Letter Cite This: Org. Lett. 2018, 20, 3741−3744
pubs.acs.org/OrgLett
Citrofulvicin, an Antiosteoporotic Polyketide from Penicillium velutinum Yong Chen,†,‡ Nan Jiang,§ Ying Jie Wei,† Xiang Li,† Hui Ming Ge,‡ Rui Hua Jiao,‡ and Ren Xiang Tan*,†,‡
Downloaded via STONY BROOK UNIV SUNY on July 6, 2018 at 12:03:57 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
†
State Key Laboratory Cultivation Base for TCM Quality and Efficacy, Nanjing University of Chinese Medicine, Nanjing 210023, China ‡ State Key Laboratory of Pharmaceutical Biotechnology, Institute of Functional Biomolecules, Nanjing University, Nanjing 210023, China § School of Pharmacy, Nanjing Medical University, Nanjing 210029, China S Supporting Information *
ABSTRACT: Citrofulvicin (1), along with its early shunt product fulvionol (2), was characterized as a skeletally unprecedented antiosteoporotic agent from a human sputum-derived fungus Penicillium velutinum. The unique citrofulvicin framework is likely formed by a nonenzymatic intermolecular Diels−Alder cycloaddition between heptaketide-based intermediates. Citrofulvicin and fulvionol were demonstrated to be osteogenic at 0.1 μM in the prednisoloneinduced osteoporotic zebrafish.
O
some processed foods such as blue cheese.15,16 We intuited that diverse fungi may reside in the lungs and the oral cavity and might be gathered in the mucus sputum. To combat the companion bacteria, some of these fungi might be obligated/ evolved to generate antibacterial molecules. Here, Penicillium velutinum CGMCC 3.7984, a mucus sputum-derived fungus, was shown to produce citrofulvicin (1) as a skeletally unprecedented polyketide with pronounced antiosteoporotic but negligible antibacterial activities. P. velutinum was cultured on the malt extract (ME) medium, followed by extraction with ethyl acetate. The extract obtained was fractionated mainly by a combination of column chromatography and HPLC refinement as detailed in the Supporting Information. Citrofulvicin (1) was afforded as yellowish crystals with its molecular formula evidenced to be C28H22O15 from its protonated molecular ion at m/z 599.1032 in its high-resolution electrospray ionization mass spectrometry (HR-ESI-MS) (C28H23O15 requires 599.1031). In line with the 18 double bond equivalents of the molecular formula, the 1H and 13C NMR spectra of 1 highlighted that its molecule has more quaternary carbons. This assumption was substantiated by the follow-up DEPT experiments, indicating its possession of twenty quaternary, three methine, three methylene, and two methyl carbons. The DEPT spectra of 1, along with its molecular formula, highlighted the presence of a total of 15 carbon-connected protons in the molecule, thereby implying its possession of seven hydroxyl groups (alcoholic/phenolic/
ur health depends on, or at least has multiple association with, the human microbiota capable of biosynthesizing structurally intriguing metabolites with various bioactivities. However, the human microbiome-derived natural products described so far are largely limited to the secondary metabolites of gut bacteria.1,2 As a matter of fact, human bodies also harbor highly diverse, as yet undefined communities of fungi whose cells are typically more than 100-fold larger than those of bacteria.3 Fungi have been found to exist in nearly all mucosal surfaces,4 although the human body temperature was suspected to be unsuitable for fungal growth.5 Like the cross-generation transmission of bacterial microbiota,6 the early mycobiota (viz., fungal microbiota) is succeeded via the mother−offspring transfer in a host− phenotype dependent manner; however, the population of human commensal fungi is influenced by fungal strains in the diet and environment.7 The cross-feeding and/or -talking mechanisms have been suggested for the microbial interaction among/between different bacterial and fungal species in microbiota.8,9 Moreover, the lateral gene transfers across microbiota members make the eukaryotic metabolism more chimeric to (help) survive the selection pressure applied to the commensal microbes.10 In the particular host niches, the antagonism or competition occurs between fungi and bacteria for the metabolic substrate and mutual tolerance.11 Therefore, fungi keep evolving to produce “special compounds” that antagonize bacterial growth and vice versa. Fungi belonging to the Penicillium genus are a rich source of antibacterial natural products such as penicillin.12 Penicillium species widely occur in the human bronchus13 and gut14 and in © 2018 American Chemical Society
Received: April 21, 2018 Published: June 21, 2018 3741
DOI: 10.1021/acs.orglett.8b01272 Org. Lett. 2018, 20, 3741−3744
Letter
Organic Letters
Figure 1. Absolute configuration assignments of citrofulvicin (1). Top: X-ray structure of (±)-1. Middle: comparison between the recorded and calculated ECD spectra. Bottom: absolute structures of (+)-1 and (−)-1.
