Colossolactone J, a highly modified lanostane triterpenoid from a natural fruiting body of Ganoderma colossus

Abstract Colossolactone J (1), an undescribed lanostane triterpenoid was isolated from a natural fruiting body of Ganoderma colossus using silica gel column chromatography and preparative HPLC. Its structure was elucidated on the basis of the spectroscopic method. The absolute configuration was determined by the combination of the modified Mosher’s method and detailed NMR data analysis. Graphical abstract


Introduction
Ganoderma colossus (synonym, Tomophagus colossus) is a unique species in the woodrot mushrooms of the genus Ganoderma, which forms yellow colored basidiocarps on the upper surface, context soft with white to cream colored when fresh and turning light brown color after drying, pore surface white to cream color when fresh and turning dark brown after drying. This characteristic basidiome morphology is different from the majority of Ganoderma, including G. lucidum (known as lingzhi). Although it is not a medicinal mushroom, several highly modified lanostanes such as colossolactones A-H (Chen et al. 2016;Kleinw€ achter et al. 2001) and colossolactones I-VIII (El Dine, El Halawany, Nakamura, et al. 2008;) have been isolated from G. colossus. Most of them are 3,4-seco-lanostanes with rearranged B-ring, and they are structurally different from the lanostanes isolated from common

Results and discussion
Colossolactone J (1) was isolated as a colorless solid. Its molecular formula was determined to be C 32 H 44 O 8 on the basis of HRESIMS and 13 C NMR analyses. The 1 H and 13 C NMR, DEPT, and HSQC data of 1 supported the presence of three carbonyl carbons, three sp 2 tetrasubstituted carbons, three olefinic methines, two oxygenated tetrasubstituted carbons, three oxymethines, two sp 3 quaternary carbons, three methines, six methylenes, and seven methyl groups. The planar structure was elucidated by analyzing the COSY and HMBC spectra ( Figure S2). The NMR spectroscopic data for the ABCD-ring were very similar to those of the known co-metabolites, colossolactones G (Lakornwong et al. 2014) and H (Isaka et al. 2019; Figure S1), while the structure of the C-20-C-27 side-chain of 1 was different from any lanostanes from G. colossus. The presence of a 2-hydroxyisopropyl group (C-28/C-4/C-29), bonded to a downfield oxygenated carbon (d C 92.7, C-5), was revealed by the HMBC correlations from H 3 -28, H 3 -29, H-1 and H-19 to C-5, and from H 3 -28 and H 3 -29 to the tertiary alcohol carbon (d C 77.5, C-4). The location of an acetoxy group was indicated by the HMBC correlation from the oxymethine proton H-15 (d H 4.87) to the carbonyl carbon of the acetyl (d C 170.3). The C-20-C-27 side-chain was also elucidated on the basis of the COSY and HMBC data. The COSY spectrum indicated the spin system for C-21/C-20/C-22/C-23/C-24/C-25/C-27. The presence of a c-lactone was revealed by the HMBC correlations from the oxymethine proton H-23 (d H 4.42), H a -24 (d H 2.44), H b -24 (d H 2.08), H-25, and H 3 -27 to the carbonyl carbon C-26 (d C 179.9). The relative configuration of C-13/C-14/ C-15/C-17/C-20, the same as the co-metabolites colossolactones G and H, was confirmed by the NOESY correlations between H 3 -30/H a -12,  Figure S3).
The remaining question was the absolute configurations of the C-22, C-23, and C-25 chiral carbons. NOESY correlations between H-23/H 3 -27 (weak), H-23/H b -24, and H b -24/H 3 -27 indicated cis-relation of H-23, H b -24, and CH 3 -27. Therefore, the absolute configuration of the c-lactone should be either 23S,25R or 23 R,25S. The absolute configuration of the secondary alcohol carbon C-22 was determined using the modified Mosher's method (Ohtani et al. 1991). Compound 1 was converted to the a-methoxya-(trifluoromethyl)phenylacetic acid (MTPA) esters 2a and 2 b ( Figure S4 Therefore, these upfield shifts were not due to the shielding by the phenyl group of the MTPA. The 3 D structure of the (23S,25R)-isomer (2a) shown in Figure S5 suggested shielding by the carbonyl group of the MTPA (Chinthanom et al, 2021). This shielding effect should be sensitive to the slight difference of the ester conformation. The irregular Dd-value for H a -24 (-0.02 ppm), also for H-23 (-0.02 ppm), can be explained by the larger or similar magnitude of shielding effect by the ester carbonyl (both for 2a and 2 b) than that by the phenyl group of (R)-MTPA in 2 b. The 3 D structure of (23S,25R)-2a ( Figure S5) also suggested that H a -16 was located outside the strong shielding zone by the phenyl group of (S)-MTPA in 2a. In addition, the close distance between H a -16 and the oxygen atom in the c-lactone ring was suggested. Although not certain, a possible explanation for the unusual Dd-value for H a -16 (þ0.03 ppm) could be the weak deshielding of H a -16 by the lactone oxygen atom for both (23S,25R)-2a and (23S,25R)-2b. All NOESY correlations observed for 2a and 2 b were also consistent with the (23 R,25S)-isomer. However, considering the above discussed reason, we propose the (23S,25R)-configuration for colossolactone J (1). In the 1 H NMR spectrum of 1, H-22 resonated as a triplet (J ¼ 7.2 Hz), coupled with H-23 and 22-OH. Therefore, the major conformer should be similar to 2a and 2 b. A plausible major conformer of (23S,25R)-1, on the basis of its NOESY correlations, is shown in Figure S3. Consequently, the structure of colossolactone J (1) was elucidated as depicted in Figure 1.
Colossolactones A-H have been previously isolated and named (Chen et al. 2016;Kleinw€ achter et al. 2001), while there were later reported another series of trivial names of colossolactones I-VIII (Roman numerals) for different compounds (El Dine, El Halawany, Nakamura, et al. 2008;. To avoid confusion, a new trivial name of colossolactone I (alphabetical) was skipped for 1.

