Tilia species (linden) exert anti-cancer effects on MIA PaCa-2 cells through the modulation of oxidative stress and inflammation

Abstract

This study investigated the anti-cancer effects of the chemically characterized Tilia species (linden) on MIA PaCa-2 cells by analyzing various cancer-triggering mechanisms, including oxidative stress and inflammation status. Extracts from the flowers, bracts, and inflorescences of T. cordata, T. platyphyllos, T. rubra, and T. tomentosa were evaluated for antioxidant activity; subsequently, their ability to mitigate inflammation was assessed through in vitro nitrite assays in LPS-induced RAW264.7 cells. The anticancer potentials of the extracts against MIA PaCa-2 pancreatic cancer cells were investigated in 2D (cytotoxic effect) and 3D (effect on spheroid growth) models in vitro. All investigated Tilia species displayed remarkable antioxidant activity and significantly inhibited LPS-induced nitrite, IL-6, and PGE₂ production. Extract from T. rubra bracts showed the highest cytotoxic activity against MIA PaCa-2 cells with an IC₅₀ value of 0.16 mg/mL, as well as the most significant delay on spheroid growth, which was further confirmed through the arrest in cell cycle. In the Annexin V cell death assays of T. rubra, cells treated with the flower extract exhibited the highest rate of necrotic population with 66.53%. Overall, our results highlight a potential use for Tilia extracts, particularly T. rubra, in pancreatic cancer treatment by modulating cell death.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-86457-2.

Introductıon

Cancer is a multi-stage disease that involves uncontrolled cell proliferation and mutations. Based on the molecular, cellular, and pathologic aspects of the transforming of a normal cell into a neoplastic lesion, the multistage processes leading to cancer development involve initiation, promotion, and progression stages¹. Oxidative stress, which is defined as an imbalance between the production of free radicals and reactive oxygen species (ROS), is one of the major causes of these three stages. ROS may damage DNA at the beginning stage by introducing structural changes and gene mutations. During the promotion phase, ROS may play a role in aberrant gene expression, disruption of intercellular communication, and alteration of second-messenger systems. These outcomes may lead to either an increase in cell division or a reduction in the apoptosis of the initial cell population^2,3. Additionally, ROS represent a link between chronic inflammation and cancer. Inflammatory cells that release ROS may also cause DNA damage. Indeed, numerous stages of the carcinogenesis process, such as cellular transformation, promotion, survival, proliferation, invasion, angiogenesis, and metastasis, have been connected to chronic inflammation and oxidative stress^1,2. The use of medicinal plants has become a limelight for their potential to mitigate oxidative stress-related diseases, alleviate inflammatory conditions, and even combat cancer, stating their promise as a natural remedy with multifaceted health benefits.

Tilia species known as linden or lime trees, are found in Türkiye as Tilia cordata Mill., Tilia tomentosa Moench, Tilia rubra DC., and Tilia platyphyllos Scop. The light to dark green-colored leaves of Tilia species are alternate and have cordate and asymmetric lamina with dentate- or serrate-shaped margins. The yellowish-colored flowers in inflorescence bloom between June and July. The peduncle (the main stalk of the inflorescence) is adherent to long and oblong-lanceolate shaped bracts from the bottom to the middle, and they grow together. The number of flowers in the inflorescence varies from two to ten, depending on the species⁴. Traditionally, brewed flowers or inflorescences comprising flowers and bracts of Tilia spp. have been used to treat common cold and cough⁵. Additionally, scientific studies have shown their sedative, antioxidant, neuroprotective, antiproliferative, antimicrobial, antitussive, and immunomodulating activities. These effects are attributed to its flavonoids (derivatives of kaempferol and quercetin), phenolic acids, volatile oil, and polysaccharide components^6–9. Astragalin and tiliroside, which were initially isolated from the leaves of T. tomentosa, are responsible for various pharmacological activities of Tilia species, including antioxidant, anti-inflammatory, hepatoprotective, neuroprotective, antithrombotic, antidiabetic, antiobesity, antiviral, antiosteoporotic, and antiproliferative activities^6,7. Additionally, quercetin derivatives such as hyperoside, quercitrin, and isoquercitrin have a broad spectrum of biological activities, including anticancer, anti-inflammatory, antibacterial, antiviral, antidepressant, and organ protective effects^10,11. Apart from flavonoids, alkaloids such as tiliine A, tiliamine B, and tilacetine A found in the mixture of T. platyphyllos and T. cordata may act as acetylcholinesterase inhibitors¹².

