key: cord-0054197-45vgoyv4 authors: Abdullah, Syahriel; Oh, Yoon Sin; Kwak, Min-Kyu; Chong, KhimPhin title: Biophysical characterization of antibacterial compounds derived from pathogenic fungi Ganoderma boninense date: 2020-12-23 journal: J Microbiol DOI: 10.1007/s12275-021-0551-8 sha: 6bfc76da663a46b3839ac1d4600c03ce7e4b90e0 doc_id: 54197 cord_uid: 45vgoyv4 There have been relatively few studies which support a link between Ganoderma boninense, a phytopathogenic fungus that is particularly cytotoxic and pathogenic to plant tissues and roots, and antimicrobial compounds. We previously observed that liquid-liquid extraction (LLE) using chloroformmethanol-water at a ratio (1:1:1) was superior at detecting antibacterial activities and significant quantities of antibacterial compounds. Herein, we demonstrate that antibacterial secondary metabolites are produced from G. boninense mycelia. Antibacterial compounds were monitored in concurrent biochemical and biophysical experiments. The combined methods included high performance thin-layer chromatography (HPTLC), gas chromatography-mass spectrometry (GC-MS), high-performance liquid chromatography (HPLC), fourier transform infrared (FTIR), and nuclear magnetic resonance (NMR) spectroscopy. The antibacterial compounds derived from mycelia with chloroform-methanol extraction through LLE were isolated via a gradient solvent elution system using HPTLC. The antibacterial activity of the isolated compounds was observed to be the most potent against Staphylococcus aureus ATCC 25923 and multidrug-resistant S. aureus NCTC 11939. GC-MS, HPLC, and FTIR analysis confirmed two antibacterial compounds, which were identified as 4,4,14α-trimethylcholestane (m/z = 414.75; lanostane, C(30)H(54)) and ergosta-5,7,22-trien-3β-ol (m/z = 396.65; ergosterol, C(28)H(44)O). With the aid of spectroscopic evaluations, ganoboninketal (m/z = 498.66, C(30)H(42)O(6)), which belongs to the 3,4-seco-27-norlanostane triterpene family, was additionally characterized by 2D-NMR analysis. Despite the lack of antibacterial potential exhibited by lanostane; both ergosterol and ganoboninketal displayed significant antibacterial activities against bacterial pathogens. Results provide evidence for the existence of bioactive compounds in the mycelia of the relatively unexplored phytopathogenic G. boninense, together with a robust method for estimating the corresponding potent antibacterial secondary metabolites. ELECTRONIC SUPPLEMENTARY MATERIAL: Supplementary material is available for this article at 10.1007/s12275-021-0551-8 and is accessible for authorized users. The intricate cross-talk between the bioactive phytochemicals and antimicrobial constituents of certain basidiomycetes has recently been highlighted, but their significance has not yet been evaluated (Schwan, 2012) . Some of basidiomycetes have attracted considerable attention owing to the widespread increase in the prevalence of natural and nosocomial infections, along with antibiotic resistance. As a major class of higher fungi, they can adapt to environmental conditions, and as mushrooms, they potentially possess excellent immunomodulatory functions (Anke, 1989; Ma et al., 2014) . Specifically, many species from the genus Ganoderma of the Ganodermataceae family have been steadily suggested to exhibit antimicrobial activities by the presence of bioactive compounds, such as triterpenoids, colossolactones, and polysaccharides (Ofodile et al., 2012) . Ganoderma boninense, also known as bracket fungi, belongs to the genus of polypore fungi Ganoderma of the Ganodermataceae (Polyporales) family (Peng et al., 2014; Richter et al., 2015) . Meanwhile, there remains undiscovered potential for bioactivity and functionality in G. boninense. This is because this phytopathogenic G. boninense is recognized as the most virulent and predominant cause of basal stem rot (Ho and Nawawi, 1985) , which is one of the most restrained diseases affecting the oil palm industry (Corley et al., 2008) . Thus, due to its notorious reputation, G. boninense has yet to have its medicinal potential sufficiently clarified. The sole evidence of most observed bioactive secondary metabolites is predominantly restricted to lanostane-type triterpenoids and their derivatives in G. boninense (Gong et al., 2019) . Thus, there is limited information regarding G. boninense metabolites such as 3,4-seco-27-norlanostane triterpenes. Lanostanes from G. boninense commonly include 3 ganoboninketals and 6 ganoboninones with a tetradecahydrobenzo [4, 5] indeno [1,7a-c] furan backbone (Ma et al., 2014 . To a greater or lesser degree, these compounds derived from G. boninense exhibit antimicrobial activities, such as anti-Candida (Daruliza et al., 2012) , anti-plasmodial , and antibacterial activity (Ismail et al., 2014) . However, most studies on G. boninense excessively emphasize its pathogenicity and its consequent negative effects on oil palm † These authors contributed equally to this work. (Chong et al., 2011 (Chong et al., , 2012 (Chong et al., , 2013 , rather than exploring its medicinal potential. For this reason, convincing experimental evidence on the biologically active secondary metabolites of G. boninense is still lacking. Despite substantial