also obtained as yellowish crystals and was shown to be a new heptaketide that we have named fulvionol (2). The molecular formula of 2 was indicated to be C14H14O8 by its protonated molecular ion at m/z 311.0759 in its HR-ESI-MS (calcd for C14H15O8, 311.0761). The structure of 2 was established by a correlative interpretation of its NMR spectra (1H and 13C NMR, DEPT, 1H−1H COSY, HSQC, and HMBC). The structure of 2 was confirmed by its single-crystal X-ray crystallographic analysis (Cu Kα) (Figure S21). However, we faced a challenge in addressing its absolute configuration, owing to the vibration of the 2-hydroxypropyl group during the X-ray diffraction analysis (Figure S21). Alternatively, the (2R)configuration of 2 was established by comparing its optical rotation ([α]20D +28.0, c 0.09, MeOH) with those of chirally inverted analogues with a single chiral carbon (Figure S1), (2′S)-2-(propan-2′-ol)-5-methyl-7-hydroxy-benzopyran-4-one ([α]24 −9.1, c 0.23, MeOH)19 and (2′R)-2-(propan-2′-ol)-5hydroxybenzopyran-4-one ([α]25D +30.0, c 0.25, MeOH), whose structure is indicative of its (2′R)-configuration (not “2′S” as indicated incorrectly therein).20 This was reinforced by the positive [α]D nature discerned with (2R)-fulvionol peracetate (2a, formed with pure acetic anhydride) and computed for (2′R)-2-(propan-2′-ol)-5-hydroxybenzopyran-4one and (2R)-fulvionol (2) (Tables S3 and S4). The coidentification of heptaketides 2, citromycetin, and fulvic acid inspired the postulation of the biosynthetic pathway of citrofulvicin (1). As illustrated in Scheme 1, the
carboxylic). Such an unusual situation made the NMR methodology incapable of determining the exact structure of 1, although some individual NMR signals could be readily ascribed to the isolated motifs such as two methyls (appearing as singlets at δH 1.24 and 1.32), two aromatic protons (corresponding to the singlets at δH 6.25 and 6.79), and three methylenes resonating as geminally coupled doublet pairs at δH 1.74 and 2.05 (J = 12.7 Hz), at δH 2.40 and 3.04 (J = 18.4 Hz), and at δH 3.81 and 4.23 (J = 10.6 Hz). Gratifyingly, this frustration was overcome by the single-crystal X-ray diffraction of 1 (Cu Kα), clarifying its structure and racemic nature (Figure 1). Finally, 1 was separated by chiral HPLC to yield (+)-1 and (−)-1, which were demonstrated to have (2S,3R,2′S,4′R,5′R,7′S)- and (2R,3S,2′R,4′S,5′S,7′R)-configurations, respectively, by comparing the recorded electronic circular dichroism (ECD) curves with those calculated for all of the optional stereoisomers (Figure 1). With 1 addressed stereochemically, we asked how it might be formed by the fungus. Scrutiny of the molecular framework of 1 suggested that its unusual carbon skeleton might result from the hybridization of appropriately structured precursors leading to the heptaketides citromycetin and/or fulvic acid.17 To address the hypothesis, the mother liquors (leftovers upon the isolation of 1) were combined and refractionated with an intention of hitting the heptaketide(s). As expected, two major heptaketides were afforded and identified to be citromycetin and fulvic acid (as a racemate).17,18 A minor companion was 3742
DOI: 10.1021/acs.orglett.8b01272 Org. Lett. 2018, 20, 3741−3744
Letter
Organic Letters Scheme 1. Plausible Biosynthetic Pathway toward Citrofulvicin (1)