Previously, the anti-cancer properties of T. platyphyllos were evaluated against different cell lines such as HeLa (cervix cancer), HepG2 (hepatocellular carcinoma), NCI-H460 (lung carcinoma), and MCF-7 (breast adenocarcinoma)¹³. Other studies examined the antiproliferative effect of T. cordata extracts on LS180 (human colon adenocarcinoma) and BW 5147 (lymphoma tumor) cells dose dependently^14,15. T. tomentosa extract has been shown to be effective in lung adenocarcinoma A549 cell line¹⁶. In the literature search, no studies were found on the anticancer activity of T. rubra. Besides, among these studies, there is no comparative evaluation of different Tilia species, including T. cordata, T. tomentosa, T. rubra, and T. platyphyllos, regarding their antioxidant activity, protective role against inflammation, and anticancer potentials. Furthermore, the effect of Tilia species on pancreatic cancer remains inconclusive. Pancreatic cancer is the seventh leading cause of cancer-related deaths in the world and is expected to be third by 2025 with poor diagnosis and occurs in both genders¹⁷.

The primary aim of this study was to assess the anti-cancer properties of the hydroethanolic extracts of different Tilia species using the MIA PaCa-2 pancreatic cancer cell line by investigating the status of various cancer-triggering mechanisms, including oxidative stress and inflammation. To study the anti-cancer potential, cytotoxicity on the cell monolayer (2D model), effect on 3D spheroid growth, and apoptosis assays were also conducted. To further characterize and quantify the main phenolic components, high-performance thin-layer chromatography (HPTLC) and high-performance liquid chromatography (HPLC) were used to determine the components underlying the aforementioned biological activities.

Materıals and methods

Chemicals and solvents

The brands of chemicals and solvents used during the analyses, along with their abbreviations, are listed in the ‘Materials and Methods’ section of Supplementary Information.

Plant materials

Tilia species were collected during the flowering stages, from May to June: T. cordata was collected from Şile-İstanbul, Türkiye (41.1749° N, 29.6096° E); T. platyphyllos from Bayramiç-Çanakkale, Türkiye (39.8095° N, 26.6120° E); T. tomentosa from Serhat Köyü-Bayramiç-Çanakkale, Türkiye (39.7397° N, 26.7064° E); and T. rubra from Turgutlu-Manisa, Türkiye (38.5002° N, 27.7084° E). The specimens were authenticated by Assoc. Prof. Gizem Emre using the Flora of Turkey and the East Aegean Islands (Davis, 1965–1985). The voucher specimens for T. cordata Mill. (MARE 22872), T. platyphyllos Scop. (MARE 22871), T. tomentosa Moench (MARE 22727), and T. rubra DC. subsp. caucasica (Rupr.) V. Engler (MARE 2272a) are stored at the Herbarium of Marmara University Faculty of Pharmacy (MARE), İstanbul, Türkiye. All plant parts were dried at room temperature and stored in appropriate conditions until analysis.

Preparation of sample and standard solutions

Five grams of powdered flowers, bracts, and inflorescence containing bracts and flowers (1:1 w/w) were extracted with an 80% ethanol_(aq) solution, and each sample was ultrasonicated for 30 min. Then the samples were filtered through a filter paper and a syringe filter (pore size 0.45 μm). Ethanol was completely evaporated in a rotary evaporator (BÜCHI Labortechnik, Flawil, Switzerland) under reduced pressure at 40°C. Finally, the remaining part was lyophilized. The sample stock solutions were prepared using 80% ethanol at a concentration of 20 mg/mL. The solutions were further diluted using the same solvent for the antioxidant assays and HPTLC and HPLC analyses.

The stock solutions of protocatechuic acid, hyperoside, isoquercitrin, astragalin, quercitrin, and tiliroside were prepared in methanol at a concentration of 300 µg/mL. Equal volumes of each stock solution were mixed to prepare a standard mixture solution (50 µg/mL) used for the HPTLC analysis.

Prepared sample test solutions and standard solutions were stored at 4 °C during the analyses.

