Abstract
Four new tricyclic macrolides, named shikinefragalides A (1), B (2), C (3) and D (4), were isolated by physicochemical (PC) screening from a static culture material of Stachybotryaceae sp. FKI-9632. Their structures were elucidated as new analogs of colletofragarones by MS and NMR analyses. Compounds 1 and 2 showed weak antimalarial activity and cytotoxicity.
Subject terms: Phenotypic screening, Malaria
Introduction
Numerous novel natural products have been discovered from fungal species. Some of them were developed as human medicines (e.g., penicillins, cephalosporins, statins and candins), animal drugs (e.g., penicillins and cephalosporins), and agrochemicals (e.g., afidopyropene) [1, 2]. Recent advances in whole-genome sequencing technology have revealed that the fungi have many biosynthetic gene clusters encoding unidentified secondary metabolites that surpass the number of metabolites identified so far [3, 4]. Thus, fungi are now re-recognized as one of the most important microorganisms as potential sources for new drugs and agrochemicals.
Physicochemical (PC) screening is a methodology to discover new natural compounds guided by physicochemical properties including molecular weight, molecular formula, UV profile and so on. Our research group has been searching for new fungal compounds by PC screening and discovered new fungal metabolites such as hatsusamides A and B [5] from Penicillium steckii FKJ-0213, pochoniolides A and B [6] from Pochonia chlamydosporia var. spinulospora FKI-7537, cipralphelin [7] from Penicillium brevicompactum FKJ-0123 and so on so far. After their initial discovery, many of them were found to exhibit useful biological activities.
During our recent PC screening on culture broths of 25 fungal strains by LC-DAD-ESI-MS analysis combined with dereplication by natural product database “Dictionary of Natural Products” [8], we selected a Stachybotryaceae sp. FKI-9632 strain as a producer of presumed new compounds. As a result of purification guided by LC-DAD-ESI-MS analysis, four new fungal tricyclic macrolides, named shikinefragalides A (1)-D (4), were isolated from a static culture of Stachybotryaceae sp. FKI-9632 (Fig. 1). In this paper, we report the isolation, structure elucidation, and biological activity of 1-4.
Fig. 1.
Structures of shikinefragalides A (1)-D (4)
Materials & methods
General experiments
The purification of 1-4 by an ODS column was conducted using YMC-gel ODS-A (150 µm, YMC Co., Kyoto, Japan). Preparative HPLC was performed using a Capcell pak C18 MG-II column (20 i.d. × 250 mm, Osaka Soda Co. Ltd., Osaka, Japan). LC-DAD-ESI-MS spectra were obtained using an AB Sciex Triple TOFTM 5600+ LC-MS/MS Systems (AB Sciex, Framingham, MA, USA). NMR spectra were obtained using a Varian XL-400 spectrometer (Agilent Technologies, CA, USA) or a JEOL JNM-ECA-500 (JEOL, Tokyo, Japan), with 1H NMR at 400 or 500 MHz and 13C NMR at 100 or 125 MHz in DMSO-d6 or CD3OD. The chemical shifts are expressed in ppm and are referred to DMSO-d6 (2.48 ppm) or CD3OD (3.31 ppm) in the 1H NMR spectra and to DMSO-d6 (39.5 ppm) or CD3OD (49.0 ppm) in the 13C NMR spectra. IR spectra (ATR) were taken on a FT-210 Fourier transform infrared spectrometer (Horiba Ltd., Kyoto, Japan). UV spectra were acquired with a Hitachi U-2800 spectrophotometer (Hitachi Ltd., Tokyo, Japan). Optical rotation was measured with a JASCO P-2200 polarimeter (JASCO Co., Tokyo, Japan). CD spectra were recorded with a J-720 circular dichroism spectrometer (JASCO).
Taxonomic studies of strain FKI-9632
Soil samples around the root of plants were collected from Shikine Island, Izu Islands, Tokyo, Japan in 2018. The soil samples were diluted with Winogradsky’s solution and spread on Czapek yeast extract agar (CYA) for cultivation. For isolating fungal species, that CYA was used with 50 mg l−1 rose bengal, 100 mg l−1 chloramphenicol and 100 mg l−1 kanamycin and kept at 25 °C for 7 days.
DNA extraction, polymerase chain reaction (PCR) amplification of the ITS region and sequencing of the strain FKI-9632 were conducted. PCR amplification products were fabricated using the QIAGEN® Fast Cycling PCR Kit (Qiagen Inc., Valencia, CA, USA). Sequencing products were purified using BigDye XTerminator Purification Kit (Applied Biosystems, Foster City, CA, USA), and samples were analyzed on an ABI PRISM 3130 Genetic Analyzer (Applied Biosystems). Contigs were assembled using the forward and reverse sequences with the SeqMan Pro program from the Lasergene 10 package (DNASTAR Inc., Madison, WI, USA).
