UHPLC-QTOF-MS/MS-based metabolomic discovery of anticancer compounds in ethanolic extracts of Ficus hispida L. f
Saengrawee Thammawithan, Jirattiporn Thanuma, Sirinya Sitthirak, Chadapohn Panplu, Thapanee Pruksatrakul, Piya Prajumwongs, Arporn Wangwiwatsin, Poramate Klanrit, Watcharin Loilome, Nisana Namwat
TL;DR
This study identifies catharanthine in Ficus hispida as a potential anticancer compound effective against cholangiocarcinoma cells.
Contribution
The study discovers catharanthine as a key anticancer compound in Ficus hispida using metabolomic analysis and cytotoxicity testing.
Findings
Catharanthine, found in bark and leaves, showed strong anticancer activity against cholangiocarcinoma cells.
Metabolomic analysis identified 82 compounds, with distinct phytochemical profiles in different plant parts.
Bark and leaf extracts had the highest cytotoxicity against CCA cells with low IC50 values.
Abstract
Ficus hispida L. f. (F. hispida) is commonly used in traditional medicine for various health problems. No comprehensive analysis of all its components has yet been undertaken, despite researchers having explored its chemical constituents and biological activities. Using untargeted metabolomics, we aimed to assess the chemical compositions and metabolic differences among the five parts of F. hispida: bark, fruits, leaves, twigs, and stalks. To discover the compounds that might be accountable for the noted efficacy, our study assessed the correlation between the identified metabolites and anticancer activities. We applied untargeted metabolomics using UHPLC-QTOF-MS/MS to confirm a total of 82 metabolites. These compounds were classified into six phytochemical groups with predominant accumulation in each part: fatty acids and their conjugates (twigs), terpenoids (stalks), phenylpropanoids…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Fig 1
Fig 2
Fig 3
Fig 4
Fig 5
Fig 6
Fig 7- —http://dx.doi.org/10.13039/501100004071Khon Kaen University
- —The coordinating center for the plant genetic conservation project under the royal initiation of her royal highness Princess Maha Chakri Sirindhorn (RSPG), Thailand
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPhytochemistry and biological activities of Ficus species · Ginger and Zingiberaceae research · Ethnobotanical and Medicinal Plants Studies
1. Introduction
Ficus hispida L. f. (F. hispida) is a medicinal plant of the Moraceae family that is commonly used in traditional medicine in China, India, Sri Lanka, Myanmar, Thailand, and Australia [1]. The root, leaf, fruit, and bark of this plant have been applied to treat ulcers, jaundice, diabetes, bleeding, and dysentery, among other medical conditions [2]. Comprehensive research has revealed an extensive variety of phytochemicals in different parts of F. hispida, comprising phenols, flavonoids, glycosides, alkaloids, steroids, sterols, saponins, and terpenes [3,4]. Terpenoids, flavonoids, and alkaloids are recognized as the most prominent and prevalent biologically active substances, potentially having a role in antioxidant and anticancer effects, among other biological activities [4].
A metabolomic approach is a potent tool for a comprehensive analysis of plant metabolites with high-throughput analytical methods such as mass spectrometry [5]. Metabolomics offer key insights into the biochemical constitution of plant material by methodically profiling metabolites in plant tissues, extracts, and derivatives [6]. We have applied UHPLC-QTOF-MS/MS, a high-throughput technique, in metabolomics to identify metabolites and classify them into each phytochemical class. Ethanol was used as an extraction solvent to obtain moderately polar metabolites with antioxidant and anticancer effects, including phenolic acids, flavonoids, alkaloids, terpenoids, and small peptides. Statistical analyses, including principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and hierarchical cluster analysis, were performed to generate metabolomics profiling that allowed for the comparison of metabolite changes in various parts of plants [7]. Moreover, specific significant metabolites can be evaluated for their correlation with biological activities such as antioxidant, antibacterial, and anticancer activities to identify specific bioactive molecules responsible for biological effects observed in an extract.
Globally, cancer remains a leading cause of death. Cholangiocarcinoma (CCA), or bile duct cancer, is an aggressive and highly lethal malignancy originating from bile duct epithelial cells that is generally associated with a poor prognosis [8]. Current treatments, such as surgery, radiation, and chemotherapy, frequently exhibited limited effectiveness against CCA due to the aggressive characteristics of this cancer and prevalent drug resistance [9,10]. Consequently, the study and development of effective natural therapeutic agents derived from medicinal plants may prove beneficial for CCA treatment.
F. hispida has been identified as a potential source of novel anticancer agents and has previously demonstrated anticancer effects on various cell lines [11]. Previous research has shown that stem extract has strong antineoplastic activity against breast cancer [12], that crude extracts of leaves and the bark have antiproliferative potency against glioblastoma (U87MG), and that methanolic fruit extract has cytotoxic activity against HL60, KB, HeLa, HT29, and HepG2 cells [13]. Additionally, a previous study identified several compounds from fruits that may serve as promising lead candidates for androgen-dependent prostate cancer [14,15]. These results indicate that F. hispida is a significant source of therapeutically relevant phytochemicals that can be developed into natural chemopreventive anticancer agents.
