Beauvetetraones A–C, phomaligadione-derived polyketide dimers from the entomopathogenic fungus, Beauveria bassiana

Seoung Rak Lee a, Michelle Küfner b, Minji Park c, Won Hee Jung c, Sang Un Choi d, Christine Beemelmanns b and Ki Hyun Kim *a
aSchool of Pharmacy, Sungkyunkwan University, Suwon 16419, Republic of Korea. E-mail:
bLeibniz Institute for Natural Product Research and Infection Biology – Hans-Knöll-Institute, Beutenbergstraße 11a, 07745 Jena, Germany
cDepartment of Systems Biotechnology, Chung-Ang University, Anseong, 17546, Republic of Korea
dKorea Research Institute of Chemical Technology, Deajeon 305-600, Republic of Korea

Received 18th September 2018 , Accepted 31st October 2018

First published on 1st November 2018

We report the isolation of two novel epimeric phomaligadione-derived polyketides, beauvetetraones A–B (1–2), from the entomopathogenic fungus Beauveria bassiana. Beauvetetraones A and B feature an unprecedented methylene-bridged phloroglucinol skeleton with a highly rearranged scaffold. In addition, a dimer of two phomaligadiones, named beauvetetraone C, was isolated for the first time from a natural source. The structures of compounds 1–3 including their absolute configurations were unambiguously assigned by NMR spectroscopic analyses, phenylglycine methyl ester (PGME) analysis, and quantum chemical ECD calculations. A putative biosynthetic pathway for beauvetetraones A–C is proposed.


Arthropods are the most species-rich group of animals, providing an enormous nutritional resource for parasites, pathogens, and predators. Amongst entomopathogenic organisms, fungi are of particular interest, as they constitute a group of more than 700 species (∼100 genera), some of which are commercially used as biocontrol agents. Entomopathogens infect and penetrate their hosts in a multiple-step process mediated by lysogenic enzymes and highly active small molecules.1

Ongoing studies of the biosynthetic potential of cosmopolitan insect pathogenic fungal species, such as Beauveria bassiana and Metarhizium robertsii (anisopliae), have revealed enormous biosynthetic potential with up to 40 to 90 core genes involved in secondary metabolite biosynthesis, respectively.1 It is estimated that the products of up to 80% of putative secondary metabolite-associated genes have not been identified. Driven by the genomic potential and ecological importance of bioactive small molecules in the infection process of insects, we selected B. bassiana, a cosmopolitan entomopathogen, as part of our continuing efforts to explore structurally and biologically novel metabolites from insect-associated microbes.2

Herein, we report the identification of three unprecedented phomaligadione-derived polyketides, beauvetetraones A–C (1–3), by LC/MS/UV-Vis based dereplication and describe their pharmacological properties. Based on this and previous research, we propose a putative biosynthetic pathway for the phomaligadione-based structures.

Results and discussion

B. bassiana sp. (Jena Microbial Research Collection (JMRC), ST000047) was first cultivated on agar plates using different media compositions (YMA or PDA) for up to three weeks.

Mycelium-covered agar plates (2 plates) were cut into pieces and extracted, and the concentrated methanolic extracts were analyzed by LC/MS. However, media-derived metabolites as well as known natural products, such as cyclodepsipeptides (beauverolides A–F),3 dominated the metabolome. We therefore used liquid static cultivation conditions, as they are known to facilitate conidia and spore formation, which are both correlated with the infection process in B. bassiana and metabolite production.4 LC/MS/UV-Vis analysis of enriched culture extracts indicated several putative novel UV-Vis/MS-target signals, which have not yet been reported by our in-house UV database or Antibase.5 For metabolite identification, we cultivated B. bassiana in 2 L standing liquid culture (PDB) (5 × 200 mL). Combined extraction of the mycelium and supernatant, followed by solvent-partitioned extraction and UV-Vis/MS-guided semi-preparative reverse-phase HPLC purification (Phenomenex Luna C18), resulted in the isolation of three new phomaligadione-derived polyketides, beauvetetraones A–C (1–3) (Fig. 1).

image file: c8qo01009a-f1.tif
Fig. 1 (A) Chemical structures of phomaligadione-derived polyketides, beauvetetraones A–C (1–3); (B) key COSY (image file: c8qo01009a-u1.tif) and HMBC (image file: c8qo01009a-u2.tif) correlations of units A and B of 1 and 3; (C) and (D) key ROESY correlations of beauvetetraones A (1) and B (2).