Fermentation
Strain FKI-9632 was grown and maintained on an agar slant consisting of 0.1% glycerol, 0.08% KH2PO4, 0.02% K2HPO4, 0.02% MgSO4·7H2O, 0.02% KCl, 0.2% NaNO3, 0.02% yeast extract and 1.5% agar (adjusted to pH 6.0 before sterilization). A loopful of spores of the strain was inoculated into 100 ml of the GP seed medium consisting of 2.0% glucose, 0.5% Hipolypepton (Nihon Pharmaceutical Co., Tokyo, Japan), 0.2% yeast extract, 0.1% KH2PO4, 0.05% MgSO4·7H2O, and 0.1% agar (adjusted to pH 6.0 before sterilization) in a 500-ml Erlenmeyer flask. The flask was incubated on a rotary shaker (210 rpm) at 27 °C for 3 days. Fifty milliliters of the seed cultures were inoculated into each of two culture bags (Ulpack 47, Hokken Co. Ltd., Tochigi, Japan) containing a production medium (1 kg of water-sodden rice and 10 g of seaweed tea powder (Ito en Ltd., Tokyo, Japan)). Static fermentation was continued at 25 °C for 15 days.
Antimicrobial activity
The following microorganisms were used for evaluation of antimicrobial activity on a paper disc method: Bacillus subtilis KB 211 (ATCC 6633), Kocuria rhizophila KB 212 (ATCC 9341), Escherichia coli KB 213 (NIHJ), Xanthomonas oryzae KB 88, Candida albicans KF 1 (ATCC 64548) and Mucor racemosus KF 223 (IFO 4581). All compounds were prepared as 1 mg ml−1 MeOH solution. Each paper disk (diameter 6 mm, thin type, Advantec, Tokyo, Japan) impregnated with 10, 3, 1, 0.3, 0.1 and 0.03 µg of 1-4 was put into an agar plate, followed by incubation for 1–2 days at 37 °C for B. subtilis, K. rhizophila and E. coli KB 213 (NIHJ) or 27 °C for X. oryzae, C. albicans and M. racemosus.
In vitro cultivation of Plasmodium falciparum and antimalarial assay
In vitro cultivation and in vitro antimalarial activities against P. falciparum FCR3 (chloroquine-sensitive) and K1 (chloroquine-resistant) strains were evaluated, using the method described previously [9]. This study was approved by the “Kitasato Institute Hospital Research Ethics Committee (No12102)” because of the donation of human erythrocytes from volunteers.
Cytotoxic assay against MRC-5 cells
Cytotoxic assay against human fetal lung fibroblast MRC-5 cells was carried out as described previously [10].
Results
Identification of a fungal FKI-9632 strain as a producer of 1-4 by PC screening
PC screening was performed as shown in Scheme S1. In this PC screening, we used 25 fungal strains that were presumed to be unknown species with homology of less than 90% compared to known species from the genetic analysis. They were selected from 600 fungal strains of FKI-9401~10000. These 25 strains were cultured on two different media to get 50 cultured broths. The obtained cultured broths were analyzed by LC-DAD-ESI-MS analysis to collect data sets of molecular weights and UV spectra. Manual dereplication by natural product database “Dictionary of Natural Products” [8], on DVD (Ver. 26.2, CRC Press) using these data sets allowed us to find 4 strains as producers of presumed new compounds (Table S1). As a result of re-culturing these strains, reproducibility was obtained only for FKI-9632 strain which produce a presumed new compound detected as a [M + H]+ ion (m/z 405.1907) and a characteristic UV profile (λmax 264 and 337 nm) by LC-DAD-ESI-MS analysis (Fig. S1).
Taxonomy of the producing strain of Stachybotryaceae sp. FKI-9632
The fungal strain FKI-9632 was isolated from a soil sample around Smilax china in Shikine Island, Izu Islands, Tokyo, Japan. This strain produced verticillium-like conidiophores (Fig. S2). The internal transcribed spacer (ITS) region including 5.8 S ribosomal RNA gene sequence of FKI-9632 was compared to sequences in the GenBank database by BLASTN 2.12.0 analysis [11]. The sequence of it was 87.4% similar to the sequence of CBS 143444 (holotype of Sirastachys cyperacearum Crous & T.I. Burgess [12], GenBank accession number MH107917). Conidiophores of FKI-9632 were different from characteristics of genus Sirastachys. The producing strain FKI-9632 was identified with the Stachybotryaceae based on sequence analysis.