Chemical characterization of Tilia species

• HPTLC analysis

Samples (20 mg/mL) and standard mixture solution (50 µg/mL) were applied on an HPTLC silica gel 60 F₂₅₄ glass plate (20 × 10 cm, Merck, Darmstadt, Germany) in 4 µL and 2 µL volumes, respectively, using a semi-automatic sample spotter Linomat 5 (Camag, Muttenz, Switzerland). The plate was developed using a developing solvent system containing ethyl acetate-formic acid-water (30:3:3, v/v/v) up to a migration distance of 7 cm in 20 min in a saturated twin trough chamber (Camag) to separate the investigated compounds in the sample test solution. After development, the plate was dried by exposure to cold air. Then, the plate was heated at 105 °C for 3 s on a TLC plate heater (Camag). Next, the plate was dipped into a solution containing 2-aminoethyl-diphenylborinate in ethyl acetate (Natural Product, NP) and polyethylene glycol (PEG) 400 in dichloromethane, respectively. Finally, the plate was documented at 366 nm using the Camag TLC visualizer. All steps were operated by the winCATS program (version 1.4.8, Camag).

• HPLC analysis

The chromatographic separations were performed using the Agilent 1260 Infinity HPLC system with a diode array detector (DAD), which has a vacuum degassing system, an automatic sampler, a thermostatted column compartment, and a quaternary pump. The Agilent Chem Station software was used for analysis. An Agilent Zorbax Plus C 18 column (4.6 mm × 250 mm) with a 5 μm particle size was employed to provide optimal conditions to separate the investigated compounds. The best separation was obtained with a 1 mL/min flowing rate at 25 °C. Ultra-purified water with 0.1% o-phosphoric acid and acetonitrile were used as mobile phases A and B, respectively. The gradient profile was as follows: 85–15% A-B (0–3 min), 82–18% A-B (3–11 min), 80–20% B (11–16 min), 70–30% A-B (16–22 min), 50–50% A-B (22–25 min), 20–80% A-B (25–26 min), and 85–15% A-B (26–30 min). A volume of 10 µL was injected for standard solutions and samples, and each injection was performed in triplicate. Different wavelengths were used to monitor the analytes. These values were selected according to their UV_max values. Protocatechuic acid, hyperoside, isoquercitrin, and quercitrin were monitored at 260 nm, whereas astragalin and tiliroside were monitored at 265 and 320 nm, respectively.

The linearity data of the calibration curve, along with the values for the LOQ and LOD, are provided in Supplementary Table S1. Repeatability and precision data for standards are presented in Supplementary Table S2, and recovery results are given in Supplementary Table S3.

• Total phenolic content estimation by the Folin–Ciocalteu method

The total phenolic content (TPC) of the samples was evaluated using the Folin–Ciocalteu reagent method described by Singleton et al. (1999), with a slight modification¹⁸. Firstly, 50 µL of different concentrations of gallic acid standard solutions (3.125–100 µg/mL), sample test solutions that belong to Tilia samples, and blank were placed into each well separately in a 96-well plate and mixed with 50 µL of 10% Folin–Ciocalteu reagent. Then, 100 µL of 7.5% (w/v) Na₂CO₃ was added to these mixtures. The prepared mixtures were kept in the dark for 30 min at room temperature. The absorbance was measured using a UV/Vis spectrophotometer (Multiscan Go, Thermo Scientific) at 760 nm. All analyses were performed in triplicate, and the TPCs of the samples were calculated according to the obtained calibration curve by gallic acid standards and expressed as mg of gallic acid equivalents (GAE) per g of hydroethanolic Tilia extract (mg GAE/g hydroethanolic Tilia extract).

• Evaluation of total flavonoid content

The total flavonoid content (TFC) in Tilia samples was determined by a slightly modified method stated by Farasat et al. (1999)¹⁹. 30 µL of each sample test solution, quercetin standard solutions, and blank (MeOH) were first placed in separate wells of a 96-well plate, and 30 µL of 10% aluminum chloride and 30 µL of 1 M sodium acetate were added, respectively. Finally, the volume was adjusted to 240 µL with distilled water. After 30 min of incubation at room temperature, the absorbance was measured at 415 nm using a UV/Vis spectrophotometer. All analyses were performed in triplicate, and the total flavonoid content of the Tilia samples was expressed as mg quercetin equivalents (QE) per g of hydroethanolic Tilia extract (mg QE/g hydroethanolic Tilia extract).