construction of the aromatic polyketide framework of 1 was initiated by repetitive decarboxylative Claisen thioester condensations of an acetyl-CoA starter unit with six malonylCoA extender units, followed by the aromatization and oxidation that gave intermediate I. The ensuing Baeyer− Villiger oxidative cleavage of I yielded intermediate II with its ε-lactone hydrolyzed to form intermediate III. An intramolecular aldol reaction of III would give fulvionol (2) and fulvic acid depending on whether its acetonyl group was reduced or intramolecularly hemiketalized. Meanwhile, III can be tautomerized into intermediate V, which might afford citromycetin, presumably via intermediate VI after dehydration and intramolecular aldol reactions. With this rationalization in mind, citrofulvicin (1) might be produced by the Diels−Alder cycloaddition of VI with VII formed presumably from IV via elimination of two molecules of water. The isolation of citrofulvicin (1) as a racemate suggested that the Diels−Alder cycloaddition involved in its biosynthesis might be nonenzymatic in view of the stereoselectivity discerned with such reactions catalyzed by enzymes.21,22 In addition, fulvic acid racemate might form from IV via a spontaneous intramolecular hemiketalization. The single enantiomers (+)-1 and (−)-1 and their racemates ((±)-1) were evaluated first for the antibacterial action using as target strains Staphylococcus aureus CMCC(B)26003, methicillin-resistant Staphylococcus aureus (MRSA) ATCC43300, Streptococcus pyogenes ATCC19615, Bacillus subtilis CICC10283, and Micrococcus luteus. However, none of them could inhibit the growth of these bacteria at 10 μM. This suggested that P. velutinum might be tolerable to bacteria.11 On the other hand, some heptaketides derived from Penicillium spp. were found to have the semaphorin 3A modulating activities,23,24 which is also involved in human osteogenic differentiation.25 This observation prompted us to evaluate (+)-1, (−)-1, and (±)-1 for antiosteoporotic action in the prednisolone-induced osteoporotic zebrafish. As shown in Figure 2, they all alleviated the bone loss of osteoporotic
Figure 2. Effect of citrofulvicin ((±)-1) and its enantiomers ((+)-1 and (−)-1) on the bone loss of zebrafishes with prednisolone-induced osteoporosis. Three days postfertilization (3 dpf), the zebrafish were treated for 5 days with DMSO (vehicle control), prednisolone (negative reference), prednisolone combined with etidronate disodium (positive control), and prednisolone combined with (±)-1, (+)-1 and (−)-1, respectively. The alizarin red staining area and integrated optical density of the anterior tip of the notochord (arrow) of zebrafish was selected as measures of bone mineralization: (A) 0.125% DMSO; (B) 25 μM prednisolone; (C) 25 μM prednisolone combined with 60 μM etidronate disodium; (D−F)/ (G−I/J−L) 25 μM prednisolone combined with (+)-1/(−)-1/(±)-1 at 0.1, 0.5, and 2.5 μM, respectively. Quantitative data presents mean ± SEM, n = 6; ## p < 0.01, ### p < 0.001 versus the DMSO-treated group; * p < 0.05, ** p < 0.01, *** p < 0.001 versus the prednisolonetreated group.
zebrafish at 0.1, 0.5, and 2.5 μM in a roughly dose-dependent manner, with an approximate order of magnitude: (+)-1 > (±)-1 > (−)-1. At those dosages, 2 was also antiosteoporotic 3743
DOI: 10.1021/acs.orglett.8b01272 Org. Lett. 2018, 20, 3741−3744
Letter
Organic Letters
(6) Clemente, J. C.; Ursell, L. K.; Parfrey, L. W.; Knight, R. Cell 2012, 148, 1258−1270. (7) Schei, K.; Avershina, E.; Oien, T.; Rudi, K.; Follestad, T.; Salamati, S.; Odegard, R. A. Microbiome 2017, 5, 107. (8) Newman, M. A.; Petri, R. M.; Grull, D.; Zebeli, Q.; MetzlerZebeli, B. U. Front. Microbiol. 2018, 9, 224. (9) He, K.; Ma, H.; Xu, H. S.; Zou, Z. Y.; Feng, M.; Li, X.; Ye, X. L. J. Funct. Foods 2017, 35, 205−215. (10) Alsmark, C.; Foster, P. G.; Sicheritz-Ponten, T.; Nakjang, S.; Martin Embley, T.; Hirt, R. P. Genome Biol. 2013, 14, R19. (11) Mille-Lindblom, C.; Fischer, H.; Tranvik, L. J. Oikos 2006, 113, 233−242. (12) Ozcengiz, G.; Demain, A. L. Biotechnol. Adv. 2013, 31, 287− 311. (13) Oshikata, C.; Watanabe, M.; Saito, A.; Yasueda, H.; Akiyama, K.; Kamata, Y.; Tsurikisawa, N. Med. Mycol. 2017, 15, 9−11. (14) Borges, F. M.; de Paula, T. O.; Sarmiento, M. R. A.; de Oliveira, M. G.; Pereira, M. L. M.; Toledo, I. V.; Nascimento, T. C.; FerreiraMachado, A. B.; Silva, V. L.; Diniz, C. G. Curr. Microbiol. 