GC-MS Profiling,Bioactivities,and Correlations of Dillenia indica L. Leaf,Stem,Unripe Fruit and Root Extracts

Abstract

Despite the widespread medicinal use of Dillenia indica L. across Asia, a comprehensive comparative analysis of its various morphological parts, specifically the root, has remained largely unexplored. This research introduces a novel comparative methodology to define the distinct chemical and biological signatures of four plant parts (leaf, stem, unripe fruit, and root). Notably, extracts from the root and stem exhibited approximately 2× higher DPPH scavenging activity and nearly 25× greater α-glucosidase inhibition compared to those from the leaf and fruit. The root extract specifically demonstrated a 5× higher hydrogen peroxide (H₂O₂) scavenging capacity than all other extracts, which strongly correlates with its elevated glyceride, phenolic, and xanthone content. Conversely, the leaf extract displayed selective α-amylase inhibition, an activity potentially attributed to its specific alcohol and terpenoid contents. Furthermore, all extracts demonstrated selective cytotoxicity against colorectal cancer cells over normal fibroblasts, with leaf extract showing the greatest potency (IC₅₀ = 297.83 µg/mL at 72 h). These findings highlight the part-specific bioactivities of D. indica, supporting the rational selection of plant parts for developing safe, targeted functional foods and herbal products for preventive healthcare and chronic disease management.

Keywords

phytochemical profiling enzyme inhibition Pearson correlation selective cytotoxicity

Introduction

Due to the rise of aging populations, there has been a sharp increase in age-related diseases worldwide, including Type 2 diabetes, neurodegeneration, cardiovascular disorders, and cancer. These diseases share common pathological pathways and cellular disturbances, such as oxidative stress, cellular senescence, metabolic dysregulation, and chronic inflammation. One of the most effective factors is reactive oxygen species (ROS), which play a central role in accelerating these processes by damaging biomolecules and disrupting normal cellular function and homeostasis. Consequently, identifying natural bioactive compounds that can modulate cellular ROS and/or prevent imbalances in ROS homeostasis can be an important and effective strategy for preventing and treating age-related health problems.¹

Dillenia indica L., commonly known as elephant apple (“Matad” in Thai), has been traditionally used as a medicine in many Southeast Asian countries for its various therapeutic properties, including anti-inflammatory, anti-diabetic, and antimicrobial effects. The use of this plant has been documented in ancient pharmacopeias, such as the Yajurveda, Sushruta Samhita, and Flora of China,² revealing a long history of medicinal applications. Despite its widespread traditional use in Thai communities for cooking and medicine, there has been limited scientific research on this plant in Thailand. This gap is likely due to a lack of documented indigenous knowledge, a weak connection between traditional practices and modern research, and not enough funding to properly study native medicinal plants. As a result, most published studies on D. indica come from other countries. These studies consistently show that the plant has various health benefits, including antioxidant, anti-diabetic, and anti-inflammatory effects.^3–5

Previous studies have primarily focused on extracts from a single plant part, such as fruits or leaves. However, they have not provided a comprehensive comparative analysis of different plant parts under the same experimental conditions. This lack of comprehensive comparison limits our understanding of the relative bioactivities and phytochemical profiles across the whole plant. Indeed, different parts of medicinal plants can accumulate unique sets of bioactive compounds, which may contribute variably to their therapeutic effects.

Thus, this present study aims to bridge this gap by systematically analyzing the phytochemical composition and biological activities of D. indica extracts from leaves, stems, unripe fruits, and roots. Notably, this is the first report to examine and contrast these four parts in parallel, especially regarding the chemical constituents of the root, which have been largely overlooked. Advanced Gas Chromatography-Mass Spectrometry (GC-MS) was used to identify key bioactive compounds in the extracts. These extracts were then evaluated for their antioxidant activity via DPPH and hydrogen peroxide (H_₂O₂) scavenging capacity, which reflects a protection against oxidative stress. Additional assays were performed to assess anti-hyperglycemic (α-amylase and α-glucosidase inhibition), hepatoprotective (β-glucuronidase inhibition), and cytotoxic effects on cancer and normal cells. Beyond individual activity profiling, correlations among plant parts, bioactivities, and phytochemical groups were analyzed to elucidate the relationships between specific chemical groups and their biological effects. This integrative approach highlights the therapeutic potential of distinct plant parts for targeted health applications. Considering the roles of oxidative stress, inflammation, metabolic dysfunction, and chronic disease, these findings scientifically validate the ethnomedicinal uses of D. indica. This also supports both the rational selection of raw materials and its potential for developing functional foods or herbal products aimed at preventive healthcare and healthy aging worldwide.

Materials and Methods

Chemicals and Instrumentations

All chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA), unless otherwise specified. For the antioxidant assays, the reagents used were 2,2-diphenyl-1-picrylhydrazyl (DPPH; Cat. No. D9132, Lot No. 0000177452), 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS; Cat. No. 11557, Lot No. BCCF0359), H₂O₂ (Cat. No. H1009, Lot No. MKCV5253), and ferrous sulfate (FeSO₄; Cat. No. 215422, Lot No. 0000139616). The enzyme inhibition studies utilized α-amylase (from Bacillus subtilis, Cat. No. A6817, Lot No. 0000483401), α-glucosidase (from Saccharomyces cerevisiae, Cat. No. G0660, Lot No. SLCC4854), and β-glucuronidase (from bovine liver, Cat. No. G0251, Lot No. 0000364254). The associated substrates and coloring reagents included soluble starch (Cat. No. S9765, Lot No. 0000297817), 3,5-dinitrosalicylic acid (DNS; Cat. No. 128848, Lot No. STBL3907), p-nitrophenyl-α-D-glucopyranoside (pNPG; Cat. No. N1377, Lot No. BCBW3619), and p-nitrophenyl-β-D-glucuronide (pNPGlc; Cat. No. 487500, Lot No. 4219135). Sodium carbonate (Na₂CO₃; Cat. No. 223530, Lot No. 0000265315) was employed as the stopping reagent. Additionally, quercetin (Cat. No. Q4951, Lot No. SLCQ4754) and acarbose (Cat. No. A8980, Lot No. WXBF0939V) were used as standard reference compounds. For the cytotoxicity evaluation, sulforhodamine B (SRB; Cat. No. S1402, Batch No. SHBR3296) was used.

All cell culture reagents, including Medium 106 (Cat. No. M10650, Lot No. 3118669), LSGS (Cat. No. S00310, Lot No. 3097265), fetal bovine serum (FBS; Cat. No. A5256701, Lot No. 2740173), 1× antibiotic-antimycotic (Cat. No. 15240062, Lot No. 265249), and 0.05% trypsin-EDTA (Cat. No. 15400054, Lot No. 3209605), were purchased from Gibco (Thermo Fisher Scientific, Grand Island, NY, USA). Dulbecco’s Modified Eagle Medium (DMEM; Cat. No. SH30003.02, Lot No. MK30804523C) was sourced from HyClone (Cytiva, Logan, UT, USA). Two cell lines were used in this study: human dermal fibroblasts-alpha (HDF-α; Cat. No. C0135C, Lot No. 2247748, Thermo Fisher Scientific, USA) as the normal cell model and SW480 colon cancer cells (Cat. No. CCL-228™, Lot No. 70043599, ATCC^→, Manassas, VA, USA) as the cancer cell model.

Plant extracts were concentrated using a rotary evaporator (Buchi, Rotavapor^→ R-300, Flawil, Switzerland). Phytochemical profiling was performed using a GC-MS system (7890A/5975C Series, Agilent Technologies Inc., Santa Clara, CA, USA). Absorbance measurements were recorded using a Varioskan™ multimode microplate reader (Thermo Fisher Scientific Inc., Waltham, MA, USA).

Preparation of Plant Samples and Extraction of Bioactive Compounds

Different parts of D. indica, including leaves, stems, unripe fruits and roots were collected from the Khlong Phikun community, Bang Prok Subdistrict, Mueang District, Pathum Thani Province, Thailand (GPS coordinates: 14.082653°N, 100.625331°E). A voucher specimen was deposited at the Department of Thai Traditional and Alternative Medicine, Bangkok, Thailand, under voucher number TTM No. 005499. The plant materials were thoroughly cleaned and then dried in a hot air oven at approximately 50°C. Each part of the plant was then extracted using ethanol. Specifically, 500 g of dried material was macerated in 95% ethanol with occasional stirring for 3 days at room temperature. The mixtures were filtered through Whatman No. 1 filter paper, and the filtrates were concentrated using a rotary evaporator at 40°C and 600 rpm until the solvent was completely removed. All crude extracts were weighed, and the percentage yields were calculated based on the initial dry weights. The extracts were stored at –20°C until further use.