The first isolated compound, beauvetetraone A (1), was obtained as an amorphous powder and its molecular formula was assigned as C28H36O11 (unsaturation degree of 11) on the basis of a distinctive pseudomolecular ion peak at m/z 571.2154 [M + Na]+ (calcd for 571.2150). The IR spectrum showed absorption bands at 1630 and 1694 cm−1, which indicated the presence of an α,β-unsaturated ketone system. The 1H NMR spectrum (Table S1), with the assistance of HSQC, indicated nine characteristic methyl groups [δH 0.82 (3H, t, J = 7.5 Hz), 0.86 (3H, t, J = 7.5 Hz), 1.01 (3H, d, J = 7.0 Hz), 1.05 (3H, d, J = 7.0 Hz), 1.38 (3H, s), 1.44 (3H, s), 1.46 (3H, s), 3.66 (3H, s) and 3.80 (3H, s)] including two methoxy groups, three methylene groups [δH 1.51 (2H, m), 1.57 (2H, m), 3.12 (1H, d, J = 13.0 Hz), and 3.30 (1H, d, J = 13.0 Hz)], two methine groups [δH 2.35 (2H, m)], and olefinic methine [δH 5.54 (1H, s)] proton signals. Analyses of HSQC and HMBC data indicated a total of 28 carbon signals comprising nine methyl (δC 11.4, 11.5, 16.8, 17.0, 18.2, 23.5, 23.6, 53.4, and 57.1) including two methoxy carbons, three methylene (δC 27.5 (×2), 32.4), three methine [δC 40.1 (×2) and 99.7], six quaternary carbons (δC 54.0, 77.6, 83.1, 143.4, 157.8, and 177.3), and seven carbonyl carbons (δC 163.2, 176.7, 176.8, 194.1, 195.2, 199.9, 206.6) including four ketone groups.

As shown in Fig. 1B, the partial structures of unit A and unit B were deduced by combined interpretation of 1H–1H COSY, HSQC, and HMBC correlations. The connectivity of the H4′/H3′/H2′/H5′ system was identified from 3JHH couplings in the 1H–1H COSY spectrum, as well as 1H–13C long-range correlations of H-4′/C-2′, H-4′/C-3′, H-5′/C-1′, H-5′/C-2′, and H-5′/C-3′ suggesting the existence of a 2-methylbutyric ester group in unit A. We observed long-range 1H–13C correlations from quaternary methyl (H3-14) to ketone carbons at δC 206.6 (C-1) and δC 194.1 (C-3) and from quaternary methyl (H3-15) to a ketone carbon at δC 206.6 (C-1) as well as δC 177.3 (C-5) and δC 54.0 (C-6). The key HMBC correlations of an olefinic proton at δH 5.54 (H-4) to δC 83.1 (C-2), δC 194.1 (C-3), δC 177.3 (C-5), and δC 54.0 (C-6) were also examined, which unambiguously led to the 2,6-dimethyl-1,3-dioxocyclohex-4-ene for unit A. The position of a methoxy group [δH 3.66 (3H, s); δC (57.1)] was verified to be C-5 (δC 177.3) by a strong HMBC correlation from the methoxy proton to olefinic quaternary carbon (C-5). Key HMBC correlations from a quaternary methyl at δH 1.51 (H3-16) to C-9 (δC 199.9), C-11 (δC 195.2), and C-10 (δC 77.6) indicated the existence of two ketone moieties. A 10-methyl-9,11-dioxocyclopent-12-ene moiety (unit B) was deduced from HMBC correlations from methylene protons [δH 3.12 (H-7a) and 3.30 (H-7b)] to olefinic quaternary carbons [C-8 (δC 157.8), C-12 (δC 143.4)] and a ketone group at δC 199.9 (C-9). The presence of methyl ester was linked to C-12 based on the distinct proton signal at δH 3.80, which showed a strong HMBC correlation with ester carbon at δC 163.2 (C-13) and distinctive chemical shifts of olefinic quaternary carbons at C-8 (δC 157.8) and C-12 (δC 143.4). Another 2-methylbutyric ester group was also identified in unit B, and the locations of two 2-methylbutyric ester groups in units A and B were deduced to be C-2 and C-10, respectively, based on distinctive down-field chemical shifts of quaternary carbon C-2 (δC 83.1) and C-10 (δC 77.6). The partial structures A and B were linked by clear HMBC correlations from the methylene protons (H2-7) to C-1 (δC 206.6), C-5 (δC 177.3), C-6 (δC 54.0), and C-15 (δC 23.5) (Fig. 1B).