Therefore, our study aimed to utilize an untargeted metabolomic approach to evaluate the chemical compositions and metabolic variations among five specific parts of F. hispida: the bark, fruits, leaves, twigs, and stalks. Furthermore, we aimed to assess the in vitro anticancer activity, specifically the cytotoxicity against CCA cells, and investigate the relationship between the identified metabolites and this anticancer activity to discover the compounds that might be responsible for the noted efficacy.
2. Materials and methods
2.1 Chemicals and reagents
Commercial-grade 99.7% ethanol was obtained from S.C. Science Co., Ltd., Thailand. Dimethyl sulfoxide (DMSO), sulforhodamine B (SRB), Dulbecco's Modified Eagle Medium (DMEM), (-)-quinic acid, (+)-Catechin (cianidanol), and procyanidin B2 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Catharanthine was purchased from MedChemExpress (Monmouth Junction, NJ, USA). Gemcitabine was purchased from Fresenius Kabi (Maharashtra, India). Trichloroacetic acid (TCA), penicillin-streptomycin, and trypsin-EDTA were obtained from Life Technologies (Grand Island, NY, USA).
2.2 Plant materials and sample preparation
The fresh F. hispida samples were collected from the Romklao Kallapruek Park, Khon Kaen University, Khon Kaen, Thailand (16°28’25.3"N 102°49’03.0"E), in January 2024. The plant material was identified by Mr. Piya Sukkharom, and a voucher specimen (S. Thammawithan 1) was deposited at the KKU Herbarium (KKU No. 27631), Khon Kaen, Thailand (check the details in S1 File). The collected samples were subsequently chopped and washed with water to eliminate the surface dirt and adhering particles. Subsequently, the samples were then dried in a hot air oven at 60°C. The dried samples were ground into fine powder and stored at 4°C until the subsequent step.
2.3 Sample extraction
The extraction process was performed according to a previous study with some modifications [16]. The powder of each sample was extracted with 99.7% ethanol. Briefly, 20 g of powder was added to 80 mL of ethanol. The mixture was subjected to sonication using an ultrasonic bath cleaner (Jeken, China) at 40 Hz for 30 min. Various previous studies [17,18] have used this sonication treatment, which effectively enhances the antioxidant compounds. The extraction was repeated three times for the high yield recovery and then combined and filtered with Whatman filter paper no. 1. A rotatory evaporator (Buchi/R-100, Buchi (Thailand) Ltd.) and SpeedVac (Labconco, Wertheim, Germany) were used for concentrating and evaporating the filtered extracts at 40°C under reduced pressure for 48 h and then kept at 4°C.
2.4 LC-MS/MS method for metabolomics analysis of F. hispida extracts
The metabolomics analysis was carried out by using ultra-high-pressure liquid chromatography (UHPLC) coupled to a QTOF impact II mass spectrometer (Bruker Daltonics, Bremen, Germany). Briefly, ethanolic extracts of the plant were diluted to a concentration of 5 mg/mL and filtered using a 0.2 μm syringe filter. Thereafter, the extracts were separated by an Intensity Solo 2 C18 chromatographic column (100 x 2.1 mm, 2.1 mm). The flow rate was set at 0.4 mL/min, with a 5 μL injection volume. The temperature of the autosampler was set at 10°C and a column oven temperature of 40°C. Mobile phases A and B were used: A was a mix of ultrapure water and methanol (99:1) with 5 mM ammonium formate and 0.01% formic acid, while B was methanol with 5 mM ammonium formate and 0.01% formic acid. The LC gradient program started with 0.1% B from 0 to 1 min, followed by a linear increase to 99.9% B at 10 min for the separation of plant metabolites. The mobile phase composition was maintained for 12 min for the column washing step, then returned to the initial conditions at 12.1 min and held until 15 min for column re-equilibration prior to the next injection. The mass spectra were acquired through data-dependent acquisition (DDA) in positive and negative polarity modes. The scan range was from 50 to 1300 m/z with a high-resolution accurate mass at 120,000. The precursor ions were fragmented at collision energies of 20, 25, and 30 eV, with a resolution of 30,000 for fragmented ions.
2.5 Anticancer activities of F. hispida extracts
Cell culture.
The CCA cell lines KKU-213A and KKU-055 were obtained from the Japanese Collection of Research Bioresources (JCRB) Cell Bank, Osaka, Japan (check the details in S2 in S1 File), and cultured in Dulbecco's modified Eagle medium (DMEM) mixture supplemented with 10% fetal bovine serum and 100 IU/mL of penicillin-streptomycin. The cultured cell lines were incubated at 37˚C in a humidified incubator with a 5% CO_2_ atmosphere.