Phytochemical Analysis by GC-MS

GC-MS was employed to analyze the phytochemical constituents and generate a chromatographic fingerprint of the D. indica extract. The analysis was conducted using an Agilent 7890A GC system with a 5975C Mass Selective Detector (MSD) equipped with a 30 m × 0.25 mm × 0.25 µm capillary column. The instrument parameters were established from the method described by Ruangsuriya et al. (2023).⁶ Phytochemical constituents were tentatively identified by comparing their retention times, peak areas, and mass spectral patterns with those in the W08N08 mass spectral library (John Wiley & Sons, Inc., USA). Additional confirmation of identified compounds was conducted through database comparison using PubChem (https://pubchem.ncbi.nlm.nih.gov/) to verify structural and chemical properties. Subsequently, the phytochemicals were classified into main and sub-groups of chemicals. Pie graphs displaying the percentage area profiles of these chemical groups within each extract were constructed and arranged in alphabetical order for enhanced comparative analysis among the extracts.

DPPH Radical Scavenging Assay

The free radical scavenging activity of the extracts was assessed using the DPPH assay according to the method described by Brand-Williams et al. (1995),⁷ with slight modifications. A stock solution of 0.2 mM DPPH was prepared by dissolving 7.89 mg of DPPH in 100 mL of analytical grade methanol. The solution was kept in an amber flask, stored at 4°C and used within 24 hours to ensure radical stability. To evaluate the radical scavenging activity, 100 µL of each extract at various concentrations was mixed with 100 µL of 0.2 mM DPPH solution in methanol. The mixture was incubated in the dark at room temperature for 30 minutes. Absorbance was measured at 517 nm using a microplate reader. The percentage of radical scavenging activity was calculated based on the absorbance values compared to a control (blank). Ascorbic acid was used as a positive control. A stock solution (1 mg/mL) was prepared in methanol and further serially diluted with methanol to obtain the working concentrations.

Anti-oxidative Stress Evaluation Using an H₂O₂-mediated Hydroxyl Radical Scavenging Assay

The H₂O₂-mediated hydroxyl radical scavenging activity of the extracts was evaluated based on the method of Wang et al.,⁸ with slight modifications. This assay employed ABTS in conjunction with FeSO₄ via the Fenton reaction to generate hydroxyl radicals for assessing antioxidant capacity. The ABTS/FeSO₄ working solution was prepared by dissolving 109.8 mg of ABTS and 2.78 mg of FeSO₄ in 100 mL of distilled water, yielding final concentrations of 2 mM ABTS and 100 μM FeSO₄. For the reaction, 100 μL of this solution was mixed with 50 μL of freshly prepared 1 mM H₂O₂ and 50 μL of the extract at various concentrations. Distilled water was used as the control (blank), and quercetin was included as a positive control. The quercetin stock solution was prepared by dissolving the powder in absolute ethanol to a concentration of 1 mg/mL, followed by serial dilution in distilled water to reach the working concentrations. The mixture was incubated in the dark at room temperature for 15 minutes, followed by measurement of absorbance at 414 nm. The H₂O₂ scavenging activity was calculated using the following equation:

% I n h i b i t i o n = [(A c o n t r o l - A s a m p l e) / A c o n t r o l] \times 100.

The effective concentration at 50% inhibition (EC₅₀) value was determined.

Inhibition of α-amylase and α-glucosidase Activity Assays

The inhibitory activities of the plant extracts against α-amylase and α-glucosidase were determined using spectrophotometric methods adapted from published protocols.⁹

Inhibition of α-amylase

Plant extracts at various concentrations (100 µL) were mixed with 100 µL of α-amylase enzyme solution (from B. subtilis, 1,380 U/mg). The enzyme was prepared in 0.02 M phosphate buffer, pH 6.9, at a working concentration of 3 U/mL to achieve a final reaction concentration of 1 U/mL. This was achieved by dissolving approximately 0.00217 mg of the enzyme powder per 1 mL of buffer using a master stock and subsequent serial dilution to get a working concentration. The mixture was incubated at 37°C for 10 minutes. Then, 100 µL of 1% soluble starch solution, prepared by dissolving 1 g of starch in 100 mL of distilled water with gentle heating, was added as a substrate. After incubation at 37°C for 15 minutes, the reaction was stopped by adding 200 µL of 3,5-dinitrosalicylic acid (DNS) reagent and boiling for 5 minutes. After cooling, absorbance was measured at 540 nm. Acarbose was used as a positive control. The acarbose stock solution was prepared by dissolving the powder in deionized water (or 0.02 M phosphate buffer) to a concentration of 1 mg/mL, which was then further diluted to a range of 10–200 µg/mL to determine the EC₅₀ values.

Inhibition of α-glucosidase

Plant extracts (100 µL) were mixed with 100 µL of α-glucosidase enzyme solution (from S. cerevisiae, 213 U/mg) prepared in phosphate buffer (0.1 M, pH 6.8). To achieve a final reaction concentration of 1 U/mL, the enzyme was prepared at a working concentration of 3 U/mL by dissolving approximately 0.014 mg of the lyophilized powder per 1 mL of buffer via precise serial dilution of a concentrated master stock. The mixture was incubated at 37°C for 10 minutes before adding 100 µL of 5 mM pNPG, which was prepared by dissolving 1.51 mg pNPG in 1 mL of 0.1 M phosphate buffer (pH 6.8). After incubation at 37°C for 20 minutes, the reaction was terminated with 1 mL of 0.1 M Na₂CO₃ (prepared by dissolving 10.6 mg of anhydrous Na₂CO₃ in 1 mL of deionized water). Absorbance of the reaction was recorded at 405 nm. Acarbose served as the positive control.

Percentage inhibition was calculated relative to controls without extracts. EC₅₀ values were derived from dose-response curves.

Indirect Hepatoprotective Evaluation by β-glucuronidase Inhibition Assay

The potential indirect hepatoprotective effect of the extracts was evaluated by assessing their inhibitory activity against β-glucuronidase, an enzyme associated with liver inflammation and xenobiotic toxicity. The assay protocol was adapted from Acharya and De (2016)¹⁰ with minor modifications. The β-glucuronidase enzyme (from bovine liver, 1,146.256 U/mg) was prepared at a stock concentration of 10,000 U/mL by dissolving approximately 8.72 mg of the lyophilized powder in 1 mL phosphate buffer (pH 7.0). Briefly, 170 μL of the plant extract (or buffer for the control) was pre-incubated with 50 μL of the β-glucuronidase enzyme solution at 37°C for 10 minutes. The reaction was initiated by adding 30 μL of 10 mM pNPGlc, which was prepared by dissolving 3.15 mg of pNPGlc in 1 mL of the same phosphate buffer. The total reaction mixture was incubated at 37°C for 60 minutes and then terminated by adding 50 μL of 0.1 M Na₂CO₃ as the stop solution. The final enzymatic activity in the total volume (300 μL) was 1,666.67 U/mL after dilution in the reaction mixture. The absorbance was measured at 405 nm. A control reaction, a buffer containing no extract, was used to define 0% inhibition, while a blank without enzyme accounted for background absorbance. Quercetin was used as a standard reference. The EC₅₀ value represents the concentration of extract required to inhibit 50% of enzyme activity. Inhibition of β-glucuronidase activity indicates a potential for the extract to attenuate liver inflammation and toxicity, supporting its role as an indirect hepatoprotective agent.

Relative Efficiency Calculation and Visualization

To facilitate a direct comparison of biological activities, all bioassay results were converted into percentage relative efficiency (%RE). The positive control for each assay was defined as the 100% efficiency benchmark: ascorbic acid for the DPPH assay, quercetin for the H₂O₂ scavenging and β-glucuronidase inhibition assays, and acarbose for both the α-amylase and α-glucosidase inhibition assays. The %RE was determined by expressing the potency of the extract as a percentage relative to the control, calculated as %RE = (EC50 control/ EC50 sample) × 100. Histograms were subsequently generated to visualize the distribution of these efficiencies across the four plant parts, providing a standardized measure of their comparative therapeutic potential.