The relative configuration of 1 was delineated using rotational nuclear overhauser effect spectroscopy (ROESY) (Fig. 1C). The strong ROESY correlations between H3-14 and H3-15 led to the confirmation of identical α-orientations of CH3-14 and CH3-15. Key ROESY correlations from the methoxy group at C-5 to H3-15 and H3-16 suggested an α-oriented position of H3-16 attached to C-10 (Fig. 1C). Based on these observations, the relative configurations of C-2, C-6, and C-10 were tentatively determined to be 2R*, 6S*, and 10R*, respectively. The absolute configurations of C-2′ and C-2′′ of both 2-methylbutyric ester groups were determined as R-forms by chemical reactions using alkaline hydrolysis and phenyl-glycine methyl ester (PGME) reagent as well as LC/MS analysis.6

The absolute configurations of C-2, C-6, and C-10 were determined by comparing the experimental ECD spectrum of 1 with the ECD data calculated for two enantiomers 1a (2R,6S,10R) and 1b (2S,6R,10S) (Fig. 2). Calculated ECD spectra were obtained by time-dependent density-functional theory (TD-DFT) at the B3LYP/def2-TZVPP//B3LYP/def-SV(P) level for all atoms.7 As depicted in Fig. 2, the experimental ECD curve of 1 matched with the calculated ECD spectrum for 1a. In addition, the gauge-including atomic orbital (GIAO) NMR chemical shift calculation was carried out to verify the absolute configuration of C-10 as R since the C7–C8 bond in compound 1 can rotate freely. The calculated 1H and 13C NMR chemical shifts of two possible diastereomers 1A (2R,6S,10R) and 1B (2R,6S,10S) were subjected to DP4+ probability analysis with the experimental values of 1, which indicated that diastereomer 1A shows a DP4+ probability score of 100% (Fig. S4). Thus, the absolute stereochemistry was assigned as 2R,6S,10R.

image file: c8qo01009a-f2.tif
Fig. 2 Experimental and calculated ECD spectra of compounds 1–3.

Beauvetetraone B (2) displayed the same molecular formula as 1 [C28H36O11, m/z 547.2184 [M − H] (calcd for 547.2185)]. 1D and 2D NMR analyses indicated that the planar structure of 2 was identical to that of 1 (Fig. 1A). However, significant differences in chemical shifts of 2 were observed compared to 1, most dominantly for C-7 [δH 2.91 and 3.54 (1H, d, J = 13.0 Hz, each); δC 29.5] compared to C-7 [δH 3.12 and 3.30 (1H, d, J = 13.0 Hz, each); δC 32.4] of 1. Thus, we anticipated that compound 2 was an epimer of 1. Additional ROESY correlations of 5-OCH3/H2-7, 5-OCH3/H3-16, and H2-7/H3-14 supported the hypothesis and led to the tentative relative configuration assignment of C-2, C-6, and C-10 as 2R*, 6R*, and 10R*, respectively (Fig. 1D). The absolute configuration was again deduced by comparative ECD analysis of 2 using calculated ECD data for two enantiomers, 2a (2R,6R,10R) and 2b (2S,6S,10S). The experimental ECD spectrum of 2 was most similar to the calculated ECD spectrum of 2a (Fig. 2). To verify the absolute configuration of C-10 as R, the DP4+ analysis was again applied to the simulated 1H and 13C NMR chemical shifts of the two possible diastereomers 2A (2R,6R,10R) and 2B (2R,6R,10S). The results showed the structural equivalence of 2 to 2A with 100% probability (Fig. S4). Thus, the absolute stereochemistry of 2 was assigned as 2R,6R,10R, and identified to be a C-6 epimer of 1.