Cytotoxicity test by sulforhodamine B (SRB) method.
The cell lines were plated into a 96-well plate at a density of 2.5 x 10^3^ cells per well. After culturing for 12–13 h, the various concentrations of crude extract were added, and the cells were cultured for a further 48 h. Gemcitabine served as a positive control. A plate containing only cell suspension for a no-growth control (day 0) was set, and cell viability was measured by the Sulforhodamine B (SRB) method [19]. Briefly, the cells were fixed in 10% trichloroacetic acid for 1 h and washed three times with DI water. Then, fifty microliters of 0.4% SRB were exposed and incubated for 45 min. After unbound dye was washed with 1% acetic acid, SRB was solubilized with 200 µl of 10 mM Tris-base pH 10.5, and the optical density was read at 540 nm with a microplate reader (Tecan, Switzerland).
2.6 Statistical analysis and data analysis
Statistical analysis of the cytotoxic test was conducted by using GraphPad Prism software. All assays were performed in three independent experiments, and data were reported as mean values with standard deviations (mean ± SD). The data were analyzed using the one-way analysis of variance (ANOVA) to evaluate the differences between the samples. A p-value of less than 0.05 was considered statistically significant. The analysis of the complex data, which included methods such as principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA) score plots, variable importance in projection (VIP) score plot analysis, a heatmap plot, and univariate analysis, was done using MetaboAnalyst 6.0.
3. Results
3.1 Metabolite identification and classification among five parts of F. hispida extract by UHPLC-QTOF-MS/MS analysis
The chemical compositions of the bark, fruits, leaves, twigs, and stalks of F. hispida extract were investigated using UHPLC-QTOF-MS/MS in the positive and negative ionization modes to identify various components. A comparison of chromatograms among the five parts of F. hispida extract is shown in S1 in S1 File. The chromatogram in positive mode revealed that the categories of metabolites among five parts were similar, while the content of many metabolites varied (S1A Fig in S1 File). It was noted that the chromatogram showed peak intensity at 325 m/z in all parts. This characteristic was annotated to be denatonium, a synthetic compound, which is used for denaturing ethanol. Therefore, denatonium was excluded from our analysis. In contrast, the types and the content of metabolites were different between the bark and other parts in negative mode (S1B Fig in S1 File). The ethanol extract of F. hispida yielded metabolites in the positive rather than the negative mode. Tentative metabolite identifications were evaluated by comparing MS/MS fragment ions, retention time, and comparability with available literature data and open-source software, including MZmine, GNPS, and SIRIUS. We confirmed a total of 82 metabolites under both ionization modes, belonging to various classes, including fatty acids and conjugates, terpenoids, phenylpropanoids, alkaloids, saccharides and conjugates, amino acids, peptides, and analogues (S1 Table). Of all the metabolites, we found 68 and 16 in the positive and negative modes, respectively. Two metabolites, procyanidin B2 and chlorogenic acid, were detected in both modes. The total number of compounds varied among all groups. There were 26, 14, 37, 47, and 45 metabolites identified in the bark, fruits, leaves, twigs, and stalks, respectively.
Fig 1 illustrates the phytochemical classes present in five parts of F. hispida. The identified compounds encompassed various fatty acids and conjugates, terpenoids, phenylpropanoids, alkaloids, saccharides and conjugates, amino acids, peptides, and analogues, which accounted for 28.4%, 28.4%, 23.7%, 10.1%, 6.51%, and 2.96% of the total (Fig 1A), respectively. When comparing the five parts of the extract, the maximum numbers of metabolites in fatty acids and conjugates and terpenoids were accumulated in the twigs and stalks. The phenylpropanoid content was found in the bark and twig extracts, respectively. In contrast, alkaloids, saccharides and conjugates, and amino acids, peptides, and analogues were predominantly accumulated in leaves, bark, and twigs together with stalks, respectively. Comparing the phytochemical classes in each part of the plant, the bark extract contained the greatest phenylpropanoid content, with 38.5% (Fig 1B). The fruit and twig extracts had higher amounts of fatty acids and conjugates than other classes, with 42.9% and 38.3%, respectively. The leaf and stalk extracts comprised the highest amount of terpenoids, with 29.7% and 33.3%, respectively. The results indicated that the metabolite content varied among the five parts of the F. hispida extract for each chemical class.
Classification of phytochemical classes in different parts of F. hispida.(A) The number of differential metabolites that accumulated predominantly in each part of F. hispida. (B) % Composition of metabolite in each phytochemical class in five parts of F. hispida.