Isolation of shikinefragalides A-D
Shikinefragalides A-D were isolated from a 15-day-old static cultured material guided UV and MS profiles using LC-DAD-ESI-MS analysis (Scheme 1). They were purified under light-shielded condition within 3 days after preparation of a cultured material due to light-sensitivity. The stationary culture (2.0 kg) was extracted with 2.0 l of MeOH. After filtration in vacuo, the filtrate was evaporated in vacuo to make a 30% MeOH aq. solution. The obtained 30% MeOH aq. solution (0.6 l) was applied to an ODS column (150 ml resin, 55 i.d. × 55 mm; YMC Co., Kyoto, Japan). The column was eluted stepwise with 40% MeOH aq. (500 ml), 60% MeOH aq. (500 ml), 80% MeOH aq. (500 ml) and 100% MeOH (500 ml). The 80% MeOH aq. fraction including shikinefragalides was concentrated in vacuo under light-shielded condition and freeze-dried. A part (125.1 mg) of the obtained material (702.1 mg) underwent HPLC using a reverse-phase column (Capcell pak C18 MG-II, 20 i.d. × 250 mm; Osaka Soda Co. Ltd., Osaka, Japan) with an isocratic solvent system of CH3CN-H2O (40:60) at a flow rate of 7.0 ml min−1 detected by UV 323 nm. The four fractions with retention times of 26–28, 30–32, 43–45, and 48–50 min were collected (Fig. S3), evaporated in vacuo, and freeze-dried under light-shielded condition to afford shikinefragalides A (1, 23.9 mg), B (2, 14.3 mg), C (3, 6.8 mg) and D (4, 5.9 mg), respectively.
Scheme 1.
Fermentation and isolation of shikinefragalides A (1)–D (4)
Structure elucidation of shikinefragalides A-D
Physico-chemical properties of 1-4 are summarized in Table S4–1. Compounds 1-4 were expected to be analogs because they had similar physico-chemical properties (UV: λmax 264-269 and 334-345, and ESI-MS: m/z 405 or 421).
The structure of shikinefragalide B (2) was elucidated at first. The molecular formula of 2 was elucidated as C22H28O7 based on a [M + H]+ ion at m/z 405.1880 (calcd. m/z 405.1907) in HR-ESI-MS with 9 degrees of unsaturation. Analyses of 1H, 13C, and HSQC spectra measured in DMSO-d6 (Table 1, Figs. S4–2–4, 5 and 7) indicated the presence of one ester carbonyl carbon, ten sp2 olefinic carbons, six sp3 methine carbons including five oxygenated methines, two sp3 methylene carbons, one oxygenated sp3 tetrasubstituted carbon and two methyl carbons.
Table 1.
1H and 13C NMR data of 1 and 2 measured in DMSO-d6
| Shikinefragalide A (1) | Sihikinefragalide B (2) | |||
|---|---|---|---|---|
| position | δC | δH (int., mult., J in Hz) | δC | δH (int., mult., J in Hz) |
| 1 | 77.6 | 5.15 (1H, dd, 10.1, 3.5) | 77.5 | 5.13 (1H, dd, 10.0, 3.6) |
| 2 | 124.4 | 5.66 (1H, dd, 9,9, 3.5) | 124.4 | 5.62 (1H, dd, 10.0, 3.6) |
| 3 | 130.4 | 5.89 (1H, dd, 9.9, 5.6) | 130.3 | 5.87 (1H, dd, 10.0, 5.6) |
| 4 | 64.5 | 4.01 (1H, dd, 5.6, 5.4) | 64.5 | 4.00 (1H, dd, 5.6, 5.4) |
| 4-OH | 4.39 (1H, d, 5.4) | 4.37 (1H, d, 5.4) | ||
| 5 | 75.4 | – | 75.4 | – |
| 5-OH | 4.12 (1H, s) | 4.10 (1H, s) | ||
| 6 | 70.6 | 3.37 (1H, dd, 9.5, 7.1) | 70.6 | 3.37 (1H, dd, 9.5,7.2) |
| 6-OH | 4.09 (1H, d, 7.1) | 4.08 (1H, d, 7.2) | ||
| 7α | 41.7 | 1.39 (1H, dd, 13.8, 9.5) | 41.8 | 1.38 (1H, dd, 13.9, 9.5) |
| 7β | 2.06 (1H, dd, 13.9, 9.3) | 2.05 (1H, dd, 13.9, 9.5) | ||
| 8 | 68.5 | 3.74 (1H, m) | 68.5 | 3.76 (1H, m) |