developments in aspects of extraction and separation techniques in the case of drug discovery studies, isolation of natural products is still an exceptionally challenging process. Undoubtedly, comprehensive structural verification data regarding the metabolites of novel drugs and/or biologically active secondary metabolites have been developed via hybrid methods such as gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), two-dimensional nuclear magnetic resonance spectroscopy (2D-NMR), or interfacing liquid chromatography with parallel NMR and mass spectrometry (LC-NMR-MS) (Schlotterbeck and Ceccarelli, 2009; Sarker and Nahar, 2012) . These biophysical approaches enable scientific and commercial applications in natural products, combinatorial chemistry, and drug metabolism studies. Full chemical structures, including the stereochemistry of new compounds, most likely need to be isolated, as this data requires highly purified compounds. For example, 2D-NMR spectra of small molecules can be obtained with less than 100 μg (Molinski, 2010; Mahrous and Farag, 2015) . Furthermore, in vitro and in vivo bioassays required for evaluating biological activities have been performed, followed by a purification process to exclude interference and impurities (Bucar et al., 2013) . Finally, reference standards for the quality control of natural product medicines largely depend on isolated compounds with documented purity. In recent years, natural products have experienced a renaissance in drug discovery programs, mainly due to their drug-like properties and their superior chemical diversity over synthetic compound libraries (Vetter et al., 2011) . In the present study, we demonstrated and evaluated techniques for the isolation, identification, and characterization of antibacterial compounds derived from G. boninense mycelia. We have previously evaluated the efficacy of liquidliquid extraction (LLE) for the isolation of potential antibacterial compounds (Abdullaha et al., 2020) . To detect and isolate these antibacterial compounds, we employed biochemical methods, including high performance thin-layer chromatography (HPTLC), GC-MS, infrared (IR), and 2D-NMR spectroscopy. The revealed antibacterial compounds were respectively tested against pathogenic Gram-positive and Gramnegative bacteria. Our findings propose the employment of a new, more robust method for estimating the potent antibacterial secondary metabolites in the mycelia of G. boninense, a relatively unexplored phytopathogenic fungal strain. All fungal and bacterial strains used in this study are listed in Supplementary data Table S1 . Mycelia and fruiting bodies were obtained from oil palm (Elaeis guineensis Jacq.) infected by G. boninense, with the infection scores of the selected palm trees being assessed according to the disease severity index of G. boninense-infected oil palms . Ganoderma boninense collection was conducted at Langkon Plantation, Sawit Kinabalu, Sabah, Malaysia. Based on our previous experiments regarding the relatively higher antibacterial properties of G. boninense mycelia (Abdullaha et al., 2020) , G. boninense mycelium cultured on potato dextrose agar (PDA), designated as GBMA, was used in this study. Ganoderma boninense was cultured to obtain mycelia as previously described, with modifications (Abdullah et al., 2018) . Ganoderma boninense mycelia were inoculated on PDA plates and were routinely cultured at 28°C for 14 days. The resulting mycelial plugs were transferred to PDA and PDB plates and were cultured for further experimentation. Mycelial plugs were loaded onto PDA agar media and were again cultured for 7 days at 28°C. Additionally, five mycelial plugs were inoculated into PDB liquid medium and were cultured for 7 days at 28°C. Unless otherwise stated, each strain for the biochemical experiments was routinely cultured in nutrient broth (NB) (Merck) at 37°C. The grown mycelia and/or fruiting bodies were harvested via centrifugation at 10,000 × g for 5 min, washed with 200 mM phosphate-buffered saline (PBS); pH 7.0, and resuspended in 1 ml 30% glycerol solution. Six bacterial strains were used for the antibacterial activity assays. Gram-positive bacteria, including S. aureus American Type Culture Collection (ATCC 25923), multidrug-resistant S. aureus (MRSA) National Collection of Type Culture (NCTC 11939), and S. pyogenes ATCC 19615, were used. In contrast, P. aeruginosa ATCC 9027, K. pneumoniae ATCC 1705, and E. coli ATCC 35218 were used as Gram-negative bacteria. ATCC bacterial strains and MRSA (NCTC 11939) were purchased from Sigma-Aldrich or Oxoid Ltd., respectively. Prior to antibacterial activity assays using solvent-extracted G. boninense, the exponential phase of the growth cycle of all bacterial strains was determined as previously described (Brannan, 2006) . To isolate unknown compounds, thin-layer chromatography (TLC) was utilized as previously reported (Rajauria and Abu-Ghannam, 2013) . A high-performance thin-layer chromatography (HPTLC) system (CAMAG) equipped with a sample applicator was used for compound purification. Approximately 100 μl of the CM and WM fractions was used for the band spraying application, with three replicates on each 10 cm × 10 cm preparative-TLC (PTLC) glass silica gel plate (Merck, Gel 60, 1 mm thick) in the stationary phase. The combination of PTLC and HPTLC was used to prepare the biological samples and for their separation. PTLC plates were developed in a twin trough glass separation chamber using a gradient solvent separation system, as summarized in Supplementary data Table S2 . The bands derived from the CM and WM fractions on the PTLC plates were visualized under visible (white), short UV (254 nm), and long UV (366 nm) lights using a visualizer (CAMAG Visualizer). The movement of each separated band from the two fractions was expressed as its retention factor (Rf). pHPTLC separation of the CM fraction was performed with 68 plate replicates. Thereafter, 18 plates were used for the pHPTLC antibacterial bioassay on the six bacterial samples whereas the other 50 plates were used for antibacterial compound isolation. The pHPTLC antibacterial bioassay protocol was adopted as proposed previously (Valle et al., 2016) , with minor modifications. Six bacterial strains grown until the exponential phase were harvested, washed, resuspended, and adjusted with NB until 2-4.5 × 10 5 (CFU/ml) was reached, according to the guidelines of the CLSI (CLSI, 2011). Approximately 1 ml of each bacterial sample was inoculated and was mixed into 30 ml of nutrient agar (NA), and this was then poured onto a PTLC plate containing both fractions. GC-MS analysis of the potential antibacterial compounds was performed as previously described (Ma et al., 2014) , with minor modifications. The separated bands that exhibited antibacterial potential (Rf = 0.4) were collected, transferred into a 15 ml polytetrafluoroethylene (PTFE) centrifuge tube, to which 3 ml of methanol:chloroform (1:1) was added. This was vortexed and centrifuged at 5,000 rpm for 5 min to separate the antibacterial compounds from PTLC silica gel. For GC-MS analysis, the isolated compounds were diluted to 200 ppm, filtered with a 0.45 μm PTFE filter, and then transferred into a 2 ml borosilicate amber glass (Agilent Technologies). GC-MS analysis of the isolated compounds was performed using an Agilent 2000 Electron Spray ionization GC-MS equipped with a fused silica 15 m × 0.2 mm ID × 1 μm capillary column. The ionization voltage was 70 eV. The raw data obtained from the GC-MS were converted into vendorindependent network common data format (NetCDF) format. MZmine software version 2.23 (http://mzmine.github.io) was used for data processing. Agilent MassHunter software with the Find Molecular Formula Algorithm function (MFG) was also used for molecular formula prediction based on mass fragmentation. For identification, the NIST MS Search-v.2.3 (2017) software with NIST 17 EI Library Version was utilized. Prior to sPHPLC analysis, lyophilized compounds from pHPTLC were dissolved in 100 mg/ml acetonitrile. sPHPLC separation was performed as previously described (Ma et al., 2014) with some modifications. A Shimadzu LC-6AD HPLC system was used for analysis, with an injection volume of 5 ml using a preparative reverse phase (RP) Shim-pack PRC-ODS (250 × 20 mm id, 5 μm, Shimadzu) with a wavelength of 282 nm. The mobile phase consisted of acetonitrile (A) and 0.1% TFA in water (B) using a gradient elution of 15−40% A at 0−10 min, 40−60% A at 10−25 min, 60−100% A at 25−45 min, and 100% A at 45−55 min. The flow rate was maintained at 3 ml/min. A Shimadzu FRC-10A Fraction Collector was utilized to collect specific compounds at specific sPHPLC retention times (Rt). Compounds identified earlier using GC-MS were spiked with standards to confirm the identity of the isolated compounds. The ample fraction was collected through a series of sPHPLC and was freeze-dried for further experimental use. The FTIR spectra of the isolated compounds from sPHPLC were acquired at room temperature using a Perkin Elmer 1600 FTIR spectrometer. The sampling station was equipped with an overhead attenuated total reflection accessory (ATR) comprising transfer optics within the chamber through which infrared radiation was directed to a detachable zinc selenide crystal mounted into a shallow trough for sample containment. The crystal geometry was a 45° parallelogram with mirrored angled faces, with 10 nominal internal reflections. Approximately 64 scans were co-added at a nominal resolution of 4/cm in the spectral region of 4,000-650/cm. Singlebeam spectra of the samples were collected and rationed against the background and presented as absorbance units. Samples were applied to the ATR crystal to obtain the total crystal coverage; the crystal was cleaned between samples using acetone and dried with a lens cleaning tissue. The cleaned crystal was carefully examined visually to ensure that no residue from the previous sampling was retained on the crystal surface and was checked with the background spectrum. FTIR spectra for known compounds based on MS analysis were cross-referenced with their respective standard for further identity confirmation. NMR spectral studies for 2D-NMR analysis on compound C were performed at the Analytical Biochemistry Research Centre (ABrC), Universiti Sains Malaysia using an NMR 700 MHz ASCEND TM spectrometer. Heteronuclear multiple-bond correlation spectroscopy (HMBC), heteronuclear