−): no cytotoxicity or changed cells observed (max. concentration 1 mg mL^−1 = 4.5 mM for 4-MeDHCou and 4.2 mM for DHCou), n.t.: not tested.
KB3.1 (ACC158)	—	—	29.8	30.2
L929 (ACC2)	—	—	41.5	33.8
A549 (ACC107)	n.t.	n.t.	18.5	21.9
A431 (ACC91)	n.t.	n.t.	41.5	38.5
PC-3 (ACC465)	n.t.	n.t.	46.0	38.0
SKOV-3 (ATCC HTB 77)	n.t.	n.t.	45.5	42.7
MCF-7 (A115)	n.t.	n.t.	19.6	27.1

---

The thermal [3.3]sigmatropic propargyl-Claisen rearrangement and subsequent CsF-mediated nucleophilic 5-exo-dig cyclization proceeded with 55% yield and gave access to the 2-methylbenzofurane 17 (Scheme 3). Ozonolysis, followed by reductive workup in the presence of triphenylphosphine and subsequent Pinnick oxidation led to formation of the O-protected 6,7-dihydroxycoumarin-5-carboxylate 19 in 54%. Again, it turned out difficult to isolate the intermediate aldehyde 18, which could only be obtained in small amounts and with certain impurities when the oxidative cleavage was done via a sequence of oxidation with trifluoroperoxoacetic acid and subsequent cleavage in the presence of sodium periodate (see ESI†). From 19 the O,O-diacetate 20 and 4-MeDHCou were obtained following the procedure established before.

Similar to DHCou, 4-MeDHCou showed no cytotoxic activity against the mammalian cervic carcinoma cell line KB3.1 and the mouse fibroblasts cell line L929 when applied at the highest concentration of 1 mg mL⁻¹ (4.2 mM).

In contrast to that and in accordance with the observed cytotoxicity of esculetin (2), 4-methylesculetin (14) exhibited cytotoxic activity against all tested mammalian cell lines (Table 2). Furthermore, while the inhibitory effects against C. albicans biofilms were comparable to that of DHCou, 4-MeDHCou inhibited the formation by 60% at 250 μg mL⁻¹ (Table 1), we could also observe activity against S. aureus biofilms. Thus, 4–MeDHCou showed inhibition effects of 75% on the formation of S. aureus biofilms at the concentration of 250 μg mL⁻¹ and of 43% at 125 μg mL⁻¹. The antibiofilm activity of 4-methylesculetin (14) was observed to be higher compared to 4-MeDHCou with 76% inhibition at 250 μg mL⁻¹ and 48% at 125 μg mL⁻¹ against C. albicans and 94% inhibition at 250 μg mL⁻¹ and 48% at 125 μg mL⁻¹ against S. aureus. Furthermore, no dispersal effects against preformed biofilms of S. aureus were observed for 4-MeDHCou and 4-methylesculetin (14).

However, considering a potential contribution of the observed cytotoxicity of 4-methylesculetin (14) to its antibiofilm activity, a significant portion of the antibiofilm activity could be retained by C5 carboxylate substitution in 4-MeDHCou.

In addition, none of the compounds (DHCou, 4-MeDHCou, esculetin (2) or 4-methylesculetin (14)) showed antimicrobial activity (see Table S1 in the ESI†) against a panel of Gram-positive and Gram-negative bacteria (B. subtilis, S. aureus, M. Smegmatis, A. baumanni, C. violaceum, E. coli, and P. aeruginosa) and different fungi (M. hiemalis, P. anomala, R. glutinis, C. albicans and S. pombe) up to a concentration of 66.7 μg mL⁻¹.

Besides their antibiofilm activity, coumarins are known for their UV fluorescence and use as dyes in chemical biology. In line with that also 4-MeDHCou and its coumarin precursors 15, 16, 17 and 19 showed fluorescence with maximal emission wavelengths between 400–450 nm (Fig. 2). Interestingly 4-MeDHCou was the most bathochromic shifted compound of this series, although fluorescence intensity at normalized concentration decreased along the synthesis route.