| 8-OH | 4.46 (1H, d, 4.2) | 4.45 (1H, d, 4.2) | ||
| 9α | 45.3 | 1.90 (1H, br. d, 14.5) | 45.3 | 1.89 (1H, br. dd, 13.9, 1.8) |
| 9β | 1.61 (1H, ddd, 14.5, 11.1, 10.0) | 1.61 (1H, ddd, 13.9, 10.0, 9.5) | ||
| 10 | 69.9 | 4.54 (1H, dqd, 11.1, 6.1, 1.9) | 69.8 | 4.53 (1H, dqd, 10.0, 6.2, 1.8) |
| 11 | 165.1 | – | 165.2 | – |
| 12 | 108.4 | – | 107.8 | – |
| 13 | 157.8 | – | 158.0 | – |
| 14 | 46.6 | 3.63 (1H, d, 10.1) | 46.6 | 3.61 (1H, d, 10.0) |
| 15 | 21.5 | 1.29 (3H, d, 6.1) | 21.5 | 1.29 (3H, d, 6.2) |
| 16 | 118.4 | 6.36 (1H, d, 15.3) | 117.5 | 6.33 (1H, d, 15.5) |
| 17 | 129.1 | 7.02 (1H, dd, 15.3, 11.6) | 134.6 | 6.62 (1H, dd, 15.5, 11.2) |
| 18 | 125.8 | 5.97 (1H, dd, 11.6, 11.2) | 129.3 | 6.24 (1H, dd, 14.9, 11.2) |
| 19 | 133.8 | 6.13 (1H, dd, 11.3, 11.2) | 137.4 | 6.44 (1H, dd, 14.9, 10.7) |
| 20 | 126.9 | 6.58 (1H, dd, 14.5, 11.3) | 131.7 | 6.14 (1H, dd, 15.0, 10.7) |
| 21 | 133.6 | 5.84 (1H, dq, 14.5, 6.9) | 132.6 | 5.83 (1H, dq, 15.0, 6.9) |
| 22 | 18.3 | 1.79 (3H, d, 6.9) | 18.2 | 1.75 (3H, d, 6.9) |
The 1H-1H COSY analysis of 2 revealed three partial structures I-III as shown in Fig. 2a, H-14 (δH 3.61)/H-1 (δH 5.13)/H-2 (δH 5.62)/H-3 (δH 5.87)/H-4 (δH 4.00)/4-OH (δH 4.37) as I, H-6 (δH 3.37)/H2-7 (δH 1.38, 2.05)/6-OH (δH 4.08)/H-8 (δH 3.76) /8-OH (δH 4.45)/H2-9 (δH 1.61, 1.89)/H-10 (δH 4.53)/H3-15 (δH 1.29) as II, and H-16 (δH 6.33)/H-17 (δH 6.62)/H-18 (δH 6.24)/H-19 (δH 6.44)/H-20 (δH 6.14)/H-21 (δH 5.83)/H3-22 (δH 1.75) as III. The geometry of three double bonds in III was determined to be all E by large 1H-1H coupling constants (15.5, 14.9 and 15.0 Hz). Based on HMBC cross peaks from H-1 to C-13 (δC 158.0) and C-12 (δC 107.8), from H-14 to C-13 and C-5 (δC 75.4), from 5-OH (δH 4.10) to C-14 (δC 46.6) and C-4 (δC 64.5), from H-4 to C-5, and from H-16 to C-13, it was suggested 2 has a dihydrofuran-fused cyclohex-2-enol ring moiety including I, which is connected to a 1,3,5-heptatrienyl moiety III at C-13 position. Finally, the correlations in HMBC from H2-7 to C-5, from H-10 to C-11 (δC 165.2), and from H-14 to C-11, suggested the presence of a 10-membered macrolactone ring including II, which was connected at C-12 and C-5 positions of a dihydrofuran-fused cyclohexene ring moiety. From all results described above, 2 was elucidated as a new analog of colletofragarones [13], and designated shikinefragalide B (Fig. 1).
Fig. 2.
Structure elucidation of shikinefragalide B (2). (a) 1H-1H COSY and key HMBC cross peaks of 2. (b) Key ROESY correlations and 1H-1H coupling constants of 2
The relative configuration of 2 was established by ROESY and 1H-1H coupling constant analyses (Fig. 2b). Key ROESY correlations between H-1/H-14, H-14/5-OH, H-4/H-6, H-6/H-10, H-8/H-10, Hβ-9/Hβ-7, Hβ-9/H3-15 and Hβ-7/H-14 and coupling constants between 1/14 (10.0 Hz), 3/4 (5.6 Hz), 6/7β (0 Hz), 6/7α (9.5 Hz), 7β/8 (9.5 Hz) and 7α/8 (0 Hz) suggested the relative configuration of 2 to be 1S*, 4S*, 5S*, 6R*, 8S*, 10S*, 14S*. The absolute configuration of 2 was elucidated by circular dichroism (CD) spectra. Compared with reported CD spectrum of colletofragarone A2 [14], 2 had same positive cotton effect around 240 nm (Fig. S4–2–16), the absolute configuration of 2 should be 1S, 4S, 5S, 6R, 8S, 10S, 14S.