single quantum coherence spectroscopy (HSQC), NMR correlation spectroscopy (COSY), nuclear Overhauser effect spectroscopy (NOESY), and UV circular dichroism (CD) for both experimental and calculated Electronic Circular Dichroism (ECD), as well as in silico structural analysis using Merck Molecular Force Field (MMFF94 variant using Conflex calculation), were subsequently performed for the elucidation of the ABrC compound structure. The antibacterial activities of the identified compounds were evaluated using a disc diffusion assay as guided by CLSI (CLSI, 2011). Each compound was prepared at 1 mg/ml in 10% DMSO to create the stock solution. Approximately 100 μl (100 μg, 1 μg/μl of concentration) of each compound was loaded into their respective discs. For the negative control discs, 10% DMSO was loaded in order to separate the discs. Chloramphenicol (30 μg) was used as the positive control. Media and bacterial preparation for the antibacterial bioassay was performed according to methods previously described (Abdullaha et al., 2020) . The plates were then incubated at 37°C for 12-48 h in an incubator, and the antibacterial activity of the identified compounds was determined by measuring the size of the inhibition zone (mm). A comparative antibacterial experiment was performed using various culture media, namely, tryptic soy agar (TSA), Muller-Hinton agar, NA, and Luria-Bertani agar (LBA). Bacterial cultures were incubated at 37°C until the exponential growth phase. Antibacterial activity was observed using conventional disc diffusion assay, as illustrated in CLSI guidelines (CLSI, 2011). Approximately 0.1 ml of bacterial culture was adjusted to 1 × 10 8 CFU/ml prior to inoculation onto nutrient agar plates. Sterilized Whatman ® No. 3 paper (6 mm diameter discs) was used for the assay. Before loading 50 μl (5 mg) of the Ganoderma extracts, the concentration of each extract sample was set to 100 mg/ml. All lyophilized extracts of the Ganoderma samples were dissolved in 10% DMSO by dilution with tripledistilled water, except for the water-extracted samples. For the positive control experiment, 50 μl of 10% DMSO was added to the discs and supplemented with tetracycline, chloramphenicol, or ampicillin; chloroform was used as the negative control. The diameter of the zone of inhibition (mm) around the disc was measured after cultivation as described above and was compared with the control. Data are expressed as the mean ± standard deviation (SD) of at least three independent experiments. The statistical significance of the differences was preliminarily evaluated using the Student's t-test in Microsoft Office Excel 2016. To evaluate the statistical significance of the experimental parameters, data were compared using one-way analysis of variance, followed by post-hoc tests such as Tukey's test, Duncan's multiple range test, and least significance difference. The level of significance was set at p = 0.05, and statistical analyses were performed using SPSS software, ver. 21 (SPSS Inc.). We have previously evaluated three culture type-dependent G. boninense, including fruiting bodies and mycelia cultured on potato dextrose agar (PDA) plates and potato dextrose broth (PDB), respectively, and the GBMA has been revealed to exhibit the most significant antibacterial activity in our previous study (Abdullaha et al., 2020) . LLE extraction shows improved extraction efficiency of secondary metabolites. This strategy for compound extraction implies that large amounts of polar substances can contribute to higher yields in more polar solvents (Seyd et al., 2012) . While the obtained two fractions consist of water-methanol (WM) and chloroform-methanol (CM) fractions, the CM contain more potent antibacterial compounds. Thus, experiments using a combination of chloroform and methanol were used to obtain appropriate and ideal solubility strength in extracting the potent antibacterial secondary metabolites in this study. Compared to WM, the CM is estimated to be a more non-polar organic solvent. This characteristics of the combined solvent suggests that the antibacterial compounds may either be non-polar or slightly polar in nature. Based on this finding, we tried to isolate the antibacterial compounds in LLE fractions derived from G. boninense mycelia. For this, the pHPTLC separation and antibacterial bioautography were preferentially performed. The combined methods of pHPTLC and bioautography have been widely used to separate bioactive secondary metabolites. These compounds include polyphenols, flavonoids, saponins, alkaloids, and steroids (Annegowda et al., 2013) , along with amino acids, proteins, peptides, hormones, and pesticides (Bhawani et al., 2012) . With the help of using different UV wavelengths, TLC separation serves as a powerful tool for qualitative preliminary metabolite identification (Rajani et al., 2001) . Based on these previous studies, fractionations using WM and CM were conducted with detection wavelengths of 254 and 366 nm through to the formation of separated bands ( Fig. 1A and B) . A relatively weak inhibition band was observed in the band (Rf = 0.4) against both MRSA and S. aureus. Non-significant visible bands were observed under visible light (white). No inhibition