Beauvetetraone C (3) was purified as a yellow amorphous powder and the molecular formula C28H36O10 was deduced from the molecular ion peak at m/z 531.2235 [M − H] (calcd for 531.2230). The 1H NMR spectrum (Table S1) of 3 showed the presence of five methyl (δH 0.88, 1.05, 1.61, 1.89, and 3.64) including a methoxy group, one methylene (δH 1.41, and 1.59), and one methine (δH 2.37) signal. The 13C NMR data (Table S1), assigned by HSQC and HMBC spectra, displayed five methyl (δC 11.0, 12.6, 17.7, 22.1, and 61.3), one methylene (δC 28.7), one methine (δC 42.4), and seven quaternary carbons (δC 86.8, 128.1, 135.2, 163.4, 177.0, 194.4, and 195.0) including three carbonyl groups. Due to differences in the atom number and m/z signal, we hypothesized that 3 consisted of a symmetrical dimer. Comparative analysis of 2D NMR data (1H–1H COSY, HSQC, and HMBC) with compounds 1 and 2 and a review of the literature led to the identification of the planar structure of 3 (Fig. 1A),8 which was supported by the strong UV absorption pattern around 370 nm, denoting the presence of a poly-conjugated system in 3. Again, the absolute configurations of two 2-methylbutyric ester groups in 3 were both assigned R-forms by chemical reactions with alkaline hydrolysis followed by PGME amidation, as well as LC/MS analysis.6 To verify the absolute configuration of 3, four possible isomers 3a (2S,10S, Δ4/12 = E-conformation), 3b (2S,10S, Δ4/12 = Z-conformation), 3c (2R,10R, Δ4/12 = E-conformation), and 3d (2R,10R, Δ4/12 = Z-conformation) were used for ECD calculations utilizing TD-DFT at the B3LYP/def2-TZVPP//B3LYP/def-SV(P) level for all atoms, and the experimental ECD spectrum was in good agreement with the predicted ECD curve of 3b (2S,10S, Δ4/12 = Z-conformation) (Fig. 2). Interestingly, the relationships of two geometric isomers (3a/3b and 3c/3d) affected the ECD curves around 290–340 nm. This phenomenon allowed us to verify the Z-conformation of the bridged Δ4/12 double bond.

A plausible scheme for the generation of beauvetetraones A–C (1–3) was proposed (Fig. 3). Both isomers of phomaligadiones A and B are presumably the precursors for the formation of beauvetetraones A–C. While phomaligol A is a result of the oxidation of the allylic α-cabonyl position of phomaligadione A, beauvetetraone C can be derived from the oxidative homodimerization of two phomaligadione B units. Both transformations are not unexpected given the likely propensity of phomaligadiones A and B to be oxidized and the likely presence of oxygenase that are required for the formation of beauvetetraones A and B.

image file: c8qo01009a-f3.tif
Fig. 3 Plausible biosynthetic pathway to produce beauvetetraones A–C (1–3).

The dimerization to beauvetetraone C could be induced via enzymatic or oxygen-mediated hydrogen abstraction at the activated allylic position of phomaligadione B via formation of allyl radical A and subsequent C–C bond formation. The dimerization sequence concludes with hydrogen abstraction at the reactive allylic position at each subunit to form the more stable conjugated double bond system of dimer 3.