Biological activities

• Free radical-scavenging activity with 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay

DPPH radical scavenging activity was measured using a slightly modified version of the method stated by Blois et al. (1957)²⁰. To perform the analysis, 20 µL of Tilia sample test solutions, trolox standard solutions in a range of 3.125–200 µg/mL, and blank (MeOH) were added in a 96-well plate separately, and each was mixed with 280 µL of DPPH solution (0.1 mM). After 30 min of incubation in a dark environment at room temperature, absorbance was measured at 520 nm using a UV/Vis spectrophotometer. The results were presented in mg Trolox equivalent (TE)/g of hydroethanolic Tilia extract (mg TE/g hydroethanolic Tilia extract).

• Metal-reducing activity by cupric-reducing antioxidant capacity (CUPRAC) assay

To perform the CUPRAC assay, 85 µL of 10 mM copper (II) sulfate pentahydrate, 85 µL of 7.5 mM neocuproine, 85 µL of ammonium acetate buffer solution (pH = 7), and 51 mL of distilled water were added to the 96-well microplate, respectively. Then, either 43 µL of each sample test solution or standard solution (3.125–200 µg/mL) or water as a blank was placed in each well. After incubation of the plate at room temperature for 30 min at 50 °C, the absorbance was read at 450 nm²¹. The results were given in mg Trolox equivalent (TE)/g of hydroethanolic Tilia extract (mg TE/g hydroethanolic Tilia extract).

• Metal-reducing activity by ferric-reducing antioxidant power (FRAP) assay

For the analysis, 20 µL of sample test solution, standard solutions in the range of 3.125–100 µg/mL, and blank (distilled water) were added to a 96-well plate separately, and each well was mixed with 280 µL of FRAP reagent [2,4,6-tri(2-pyridyl)-s-triazine (TPTZ; 0.01 M) solution, acetate buffer (pH = 3.6) solution, and 0.02 M iron (III) chloride hexahydrate solution were mixed at a 10:1:1 ratio]. After 6 min, the absorbances were determined at 595 nm²². The results were expressed in mg Trolox equivalent (TE)/g of hydroethanolic Tilia extract (mg TE/g hydroethanolic Tilia extract).

Cell culture studies

Evaluation of the protective effect against inflammatory response

• Cell viability

RAW264.7 mouse macrophage cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco, NY, USA) enriched with 1% penicillin (10.000 units/mL) and streptomycin (10.000 µg/mL) and 10% fetal bovine serum (FBS) (Gibco™, USA) at 37 °C in a humidified atmosphere of 5% CO₂.

The MTT colorimetric assay was used for the cytotoxicity analysis according to the method described by Sipahi et al. (2022)²³. The plated cells were treated with four different concentrations of Tilia species extracts (0.125–1 mg/mL). Indomethacin (IND) and L-NAME (Nv-Nitro-L-arginine methyl ester hydrochloride) (100 µM) (Sigma Aldrich, St. Louis, MI, USA) was used as a positive control for evaluation of the protective effect against inflammatory response and analgesic activity. The plates with treated cells were then incubated for 24 h. After removal of the medium, MTT reagent (0.5 mg/mL) was applied, and the cells were incubated for another 2 h. The medium was then removed, and formazan crystals were dissolved in 100 µL of isopropanol. The absorbance was measured with a microplate reader (Thermo Multiscan Spectrum, Vantaa, Finland) at 570 nm wavelength. The test procedure was performed in triplicate. The cell viability in percentages was determined with the following equation:

Viability of cell (%) = 100 × OD_570test/OD_570control.

OD_570test: Mean value of measured optical density of the tested substance.

OD_570control: Mean value of measured optical density in the negative control group.

Further experiments were conducted at safe doses of extracts where the determined percent cell viability was 70% or higher compared with the control.

• Determination of nitrite levels

RAW264.7 cells seeded in a 96-well plate at a density of 5 × 10⁴ cells/well were incubated at 37 °C in 5% CO₂ for 24 h. Cells were then pretreated with different concentrations of Tilia extracts (0.125–1 mg/mL) for 2 h. After that, cells were stimulated with LPS (1 µg/mL) for 22 h. The supernatant was collected, and 50 µL of the supernatant was mixed with 50 µL of the Griess reagent. Then, the mixture was incubated for 10 min in the dark. The absorbance was determined at 540 nm spectrophotometrically with a microplate reader (Multiscan Ascent, Vantaa, Finland)²⁴.