was also observed in other bacterial samples for the pHPTLC antibacterial bioautography (not shown). Our data are based on the fact that different properties between compounds are present in both fractions, as previously demonstrated by the different fluorescence profiles of mixed compounds (Marston and Hostettmann, 1999; Marston, 2007) . However, the relatively lower fluorescence intensity observed in bands was presumably thought to be caused by low amounts of particular compounds (Rajani et al., 2001) . Although TLC did not provide specific measurements, it can be very effective when used in combination with other techniques such as LC, GC, MS, FTIR, and NMR. Antibacterial bioautography assays were performed to demonstrate the antibacterial properties of compounds separated via pHPTLC ( Fig. 1C and D) . The soluble CM fraction displayed antibacterial activity (Rf = 0.4) against MRSA NCTC 11939 and S. aureus ATCC 25923, whereas no inhibitory activity was observed against other pathogenic bacterial strains. Although there are several preparative strategies for isolating bioactive compounds, a major disadvantage in using TLC is suggested to be its compounds-pooling power compared to other preparative chromatographic methods, mainly due to its relatively low retention capacity (Bucar et al., 2013) . For example, the relatively weak activity of the CM fraction at Rf = 0.4 against two kinds of gram-positive bacteria was due to a low amount of antibacterial compounds in the separated band. Nevertheless, owing to its greater resolution and separation capacity, TLC has been used to separate bioactive compounds from complex mixtures (Bucar et al., 2013) . For further antibacterial compound isolation, the CM fraction (Rf = 0.4) was collected from 50 replicate plates so as to collect as many antibacterial compounds as possible. In addition, the pHPTLC of the CM fraction at Rf = 0.4 yielded 5 mg. Thus, isolated compounds from the pHPTLC of the CM fraction at Rf = 0.4 were designated as CMRF04. Based on our data (Fig. 1) , the methodological combination of using GC-MS with MS data processing software (i.e., MZmine and MassHunter) was used to achieve high-throughput and highly accurate experiments for compound identification (Misra et al., 2017) . To enhance the efficiency of the isolation and identification processes for the possible antibacterial compounds, this present study employed combined experiments using GC-MS analysis. Three distinct peaks were observed and were designated as compounds A, B, and C, and were correlated with their respective Rt (min). GC-MS values were also obtained and identified (Fig. 2) . We identified two out of three possible antibacterial compounds from CMRF04 with the aid of NIST Fig. S1B ). The described identity ratio (%) was determined based on the m/z fragmentation pattern when compared with the reference fragmentation pattern of similar compounds in the NIST 17 EI Library Version database. Both compounds are generally derived from triterpenes, which are essential to fungal lipid metabolism (Ríos and Andújar, 2017) . Meanwhile, in the case of compound C (Rt = 29.89), the m/z fragmentation pattern was successfully deduced from the MS data with mass and molecular formula (generated using MFG, MassHunter), which were revealed to be 498.66 and C 30 H 42 O 6 (98.9% accuracy based on MFG prediction), respectively (Supplementary data Fig. S1B ). However, we failed to identify compound C using either the NIST MS Search version 2.3 or NIST 17 EI Library databases (Supplementary data Fig. S1C ). Thus, a reliable identification for compound C based on its fragmentation pattern could not be obtained in this experiment. Based on our earlier results ( Fig. 2 and Supplementary data Fig. S1 ), we employed the sPHPLC method, which has been used as an excellent tool for separating and collecting particular compounds from a mixture due to its high sensitivity and isolation capacity (Marston and Hostettmann, 1999; Marston, 2007) . In this study, sPHPLC (Snyder et al., 2012) was used to isolate a single component from a considerably complex mixture. In this experiment, compounds A and B were spiked with their respective standard (A= lanostane, Sigma Aldrich; B = ergosterol, Sigma Aldrich) to validate the identities of peaks A and B, respectively. Three distinctive peaks with their respective Rt were again observed in CMRF04, as indicated (Supplementary data Fig. S2 ). Three peaks were observed, and compounds A and B were spiked with their respective standards. Physically, N 2 -dried compound A was a yellowish oil, while compounds B and C were colorless and darker yellowish oils, respectively. As demonstrated by GC-MS analysis, compound A (red, Rt = 39.21) and B (purple, Rt = 38.55) were characterized as lanostane and ergosterol, respectively. As the peaks corresponding to compounds A and B were positively spiked with their respective standards, they showed increases in the relative absorbance unit (mAU) at the same Rt without compromising chromatogram resolution. Because these were peaks formerly tested, compounds A and B were prominently determined in CMRF04; on the other hand, the third peak, compound C (green, Rt = 39.97), remained unidentified. However, its