The formation of beauvetetraones A and B suggests furthermore the formation of intermediates C and/or E, although efforts to identify intermediates have so far been unsuccessful. Here, we hypothesize that phomaligadione B undergoes a series of oxidative rearrangements either via an epoxide intermediate C or via a sequence of oxidative C–C cleavage and condensation steps to yield substrate D and/or E. The existence of both stereoisomers (1 and 2) suggests that the key addition step in the biosynthesis occurs spontaneously, in a manner similar to the first key step of the homodimericin A formation.9 Acid or base-catalyzed Michael-type addition of intermediate G to F can result in the unselective formation of both epimers (1 and 2).

In a last step, compounds 1–3 were evaluated for their in vitro cytotoxicity against four human breast cancer cell lines (Bt549, HCC70, MDA-MB-231 and MDA-MB-468) by the SRB method,10 wherein compounds 1–3 exhibited weak cytotoxicity against all four cell lines, with IC50 values in the range of 61.87–82.37 μM (Table S2). In addition, these compounds were tested for their antifungal activity against five human fungal pathogens such as Cryptococcus neoformans H99, Candida albicans SC5314, C. glabrata KCTC 7219, C. parapsilosis KCTC 7214 and C. tropicalis KCTC 7212 (Table S3), but none of them were active in the tested concentration range (MIC > 120 μM).

Although our assays have failed to identify any significant bioactivity, the intriguing redox-active structures of phomaligadiones A and B and beauvetetraones A–C might point to roles in oxidative stress resistance of the producer organism B. bassiana and the isolated natural products are representative intermediates or end products of redox activities occurring within the cellular environment.


In conclusion, we have reported three novel phomaligadione-derived polyketides, beauvetetraones A–C (1–3), from the entomopathogenic fungus B. bassiana. Beauvetetraones A and B possess an unprecedented methylene-bridged phloroglucinol skeleton with a highly rearranged scaffold, and beauvetetraone C is a dimer of two phomaligadione B, isolated for the first time from a natural source. Their relative and absolute configurations were determined by 1D and 2D NMR spectroscopy, PGME analysis, and ECD calculations. The proposed biosynthesis pathway explains the formation of epimeric phomaligadione-derived polyketides reasonably. Beauvetetraones A–C were shown to exhibit weak cytotoxicity against human breast cancer cells.

Conflicts of interest

There are no conflicts to declare.


This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2018R1A2B2006879).