Cell Culture and Cytotoxicity Assay

HDF-α and SW480 were cultured under standard incubation conditions at 37°C in a humidified atmosphere with 5% CO₂. HDF-α cells were maintained in Medium 106 supplemented with Low Serum Growth Supplement (LSGS) and 1× antibiotics-antimycotics. SW480 cells were cultured in DMEM supplemented with 10% FBS and 1× antibiotics-antimycotics. The culture medium was replaced every 2–3 days. When cells reached 70%–80% confluence, sub-culturing was performed using 0.05% trypsin-EDTA. The cell lines were authenticated by their respective suppliers (Thermo Fisher Scientific and ATCC) and were certified to be free of mycoplasma contamination, as verified in the manufacturer’s Certificate of Analysis (COA). Cells between passage numbers 3–5 were used for this experiment, where passage one was defined as the first subculture following receipt recovery from the vendor.

Cytotoxicity effects of the plant extracts were assessed in both HDF-α cells and SW480 using the SRB assay. HDF-α cells served as a normal cell control for comparison with cancer cells. Cells were seeded in 96-well microplates at a density of 7 × 10⁴ cells/mL using a plating volume of 100 µL/well (providing a final density of 7,000 cells/well). After a 24-hour attachment period, cells were treated with various concentrations of plant extracts or quercetin (a standard reference). Cell viability was evaluated at both 24 and 72 hours of treatment using the SRB assay.¹¹ The half-maximal inhibitory concentration (IC₅₀) values for both the plant extracts and quercetin were calculated using the PriProbit Program version 1.63.¹² The selective index (SI) was calculated using the formula: SI = IC50(normal cell)/IC50(cancer cell). An SI value > 1 indicates that the compound is more toxic to cancer cells than to normal cells, which is considered desirable for anticancer agents. Conversely, an SI value < 1 suggests higher toxicity to normal cells, which is unfavorable. Higher SI values reflect better selectivity and imply a safer and more effective therapeutic potential.⁶

Comparative Bioactivity Profiling via Normalized Radar Plot Analysis

Radar plots were utilized to compare the bioactivities across different extracts. Data processing involved two sequential stages: (a) Normalization, where the raw data from each assay were converted into %RE by comparing them against their respective reference controls; and (b) Scaling, where these %RE values were divided by the minimum %RE observed within each assay (Scaling = Extract%RE / Min%RE). This established a uniform baseline of one for the least active sample, allowing for a direct and clear visualization of the relative potency across all plant parts.

Correlation Analysis Between Phytochemical Groups and Biological Activities

Phytochemical groups identified by GC-MS in each extract were categorized into chemical groups, and the relative abundance of each group was expressed as a percentage of the total area (% area). These % area values were then correlated with the normalized bioactivity data (as described in the above section) for each extract using Pearson’s correlation coefficient in Microsoft Excel, and a heatmap was generated. The resulting correlation coefficients, which range from –1 (indicating a perfect negative linear relationship) to +1 (indicating a perfect positive linear relationship), with 0 representing no linear association, were visualized through the heatmap to highlight significant relationships between phytochemical classes and biological activities. Given the small sample size inherent to this four-partcomparative design (n = 4), Pearson correlations are presented as descriptive and exploratory. Statistical significance testing was not applied, and results should be interpreted accordingly.

Statistical Analysis

All bioassays were performed in five replicates. Data are presented as mean ± standard deviation (SD). Statistical analyses were conducted using SPSS and Microsoft Excel. Differences among groups were evaluated using one-way analysis of variance (ANOVA), followed by Tukey’s honestly significant difference (HSD) post-hoc test. A p < .05 was considered statistically significant.

Results

Extraction Yield and Phytochemical Composition of D. indica Extracts

The ethanolic extraction yields of D. indica showed that the leaf extract reached the highest yield at 16.16%, followed by the unripe fruit extract at 9.13%. Root and stem extracts exhibited lower yields of 3.59% and 4.20%, respectively. These differences suggest that leaves and unripe fruits contain higher amounts of extractable phytochemicals compared to roots and stems under the extraction conditions used.

GC-MS analysis was performed to identify and quantify phytochemicals present in the leaf, stem, unripe fruit, and root extracts of D. indica. The identified compounds with a match quality of ≥80% are summarized in Tables 1–4. Chromatograms and retention times were provided in Supplementary Information (Figures S1–S4 and Tables S1–S4). The phytochemicals were then grouped into chemical classes, and their relative abundances (% area) were illustrated as pie charts for each extract (Figure 1).

Figure 1.

Relative Proportion (% Area) of Phytochemical Groups in Dillenia indica Extracts. Pie Charts Show the Percentage Area of Each Chemical Group Identified in the (A) Leaf, (B) Stem, (C) Unripe Fruit, and (D) Root Extracts.

A notable finding was that esters were the major identified phytochemical group across all four extracts (Figure 1). The highest proportion was found in the unripe fruit extract (57.2%), followed by the stem (47.01%), root (41.96%), and leaf (26.73%) extracts, respectively.

While esters were dominant, the extracts also displayed unique secondary and minor phytochemical profiles. The leaf extract was characterized by significant proportions of alcohols (18.24%) and terpenoids (6.49%) (Figure 1A). Other identified groups in the leaf extract, each contributing less than 1%, included carboxylic acids, heterocyclic compounds, alkenes, ketones, phenolic compounds, and lactones. A considerable fraction of this extract (44.09%) was composed of unidentified miscellaneous compounds.

In contrast, the stem extract contained a diverse profile, with carboxylic acids (14.14%), glycerides (5.21%), and aldehydes (4.95%) as its most prominent secondary groups (Figure 1B). It also contained 4.0% phenolic compounds, as well as minor amounts of xanthones (0.77%), thiochromenes (0.58%), and other minor groups. Unidentified compounds made up 22.17% of the stem extract.

The unripe fruit extract stood out for its high proportion of aldehydes (24.49%) as the second most abundant group (Figure 1C). This was followed by a smaller fraction of unidentified compounds (8.36%) and carboxylic acids (4.53%). Other identified phytochemicals included phenolic compounds (0.84%), terpenoids (0.65%), along with minor contributions from furanones, heterocyclic compounds, ketones, alcohols, and pyrone.

Finally, the root extract contained a substantial amount of phenolic compounds (9.95%) and carboxylic acids (9.93%) (Figure 1D). It also contained the highest concentration of xanthones (1.49%) among all extracts, with glycerides (6.94%) also being a notable component. The remaining 27.01% was classified as unidentified compounds and miscellaneous.

Bioactivities of D. indica Extracts

DPPH Radical Scavenging Activity

The DPPH radical scavenging activity of D. indica extracts at 100 µg/mL is shown as % relative efficiencies compared to ascorbic acid (Figure 2A). Among the extracts, the stem showed the highest scavenging activity, followed by the root extract. The leaf extract demonstrated moderate activity, whereas the unripe fruit exhibited the lowest.

Figure 2.

Percentage Relative Efficiencies of the Ethanolic Extracts from Dillenia indica Compared to Standard Substances in (A) DPPH Scavenging Activity, (B) H₂O₂ Scavenging Activity, (C) α-Amylase Inhibition, (D) α-Glucosidase Inhibition, and (E) β-Glucuronidase Inhibition. Different Letters (a, b, c, d) Above the Bars Indicate Statistically Significant Differences Among Means (p < .05).

Anti-oxidative Stress Evaluation via H₂O₂ Scavenging Activity

The abilities of D. indica extracts to scavenge H₂O₂ were assessed and expressed as % relative efficiencies compared to quercetin (Figure 2B). The results revealed that the root extract exhibited the most potent H₂O₂ scavenging activity, followed by the stem, unripe fruit, and leaf extracts.

Inhibition of α-amylase and α-glucosidase Activity

The inhibitory activities of D. indica extracts against α-amylase and α-glucosidase were assessed. The % relative efficiencies of these two enzymes, in comparison to acarbose (Figure 2C and 2D). The most potent α-amylase inhibitory activity was observed in the leaf extract, as indicated by its highest relative efficiency, followed by the stem, root, and unripe fruit extracts. In contrast, the stem and root extracts exhibited the strongest α-glucosidase inhibitory activities, with potencies nearly comparable to acarbose, followed by the leaf and unripe fruit extracts.

Indirect Hepatoprotective Evaluation by β-glucuronidase Inhibition

The β-glucuronidase inhibitory activities of D. indica extracts were evaluated and expressed as % relative efficiency (Figure 2E). The stem extract exhibited the strongest inhibition, followed by the root, leaf, and unripe fruit extracts. The inhibitory activities of the leaf and unripe fruit extracts were not significantly different.

The raw data, % inhibition and EC₅₀ of all bioactivity assays were provided in Supplementary Information Tables S5–S8.