molecular mass and formula were confirmed using GC-MS and MS data processing software. Experiments suggested that the observed different chromatogram profiles of CMRF04 between GC-MS and sPHPLC, in terms of compounds A, B, and C, originate from the column loaded and the GC-MS separation mechanisms which are mainly dependent on volatility and polarity (Marston and Hostettmann, 1999; Marston, 2007) . This behavior explains why GC separation shows lanostane first, before showing ergosterol. Studies have suggested that the presence of hydroxyl groups in ergosterol affects the volatility of the compound by exerting a hydrogen bond effect, through an increase in the compound dipole moment (Bucar et al., 2013) . Based on our results, we predicted that compound C has the lowest volatility and polarity compared to lanostane and ergosterol, causing it to elute last in both analyses. Utilization of IR spectroscopy for the identification of molecular structures commonly relies on the possibility of properly assigning absorption bands associated with their functional groups. Hence, most peaks and corresponding shoulders of the IR spectrum are attributable to specific functional groups (Bendini et al., 2007) . With the aid of an already known compound and/or standard, identification of unknown compounds using IR and using additional information for related compounds have been made available in biological samples (Coates, 2006) . Considering compounds A and B, these molecules were cross-referenced with lanostane and ergosterol, respectively. Figure 3 shows the matching FTIR spectra profiles for both compounds A and B with their respective standard, validating the identity of both compounds as lanostane and ergosterol, respectively. Therefore, lanostane and ergo-sterol standards will be used for further experimental studies of compounds A and B, respectively. The FTIR spectra of compound C (Fig. 3C ) displayed a spectroscopic profile with a variety of functional groups. A strong and sharp transmittance peak (3,335.91/cm) was observed, which is typically due to the presence of an alkyne terminal (= C-H). The presence of an Sp 2 carbon, which has a trigonal planar structure, is commonly shown as a transmittance peak at the 3,100-3,000/cm region. At the observed FTIR spectrum of compound C, in this region, successive small peaks indicated the presence of Sp 2 carbons in this compound. Meanwhile, Sp 3 carbons, which have tetrahedral structures, were detected in compound C, indicated by the presence of transmittance peaks at 2,960-2,850/cm. The strong band observed at 1,720.85/cm was due to the presence of a carbonyl group (C=O). A number of bands was observed at 1,242.73, 1,109.91, and 1,038.67/cm, which are typically attributed to the presence of aliphatic ester or ether groups and C-O-C stretching. Bands at 1,458.79/cm and 1,381.49/cm were observed for CH 3 and methyl symmetrical C-H, respectively. The peak observed at 1,644.23/cm suggested the presence of either a carbonyl group or amide moieties (Kavyanifard et al., 2016) . Based on the FTIR results (Fig. 3) , the structures of the newly isolated compounds can be supported through the identification of natural product compounds along with NMR spectroscopy (Marshall, 2013) . In this experiment, the functional groups of compound C were determined using FTIR spectra. For example, the identified functional groups from compound C included an alkyne terminal, as well as the number of Sp 2 (at least six of them) and Sp 3 carbons (at least three of them), the number of aliphatic esters or ethers (at least three), and the number of carbonyl groups. The obtained data allowed us to elucidate the structure of compound C with respect to the NMR spectroscopy data. The molecular formula of compound C was established as C 30 H 42 O 6 with 10 degrees of unsaturation. Herein, we performed NMR spectroscopy for the structural confirmation of compound C (Figs. 4 and 5). Analysis of the 1 H and 13 C NMR data for compound C (Supplementary data Table S3 ) revealed seven methyl groups (one methoxy), ten methylenes, (one oxymethylene), three methines (one oxymethine), five quaternary carbons including one doubly oxygenated Sp3 quaternary carbon (δ C 112.1), four olefinic carbons including one terminal olefinic carbon, one carboxylic carbon (δ C 174.), and one carbonyl carbon (δ C 202.7). Five isolated spin systems were established by 1 H− 1 H COSY and HSQC correlations, as illustrated in Fig. 5A . In the HMBC spectrum of compound C (Figs. 4A and 5A), correlations of H-7/C-6, C-8, C-9, and C-10, H 3 -19/C-1, C-5, C-9, and C-10, H 3 -29/ C-8, C-13, C-14, and C-15, and H 2 -12/C-9, C-11, C-13, C-14, C-17, and C-18 furnished a 6-6-5 ring system. HMBC correlations from H-1 to C-2, C-3, C-5, C-9, C-10, and C-19, with the correlation of H 3 -30 to C-3 not only confirmed the presence of the methyl propionate moiety but also connected C-1 with C-10. HMBC correlations from H 2 -27 (H 3 -28) to C-4 and C-5 located the propenyl group at C-5, whereas HMBC correlations located the carbonyl group at C-7. Furthermore, HMBC correlations from H 3 -21 to C-17, C-20, and C-22, and from H 2 -25 to C-23, C-24, and C-26 estab- lished the chain of C21−C26 (via C-20), with C-20 