Notes and references

  1. (a) D. M. Gibson, B. G. G. Donzelli, S. B. Krasnoff and N. O. Keyhani, Nat. Prod. Rep., 2014, 31, 1287–1305 RSC ; (b) I. Molnar, D. M. Gibson and S. B. Krasnoff, Nat. Prod. Rep., 2010, 27, 1241–1275 RSC .
  2. (a) T. P. Wyche, A. C. Ruzzini, C. Beemelmanns, K. H. Kim, J. L. Klassen, S. Cao, M. Poulsen, T. S. Bugni, C. R. Currie and J. Clardy, Org. Lett., 2017, 19, 1772–1775 CrossRef CAS ; (b) C. Beemelmanns, T. R. Ramadhar, K. H. Kim, J. L. Klassen, S. Cao, T. P. Wyche, Y. Hou, M. Poulsen, T. S. Bugni, C. R. Currie and J. Clardy, Org. Lett., 2017, 19, 1000–1003 CrossRef CAS ; (c) H. R. Kang, D. Lee, R. Benndorf, W. H. Jung, C. Beemelmanns, K. S. Kang and K. H. Kim, J. Nat. Prod., 2016, 79, 3072–3078 CrossRef CAS ; (d) K. H. Kim, T. R. Ramadhar, C. Beemelmanns, S. Cao, M. Poulsen, C. R. Currie and J. Clardy, Chem. Sci., 2014, 5, 4333–4338 RSC .
  3. (a) D. Matsuda, I. Namatame, H. Tomoda, S. Kobayashi, R. Zocher, H. Kleinkauf and S. Omura, J. Antibiot., 2004, 57, 1–9 CrossRef CAS ; (b) A. Jegorov, P. Sedmera, V. Matha, P. Simek, H. Zahradnickova, Z. Landa and J. Eyal, Phytochemistry, 1994, 37, 1301–1303 CrossRef CAS ; (c) M. Keiko, O. Ken, T. Harumi, S. Yoshikazu, N. Shigeru, M. Eiichi and Y. Shosuke, Bull. Chem. Soc. Jpn., 1993, 66, 3041–3046 CrossRef .
  4. (a) M. J. Chong-Rodríguez, M. G. Maldonado-Blanco, J. J. Hernández-Escareño, L. J. Galán-Wong and C. F. A. Sandoval-Coronado, J. Bio-technol., 2011, 10, 5736–5742 Search PubMed ; (b) M. C. Rombach, R. M. Aguda and D. W. Rob-erts, Entomophaga, 1988, 33, 315–324 CrossRef ; (c) F. E. Vega, M. A. Jackson, G. Mercadier and T. J. Poprawski, World J. Microbiol. Biotechnol., 2003, 19, 363–368 CrossRef CAS ; (d) M. J. Bidochka, T. A. Pfeifer and G. G. Khachatouri-ans, Mycopathologia, 1987, 99, 77–83 CrossRef .
  5. H. Laatsch, AntiBase 2014: The Natural Compound Identifer, John Wiley & Sons, New York, 2014 Search PubMed .
  6. (a) J. K. Woo, C. K. Kim, C. H. Ahn, D. C. Oh, K. B. Oh and J. Shin, J. Nat. Prod., 2015, 78, 218–224 CrossRef CAS ; (b) C. Chepkirui, K. T. Yuyama, L. A. Wanga, C. Decock, J. C. Matasyoh, W. Abraham and M. Stadler, J. Nat. Prod., 2018, 81, 778–784 CrossRef CAS .
  7. (a) C. S. Kim, M. Bae, J. Oh, L. Subedi, W. S. Suh, S. Z. Choi, M. W. Son, S. Y. Kim, S. U. Choi, D. C. Oh and K. R. Lee, J. Nat. Prod., 2017, 80, 471–478 CrossRef CAS ; (b) K. Koyama, Y. Hirasawa, A. E. Nugroho, T. Hosoya, T. C. Hoe, K. L. Chan and H. Morita, Org. Lett., 2010, 12, 4188–4191 CrossRef CAS ; (c) K. B. Kang, H. W. Kim, J. W. Kim, W. K. Oh, J. Kim and S. H. Sung, J. Nat. Prod., 2017, 80, 1048–1054 CrossRef CAS ; (d) S. Tanaka, Y. Honmura, S. Uesugi, E. Fukushi, K. Tanaka, H. Maeda, K. Kimura, T. Nehira and M. J. Hashimoto, Org. Chem., 2017, 82, 5574–5582 CrossRef CAS .
  8. M. S. C. Pedras, V. M. Morales and J. L. Taylor, Tetrahedron, 1993, 49, 8317–8322 CrossRef .
  9. E. Mevers, J. Saurí, Y. Liu, A. Moser, T. R. Ramadhar, M. Varlan, R. T. Williamson, G. E. Martin and J. Clardy, J. Am. Chem. Soc., 2016, 138, 12324–12327 CrossRef CAS .
  10. P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. MaMahon, D. Vis-tica, J. T. Warren, H. Bokesch, S. Kenney and M. R. Boyd, J. Natl. Cancer Inst., 1990, 82, 1107–1112 CrossRef CAS .


Electronic supplementary information (ESI) available: Detailed descriptions of the experimental procedure, biological activity, and HRESIMS and NMR spectra of 1–3. See DOI: 10.1039/c8qo01009a

This journal is © the Partner Organisations 2019