The molecular formula of shikinefragalide A (1) was determined to be C22H28O7, which was same to that of 2, based on a [M + H]+ ion at m/z 405.1866 (calcd. m/z 405.1907) in HR-ESI-MS. 1H, 13C, and HMQC spectra of 1 measured in DMSO-d6 (Table 1, Figs. S4–1–4, 5 and 7), resembled those of 2, except for the proton and carbon signals of a 1,3,5-heptatrienyl moiety. The interpretation of 1H-1H COSY and HMBC of 1 suggested 1 has a same planar structure to 2. (Fig. 3a). The difference was found in geometry of three double bonds, which was determined to be 16E, 18Z and 20E by 1H-1H coupling constants between 16/17 (15.3 Hz), 18/19 (11.2 Hz), and 20/21 (14.5 Hz), respectively. The relative configuration of 1 was established by ROESY and 1H-1H coupling constant analyses (Fig. 4a) same as 2. The absolute configuration of 1 was also elucidated to be 1S, 4S, 5S, 6R, 8S, 10S, 14S by CD spectrum (Fig. S4–1–16). From these observations described above, 1 was elucidated as a new analog, which was a 18Z isomer of 2 and designated shikinefragalide A (Fig. 1).
Fig. 3.
1H-1H COSY and key HMBC cross peaks of shikinefragalides A (1), C (3) and D (4). (a) shikinefragalide A (1), (b) shikinefragalide C (3) and shikinefragalide D (4)
Fig. 4.
Key ROESY correlations and 1H-1H coupling constants of shikinefragalides A (1), C (3) and D (4). (a) shikinefragalide A (1), (b) shikinefragalide C (3) and shikinefragalide D (4)
The molecular formula of shikinefragalide C (3) was determined to be C22H28O8, based on a [M + H]+ ion at m/z 421.1839 (calcd. m/z 421.1856) in HR-ESI-MS with 9 degrees of unsaturation, indicating the presence of an additional oxygen atom compared with that of 1. 1H, 13C, and HSQC spectra of 3 measured in DMSO-d6 (Table 2, Fig. S4–3–4, 5 and 7), resembled those of 1, except for the proton and carbon signals of a partial structure I. The interpretation of 1H-1H COSY and HMBC of 3 (Fig. 3b), suggested the 3 has a 3-hydroxylcyclohexan-1-one ring moiety, instead of a cyclohex-2-enol ring moiety of 1.
Table 2.
1H and 13C NMR data of 3 and 4 measured in DMSO-d6
| Shikinefragalide C (3) | Sihikinefragalide D (4) | |||
|---|---|---|---|---|
| position | δC | δH (int., mult., J in Hz) | δC | δH (int., mult., J in Hz) |
| 1 | 83.4 | 4.84 (1H, dd, 8.9, 3.3) | 83.8 | 4.81 (1H, dd, 8.9, 3.4) |
| 2 | 67.9 | 4.39 (1H, br. s) | 67.9 | 4.34 (1H, br. d, 7.1) |
| 2-OH | 4.08 (1H, br. d) | Not observed | ||
| 3α | 44.8 | 2.70 (1H, dd, 17.3, 7.1) | 44.8 | 2.64 (dd, 17.3, 7.1) |
| 3β | 2.90 (1H, d, 17.3) | 2.88 (d, 17.3) | ||
| 4 | 210.9 | – | 210.8 | – |
| 5 | 82.1 | – | 82.1 | – |
| 5-OH | 4.14 (1H, br. s) | Not observed | ||
| 6 | 68.7 | 3.28 (1H, dd, 9.3, 4.0) | 68.7 | 3.26 (1H, d, 8.9) |
| 6-OH | 4.25 (1H, br. d 4.0) | Not observed | ||
| 7α | 41.9 | 1.49 (1H, dd, 13.9, 9.3) | 41.9 | 1.46 (1H, dd, 14.2, 8.9) |
| 7β | 2.09 (1H, dd, 13.9, 9.3) | 2.09 (1H, dd, 14.2, 9.5) | ||
| 8 | 68.3 | 3.72 (1H, m) | 68.2 | 3.70 (1H, m) |
| 8-OH | 4.55 (1H, br. s) | Not observed | ||
| 9α | 42.3 | 1.89 (1H, br. d, 14.6) | 41.9 | 1.87 (1H, br. d, 14.4) |
| 9β | 1.63 (1H, ddd, 14.6, 8.0, 8.0) | 1.61 (1H, ddd, 14.4, 8.0, 8.0) | ||
| 10 | 70.2 | 4.80 (1H, m) | 70.1 | 4.78 (1H, m) |
| 11 | 163.8 | – | 163.8 | – |
| 12 | 103.5 | – | 102.8 | – |
| 13 | 163.8 | – | 163.9 | – |