2018, 75, 726−735. (15) Garcia-Estrada, C.; Martin, J. F. Appl. Microbiol. Biotechnol. 2016, 100, 8303−8313. (16) Bunkova, L.; Bunka, F. Crit. Rev. Food Sci. Nutr. 2017, 57, 2392−2403. (17) Capon, R. J.; Stewart, M.; Ratnayake, R.; Lacey, E.; Gill, J. H. J. Nat. Prod. 2007, 70, 1746−1752. (18) Fujita, K.; Nagamine, Y.; Ping, X.; Taniguchi, M. J. Antibiot. 1999, 52, 628−634. (19) Kashiwada, Y.; Nonaka, G.; Nishioka, I. Chem. Pharm. Bull. 1984, 32, 3493−3500. (20) Teles, H. L.; Silva, G. H.; Castro-Gamboa, I.; da Silva Bolzani, V.; Pereira, J. O.; Costa-Neto, C. M.; Haddad, R.; Eberlin, M. N.; Young, M. C.; Araujo, A. R. Phytochemistry 2005, 66, 2363−2367. (21) Jeon, B. S.; Ruszczycky, M. W.; Russell, W. K.; Lin, G. M.; Kim, N.; Choi, S. H.; Wang, S. A.; Liu, Y. N.; Patrick, J. W.; Russell, D. H.; Liu, H. W. Proc. Natl. Acad. Sci. U. S. A. 2017, 114, 10408−10413. (22) Cogan, D. P.; Hudson, G. A.; Zhang, Z.; Pogorelov, T. V.; van der Donk, W. A.; Mitchell, D. A.; Nair, S. K. Proc. Natl. Acad. Sci. U. S. A. 2017, 114, 12928−12933. (23) Axelrod, A.; Eliasen, A. M.; Chin, M. R.; Zlotkowski, K.; Siegel, D. Angew. Chem., Int. Ed. 2013, 52, 3421−3424. (24) Kumagai, K.; Hosotani, N. WO 2003062243, 2003. (25) Liu, L.; Wang, J.; Song, X.; Zhu, Q.; Shen, S.; Zhang, W. Exp. Ther. Med. 2018, 15, 3489−3494. (26) Villa, J. K. D.; Diaz, M. A. N.; Pizziolo, V. R.; Martino, H. S. D. Crit. Rev. Food Sci. Nutr. 2017, 57, 3959−3970. (27) Pavone, V.; Testa, G.; Giardina, S. M. C.; Vescio, A.; Restivo, D. A.; Sessa, G. Front. Pharmacol. 2017, 8, 803.
whereas citromycetin and fulvic acid were inactive even at 12.5 μM (Figure S36). In conclusion, P. velutinum, a human sputum-derived fungus, is a producer of citrofulvicin (1) that coexists with fulvionol (2) as its early shunt product. The unprecedented carbon skeleton of 1 is likely constructed through a nonenzymatic intermolecular Diels−Alder cycloaddition between heptaketide-based intermediates. The antiosteoporotic property of 1 and 2 in the zebrafish model seems interesting, although the antibacterial activity is insignificant. Osteoporosis is associated with decreased quality of life and is recognized as an important public health issue due to the aging global population; more drugs are desired for antiosteoporosis medications.26,27 The present characterization of citrofulvicin as a skeletally unprecedented molecule may be of value for the optimization and development of new osteoporosis-treating agents, and its unique architecture (rich in quaternary/chiral carbons) may pose a new topic for this synthetic endeavor. However, it is challenging to address the detailed structure−bioactivity relationship of 1 and 2, which are similar in substructure to each other and to their inactive congeners citromycetin and fulvic acid.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.8b01272. Material and methods; MS, 1D, and 2D NMR data for 1 and 2; X-ray structures of 2 (PDF) Accession Codes
CCDC 1828498 and 1828499 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
[email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB21EZ, UK; fax: + 44 1223 336033.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Hui Ming Ge: 0000-0002-0468-808X Ren Xiang Tan: 0000-0001-6532-6261 Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS The research was cofinanced by the NSFC (Grant Nos. 81530089, 21672101, 81573833, and 21661140001). REFERENCES
(1) Donia, M. S.; Fischbach, M. A. Science 2015, 349, 1254766. (2) Wilson, M. R.; Zha, L.; Balskus, E. P. J. Biol. Chem. 2017, 292, 8546−8552. (3) Underhill, D. M.; Iliev, I. D. Nat. Rev. Immunol. 2014, 14, 405− 416. (4) Limon, J. J.; Skalski, J. H.; Underhill, D. M. Cell Host Microbe 2017, 22, 156−165. (5) Hallen-Adams, H. E.; Suhr, M. J. Virulence 2017, 8, 352−358. 3744
DOI: 10.1021/acs.orglett.8b01272 Org. Lett. 2018, 20, 3741−3744