attached to C-17. Finally, considering the chemical shifts of C-18 (δ C 66.4), C-20 (δ C 87.60), and C-24 (δ C 112.10), two C-24 bonded oxygen atoms were attached to C-18 and C-20, respectively to provide the complete structure and to satisfy the unsaturation requirement. (Fig. 5B) . The absolute configuration of compound C was deduced by comparing the experimental and calculated ECD curves generated by the time-dependent density functional theory (TDDFT) (Fig. 5B) . Considering the relative configuration determined above, one of the two isomers, 2a (5S, 10S, 13S, 14R, 17S, 20S, 24R) or 2b (5R, 10R, 13R, 14S, 17R, 20R, 24S), should represent the actual absolute configuration of compound C. A systematic conformational analysis was performed for 2a and 2b using the molecular operating environment (MOE) software package and the MMFF-94 molecular mechanics force field calculation. The MMFF-94 conformational search followed by reoptimization using TDDFT obtained three lowest-energy conformers for enantiomers 2a and 2b, respectively (Table 1) The molecular models compound C in minimal energy were generated by Conflex calculations in silico using MMFF94's force field. (B) Experimental CD spectra of compound C in methanol and the calculated ECD spectra of 2a and 2b with their respective configurations represent possible stereoisomers for compound C. spectra were generated by a Boltzmann weighting of the conformers. Comparison of the experimental and calculated ECD curves of 2a and 2b (Fig. 5B ) allowed the unambiguous assignment of the absolute configuration of compound C as 5S, 10S, 13S, 14R, 17S, 20S, 24R. All of this data indicates that compound C belongs to the 3,4-seco-27-norlanostane triterpene group and the suggested IUPAC name is described in Table 1 . In nature, 27-norlanostane triterpenes are rare. Examples include eight 27-norlanostane saponins from Muscari paradoxam and 14 and 10 norlanostane saponins from Scilla scilloides (Kuroda et al., 2004; Ono et al., 2012 Ono et al., , 2013 Ma et al. (2014) had done NMR analysis of almost similar compound (based on NMR spectroscopy analysis), with proposed name Ganoboninketal able anti-plasmodial activity and suggested ganoboninketal as the name of the compound. Ganoboninketals are new members of the 27-norlanostane triterpenes. We thereby hypothesize that ganoboninketal is produced as a secondary byproduct during lipid metabolism, especially in the squalene and sterol metabolisms of G. boninense (Ma et al., 2014 . A possible biosynthetic pathway for ganoboninketal in this study has been proposed (Supplementary data Fig. S3 ). Accordingly, compound C is referred to as ganoboninketal throughout the remainder of this study. The identified compounds, including lanostane, ergosterol, and ganoboninketal, were evaluated for their antibacterial activity against the six bacterial samples. Both ergosterol and ganoboninketal displayed significant antibacterial activity compared to lanostane (Fig. 6) , and exerted the most potent activity against S. aureus ATCC 25923 (ergosterol, 14.67 ± 0.33 mm; ganoboninketal 15.33 ± 0.58 mm), S. pyogenes ATCC 19615 (ergosterol, 14.33 ± 1.15 mm; ganoboninketal, 14.67 ± 0.58 mm), and MRSA and NCTC 11939 (ergosterol, 14.33 ± 0.58 mm; ganoboninketal 15.00 ± 1.00 mm). Meanwhile, both ergosterol and ganoboninketal showed significantly lower antibacterial activity against Gram-negative bacteria. The lowest activity was observed against K. pneumoniae ATCC 1705 (ergosterol, 10.33 ± 0.58 mm; ganoboninketal, 11.00 ± 1.00 mm), although it was not significantly lower than that of P. aeruginosa ATCC 9027 (ergosterol, 11.33 ± 0.58 mm; ganoboninketal, 12.00 ± 1.00 mm). Arrangement of carbon-harboring pi electrons, hydroxyl groups, and carbonyl groups contributed to their hydrophobicity and lipophilic characteristics, which play important roles in inducing the antibacterial effect of a particular compound (Ben Arfa et al., 2006) . Carboxyl, hydroxyl, and carbonyl groups are essential for antibacterial activity because they mediate binding to the DNA-gyrase complex through hydrogen bonds or through electron disruptive formation (Tomasǐč et al., 2015) . The groups were also able to induce chelating ability, which is an essential requirement for transport into bacterial cells (Cui et al., 2016) . In this case, ergosterol contains one hydroxyl group at carbon number three while ganoboninketal contains several carbonyl groups. Substantive aliphatic structures (both saturated and unsaturated) in a compound induce the compound's hydrophobicity, thus assisting its ability to insert itself into bacterial cell membranes (Desbois and Smith, 2010) . Hydrophobic compounds are generally reported to be most toxic to cytoplasmic membranes, as they often disrupt and disintegrate membrane structures (Ben Arfa et al., 2006) . With these in mind, two hypotheses were made on ergosterol and ganoboninketal: 1) both compounds induce antibacterial effects through bacterial membrane disintegration, and 2) both compounds easily penetrate bacterial cell barriers with the aid of hydrophobic regions. Thus, the antibacterial effect was exerted by the hydroxyl or carbonyl group through metabolic or molecular inhibition. In the case of lanostane, although