3.2 The metabolic variations among the five parts of F. hispida
To differentiate the various metabolite profiles across the five parts of the ethanolic F. hispida extract, multivariate analyses were conducted, including unsupervised principal component analysis (PCA) score plots and supervised partial least squares discrimination analysis (PLS-DA) score plots, utilizing UHPLC-QTOF-MS/MS data in both positive and negative modes. Results are shown in Fig 2. The PCA, unsupervised multivariate analysis, was employed to evaluate natural sample variance, clustering, and outlier identification. In the positive mode, PC1 and PC2 explained 44.3% and 39.1% of the variance, respectively (Fig 2A). The initial two components accounted for 83.4% of the overall data variance. As a result, metabolite profiles among five parts were grouped into 3 distinct clusters, namely bark, leaves, and overlapping of fruits, twigs, and stalks. Bark, fruits, twigs, and stalks exhibited a close correlation, whereas these four parts were separated from leaf extract. On the other hand, PCA score plots for the negative data analysis demonstrated a distinct division between the bark extract and other groups (Fig 2B). PC1 and PC2 accounted for 67.1% and 14.2% of the variance, respectively. These first two factors explained 81.3% of the variation in the data. In addition, PCA score plots for both modes showed that all samples in the score plots were within the 95% Hotelling’s T-squared ellipse. These results indicated no outlier among the analyzed samples. To further investigate differences in metabolic variance among the five parts, a supervised analysis of PLS-DA was employed. In positive mode, the PLS-DA revealed that the samples could be divided into five distinct clusters (Fig 2C). As a result, there were differences in the metabolic variations of the five parts in positive mode. In particular, leaf extract was discriminated from other groups. The PLS1 explained 35.8% of the variation, while PLS2 explained 14.0%, according to the PLS-DA score plots. The quality metrics for PLS-DA were indicated by R^2^ and Q^2^ to evaluate modeling and predictive performance. The values of R^2^ and Q^2^ were 0.734 and 0.660, respectively. This model was considered reliable and showed no indications of overfitting. The PLS-DA score plot in negative mode showed clear differences in the metabolite patterns between the bark extract and the others (Fig 2D). PLS1 and PLS2 explained 66.8% and 12.9%, respectively. The model showed R^2^ and Q^2^ values were 0.758 and 0.701, respectively. Accordingly, R^2^ and Q^2^ confirmed that the model was effective with satisfactory predictable accuracy.
Multivariate data analysis of the metabolites.Principal component analysis (PCA) scores plots and partial least square-discriminant analysis (PLS-DA) scores plots of bark, leaf, fruit, twig, and stalk extracts. PCA score plots in positive mode (A) and negative mode (B). PLS-DA scores plot for positive mode (C) and negative mode (D). Shaded areas are the 95% confidence regions of each group.
3.3 The distribution of metabolites in the five parts of F. hispida
The hierarchical cluster analysis and heatmap plot were performed to cluster and visualize the content profile of the metabolites across five parts of F. hispida extract. Fig 3 displayed the distribution of different metabolites among the five parts. In positive mode, the hierarchical cluster showed that bark, fruits, leaves, twigs, and stalks were separated into five clusters (Fig 3A). The leaf extract was sorted out from the other groups. Following the deeper colors on the heatmap plot (Fig 3B), each part of the extract exhibited different metabolite expression levels. The twig extract had the greatest metabolite expression, followed by the leaf and stalk extracts, whereas bark and fruit extracts exhibited lower concentration levels. Conversely, differential metabolites in negative mode exhibited a clear clustering pattern into two groups, comprising bark and other components (Fig 3C). The heatmap plot distinctly illustrated two variations in metabolite distribution characterized by high and low expression levels (Fig 3D). The main distribution of metabolites was found in the bark extract, while extracts from fruits, leaves, twigs, and stalks exhibited minimal distribution and contents of metabolites.
Hierarchical cluster and heat map illustrating the distribution of metabolites across the five parts of F. hispida.(A) Hierarchical cluster analysis between five parts and (B) heat map of metabolite distribution in positive mode. (C) Hierarchical cluster analysis between five parts and (D) heat map of metabolite distribution in negative mode. Each sample is represented by a column, and each metabolite is represented by a row. The blue and red colors refer to low and high abundance, respectively.
3.4 The significantly different metabolites among the five parts
The metabolites that significantly differed among the five parts were determined by an ANOVA statistical assay with p-values less than 0.05 and PLS-DA based on the values of variable importance in the projection (VIP) more than 1. Thirteen metabolites significantly differed among the five parts of F. hispida extract, including phytosphingosine, 9-Oxo-10E,12Z- octadecadienoic acid, delta-9-tetrahydrocannabinol, catharanthine, 2-[[(2R,3S)-2-(3,4-dihydroxyphenyl)-3,5-dihydroxy-3,4-dihydro-2H-chromen-7-yl]oxy]oxane-3,4,5-triol, 13S-hydroxy-9Z,11E,15Z-octadecatrienoic acid, pheophorbide A, 1-hexadecanoyl-sn-glycerol, harmol, quinic acid, embelin, procyanidin B2, and cianidanol (Fig 4A and 4B). Among these, phytosphingosine, 9-Oxo-10E,12Z-octadecadienoic acid, 2-[[(2R,3S)-2-(3,4-dihydroxyphenyl)-3,5-dihydroxy-3,4-dihydro-2H-chromen-7-yl]oxy]oxane-3,4,5-triol, 13S-hydroxy-9Z,11E,15Z-octadecatrienoic acid, pheophorbide A, and 1-hexadecanoyl-sn-glycerol had the highest expression in twigs. Delta-9-tetrahydrocannabinol and catharanthine showed maximum expression in stalks and leaves, respectively, while harmol, quinic acid, embelin, procyanidin B2, and cianidanol exhibited maximum accumulation in bark.