Cytotoxicity of D. indica Extracts

The cytotoxic effects of D. indica extracts on HDF-α and SW480 were evaluated at 24- and 72-hours using IC₅₀ values (Table 5). After 24 hours, leaf extract exhibited moderate cytotoxicity on SW480 cells while maintaining low toxicity on HDF-α. The unripe fruit extract showed weak cytotoxicity on SW480 cells and minimal effects on HDF-α. In contrast, the root and stem extracts had very low cytotoxic effects on HDF-α cells, suggesting negligible toxicity, even at high concentrations. However, after 72 hours, all extracts demonstrated increased cytotoxicity, especially on SW480 cells. Notably, the leaf extract exhibited enhanced potency against SW480 while still showing limited toxicity toward HDF-α. The root extract remained the least cytotoxic to normal cells, while showing moderate activity on cancer cells. These results indicated that the selective cytotoxic potential of D. indica, particularly the leaf extract, toward cancer cells than normal cells. In addition, the reference compound, quercetin, exhibited strong time-dependent cytotoxicity against SW480 cells, as evidenced by the markedly reduced IC₅₀ value at 72 hours. However, quercetin also affected HDF-α cells, particularly after prolonged exposure. Furthermore, the SI of all D. indica extracts, as well as quercetin, exhibited SI values greater than 1.

Table 1.

Phytochemicals Identified by GC-MS in the Dillenia indica Leaf Extract. Retention times, match quality, and corresponding GC–MS chromatogram are provided in the Supplementary Information.

No.	Compound Name	Formula	MW (g/mol)	Categorize		Area (%)
No.	Compound Name	Formula	MW (g/mol)	Main Group	Subgroup	Area (%)
1	Trans-phytol	C₂₀H₄₀O	296.5	Alcohol	Diterpenoid alcohol	16.07
2	Linolenic acid ethyl ester	C₂₀H₃₆O₂	306.5	Ester	Unsaturated fatty acid ethyl ester	12.21
3	Palmitic acid ethyl ester	C₁₈H₃₆O₂	284.5	Ester	Fatty acid ethyl ester	11.34
4	Neophytadiene	C₂₀H₃₈	278.5	Terpenoid	Diterpene	6.49
5	Stearic acid, ethyl ester	C₂₀H₄₀O₂	312.5	Ester	Saturated fatty acid ethyl ester	3.18
6	Isophytol	C₂₀H₄₀O	296.5	Alcohol	Diterpenoid alcohol	2.17
7	Benzimidazol-2-one	C₇H₄N₂O	132.12	Heterocyclic compound	Benzimidazolone	0.89
8	1-Octadecene	C₁₈H₃₆	252.5	Alkene	Aliphatic hydrocarbon	0.88
9	6,10,14-Trimethylpenta decan-2-one	C₁₈H₃₆O	268.5	Ketone	Aliphatic ketone	0.81
10	Palmitic acid	C₁₆H₃₂O₂	256.42	Carboxylic acid	Saturated fatty acid	0.81
11	4-Heptyloxyphenol	C₁₃H₂₀O₂	208.3	Phenolic compound	Alkyl aryl ether	0.61
12	5-Methyl-5-(4,8,12-trimethyltridecyl)oxolan-2-one	C₂₁H₄₀O₂	324.5	Lactone	Macrocyclic lactone	0.45
13	Miscellaneous	–	–	–	–	44.09

Table 2.

Phytochemicals Identified by GC-MS in the Dillenia indica Stem Extract. Retention Times, Match Quality, and Corresponding GC–MS Chromatogram Are Provided in the Supplementary Information.

No.	Compound Name	Formula	MW (g/mol)	Categorize		Area (%)
No.	Compound Name	Formula	MW (g/mol)	Main Group	Subgroup	Area (%)
1	Linoleic acid ethyl ester	C₂₀H₃₆O₂	308.5	Ester	Unsaturated fatty acid ethyl ester	19.70
2	Palmitic acid ethyl ester	C₁₈H₃₆O₂	284.5	Ester	Saturated fatty acid ethyl ester	13.31
3	Oleic acid ethyl ester	C₂₀H₃₈O₂	310.5	Ester	Unsaturated fatty acid ethyl ester	10.13
4	2-Palmitoylglycerol	C₁₉H₃₈O₄	330.5	Glyceride	Monoglyceride	5.21
5	Palmitic acid	C₁₆H₃₂O₂	256.42	Carboxylic acid	Saturated fatty acid	4.10
6	Oleic acid	C₁₈H₃₄O₂	282.5	Carboxylic acid	Unsaturated fatty acid	3.74
7	(7Z,10Z)-Hexadeca-7,10-dienal	C₁₆H₂₈O	236.39	Aldehyde	Unsaturated aldehyde	2.87
8	Homovanillic acid	C₉H₁₀O₄	182.17	Carboxylic acid	Phenolic acid	2.80
9	Linoleic acid	C₁₈H₃₂O₂	280.4	Carboxylic acid	Unsaturated fatty acid	2.77
10	Stearic acid ethyl ester	C₂₀H₄₀O₂	312.5	Ester	Saturated fatty acid ethyl ester	2.57
11	5-Hydroxymethylfurfural	C₆H₆O₃	126.11	Aldehyde	Furan aldehyde	2.08
12	(E)-Isoeugenol	C₁₀H₁₂O₂	164.2	Phenolic compound	Phenylpropanoid	1.53
13	2-Methoxybenzene-1,4-diol	C₇H₈O₃	140.14	Phenolic compound	Methoxyphenol	1.31
14	4-Vinylguaiacol	C₉H₁₀O₂	150.17	Phenolic compound	Phenylpropanoid	1.16
15	5-Chloro-1,8-dihydroxy-4-methoxy-3-methyl-9H-xanthene-9-one	C₁₅H₁₁ClO₅	306.7	Xanthone	Dihydroxyxanthone	0.77
16	Elaidic acid ethyl ester	C₂₀H₃₈O₂	310.5	Ester	Unsaturated fatty acid ethyl ester	0.74
17	(13Z,16Z)-Docosa-13,16-dienoic acid	C₂₂H₄₀O₂	336.6	Carboxylic acid	Unsaturated fatty acid	0.73
18	7-Methoxy-2,2-dimethylthiochromene	C₁₂H₁₄OS	206.31	Thiochromene	Heterocyclic compound	0.58
19	Heptadecanoic acid, ethyl ester	C₁₉H₃₈O₂	298.5	Ester	Saturated fatty acid ethyl ester	0.56
20	Homovanillyl alcohol	C₉H₁₂O₃	168.19	Alcohol	Phenolic alcohol	0.47
21	Cyclopent-4-ene-1,3-dione	C₅H₄O₂	96.08	Ketone	Cyclic diketone	0.37
22	2H-Furan-5-one	C₄H₄O₂	84.07	Furanone	Cyclic ether ketone	0.33
23	Miscellaneous	–	–	–	–	22.17

Table 3.

Phytochemicals Identified by GC-MS in the Dillenia indica Unripe Fruit Extract. Retention Times, Match Quality, and Corresponding GC–MS Chromatogram Are Provided in the Supplementary Information.