| 14 | 52.8 | 3.80 (1H, d, 8.9) | 52.7 | 3.77 (1H, d, 8.9) |
| 15 | 20.9 | 1.29 (3H, d, 6.4) | 20.9 | 1.28 (3H, d, 6.4) |
| 16 | 117.7 | 6.77 (1H, d, 15.3) | 116.7 | 6.74 (1H, 15.3) |
| 17 | 133.1 | 7.18 (1H, dd, 15.3, 11.6) | 138.6 | 6.75 (1H, dd, 15.3, 9.0) |
| 18 | 125.6 | 6.02 (1H, dd, 11.6, 11.1) | 129.1 | 6.29 (1H, dd, 15.0, 9.9) |
| 19 | 136.0 | 6.24 (1H, dd, 11.4, 11.1) | 139.8 | 6.54 (1H, dd, 15.0, 10.8) |
| 20 | 127.0 | 6.63 (1H, dd, 15.1, 11.4) | 131.7 | 6.20 (1H, dd, 15.1, 10.8) |
| 21 | 135.0 | 5.93 (1H, dq, 15.1, 6.8) | 134.3 | 5.90 (1H, dq, 15.1, 6.9) |
| 22 | 18.4 | 1.79 (3H, d, 6.8) | 18.4 | 1.76 (3H, d, 6.9) |
The relative configuration of 3 was established by ROESY and 1H-1H coupling constant analyses (Fig. 4b). Key ROESY correlations between H-1/H-14, H-14/5-OH, H-6/H-10, H-8/H-10, Hβ-9/Hβ-7, Hβ-9/H3-15 and Hβ-7/H-14 and coupling constants between 1/14 (8.9 Hz), 1/2 (3.3 Hz), 2/3α (7.1 Hz), 2/3β (0 Hz), 6/7β (0 Hz), 6/7α (9.3 Hz), 7β/8 (9.3 Hz) and 7α/8 (0 Hz) suggested the relative configuration of 3 to be 1R*, 2R*, 5S*, 6R*, 8S*, 10S*, 14S*. The absolute configuration of 3 was also elucidated by the comparison with reported CD spectrum of colletoin B [14] to be 1R, 2R, 5S, 6R, 8S, 10S, 14S (Fig. S4–3–10). From all results described above, 3 was elucidated as a new analog, and designated shikinefragalide C (Fig. 1).
The molecular formula of shikinefragalide D (4) was determined to be C22H28O8, which was same to that of 3, based on a [M + H]+ ion at m/z 421.1845 (calcd. m/z 421.1856) in HR-ESI-MS. 1H, 13C, and HMQC spectra of 4 measured in DMSO-d6 (Table 2, Figs. S4–4–4, 5 and 7), resembled those of 3, except for the proton and carbon signals of a 1,3,5-heptatrienyl moiety. The interpretation of 1H-1H COSY and HMBC of 4 suggested 4 has a same planar structure to 3. (Fig. 3c). The difference was observed in geometry of three double bonds, which was determined to be all E by large 1H-1H coupling constants between 16/17 (15.3 Hz), 18/19 (15.0 Hz), and 20/21 (15.1 Hz), respectively. The relative configuration of 4 was established by ROESY and 1H-1H coupling constant analyses (Fig. 4c) same as 3. The absolute configuration of 4 was also elucidated to be 1R, 2R, 5S, 6R, 8S, 10S, 14S by CD spectrum (Fig. S4–4–10). From these results described above, 4 was elucidated as a new analog, which was a 18E isomer of 3 and designated shikinefragalide D (Fig. 1).
Biological activity
Compounds 1 and 2, isolated with over 10 mg, were tested for in vitro antimalarial activity against both a chloroquine-sensitive FCR3 strain and a chloroquine-resistant K1 strain of P. falciparum, as well as for cytotoxicity in human MRC-5 cells (Table 3, Fig S6). Compound 1 displayed in vitro weak antimalarial activity against both a chloroquine-sensitive P. falciparum FCR3 strain and a chloroquine-resistant P. falciparum K1 strain with IC50 values of 62.7 and 186.6 µM, respectively. Compound 2 showed weaker antimalarial activity with IC50 values of 104.5 and 209.4 µM than 1. In addition, 1 and 2 showed weak cytotoxicity against human MRC-5 cells, with IC50 values of 48.2 and 76.3 µM, respectively, indicating selectivity indices (IC50 values against MRC-5 cells/IC50 values against P. falciparum strains) ranging from 0.3 to 0.8.