VIP score plot and significantly differential metabolites among the five parts of F. hispida.VIP scores are derived from PLS-DA analysis at different parts (bark; B, fruits; F, leaves; L, stalks; ST, twigs; TW) of F. hispida in (A) positive mode and (B) negative mode. The column to the right of each figure indicates variations in metabolite peak intensities. * is added to the significant metabolites with VIP scores >1 and p < 0.05.
3.5 Cytotoxicity of crude extract against CCA cell lines
The viability of KKU-213A and KKU-055 cells was expressed in Fig 5. The results showed that all crude extracts reduced a percentage of cell viability on the two CCA cell lines in a dose-dependent manner. Bark and leaf extracts significantly inhibited CCA cell viability more than fruit, twig, and stalk. The IC_50_ values (S2 in S1 Table) of bark, leaves, fruits, twigs, stalks, and gemcitabine on KKU-213A were 0.71 ± 0.17, 0.82 ± 0.22, 7.96 ± 3.29, 7.53 ± 4.24, 8.51 ± 1.59, and 0.02 ± 0.00 µg/mL, respectively. The IC_50_ values of bark, leaves, fruits, twigs, stalks, and gemcitabine on KKU-055 were 0.78 ± 0.13, 1.03 ± 0.21, 11.5 ± 3.18, 9.98 ± 7.02, 11.9 ± 1.72, and 0.05 ± 0.00 µg/mL, respectively. These revealed that all crude extracts showed potential inhibition of CCA cell lines. In particular, the bark and leaf extracts had the lowest IC_50_ values on the two CCA cell lines. Compared to gemcitabine, the reference drug, the cytotoxicity of those extracts was lower.
Cytotoxic effect of various parts of F. hispida on CCA cell viability.(A and B) KKU-213A and (C and D) KKU-055 after treatment with each extract for 48 h ( and ** indicate statistical significance; p < 0.05 and p < 0.001, respectively).*
3.6 Identification of key metabolites correlated to the anticancer activities of F. hispida extract
The univariate analysis with Pearson's correlation coefficient was determined between thirteen of the significant metabolites and the IC_50_ values. This correlation can evaluate the extent to which one variable may impact another, potentially indicating a positive or negative correlation. A heatmap of three colors was plotted to observe the correlation data: the shades of red represent positive correlations (0 < r < 1), whereas shades of blue represent negative correlations (−1 < r < 0). A significant correlation was considered at a p-value less than 0.05. Among high-level metabolites, catharanthine showed a strong negative correlation as well as p-values less than 0.01 with the IC_50_ values (r = −0.615 (KKU-213A), −0.634 (KKU-055)) (Fig 6 and S3 in S1 Table). Moreover, there were 4 metabolites that showed a moderate negative correlation (r = (−0.40) – (−0.59)) as well as p-values less than 0.05 with the cytotoxicity test (IC_50_ value) in the following order: cianidanol > procyanidin B2 > quinic acid > embelin (Fig 6 and S3 in S1 Table).
*Correlation patterns of metabolites and anticancer activities.Correlation map of the metabolites analyzed by UHPLC-QTOF-MS/MS with cytotoxicity (IC50). The shades of red represent positive correlations (0 < r < 1), whereas shades of blue represent negative correlations (−1 < r < 0). Each square indicates Pearson’s correlation coefficient of a pair of metabolites and IC50. *, * indicates p-values less than 0.05 with a strong correlation (r = (−0.6) – (−0.79)) and a moderate correlation (r = (−0.40) – (−0.59)), respectively.
3.7 Cytotoxic effect of key metabolites on CCA cell lines
The cytotoxic effect of key metabolites on CCA cells was performed using the SRB assay. The IC_50_ values were assessed after KKU-213A and KKU-055 cells were treated with catharanthine, cianidanol, procyanidin B2, and quinic acid for 72 h. The percentage of cell viability is shown in Fig 7. Catharanthine, cianidanol, and procyanidin B2 decreased the cell viability of KKU-213A and KKU-055 in dose-dependent manners. The IC_50_ values of catharanthine, cianidanol, and procyanidin B2 on KKU-213A were 41.0 ± 0.99, 812 ± 166, and >500 μM (S4 in S1 Table), respectively. The IC_50_ values of catharanthine, cianidanol, and procyanidin B2 on KKU-055 were 47.4 ± 4.34, 474 ± 155, and >500 μM. Nevertheless, quinic acid at the concentration of 0–1000 μM did not inhibit the viability of KKU-213A and KKU-055. The result indicated that catharanthine and cianidanol exhibited a cytotoxicity effect against KKU-213A and KKU-055 cells.