Fig. 2 UV/Vis absorption and fluorescence emission of 4-MeDHCou and its precursors 15, 16, 17 and 19 at normalized concentration 2.125 × 10⁻⁵ mol L⁻¹.

Currently different approaches to further optimize the structure for higher antibiofilm activity and incorporate 4-MeDHCou to artificial siderophores and antimicrobial drug hybrids are under investigation to explore its potential to serve as mediator for antibiofilm activity (as outlined in Scheme 1).

Conclusions

We synthesized two novel 6,7-dihydroxycoumarin-5-carboxylates, namely DHCou and 4-MeDHCou. In contrast, to their non-carboxylated parent 6,7-dihydroxycoumarins esculetin (2) and 4-methylesculetin (14), these compounds lack any cytotoxic activity towards different mammalian cell lines, while retaining an antibiofilm activity. DHCou displayed inhibitory effects against the early stage of C. albicans biofilm formation but showed no activity against S. aureus biofilms. Furthermore, 4-MeDHCou exhibited antibiofilm activity against both, the formation of S. aureus and C. albicans biofilms. Although the structure activity relationships need to be further investigated to fully understand the observed effects and improve the antibiofilm activity of the compounds, a proof-of-principle for the design of non-cytotoxic hydroxycoumarins retaining antibiofilm activity has been made, holding potential to overcome a major limitation for the application of coumarins as biofilm disruptors. In addition, these moieties might be able to transfer their antibiofilm activity by conjugation to other entities such as antimicrobials drugs or siderophores. Further investigations are ongoing to explore their potential.

Author contributions

Conceptualization, funding acquisition, project administration and writing of original draft: PK; supervision: PK, HS; writing – review and editing: PK, RZ, AC, APL, HS and HZ. Investigation: RZ, AC, APL, HS, HZ and WC. Methodology: PK, RZ, AC and APL; resources: PK, AC, RZ, and APL; visualization: PK, HS and APL.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Parts of the work have been carried out within the framework of the Centre of Antimicrobial Resistance Research (CARe) in Gothenburg and the SMART BIOTECS alliance between the Technische Universität Braunschweig and the Leibniz Universität Hannover. This initiative is supported by the Ministry of Science and Culture (MWK) of Lower Saxony (PK), Germany. Financial support by the Deutsche Forschungsgemeinschaft (DFG, grant KL 3012/2-1 (PK)) and the Fonds der Chemischen Industrie (FCI, PK) is gratefully acknowledged. HZ is grateful for a personal PhD stipend from the “Drug Discovery and Cheminformatics for New Anti-Infectives (iCA)” and is financially supported by the Ministry for Science & Culture of the German State of Lower Saxony (MWK no. 21-78904-63-5/19). The content of this work is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies.