No.	Compound Name	Formula	MW (g/mol)	Categorize		Area (%)
No.	Compound Name	Formula	MW (g/mol)	Main Group	Subgroup	Area (%)
1	5-Hydroxymethylfurfural	C₆H₆O₃	126.11	Aldehyde	Furan aldehyde	23.43
2	Ethyl Oleate	C₂₀H₃₈O₂	310.5	Ester	Unsaturated fatty acid ethyl ester	22.72
3	Ethyl palmitate	C₁₈H₃₆O₂	284.5	Ester	Saturated fatty acid ethyl ester	14.46
4	Ethyl linoleate	C₂₀H₃₆O₂	306.5	Ester	Unsaturated fatty acid ethyl ester	12.47
5	Diethyl malate	C₈H₁₄O₅	190.19	Ester	Dicarboxylic acid ester	2.62
6	Oleic acid	C₁₈H₃₄O₂	282.5	Carboxylic acid	Unsaturated fatty acid	2.54
7	Ethyl stearate	C₂₀H₄₀O₂	312.5	Ester	Saturated fatty acid ethyl ester	2.50
8	Palmitic acid	C₁₆H₃₂O₂	256.42	Carboxylic acid	Saturated fatty acid	1.99
9	Methyl furan-2-carboxylate	C₆H₆O₃	126.11	Ester	Furan ester	1.41
10	5-Methylfurfural	C₆H₆O₂	110.11	Aldehyde	Furan aldehyde	1.06
11	7-Phenylindeno[2,1-c] pyridin-9-one	C₁₈H₁₁NO	257.3	Heterocyclic compound	Indeno[2,1-c]pyridinone	0.91
12	3H-furan-2-one	C₄H₄O₂	84.07	Furanone	Lactone	0.82
13	Isopimara-7,15-dien-3beta-ol	C₂₀H₃₂O	288.475	Alcohol	Diterpenoid alcohol	0.80
14	Cyclopent-4-ene-1,3-dione	C₅H₄O₂	96.08	Ketone	Cyclic diketone	0.75
15	Ethyl myristate	C₁₆H₃₂O₂	256.42	Ester	Saturated fatty acid ethyl ester	0.67
16	Epoxycembrene A	C₂₀H₃₂O	288.5	Terpenoid	Diterpenoid	0.65
17	Syringol	C₈H₁₀O₃	154.16	Phenolic compound	Dimethoxyphenol	0.45
18	2-Methyl-2H-furan-5-one	C₅H₆O₂	98.1	Furanone	Lactone	0.41
19	Ethyl heptadecanoate	C₁₉H₃₈O₂	298.5	Ester	Saturated fatty acid ethyl ester	0.35
20	Pyrocatechol (catechol)	C₆H₆O₂	110.11	Phenolic compound	Diphenol	0.29
21	Maltol	C₆H₆O₃	126.11	Pyrone	Cyclic ketone compound	0.24
22	Phenol	C₆H₆O	94.11	Phenolic compound	Monophenol	0.10
23	Miscellaneous	–	–	–	–	8.36

Table 4.

Phytochemicals Identified by GC-MS in the Dillenia indica Root Extract. Retention Times, Match Quality, and Corresponding GC–MS Chromatogram Are Provided in the Supplementary Information.

No.	Compound Name	Formula	MW (g/mol)	Categorize		Area (%)
No.	Compound Name	Formula	MW (g/mol)	Main Group	Subgroup	Area (%)
1	Oleic acid ethyl ester	C₂₀H₃₈O₂	310.5	Ester	Unsaturated fatty acid ethyl ester	13.47
2	Linoleic acid ethyl ester	C₂₀H₃₆O₂	308.5	Ester	Unsaturated fatty acid ethyl ester	12.65
3	Palmitic acid ethyl ester	C₁₈H₃₆O₂	284.5	Ester	Saturated fatty acid ethyl ester	12.48
4	2-Palmitoylglycerol	C19H38O4	330.5	Glyceride	Monoglyceride	6.94
5	Palmitic acid	C₁₆H₃₂O₂	256.42	Carboxylic acid	Saturated fatty acid	4.18
6	Oleic acid	C₁₈H₃₄O₂	282.5	Carboxylic acid	Unsaturated fatty acid	3.74
7	Pyrogallol	C₆H₆O₃	126.11	Phenolic compound	Triphenol	3.33
8	Stearic acid ethyl ester	C20H40O2	312.5	Ester	Saturated fatty acid ethyl ester	2.58
9	2-Methoxybenzene-1,4-diol	C₇H₈O₃	140.14	Phenolic compound	Methoxyphenol	2.14
10	Linoleic acid	C₁₈H₃₂O₂	280.4	Carboxylic acid	Unsaturated fatty acid	2.01
11	Phenol	C₆H₆O	91.44	Phenolic compound	Monophenol	1.91
12	5-Chloro-1,8-dihydroxy-4-methoxy-3-methyl-9H-xanthene-9-one	C₁₅H₁₁ClO₅	306.7	Xanthone	Dihydroxyxanthone	1.49
13	4-vinylguaiacol	C₉H₁₀O₂	150.17	Phenolic compound	Phenylpropanoid	1.29
14	Pyrocatechol (catechol)	C₆H₆O₂	110.11	Phenolic compound	Diphenol	1.28
15	Furan-2-ylmethanol	C₅H₆O₂	98.1	Alcohol	Furfuryl alcohol	1.06
16	(E)-Octadec-5-ene	C₁₈H₃₆	252.5	Alkene	Aliphatic hydrocarbon	0.96
17	Linolelaidic acid, methyl ester	C19H34O2	294.5	Ester	Unsaturated fatty acid ethyl ester	0.78
18	Cyclopent-4-ene-1,3-dione	C₅H₄O₂	96.08	Ketone	Cyclic diketone	0.7
19	Miscellaneous	–	–	–	–	27.01

Table 5.

Cytotoxicity (IC₅₀ Values in µg/mL) of Dillenia indica Extracts on Human Dermal Fibroblasts (HDF-α) and Colorectal Cancer Cells (SW480) at 24 and 72 Hours.

Extracts/Substances	24 Hours		SI	72 Hours		SI
	HDF-α	SW480		HDF-α	SW480
	IC₅₀ (µg/mL)			IC₅₀ (µg/mL)
Leaf	95,679.04	504.65	189.59	1,424.27	297.83	4.78
Stem	∞	1,468.47	∞	4,572.63	1,071.80	4.27
Unripe fruit	97,823.55	3,778.30	25.89	1,714.48	622.95	2.75
Root	∞	7,159.40	∞	5,454.11	3,952.01	1.38
Quercetin	1,636.27	525.82	3.11	153.64	28.02	5.48

Note: ∞ indicates that 50% inhibition of cell viability was not achieved at the highest tested concentration; therefore, an IC₅₀ value could not be calculated. This indicates negligible cytotoxicity toward HDF-α normal cells at the concentrations evaluated.

Radar Plot Analysis of D. indica Extracts and Bioactivities

The bioactivity of the D. indica extracts was assessed across five functional assays and illustrated in Figure 3. The stem and root extracts generally demonstrated the highest and most consistent bioactivity. Both extracts showed moderate radical scavenging activity against DPPH (2.08 and 2.04, respectively), with the root extract exhibiting the highest activity in the H₂O₂ scavenging assay (10.10). In enzyme inhibition, the root extract was the strongest inhibitor of α-glucosidase (32.14), closely followed by the stem extract (28.56). The stem extract also demonstrated the most potent inhibition of β-glucuronidase (1.59). In contrast, the unripe fruit extract showed consistently weak bioactivity, with negligible activity in DPPH scavenging, H₂O₂ scavenging, α-amylase inhibition, and β-glucuronidase inhibition (all approximately 1.00). The leaf extract presented a more specific profile; while it showed moderate DPPH scavenging (1.60) and was the most potent inhibitor of α-amylase (2.69), its activity in other assays, including H₂O₂ scavenging and α-glucosidase inhibition, was minimal. Cytotoxic activity was not included in the radar plot because the cytotoxicity assay employed a distinct cell-based endpoint (IC₅₀ via SRB assay) that is not directly comparable to the enzyme- and radical-based %RE values used for the other five assays. This precluded meaningful normalization and integration into the radar plot framework.

Figure 3.

Radar Plot Exhibits the Relative Bioactivity of Different Extracts of Dillenia indica Across Five Biochemical Assays: DPPH Radical Scavenging, H₂O₂ Scavenging, Inhibition of α-Amylase, α-Glucosidase, and β-Glucuronidase. The Values in the Radar Plot Represent the Relative Potency Scores of Leaf, Stem, Unripe Fruit, and Root Extracts, Calculated Based on Their Normalized %RE Values. Higher Scores Indicate Greater Bioactivity, Representing the Fold-Increase in Potency Relative to the Least Active Extract (Baseline = 1). Among the Samples, Stem and Root Extracts Exhibit the Highest α-Glucosidase Inhibition, While Fruit Extract Shows Minimal Relative Activity Across All Assays. Cytotoxic Activity Was Not Included in the Radar Plot Because the Cytotoxicity Assay Employed a Distinct Cell-Based Endpoint.

Correlation Analysis and Heatmap of Phytochemical Constituents and Bioactivities

Pearson correlation analysis revealed clear associations between chemical groups and bioactivities (Figure 4). Glycerides showed strong positive relationships with antioxidant capacity and enzyme inhibition, particularly α-glucosidase. Carboxylic acids, phenolic compounds, and xanthones were also positively associated with antioxidant and anti-diabetic activities, indicating their major contribution to the observed bioactivities. Thiochromenes were mainly linked to β-glucuronidase inhibition.

Figure 4.