Cytotoxic effect of key metabolites on CCA cell viability.(A) KKU-213A and (B) KKU-055 after treatment with 0, 1, 10, 100, 500, and 1,000 μM of catharanthine, cianidanol, procyanidine B2, and quinic acid for 72 h.
4. Discussion
Our study revealed a metabolite profile of five parts of F. hispida (bark, fruits, leaves, twigs, and stalks) using UHPLC-QTOF-MS/MS in both positive and negative ionization modes, demonstrating 82 metabolites across various phytochemical classes, including fatty acids and conjugates, terpenoids, phenylpropanoids, alkaloids, saccharides and conjugates, amino acids, peptides, and their analogues. The majority of metabolites in positive ionization mode corresponded to previous findings indicating that ethanol extracts of Ficus species prefer the ionization of non-polar or moderately polar compounds, such as terpenoids and fatty acids [4,20]. Consistent with earlier reports, terpenoids, phenylpropanoids, and flavonoids were identified as the main chemical components contributing to the antioxidant and anti-inflammatory properties of F. hispida [2]. Bark extract contained a high proportion of phenylpropanoids (38.5%). It is rich in phenolic compounds that may contribute to its traditional use in skin and liver diseases. Terpenoids, which are known as photoprotective and cytoprotective agents, were abundant in leaves and stalks [21]. The abundance of fatty acids and conjugates in twigs and fruits, such as linoleic, linolenic, and 13-hydroxy-octadecatrienoic acids, indicates their functions in structural and energy storage metabolism, supporting the tissue-specific specialization observed in other Ficus taxa [22,23]. In addition, the discovery of alkaloids, such as catharanthine and harmol, in various regions enriches the chemical profile of F. hispida and aligns with evidence suggesting that phenanthroindolizidine alkaloids exhibit antiproliferative effects [24,25]. Our results demonstrate more chemical diversity and clearer patterns of accumulation for each part compared to previous studies that primarily examined single parts or isolated compound groups. This comprehensive profiling highlights F. hispida as a chemically versatile medicinal species, revealing previously unrecognized insights into its tissue-specific metabolite biosynthesis and affirming its ethnomedicinal significance for antioxidant, anti-inflammatory, and anticancer applications.
Thereafter, we used PCA, PLS-DA, the hierarchical cluster, and heatmaps to evaluate the similarities and differences in metabolite profiles between five parts of F. hispida extract. In positive mode, the results indicated distinct clustering patterns among the five groups, with the leaf extract separated from other groups. For the negative mode, there was some overlapping of metabolite profiles, but bark extract separated from the other four parts. The hierarchical cluster and heatmaps of metabolites confirmed separation of each group. The results showed that the types and the content of metabolites were different in each group. The total number and the contents of compounds were found to be the highest in the ethanolic twig extract for positive mode, while negative mode indicated the highest metabolite distribution only in bark extract. Furthermore, the fruit, leaf, twig, and stalk extracts exhibited a very low number of metabolites (1–3) in the negative mode. This phenomenon explains why the metabolite profiles in the four different parts overlapped in negative mode.
Subsequently, the VIP score highlighted significant metabolites based on relative variables, facilitating the identification of potential metabolite markers. This study identified thirteen metabolites that vary significantly in content across the five parts of the F. hispida extract. Among thirteen metabolites, 9-oxo-10E,12Z-octadecadienoic acid, delta-9-tetrahydrocannabinol, catharanthine, 2-[[(2R,3S)-2-(3,4-dihydroxyphenyl)-3,5-dihydroxy-3,4-dihydro-2H-chromen-7-yl] oxy]oxane-3,4,5-triol, 13S-hydroxy-9Z,11E,15Z-octadecatrienoic acid, 1-hexadecanoyl-sn-glycerol, harmol, and embelin are plant metabolites that have not been reported in any Ficus species. These metabolites were initially discovered in F. hispida, which may contribute to the medicinal properties of this plant. In contrast, phytosphingosine, pheophorbide A, and procyanidin B2 have not been reported in F. hispida; however, they are found in other Ficus species [26–29]. Of these, quinic acid and cianidanol were reported in F. hispida extract in a previous study [3].