References

1. D. J. Krysan, Virulence, 2017, 8, 135–137.
2. E. Ksiezopolska and T. Gabaldón, Genes, 2018, 9, 461.
3. P. Klahn and M. Brönstrup, Nat. Prod. Rep., 2017, 34, 832–885.
4. P. Klahn and M. Brönstrup, Curr. Top. Microbiol. Immunol., 2016, 389, 365–417.
5. M. A. Cook and G. D. Wright, Sci. Transl. Med., 2022, 14, eabo7793.
6. N. Mobarki, B. Almerabi and A. Hattan, Int. J. Med. Dev. Countries, 2019, 40, 561–564.
7. T. M. Privalsky, A. M. Soohoo, J. Wang, C. T. Walsh, G. D. Wright, E. M. Gordon, N. S. Gray and C. Khosla, J. Am. Chem. Soc., 2021, 143, 21127–21142.
8. A. Vetrivel, M. Ramasamy, P. Vetrivel, S. Natchimuthu, S. Arunachalam, G.-S. Kim and R. Murugesan, Biologics, 2021, 1, 312–336.
9. N. Høiby, T. Bjarnsholt, M. Givskov, S. Molin and O. Ciofu, Int. J. Antimicrob. Agents, 2010, 35, 322–332.
10. V. Silva, L. Almeida, V. Gaio, N. Cerca, V. Manageiro, M. Caniça, J. L. Capelo, G. Igrejas and P. Poeta, Pathogens, 2021, 10, 970.
11. M. H. Muhammad, A. L. Idris, X. Fan, Y. Guo, Y. Yu, X. Jin, J. Qiu, X. Guan and T. Huang, Front. Microbiol., 2020, 11, 1–20.
12. L. Hall-Stoodley, J. W. Costerton and P. Stoodley, Nat. Rev. Microbiol., 2004, 2, 95–108.
13. F. Hemmati, M. A. Rezaee, S. Ebrahimzadeh, L. Yousefi, R. Nouri, H. S. Kafil and P. Gholizadeh, Mol. Biotechnol., 2021, 63, 569–586.
14. S. Fanning and A. P. Mitchell, PLoS Pathog., 2012, 8, e1002585.
15. J. V. Desai, A. P. Mitchell and D. R. Andes, Cold Spring Harbor Perspect. Med., 2014, 4, a019729.
16. M. Gulati and C. J. Nobile, Microbes Infect., 2016, 18, 310–321.
17. A. Boudet, P. Sorlin, C. Pouget, R. Chiron, J.-P. Lavigne, C. Dunyach-Remy and H. Marchandin, Front. Microbiol., 2021, 12, 750489.
18. O. Ciofu and T. Tolker-Nielsen, Front. Microbiol., 2019, 10, 913.
19. M. T. T. Thi, D. Wibowo and B. H. A. Rehm, Int. J. Mol. Sci., 2020, 21, 8671.
20. F. F. Tuon, L. R. Dantas, P. H. Suss and V. S. Tasca Ribeiro, Pathogens, 2022, 11, 300.
21. E. Banin, M. L. Vasil and E. P. Greenberg, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 11076–11081.
22. C. Tsui, E. F. Kong and M. A. Jabra-Rizk, Pathog. Dis., 2016, 74, ftw018.
23. T. Atriwal, K. Azeem, F. M. Husain, A. Hussain, M. N. Khan, M. F. Alajmi and M. Abid, Front. Microbiol., 2021, 12, 638309.
24. M. del M. Cendra and E. Torrents, Biotechnol. Adv., 2021, 49, 107734.
25. A. Ghosh, N. Jayaraman and D. Chatterji, ACS Omega, 2020, 5, 3108–3115.
26. A. Reza, J. M. Sutton and K. M. Rahman, Antibiotics, 2019, 8, 229.
27. M. G. da Cunha, J. de Cássia Orlandi Sardi, I. A. Freires, M. Franchin and P. L. Rosalen, Microb. Pathog., 2020, 139, 103855.
28. S. K. Roy, N. Kumari, S. Pahwa, U. C. Agrahari, K. K. Bhutani, S. M. Jachak and H. Nandanwar, Fitoterapia, 2013, 90, 140–150.
29. J. H. Lee, J. H. Park, H. S. Cho, S. W. Joo, M. H. Cho and J. Lee, Biofouling, 2013, 29, 491–499.