General experimental procedures
Optical rotations were determined using a JASCO P-2000 digital polarimeter. UV spectra were recorded on a JASCO V-730 spectrophotometer. FTIR spectra were acquired using a Bruker ALPHA spectrometer. NMR spectra were recorded on Bruker Avance III HD 400 MHz and 500 MHz spectrometers. ESITOF mass spectra were measured using a Bruker micrOTOF mass spectrometer. Merck Silica gel 60H (particle size, 90% <45 mm) was used for column chromatography.

Fungal material
The mushroom specimen was collected from a dead oil palm (Elaeis guineensis) trunk in a private plantation area, Nuea Khlong District, Krabi Province, Thailand, on May 16, 2016. The voucher mushroom collection was deposited in the BIOTEC Bangkok Herbarium & Fungarium as BBH 41152. This mushroom was identified as Ganoderma colossus by one of the authors (T.B.) based on the basidiocarps and culture morphology, and it was further confirmed by the sequence data of the ITS rDNA (GenBank accession numbers: MK007288 and ON307216).

Biological assays
The activity assay against Plasmodium falciparum K1 (multidrug-resistant strain) was performed in duplicate using the microculture radioisotope technique (Desjardins et al. 1979). Dihydroartemisinin (IC 50 0.0028 mM) and mefloquine (IC 50 0.059 mM) were used as positive controls. The antimycobacterial activity against Mycobacterium tuberculosis H37Ra was evaluated using the green fluorescent protein microplate assay (Changsen et al. 2003). Rifampicin (MIC 0.025 mg/mL) and isoniazid (MIC 0.0469 mg/mL) were used as positive controls. The cytotoxicity to Vero cells was also evaluated using the green fluorescent protein microplate assay (Hunt et al. 1999). Ellipticine was used as a standard compound for the assay (IC 50 3.2 mg/mL).

Conclusion
A new highly modified lanostane triterpenoid, colossolactone J (1), and sixteen known compounds were isolated from a natural fruiting body of G. colossus. To our knowledge, the C-20-C-27 side-chain of 1 is novel in the Ganoderma lanostanoids.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
Financial support from the Thailand Research Fund (grant No. DBG6280008) is gratefully acknowledged.