The results showed that five parts of F. hispida have potent anticancer activity to CCA cell lines of KKU-213A and KKU-055 with IC_50_ values of 0.71 ± 0.17–11.9 ± 1.72 µg/mL. This value conformed to the anticancer potency set by the National Cancer Institute standard for the plant extracts with an IC_50_ < 30 µg/mL [30]. Previous studies on the cytotoxicity test of bark and leaf extracts against human cancer cell lines (U87MG, A549, HT-29) showed that the IC_50_ values were higher than those in this study [31]. The cytotoxicity of fruit extract was in agreement with a previous study by Jie Zhang et al., 2018. They reported the IC_50_ values of fruit extract on human cancer cell lines (HL60, A549, SKBR3, KB, Hela, HT29, HepG2, and LO2) were in the range of 0.1–100 μg/mL [13]. Furthermore, our results revealed that bark and leaf extracts had the greatest potential cytotoxicity against the tested CCA cells. The results related to the variation of metabolites is that the main metabolite in bark extract consists of phenolic compounds. Many studies define phenolic compounds as potential anticancer agents against several types of cancer [32]. The leaf extract showed the maximum number of metabolites in the alkaloids class compared to other extracts. Numerous research studies have confirmed the antiproliferative and anticancer effects of the alkaloids [33]. Therefore, phenolic compounds and alkaloid metabolites may play an important role in anticancer activities of bark and leaf extracts.
After evaluation of the cytotoxicity assay, univariate analysis was performed to identify key metabolites that are responsible for anticancer capability. As shown in previous analyses, thirteen metabolites were identified that significantly differed among five parts. The correlation analysis between significant metabolites with cytotoxicity (IC_50_ values) revealed a strong negative correlation to catharanthine that exhibited high accumulation in bark and leaf extracts. Catharanthine is a pharmaceutically important antitumor alkaloid that is found in Catharanthus roseus [34]. It was reported to have an anticancer effect on liver carcinoma cells (HepG2) [35], leukemia cells (HL60, K562) [36], and colorectal carcinoma cells (HCT 116) [37]. Moreover, a moderate negative correlation appeared in four metabolites, including cianidanol, procyanidin B2, quinic acid, and embelin, that exhibited the highest accumulation in bark extract. Cianidanol, or (+)-catechin, a polyphenolic antioxidant plant metabolite, occurs in green tea, red wine, chocolate, and cocoa [38]. The metabolite shows various medicinal effects, such as antioxidant, antimicrobial, anticancer, and anti-inflammatory properties [39]. Procyanidin B2, a flavonoid compound, is a potent natural antioxidant that is present in grape seeds, apples, and cacao beans [40]. It has properties that fight cancer, viruses, inflammation, and free radicals [41]. Additionally, prior studies indicated that procyanidin B2 suppressed the proliferation of liver, breast, and gastric cancer cells [42–44]. Quinic acid is a naturally occurring phenolic compound found in many plants, such as coffee, bilberries, prunes, cranberries, and kiwifruit [45–47]. These compounds exhibited various biological activities, such as antioxidant, antidiabetic, anticancer, antimicrobial, antiviral, aging-protective, antinociceptive, and analgesic effects [39]. Embelin is a plant-based benzoquinone alkaloid that displays various biological activities, including antioxidant, antidiabetic, analgesic, antibacterial, antitumor, and anti-inflammatory activities [48,49]. These compounds exhibited various biological activities, such as antioxidant, antidiabetic, antimicrobial, antiviral, aging-protective, antinociceptive, and anticancer effects [50].
Subsequently, we confirmed the anticancer effect of the metabolites on CCA cells. Among four potential metabolites, including catharanthine, cianidanol, procyanidin B2, and quinic acid, our result showed that catharanthine, cianidanol, and procyanidin B2 had cytotoxicity on CCA cells. Nonetheless, catharanthine exhibited the most significant inhibitory effect on CCA cell viability. Catharanthine is a precursor to the potent chemotherapeutic vinca alkaloids, such as vinblastine and vincristine [51]. This finding aligns with previous research, indicating that alkaloids play a role in the anticancer activity of F. hispida (bark and leaves) [18]. A previous study has shown that hispiloscine, a phenanthroindolizidine alkaloid isolated from the stems, bark, and leaves of the Malaysian F. hispida, showed appreciable antiproliferative activities against MDA-MB-231, MCF-7, A549, HCT-116, and MRC-5 cell lines [25]. O-methyltylophorinidine was isolated from the leaves and stalks of F. hispida and showed a potent cytotoxic effect on Lu1 (lung) and Col2 (colon) cancer cells [52]. A comparison of IC_50_ values to the crude extracts, catharanthine (41–47 µM or 13–16 µg/mL) was found to be higher than the bark and leaf extracts (0.7–1 µg/mL). Due to plants containing plenty of phytochemical substances and targeting several biological pathways, they can occasionally work synergistically to increase overall potency and solubility. Thus, these substances could sometimes be more potent than the pure substance [53,54]. In addition, extracts work in concert to enhance activity, potentially to lessen side effects and drug resistance, while pure compounds offer precision in targeting cancer cells.