30. T. Das, M. C. Das, A. Das, S. Bhowmik, P. Sandhu, Y. Akhter, S. Bhattacharjee and U. C. De, World J. Microbiol. Biotechnol., 2018, 34, 170.
31. J.-H. J. H. Lee, Y. G. Kim, H. S. Cho, S. Y. Ryu, M. H. Cho and J.-H. J. H. Lee, Phytomedicine, 2014, 21, 1037–1042.
32. Y. Zhang, A. Sass, H. Van Acker, J. Wille, B. Verhasselt, F. Van Nieuwerburgh, V. Kaever, A. Crabbé and T. Coenye, Front. Microbiol., 2018, 9, 1–10.
33. F. J. Reen, J. A. Gutiérrez-Barranquero, M. L. Parages and F. O'Gara, Appl. Microbiol. Biotechnol., 2018, 102, 2063–2073.
34. K. Xu, J. L. Wang, M. P. Chu and C. Jia, J. Mycol. Med., 2019, 29, 28–34.
35. F. A. Qais, M. S. Khan, I. Ahmad, F. M. Husain, R. A. Khan, I. Hassan, S. A. Shahzad and W. AlHarbi, ACS Omega, 2021, 6, 18823–18835.
36. Z. He, W. Jiang, Y. Jiang, J. Dong, Z. Song, J. Xu and W. Zhou, J. Oral Microbiol., 2022, 14, 2055523.
37. S.-M. Yu, D.-H. Hu and J.-J. Zhang, Mol. Med. Rep., 2015, 12, 3869–3873.
38. L. Zhang, Q. Xie and X. Li, Phytother. Res., 2022, 36, 279–298.
39. J. Y. Yang, M. A. Della-Fera, D. L. Hartzell, C. Nelson-Dooley, D. B. Hausman and C. A. Baile, Obesity, 2006, 14, 1691–1699.
40. C. Y. Chu, Y. Y. Tsai, C. J. Wang, W. L. Lin and T. H. Tseng, Eur. J. Pharmacol., 2001, 416, 25–32.
41. R. Zscherp, J. Coetzee, J. Vornweg, J. Grunenberg, J. Herrmann, R. Müller and P. Klahn, Chem. Sci., 2021, 12, 10179–10190.
42. P. Klahn, R. Zscherp and C. C. Jimidar, Synthesis, 2022, 54, 3499–3557.
43. C. Rohrbacher, R. Zscherp, S. C. Weck, P. Klahn and C. Ducho, Chem. – Eur. J., 2022, e202202408.
44. B. Lake, Food Chem. Toxicol., 1999, 37, 423–453.
45. A. Stefanachi, F. Leonetti, L. Pisani, M. Catto and A. Carotti, Molecules, 2018, 23, 250.
46. M. Kawase, H. Sakagami, N. Motohashi, H. Hauer, S. S. Chatterjee, G. Spengler, A. V. Vigyikanne, A. Molnár and J. Molnár, In Vivo, 2005, 19, 705–711.
47. H. Ishii, T. Ishikawa, S. Takeda, S. Ueki and M. Suzuki, Chem. Pharm. Bull., 1992, 40, 1148–1153.
48. H. Ishii, T. Ishikawa, H. Wada, H. Miyazaki, Y. Kaneko and T. Harayama, Chem. Pharm. Bull., 1992, 40, 2614–2619.
49. K. T. Yuyama, L. Wendt, F. Surup, R. Kretz, C. Chepkirui, K. Wittstein, C. Boonlarppradab, S. Wongkanoun, J. Luangsa-ard, M. Stadler and W.-R. Abraham, Biomolecules, 2018, 8, 129.
50. K. T. Yuyama, C. Chepkirui, L. Wendt, D. Fortkamp, M. Stadler and W.-R. Abraham, Microorganisms, 2017, 5, 80.
51. C. Chepkirui, K. T. Yuyama, L. A. Wanga, C. Decock, J. C. Matasyoh, W.-R. R. Abraham and M. Stadler, J. Nat. Prod., 2018, 81, 778–784.
52. K. Becker, A. Wessel, J. J. Luangsa-ard and M. Stadler, Biomolecules, 2020, 10, 805.
53. P. Klahn, V. Fetz, A. Ritter, W. Collisi, B. Hinkelmann, T. Arnold, W. Tegge, K. Rox, S. Hüttel, K. I. Mohr, J. Wink, M. Stadler, J. Wissing, L. Jänsch and M. Brönstrup, Chem. Sci., 2019, 10, 5197–5210.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ob00303e

‡ These authors contributed equally.

---