Table 3.
In vitro antimalarial activity, cytotoxicity, and selectivity index of 1, 2 and chloroquine (an antimalarial drug)
| IC50 [µM] | |||||
|---|---|---|---|---|---|
| Compound | Antimalarial activity | Cytotoxicity | Selectivity index (SI) | ||
| K1 strain* | FCR3 strain** | (MRC-5) | MRC-5/K1 | MRC-5/FCR3 | |
| Shikinefragalide A (1) | 186.6 | 62.7 | 48.2 | 0.3 | 0.8 |
| Shikinefragalide B (2) | 209.4 | 104.5 | 76.3 | 0.4 | 0.7 |
| Chloroquine*** | 0.577 | 0.047 | 58.2 | 101 | 1,238 |
*chloroquine-resistant strain
**chloroquine-sensitive strain
***drug commonly used to treat malaria
Compounds 1–4 demonstrated no antimicrobial activity against B. subtilis, K. rhizophila, E. coli, X. oryzae, C. albicans and M. racemosus at 10 µg on a paper disc method.
Discussion
In this study, four new fungal macrolides, named shikinefragalides A (1)–D (4), were discovered from a static cultured material of a Stachybotryceae sp. FKI-9632 by PC screening. To data, there have been several analogs reported such as colletofragarones [13], colletoins [14], and dictyosphaeric acids [15] (Fig. S5). To our best knowledge, the biosynthetic pathway of colletofragarone-related compounds has not been clarified yet. The main differences among colletofragarone-related compounds and 1-4 are degrees of oxidation at C-2, C-3, C-4, C-5, C-6, and C-8 positions, the decipher and comparison of their biosynthetic pathway would reveal responding enzymes for oxidation.
Colletofragarone A2 and colletoin A were reported to show cytotoxicity against Saos-2 (p53R175H) cells, with IC50 values of 0.35 and 0.36 µM, respectively, whereas colletoins B and C were less active, with IC50 values of 21 and 12 µM, respectively [14]. Dictyosphaeric acid A was reported to have antibacterial activity against methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococci, and C. albicans, while dictyosphaeric acid B was reported to show no significant antibacterial activity [15, 16]. These reports suggest that an α, β-unsaturated carbonyl group in their six-membered ring is essential for their activity (Fig. S5) [14, 16]. Compounds 1-4 without an α, β-unsaturated carbonyl group, showed no antibacterial activity, which also support its importance.
This is the first report of antimalarial activity of colletofragarone-related compounds against P. falciparum. Compounds 1 and 2 showed weak in vitro antimalarial activity. Recently, dictyosphaeric acid A was passed in silico screening against COVID-19 and was reported to be a potential inhibitor of the host enzyme, transmembrane protease serine 2 (TMPRSS2) [17]. Since 1-4 are sensitive to light and gradually degraded even under light-shielded condition, they are not suitable for long-term storage, and it would be difficult to evaluate their activity against various targets. Therefore, it is necessary to chemically modify colletofragarone-related compounds including 1-4 to improve their stability for pharmaceutical applications by inducing them.
In this study, we used 25 unidentified fungi having less than 90% homology with known species for sources of PC screening. As a result, we could select 4 producers of presumed new compounds with higher rate of 16% than those of identified species (less than 5%) in our research team, suggesting that fungal strains with high novelty are very good sources for PC screening to improve efficacy of discovery new fungal secondary metabolites.
Supplementary information
Acknowledgements
We are grateful to Distinguished Emeritus Professor Satoshi Ōmura (Kitasato University) for his helpful support and valuable guidance and suggestions. We thank Dr. K. Nagai and Ms. N. Sato (School of Pharmacy, Kitasato University) for various instrumental analyses, and Mr. T. Tokiwa for taking fungal photo and support on paper disc evaluation. This study was also partially supported by the Platform Project for Supporting Drug Discovery and Life Science Research (Basis for Supporting Innovative Drug Discovery and Life Science Research (BINDS)) from the Japan Agency for Medical Research & Development (AMED) under Grant Numbers JP19am0101096.
Compliance with ethical standards
Conflict of interest
The authors declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Rima Koike, Yoshihiro Watanabe
Supplementary information
The online version contains supplementary material available at 10.1038/s41429-022-00512-y.