We have shown, for the first time, that the anticancer effect attributed to catharanthine on CCA cells is present in the bark and leaves of F. hispida. Because there are no previous reports on the use of catharanthine against CCA, we can only compare the IC_50_ value to reported results for other cancers. We found that the IC_50_ value of catharanthine for CCA cells in this study was lower than that observed for HepG2 cells (IC_50_ at 130–135 µM) in the past. They demonstrated that catharanthine-triggered apoptosis was linked to increased autophagy signaling pathways by interacting with caspase-8, inhibiting the PI3K/Akt/mTOR pathway, upregulating Beclin-1, LC3, and ULK1, and enhancing SIRT1 expression [35]. In CCA cells, they may be cytotoxically affected by catharanthine via a synergistic anti-cancer effect through a multimodal mechanism of action typical of vinca alkaloids. Hence, it is necessary to further clarify the biological mechanisms of catharanthine that underlie its anti-CCA activity. Furthermore, in comparison to chemotherapy drugs, such as gemcitabine, the cytotoxic potency of catharanthine is typically lower in vitro and requires higher dosages to impede growth as much. Despite this, its different mechanisms of action, such as activating autophagy signaling pathways, highlight the potential of catharanthine for development as an adjuvant or combination therapy. This could potentially enhance the efficacy of treatment or help overcome drug resistance in cancer cells. Nonetheless, our study has limitations, as all findings were derived from in vitro experiments, which fail to adequately represent the complexity of tumor-stromal interactions or pharmacokinetic constraints in vivo. In addition to this, data on the stability, bioavailability, and metabolic pathway of catharanthine are lacking, limiting clinical interpretation. Concurrently with clarifying the mechanisms of action in CCA cells, further studies should focus on validating results in animal models, analyzing pharmacokinetics, and evaluating synergistic effects with standard chemotherapeutic agents to more accurately determine the therapeutic potential of catharanthine in cholangiocarcinoma.
5. Conclusion
Our findings show that F. hispida contains a wide range of metabolites and has a pharmacological potential that provides a valuable source for natural product development. The identification of catharanthine as a major cytotoxic metabolite provides new information about the bioactive compounds in this species, and establishes a promising basis for future mechanistic and therapeutic advancements against CCA. Further research should focus on elucidating the exact anticancer mechanisms by which catharanthine exerts anti-CCA effects, paving the way for the development of alternative chemotherapeutic agents.
Supporting information
S1 FilePlant material.Detailed information on plant identification.(PDF)
S1 TableMetabolites identified and phytochemical classes in various parts of F. hispida extract by UHPLC-QTOF-MS/MS analysis.(ZIP)
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Ripu MK, Bussmann R. Ficus (Fig) species in Nepal: A review of diversity and indigenous uses. J Ecol Appl. 2006;11:85–97.
- 2Ali M, Chaudhary N. Ficus hispida Linn.: A review of its pharmacognostic and ethnomedicinal properties. Pharmacogn Rev. 2011;5(9):96–102. doi: 10.4103/0973-7847.79104 22096323 PMC 3210011 · doi ↗ · pubmed ↗
- 3Kumar S, Singh A, Kushwaha AK, Tiwari R, Chaudhary LB, Srivastava M, et al. The UPLC–ESI–Qq Q LIT –MS/MS method for quantitative determination of phytochemicals in ethanolic extracts of different parts of eight Ficus species: Development and validation. International Journal of Food Properties. 2018;21(1):328–44. doi: 10.1080/10942912.2017.1414838 · doi ↗
- 4Cheng J-X, Zhang B-D, Zhu W-F, Zhang C-F, Qin Y-M, Abe M, et al. Traditional uses, phytochemistry, and pharmacology of Ficus hispida L.f.: A review. J Ethnopharmacol. 2020;248:112204. doi: 10.1016/j.jep.2019.112204 31669442 · doi ↗ · pubmed ↗
- 5Shen S, Zhan C, Yang C, Fernie AR, Luo J. Metabolomics-centered mining of plant metabolic diversity and function: Past decade and future perspectives. Mol Plant. 2023;16(1):43–63. doi: 10.1016/j.molp.2022.09.007 36114669 · doi ↗ · pubmed ↗
- 6Manickam S, Rajagopalan VR, Kambale R, Rajasekaran R, Kanagarajan S, Muthurajan R. Plant Metabolomics: Current Initiatives and Future Prospects. Curr Issues Mol Biol. 2023;45(11):8894–906. doi: 10.3390/cimb 45110558 37998735 PMC 10670879 · doi ↗ · pubmed ↗
- 7Plumb RS, Gethings LA, Rainville PD, Isaac G, Trengove R, King AM, et al. Advances in high throughput LC/MS based metabolomics: A review. Tr AC Trends in Analytical Chemistry. 2023;160:116954. doi: 10.1016/j.trac.2023.116954 · doi ↗
- 8Brindley PJ, Bachini M, Ilyas SI, Khan SA, Loukas A, Sirica AE, et al. Cholangiocarcinoma. Nat Rev Dis Primers. 2021;7(1):65. doi: 10.1038/s 41572-021-00300-2 34504109 PMC 9246479 · doi ↗ · pubmed ↗