Heatmap Illustrates the Pearson Correlation Between Phytochemical Groups and Bioactivities. The Color Scale Indicates the Strength and Direction of the Correlation Coefficient (r). Green Indicates a Positive Correlation, Red Indicates a Negative Correlation, and Yellow Indicates a Weak or Negligible Correlation. The color scale reflects the strength and direction of correlation (r values). Given the exploratory nature of the analysis (n = 4), these correlations are presented as hypothesis-generating and should be interpreted with appropriate caution.

In contrast, heterocyclic compounds, aldehydes, furanones, and pyrones exhibited negative associations with antioxidant and enzyme inhibitory activities. Alcohols and terpenoids showed mixed effects, with positive associations limited to α-amylase inhibition, while esters were negatively correlated with this activity and showed minimal influence on other assays.

Overall, the results demonstrate that specific chemical groups distinctly contribute to the bioactivities of D. indica extracts, with glycerides, phenolics, and xanthones as key positive contributors.

Discussion

Although numerous studies have reported the pharmacological properties of D. indica, including antioxidant, anti-inflammatory, and anti-hyperglycemic activities, few have systematically compared different plant parts under identical conditions or linked their phytochemical profiles to bioactivity, particularly for the root, where data remain scarce. This study addressed that gap by analyzing leaves, stems, unripe fruits, and roots side by side, integrating GC-MS profiling with bioactivity assays and chemometric tools such as relative efficiency bar charts, radar plots, and correlation heatmaps. These approaches allowed rapid visualization of trends and associations, consistent with current practices in natural product research.¹³

Our extraction yield data showed significant variation among plant parts, with leaves producing the highest yield, followed by unripe fruits, roots, and stems. This reflects differences in ethanol-soluble phytochemical content and aligns with GC-MS findings showing alcohols and esters as major components in leaves and fruits. Ethanol’s polarity and tissue penetration capacity make it a suitable solvent for diverse bioactive compounds, including phenolics and flavonoids.^14,15 However, a higher yield did not necessarily correspond to stronger bioactivity.

Previous studies have comprehensively characterized the antioxidant, anti-diabetic, anti-inflammatory, and cytotoxic properties of D. indica extracts.^2–5 Our findings are consistent with these reports, both in terms of biological activities and the types of phytochemicals identified across plant parts.^4,16 However, unlike earlier work, which focused on single plant parts, our study provides a direct comparative analysis under identical conditions and integrates chemical composition data with bioactivity outcomes.

The bioactivity data demonstrates clear part-specific differences linked to phytochemical composition. Stem and root extracts showed superior DPPH scavenging activity, correlating with phenolic compounds, glycerides, and xanthones, all known potent antioxidants.^17–19 A similar pattern appeared in H₂O₂ scavenging, where phenolics and xanthones strongly correlated with activity.^20,21 The lower activity of the leaf extract may result from its high diterpenoid alcohol and terpenoid content, which can elevate ROS and induce apoptosis.^22,23 These findings indicate that antioxidant properties are largely driven by synergistic actions of phenolics, glycerides, and xanthones.^20,24

Enzyme inhibition analysis revealed distinct activity profiles. The leaf extract exhibited the strongest α-amylase inhibition, correlating with diterpenoid alcohols and terpenoids, both widely reported as carbohydrate-hydrolyzing enzyme inhibitors.^25,26 The negative correlation with esters may explain the weaker activity of other extracts.²⁷ In contrast, stem and root extracts showed stronger α-glucosidase inhibition, strongly correlated with glycerides, xanthones, carboxylic acids, and phenolics. These groups are well documented as α-glucosidase inhibitors and glucose absorption modulators^28–31 while thiochromenes, also present, are known for anti-diabetic potential.³²

All extracts exhibited β-glucuronidase inhibition, with stem and root extracts being most potent. This enzyme, critical to xenobiotic metabolism, is linked to liver injury, colon carcinogenesis, and drug toxicity when overactive.³³ Glycerides and carboxylic acids were strongly correlated with inhibition, consistent with their known enzyme-modulating roles.^34–36 Cyclopent-4-ene-1,3-dione in the root extract may further contribute by covalently binding to β-glucuronidase.³⁷ D. indica leaf extract has also been shown to protect against liver injury in vivo.³⁸ Together, these findings support its traditional use in promoting detoxification and reducing hepatic and intestinal damage.

However, negative correlations were observed between β-glucuronidase inhibition and ketones and heterocyclic compounds, indicating potential hepatotoxicity or pro-inflammatory effects. The leaf extract contained 6,10,14-trimethylpentadecan-2-one, a sesquiterpenoid ketone linked to hepatotoxicity,³⁹ and benzimidazol-2-one, structurally related to the hepatotoxic HSP90 inhibitor MEY.⁴⁰ Such findings emphasize the need for part-specific selection to balance efficacy and safety. At the same time, β-glucuronidase inhibition underscores the extracts’ anti-inflammatory potential, consistent with the roles of phenolics and flavonoids in reducing oxidative stress and cytokines like TNF-α and IL-6.^4,41

The extracts also showed selective cytotoxicity against SW480 colorectal cancer cells, with minimal effects on HDF-α cells, reflected by SI > 1, a key indicator of therapeutic potential.⁶ Esters, abundant in all extracts, may contribute by modulating PI3K/Akt and AMPK/mTOR signaling.^42,43 The leaf extract showed the strongest cytotoxicity, linked to diterpenoid alcohols (trans-phytol, phytol) and terpenoids (neophytadiene), which modulate NF-κB, PI3K/Akt, and MAPK pathways.^44,45 These compounds induce selective oxidative stress and mitochondrial dysfunction in cancer cells.^22,23 Root and stem extracts, rich in phenolics and xanthones, exhibited moderate but selective cytotoxicity, possibly due to 2-palmitoylglycerol.⁴⁶ In contrast, the unripe fruit extract showed minimal activity, likely due to its aldehyde and furanone content.⁴⁷

Collectively, the cytotoxicity profiles mirror phytochemical composition: alcohols and terpenoids appear to drive strong activity in leaf extracts, while phenolics and xanthones underlie the safer, moderate effects of root and stem extracts. These results validate the ethnomedicinal relevance of D. indica and highlight its potential as a source of selective anticancer agents. Furthermore, the bioactivities observed in this study are closely linked to the presence of specific chemical groups identified in the GC-MS profile. While alkaloids represent a significant class of natural products, they were not the focus of this study as our preliminary screening yielded negative results. This is consistent with the findings of Shipra et al. (2018)⁴⁸ and Saikia et al. (2023),⁴ who noted that although alkaloids may be detected in specific parts like the bark or fruit pulp under certain extraction conditions, they are often absent or found in negligible amounts in other tissues.

The multi-faceted bioactivities observed in this study provide a scientific rationale for the traditional use of D. indica across Southeast Asia. The potent anti-diabetic activity (α-glucosidase inhibition) found in all extracts validates its long-standing reputation in ancient pharmacopeias for managing metabolic disorders. Furthermore, the anti-inflammatory potential of the plant is supported by its robust antioxidant capacity and the presence of specific bioactive markers identified via GC-MS, such as triterpenoids and phenolic derivatives, which are known to mitigate oxidative stress, a primary driver of chronic inflammation. Crucially, our findings support the safety of traditional consumption; the root and stem extracts exhibited moderate to low cytotoxicity, confirming that their use in Thai cooking and traditional remedies is biologically sound and poses minimal risk to cellular health.

In conclusion, this study establishes direct links between phytochemical composition and part-specific bioactivity in D. indica, highlighting promising antioxidant, anti-diabetic, hepatoprotective, anti-inflammatory, and anticancer effects. While certain phytochemicals may pose toxicity risks, particularly in leaf extracts, part-specific evaluation facilitates a safer approach to therapeutic application. Future research should prioritize bioassay-guided fractionation and explore alternative solvent systems to uncover additional bioactive components. Moreover, the employment of advanced analytical techniques, such as HPLC and LC-MS, alongside mechanistic studies and in vivo validation, is essential to further confirm these findings and investigate potential synergistic effects between combined plant parts.

Conclusion

This study provides a comprehensive analysis of the phytochemical compositions and bioactivities of different parts of D. indica. The results conclusively demonstrate that the biological activities of the extracts are directly dependent on the specific plant part used. This is the first report on the phytochemical profile of the D. indica root extract, which was found to be particularly rich in esters, glycerides, phenolic compounds, and xanthones. The stem and root extracts emerged as the most promising fractions, exhibiting the highest and most consistent antioxidant and anti-diabetic activities. These effects were strongly correlated with the presence of the key bioactive compounds identified. Furthermore, the extracts, especially the leaf extract, showed favorable selective cytotoxicity against colorectal cancer cells, with minimal effects on normal cells. In summary, our findings provide scientific support for the traditional use of D. indica and highlight its potential as a source of bioactive compounds with antioxidant, anti-diabetic, and anticancer–related activities. The results also indicate part-specific differences in both beneficial effects and potential risks, which may inform the rational selection of plant parts for the future development of safe functional products and herbal formulations aimed at promoting healthy aging.