References
- 1.Newman DJ, Cragg GM. Natural products as sources of new drugs over the nearly four decades from 01/1981 to 09/2019. J Nat Prod. 2020;83:770–803. doi: 10.1021/acs.jnatprod.9b01285. [DOI] [PubMed] [Google Scholar]
- 2.Kandasamy R, London D, Stam L, von Deyn W, Zhao X, Salgado VL, et al. Afidopyropen: new and potent modulator of insect transient receptor potential channels. Insect Biochem Mol. 2017;84:32–9. doi: 10.1016/j.ibmb.2017.03.005. [DOI] [PubMed] [Google Scholar]
- 3.Meng X, Fang Y, Ding M, Zhang Y, Jia K, Li Z, et al. Developing fungal heterologous expression platforms to explore and improve the production of natural products from fungal biodiversity. Biotechnol Adv. 2021;54:107866. doi: 10.1016/j.biotechadv.2021.107866. [DOI] [PubMed] [Google Scholar]
- 4.Keller NP. Fungal secondary metabolism: regulation, function, and drug discovery. Nat Rev Microbiol. 2019;17:167–80. doi: 10.1038/s41579-018-0121-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Matsuo H, Hokari R, Ishiyama A, Iwatsuki M, Higo M, Nonaka K, et al. Hatsusamides A and B: two new metabolites produced by the deep-sea-derived fungal strain Penicillium steckii FKJ-0213. Mar Drugs. 2020;18:513. doi: 10.3390/md18100513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Miyano R, Matsuo H, Nonaka K, Mokudai T, Niwano Y, Shiomi K, et al. Pochoniolides A and B, new antioxidants from the fungal strain Pochonia chlamydosporia var. spinulospora FKI-7537. J Biosci Bioeng. 2018;126:661–6. doi: 10.1016/j.jbiosc.2018.05.003. [DOI] [PubMed] [Google Scholar]
- 7.Matsuo H, Mokudai T, Higo M, Nonaka K, Nagano Y, Hagahama T, et al. Cipralphelin, a new anti-oxidative N-cinnamoyl tripeptide produced by the deep sea-derived fungal strain Penicillium brevicompactum FKJ-0123. J Antibiot. 2019;72:775–8. doi: 10.1038/s41429-019-0208-6. [DOI] [PubMed] [Google Scholar]
- 8.Dictionary of Natural Products, on DVD (Ver. 26.2) CRC Press 2021.
- 9.Otoguro K, Kohana A, Manabe C, Ishiyama A, Ui H, Shiomi K, et al. Potent antimalarial activities of polyether antibiotic, X-206. J Antibiot. 2001;54:658–63. doi: 10.7164/antibiotics.54.658. [DOI] [PubMed] [Google Scholar]
- 10.Otoguro K, Ui H, Ishiyama A, Arai N, Kobayashi M, Takahashi Y, et al. In vitro antimalarial activities of the microbial metabolites. J Antibiot. 2003;56:322–4. doi: 10.7164/antibiotics.56.322. [DOI] [PubMed] [Google Scholar]
- 11.Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997;25:3389–402. doi: 10.1093/nar/25.17.3389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Crous PW, Schumacher RK, Wingfield MJ, Akulov A, Denman S, Roux J, et al. New and interesting fungi. 1. Fungal Syst Evol. 2018;1:169–216. doi: 10.3114/fuse.2018.01.08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.lnoue M, Takenaka H, Tsurushima T, Miyagawa H, Ueno T. Colletofragarones A1 and A2, novel germination self-inhibitors from the Fungus Colletotrichum fragariae. Tetrahedron Lett. 1996;37:5731–4. doi: 10.1016/0040-4039(96)01212-9. [DOI] [Google Scholar]
- 14.Sadahiro Y, Hitora Y, Tsukamoto S. Colletofragarone A2 and colletoins A–C from a fungus Colletotrichum sp. decrease mutant p53 levels in cells. J Nat Prod. 2021;84:3131–7. doi: 10.1021/acs.jnatprod.1c00913. [DOI] [PubMed] [Google Scholar]
- 15.Bugni TS, Janso JE, Williamson RT, Feng X, Bernan VS, Greenstein M, et al. Dictyosphaeric acids A and B: new decalactones from an Undescribed Penicillium sp. obtained from the alga Dictyosphaeria versluyii. J Nat Prod. 2004;67:1396–9. doi: 10.1021/np049973t. [DOI] [PubMed] [Google Scholar]
- 16.Burns AR, McAllister GD, Shanahan SE, Taylor RJK. Total synthesis and structural reassignment of (+)-dictyosphaeric acid A: A tandem intramolecular michael addition/alkene migration approach. Angew Chem Int Ed. 2010;49:5574–7. doi: 10.1002/anie.201002416. [DOI] [PubMed] [Google Scholar]
- 17.Rahman N, Basharat Z, Yousuf M, Castaldo G, Rastrelli L, Khan H. Virtual screening of natural products against type II transmembrane serine protease (TMPRSS2), the priming agent of Coronavirus 2 (SARS-CoV-2) Molecules. 2020;25:2271. doi: 10.3390/molecules25102271. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.