the compound possesses great hydrophobicity due to the presence of a substantive hydrocarbon structure, its antibacterial activity was least observed. This phenomenon is best explained in non-disrup-tive hydrophobic compounds towards bacterial membrane cells (Sikkema et al., 1995) . The hydrophobic region enables a particular compound to insert itself into bacterial cell membranes, but cannot disrupt the membrane cells' structure because its own structure is not massive enough (Sikkema et al., 1995) . Nonetheless, a hydrophobic disruptive effect by lanostane was observed in S. aureus ATCC 25923, S. pyogenes ATCC 19615, and E. coli ATCC 35218, but this effect was very minimal. Ganoderma boninense mycelia for phytochemicals and secondary metabolites with antibacterial activity Isolation and characterization of antibacterial compounds from Ganoderma boninense against methicillin-resistant Staphylococcus aureus (MRSA) Basidiomycetes: a source for new bioactive secondary metabolites TLC-bioautography-guided isolation, HPTLC and GC-MS-assisted analysis of bioactives of Piper betle leaf extract obtained from various extraction techniques: in vitro evaluation of phenolic content, antioxidant and antimicrobial activities Antimicrobial activity of carvacrol related to its chemical structure Preliminary evaluation of the application of the FTIR spectroscopy to control the geographic origin and quality of virgin olive oils Thin-layer chromatography of amino acids: a review Cosmetic Microbiology: A Practical Approach Natural product isolation-how to get from biological material to pure compounds Molecular fingerprint of Ganoderma spp. from Sabah, Malaysia The role of syringic acid in the interaction between oil palm and Ganoderma boninense, the causal agent of basal stem rot Relation of Ganoderma ergosterol content to Basal Stem Rot disease severity index HPLC Fingerprints and in vitro antimicrobial activity of syringic acid, caffeic acid and 4-hydroxybenzoic acid against Ganoderma boninense Performance standards for antimicrobial susceptibility testing; 21st informational supplement. CLSI document M100-S21 Interpretation of infrared spectra, a practical approach The oil palm Novel 3-aminothiazolquinolones: design, synthesis, bioactive evaluation, SARs, and preliminary antibacterial mechanism Anti-Candida activity and brine shrimp toxicity assay of Ganoderma boninense Antibacterial free fatty acids: activities, mechanisms of action and biotechnological potential Chemical components of Ganoderma Ganoderma boninense Pat. from based stem rot of oil palm Screening for potential antimicrobial compounds from Ganoderma boninense against selected food borne and skin disease pathogens Structure characterization of a methylated ester biosurfactant produced by a newly isolated Dietzia cinnamea KA 1 Steroidal glycosides from the bulbs of Ornithogalum thyrsoides Six new 3, 4-seco-27-norlanostane triterpenes from the medicinal mushroom Ganoderma boninense and their antiplasmodial activity and agonistic activity to LXRβ Ganoboninketals A-C, antiplasmodial 3, 4-seco-27-norlanostane triterpenes from Ganoderma boninense pat Two dimensional NMR spectroscopic approaches for exploring plant metabolome: a review Fourier, hadamard, and hilbert transforms in chemistry Role of advances in chromatographic techniques in phytochemistry Biological and chemical evaluation of plant extracts and subsequent isolation strategy Review of emerging metabolomic tools and resources NMR of natural products at the 'nanomole-scale' Antibacterial compounds from the mushroom Ganoderma colossum from Nigeria Five new nortriterpenoid glycosides from the bulbs of Scilla scilloides Two new nortriterpenoid glycosides and a new phenylpropanoid glycoside from the bulbs of Scilla scilloides Hepatoprotective effects of triterpenoids from Ganoderma cochlear HPTLC-aided phytochemical fingerprinting analysis as a tool for evaluation of herbal drugs. A case study of Ushaq (ammoniacum gum) Isolation and partial characterization of bioactive fucoxanthin from Himanthalia elongata brown seaweed: a TLC-based approach An assessment of the taxonomy and chemotaxonomy of Ganoderma Lanostanoids from fungi as potential medicinal agents An introduction to natural products isolation LC-SPE-NMR-MS: a total analysis system for bioanalysis Mushrooms: an untapped reservoir for nutraceutical antibacterial applications and antibacterial compounds Modelling of yield and distribution coefficient in a liquid-liquid extraction: Effect of the concentration of ligand Mechanisms of membrane toxicity of hydrocarbons Practical HPLC method development ] thiazoles as novel DNA gyrase inhibitors targeting the ATP-binding site Thin layer chromatography-bioautography and gas chromatographymass spectrometry of antimicrobial leaf extracts from Philippine Piper betle L. against multidrug-resistant bacteria Venomics: a new paradigm for natural products-based drug discovery This research was supported by the Ministry of Education of Malaysia (FRGS0384-SG-2/2014). Syahriel Abdullah and Yoon Sin Oh-Two contributed equally to this work and should be considered co-first authors. The authors declare that there are no competing interests associated with the manuscript.