Supplemental Material

Supplemental material for this article is available online.

Supplemental Material

Supplemental material for this article is available online.

Footnotes

Acknowledgements

The authors would like to acknowledge the Center of Excellence on Applied Thai Traditional Medicine Research, Faculty of Medicine, Thammasat University for its support and contribution to this study.

Authors’ Contribution

All authors made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; took part in drafting the article or revising it critically for important intellectual content; agreed to submit to the current journal; gave final approval of the version to be published; and agree to be accountable for all aspects of the work. All the authors are eligible to be an author as per the International Committee of Medical Journal Editors (ICMJE) requirements/guidelines.

Consent to Participate

Not applicable.

Consent for Publication

Not applicable.

Data Availability Statement

All data generated and analyzed are included in this research article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

No human or animal subjects were involved; therefore, ethical approval was not required.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Faculty of Science and Technology, Thammasat University (Contract No. SciGR 34/2567).

Informed Consent

Not applicable.

Publisher’s Note:

All claims expressed in this article are solely those of the authors and do not necessarily represent those of the publisher, the editors and the reviewers. This journal remains neutral with regard to jurisdictional claims in published institutional affiliation.

Use of Artificial Intelligence-assisted Tools:

The authors declared that the manuscript was drafted and revised by the authors. An artificial intelligence (AI)-assisted tool was exclusively for grammar and language refinement. It was not used for data generation, analysis, interpretation or scientific content development. No images were manipulated using AI. The revised text was subsequently reviewed by a native English-speaking co-author, and all authors approved the final version. The authors take full responsibility for the scientific content of this work.

References

Salehi

, Azzini

, Zucca

, . Plant-derived bioactives and oxidative stress-related disorders: A key trend towards healthy aging and longevity promotion. Appl Sci, 2020; 10(3): 947. DOI: 10.3390/app10030947.

Kamboj

, Talukdar

and Banerjee

. Therapeutic benefit of Dillenia indica in diabetes and its associated complications. J Diabetes Res, 2019; 2019(1): 4632491. DOI: 10.1155/2019/4632491.

Gandhi

and Mehta

. Dillenia indica Linn. and Dillenia pentagyna Roxb.: Pharmacognostic, phytochemical and therapeutic aspects. J Appl Pharm Sci, 2013; 3(11): 134–142. DOI: 10.7324/JAPS.2013.31124.

Saikia

, Kesavan

, Inbaraj

, . Bioactive compounds and health-promoting properties of elephant apple (Dillenia indica L.): A comprehensive review. Foods, 2023; 12(16): 2993. DOI: 10.3390/foods12162993.

Gogoi

, Kumar

, Dutta

, . α-amylase and α-glucosidase enzyme inhibition, molecular docking, and pharmacokinetic studies of natural products from Dillenia indica L. barks. J Biol Act Prod Nat, 2023; 13(3): 265–283. DOI: 10.1080/22311866.2023.2234334.

Ruangsuriya

, Sichaem

, Tantraworasin

, . Phytochemical profiles and anticancer effects of Calophyllum inophyllum L. extract relating to reactive oxygen species modulation on patient-derived cells from breast and lung cancers. Scientifica, 2023; 2023(1): 6613670. DOI: 10.1155/2023/6613670.

Brand-Williams

, Cuvelier

and Berset

. Use of a free radical method to evaluate antioxidant activity. LWT Food Sci Technol, 1995; 28(1): 25–30. DOI: 10.1016/S0023-6438(95)80008-5.

Wang

, Wang

, Qiu

, . Multi-wavelength spectrophotometric determination of hydrogen peroxide in water by oxidative coloration of ABTS via Fenton reaction. Environ Sci Pollut Res, 2019; 26: 27063–27072. DOI: 10.1007/s11356-019-05884-7.

Nisar

, Shah

, Akram

, . Phytochemical screening, antioxidant, and inhibition activity of Picrorhiza kurroa against α-amylase and α-glucosidase. Dose Response, 2022; 20(2): 15593258221095960. DOI: 10.1177/15593258221095960.

10.

Acharya

and De

. Bioactivity-guided fractionation to identify β-glucuronidase inhibitors in Nymphaea pubescens flower extract. Cogent Food Agric, 2016; 2(1): 1134379. DOI: 10.1080/23311932.2015.1134379.

11.

Vichai

and Kirtikara

Sulforhodamine B colorimetric assay for cytotoxicity screening. Nat Protoc, 2006; 1(3): 1112–1116. DOI: 10.1038/nprot.2006.179.

12.

Sakuma

. Probit analysis of preference data. Appl Entomol Zool, 1998; 33(3): 339–347.

13.

, Fu

, Wang

, . Comparison of physicochemical, phytochemical, and volatile profiles, and bioactivities of citrus wines brewed from different varieties. Food Chem, 2025; 29: 102726. DOI: 10.1016/j.fochx.2025.102726.

14.

Akullo

, Kiage-Mokua

, Nakimbugwe

, . Phytochemical profile and antioxidant activity of various solvent extracts of two varieties of ginger and garlic. Heliyon, 2023; 9(8): e18806. DOI: 10.1016/j.heliyon.2023.e18806.

15.

Sharma

, Bhardwaj

, Gaba

, . Natural antioxidants: Assays and extraction methods/solvents used for their isolation. In: Nayik

and Gull

(eds) Antioxidants in fruits: Properties and health benefits. Singapore: Springer, 2020, pp. 1–33. DOI: 10.1007/978-981-15-7285-2_1.

16.

Bate-Smith

and Harborne

. Differences in flavonoid content between fresh and herbarium leaf tissue in Dillenia. Phytochemistry, 1971; 10(5): 1055–1058. DOI: 10.1016/S0031-9422(00)89938-0.

17.

, Tang

and Zhang

. Theoretical elucidation of DPPH radical-scavenging activity difference of antioxidant xanthones. QSAR Comb Sci, 2005; 24(7): 826–830. DOI: 10.1002/qsar.200430917.

18.

Tiranakwit

, Puangpun

, Tamprasit

, . Phytochemical screening on phenolic, flavonoid contents, and antioxidant activities of six indigenous plants used in traditional Thai medicine. Int J Mol Sci, 2023; 24(17): 13425. DOI: 10.3390/ijms241713425.

19.

Sari

, Chung

, Agatha

, . Evaluation of antioxidant and antimicrobial activity of phenolic lipids produced by the transesterification of 4-hydroxyphenylacetic acid and triglycerides. Appl Biol Chem, 2019; 62(1): 5. DOI: 10.1186/s13765-019-0419-3.

20.

Francik

, Szkaradek

, Zelaszczyk

, . Antioxidant activity of xanthone derivatives. Acta Pol Pharm, 2016; 73(6): 1505–1509.

21.

Özyürek

, Bektaşoğlu

, Güçlü

, . A novel hydrogen peroxide scavenging assay of phenolics and flavonoids using cupric reducing antioxidant capacity (CUPRAC) methodology. J Food Compos Anal, 2010; 23(7): 689–698. DOI: 10.1016/j.jfca.2010.02.013.

22.

Rostamabadi

, Bahraseman

MRS

and Esmaeilzadeh-Salestani

. Froriepia subpinnata leaf extract-induced apoptosis in the MCF-7 breast cancer cell line by increasing intracellular oxidative stress. Iran J Pharm Res, 2023; 22(1): e136643. DOI: 10.5812/ijpr-136643.

23.

Thakor

, Subramanian

, Thakkar

, . Phytol induces ROS mediated apoptosis by induction of caspase 9 and 3 through activation of TRAIL, FAS and TNF receptors and inhibits tumor progression factor glucose 6 phosphate dehydrogenase in lung carcinoma cell line (A549). Biomed Pharmacother, 2017; 92: 491–500. DOI: 10.1016/j.biopha.2017.05.066.

24.

Yuan

and Zeng

An overview of multifaceted applications and the future prospects of glyceroglycolipids in plants. J Agric Food Chem, 2024; 72(41): 22420–22432. DOI: 10.1021/acs.jafc.4c05923.

25.

Ajayi

, Balogun

, Olawuni

, . Alpha amylase inhibition and antioxidant activities of bicyclic diterpenoid lactones from Andrographis paniculata. Trop J Nat Prod Res, 2021; 5(6): 1110–1117. DOI: 10.26538/tjnpr/v5i6.22.

26.

Etsassala

, Badmus

, Waryo

, . Alpha-glucosidase and alpha-amylase inhibitory activities of novel abietane diterpenes from Salvia africana-lutea. Antioxidants, 2019; 8(10): 421. DOI: 10.3390/antiox8100421.

27.

Teng

and Chen

α-Glucosidase and α-amylase inhibitors from seed oil: A review of liposoluble substance to treat diabetes. Crit Rev Food Sci Nutr, 2017; 57(16): 3438–3448. DOI: 10.1080/10408398.2015.1129309.

28.

Kumar

, Kumar

and Prakash

Enzymes inhibition and antidiabetic effect of isolated constituents from Dillenia indica. Biomed Res Int, 2013; 2013(1): 382063. DOI: 10.1155/2013/382063.

29.

Ćorković

, Gašo-Sokač

, Pichler

, . Dietary polyphenols as natural inhibitors of α-amylase and α-glucosidase. Life, 2022; 12(11): 1692. DOI: 10.3390/life12111692.

30.

Nguyen

, Huyen

, Thi

, . Impact of different drying temperatures on in vitro antioxidant and antidiabetic activities and phenolic compounds of wild guava leaves collected in the central highland of Vietnam. Nat Prod Commun, 2022; 17(4): 1934578X221095349. DOI: 10.1177/1934578X221095349.

31.

Shataer

, Chen

, Wu

, . Edible plant-derived xanthones as functional food components for metabolic syndrome mitigation: Bioactivities and mechanisms. Foods, 2005; 14(13): 2344. DOI: 10.3390/foods14132344.

32.

Fan

, Zhong

, Huang

, . Synthesis, biological evaluation and molecular docking studies of chromone derivatives as potential α-glucosidase inhibitors. J Mol Struct, 2023; 1274(1): 134575. DOI: 10.1016/j.molstruc.2022.134575.

33.

Allam

, Rudayni

, Ahmed

, . β-glucuronidase inhibition in drug development: Emerging strategies for mitigating drug-induced toxicity and enhancing therapeutic outcomes. Drug Dev Res, 2025; 86(4): e70118. DOI: 10.1002/ddr.70118.

34.

Awolade

, Cele

, Kerru

, . Therapeutic significance of β-glucuronidase activity and its inhibitors: A review. Eur J Med Chem, 2020; 187: 111921. DOI: 10.1016/j.ejmech.2019.111921.

35.

Alqhtani

, Othman

, Alkhayl

, . Inhibitory mechanisms of β-glucuronidase by Hibiscus syriacus phenolics: Integrating computational and experimental approaches. ChemistrySelect, 2025; 10(1): e202402984. DOI: 10.1002/slct.202402984.

36.

Bai

, Chen

, Cao

, . Beta-glucuronidase inhibition by constituents of mulberry bark. Planta Med, 2021; 87(08): 631–641. DOI: 10.1055/a-1402-6431.

37.

, Cheng

, Wu

, . Diverse diterpenoids with α-glucosidase and β-glucuronidase inhibitory activities from Euphorbia milii. Phytochemistry, 2022; 196: 113106. DOI: 10.1016/j.phytochem.2022.113106.

38.

Padhya

, Choudhary

, Padhy

, . Effect of Dillenia indica leaves against carbon tetrachloride induced hepatotoxicity. J Pharm Chem, 2008; 2(4): 24–27.

39.

Zárybnický

, Boušová

, Ambrož

, . Hepatotoxicity of monoterpenes and sesquiterpenes. Arch Toxicol, 2018; 92(1): 1–3. DOI: 10.1007/s00204-017-2062-2.

40.

Mazri

, Bourougaa

, Zekri

, . Discovery of N-aryl-benzimidazolone analogs as novel potential HSP90 inhibitors: A computational approach. Appl Sci, 2024; 14(23): 10817. DOI: 10.3390/app142310817.

41.

Rodríguez-Yoldi

. Anti-inflammatory and antioxidant properties of plant extracts. Antioxidants, 2021; 10(6): 921. DOI: 10.3390/antiox10060921.

42.

Zhao

, Li

, Wei

, . The ester derivatives obtained by C-ring modification of podophyllotoxin induce apoptosis and inhibit proliferation in PC-3M cells via down-regulation of PI3K/Akt signaling pathway. Bioorg Med Chem Lett, 2021; 46: 128174. DOI: 10.1016/j.bmcl.2021.128174.

43.

Zheng

, Li

, Xu

, . Sinomenine ester derivative inhibits glioblastoma by inducing mitochondria-dependent apoptosis and autophagy via PI3K/AKT/mTOR and AMPK/mTOR pathway. Acta Pharm Sin B, 2021; 11(11): 3465–3480. DOI: 10.1016/j.apsb.2021.05.027.

44.

Siraj

, Islam

, Al Fahad

, . Cancer chemopreventive role of dietary terpenoids by modulating Keap1-Nrf2-ARE signaling system—a comprehensive update. Appl Sci, 2021; 11(22): 10806. DOI: 10.3390/app112210806.

45.

El-Baba

, Baassiri

, Kiriako

, . Terpenoids’ anti-cancer effects: Focus on autophagy. Apoptosis, 2021; 26(9): 491–511. DOI: 10.1007/s10495-021-01684-y.

46.

Molina

, Morán-Valero

, Martín

, . Antiproliferative effect of alkylglycerols as vehicles of butyric acid on colon cancer cells. Chem Phys Lipids, 2013; 175: 50–56. DOI: 10.1016/j.chemphyslip.2013.07.011.

47.

Cimmino

, Scafato

, Mathieu

, . Natural and synthetic furanones with anticancer activity. Nat Prod Commun, 2016; 11(10): 1934578X1601101013. DOI: 10.1177/1934578X1601101013.

48.

Shipra

, Pandey

and Kalawati

Preliminary phytochemical analysis and pharmacognostical studies of different parts of Dillenia indica Linn. Int J Sci Res Publ, 2018; 8(6): 175–185. DOI: 10.29322/IJSRP.8.6.2018.p7825.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

3.58 MB

GC-MS Profiling,Bioactivities,and Correlations of Dillenia indica L. Leaf,Stem,Unripe Fruit and Root Extracts

Abstract

Keywords

Introduction

Materials and Methods

Chemicals and Instrumentations

Preparation of Plant Samples and Extraction of Bioactive Compounds

Phytochemical Analysis by GC-MS

DPPH Radical Scavenging Assay

Anti-oxidative Stress Evaluation Using an H2O2-mediated Hydroxyl Radical Scavenging Assay

Inhibition of α-amylase and α-glucosidase Activity Assays

Inhibition of α-amylase

Inhibition of α-glucosidase

Indirect Hepatoprotective Evaluation by β-glucuronidase Inhibition Assay

Relative Efficiency Calculation and Visualization

Cell Culture and Cytotoxicity Assay

Comparative Bioactivity Profiling via Normalized Radar Plot Analysis

Correlation Analysis Between Phytochemical Groups and Biological Activities

Statistical Analysis

Results

Extraction Yield and Phytochemical Composition of D. indica Extracts

Relative Proportion (% Area) of Phytochemical Groups in Dillenia indica Extracts. Pie Charts Show the Percentage Area of Each Chemical Group Identified in the (A) Leaf, (B) Stem, (C) Unripe Fruit, and (D) Root Extracts.

Bioactivities of D. indica Extracts

DPPH Radical Scavenging Activity

Anti-oxidative Stress Evaluation via H2O2 Scavenging Activity

Inhibition of α-amylase and α-glucosidase Activity

Indirect Hepatoprotective Evaluation by β-glucuronidase Inhibition

Cytotoxicity of D. indica Extracts

Radar Plot Analysis of D. indica Extracts and Bioactivities

Correlation Analysis and Heatmap of Phytochemical Constituents and Bioactivities

Discussion

Conclusion

Supplemental Material

Supplemental Material

Footnotes

Acknowledgements

Authors’ Contribution

Consent to Participate

Consent for Publication

Data Availability Statement

Declaration of Conflicting Interests

Ethical Approval

Funding

Informed Consent

Publisher’s Note:

Use of Artificial Intelligence-assisted Tools:

References

Supplementary Material

Anti-oxidative Stress Evaluation Using an H₂O₂-mediated Hydroxyl Radical Scavenging Assay

Anti-oxidative Stress Evaluation via H₂O₂ Scavenging Activity