• IL-6 releasing ınhibition assay

IL-6 levels were tested at the highest non-cytotoxic extract dose. The manufacturer’s recommendations were followed while utilizing a commercially available quantitative enzyme-linked immunosorbent assay (ELISA) kit (Invitrogen, USA) to measure the quantity of IL-6 in cell culture supernatant. Given the high IL-6 value of the samples, dilution was performed five times²⁵. The results were expressed as pg/mL.

• Evaluation of analgesic activity with prostaglandin E2 (PGE ₂ ) levels

PGE₂ levels were tested at the highest non-cytotoxic extract dose. Using an ELISA kit from Abcam (USA), the release of PGE₂ from cell supernatants was measured in duplicates in accordance with the manufacturer’s instructions. The results were expressed as pg/mL²⁶.

Anticancer activity

• Two-dimensional (2D) cell cytotoxicity assay

Human pancreatic cancer cell line MIA PaCa-2 (CRM-CRL-1420) and human dermal fibroblast cell line HDF (PCS-201-012) were obtained from the American Type Culture Collection (ATCC, USA). Cell lines were cultured in Dulbecco’s Modified Eagle’s Medium supplemented with 10% FBS, 1% penicillin (10.000 units/mL), and streptomycin (10.000 µg/mL) at 37 °C under a humidified atmosphere of 5% CO₂²⁷.

To determine 2D anticancer activity, MIA PaCa-2 and HDF cells (10⁴ cells/well) were seeded in a 96-well cell culture plate and allowed to attach for 24 h. To determine non-cytotoxic concentrations, cells were treated with Tilia extracts at four different concentrations (0.125, 0.25, 0.5, and 1 mg/mL) and incubated for 24 h. The culture medium was then removed, the MTT assay reagent (0.5 mg/mL) was added to all wells, and the cells were incubated for 2 h at 37 °C. Next, the culture medium was removed, and, in order to dissolve formazan crystals, 100 µL of isopropanol was added into the wells. The absorbance was measured at 570 nm wavelength with an ELISA microplate reader (Thermo Multiskan Spectrum, Finland). Cell viability was calculated as percentages using the following equation:

Viability% = (Absorbance_{treatment group})/(Absorbance_ontrol) × 100%.

The IC₅₀ values of Tilia extracts were calculated via the ECIC function code inserted into Microsoft Excel.

• Three-dimensional (3D) spheroid formation/growth assay

For 3D spheroid formation, MIA PaCa-2 and HDF cells were used according to the magnetic 3D Bioprinting method²⁸. Human fibroblasts were mixed equally with pancreatic cancer cells to better stimulate the tumor environment. After cells reached 70–80% confluency, cells were treated with biocompatible NanoShuttle^TM-PL (Bioprinting Kit, Greiner Bio-One, Germany) and incubated overnight. Then, the cells were resuspended and seeded into ultra-low attachment 96-well culture plates with a 100 µL (2.000 pancreatic cancer cells and 2.000 human fibroblasts per well) volume²⁹. Until spheroids were formed, the culture plate was placed on a magnetic drive and incubated in a humidified atmosphere containing 5% CO₂ at 37 °C for 48 h. Two days after the incubation period, the spheroids were photographed using light microscopy (Carl Zeiss Primo Vert, Germany). After that, spheroids were treated with flower, bract, and inflorescence of T. cordata, T. tomentosa, T. rubra, and T. platyphyllos extracts with the same concentration interval tested in RAW264.7 cells (0.125–1 mg/mL). For 10 days, treatment media were replaced with fresh media, and images of spheroids were captured every 3 days. The effect of Tilia extracts on spheroid growth was assessed by measuring the size change of spheroids using ImageJ software (Image J.2.0 software, USA). Doxorubicin (DOX) was used as a comparator (10 µM). All the measurements were performed in triplicate.

Cell cycle arrest

MIA PaCa-2 human pancreatic carcinoma cells were seeded in 6-well plates at a density of 3 × 10⁵ cells/well. After 24 h, cells were subjected to serum-depleted media for 4 h prior to treatment with T. rubra inflorescence, bract, and flower extracts for 16 h. Following the incubation period, cells were collected, fixed with a 4% ethanol solution, and the cell cycle analysis protocol was applied by incubating samples in a mixture of 0.1% (v/v) triton X-100, 0.5 mg/mL RNaseA, and 5 µg/mL PI solution. 20.000 events were analyzed by Guava easyCyte Flow Cytometer (Merck Millipore, Germany).

Cell death analysis by annexin V assay

The relative percentage of apoptotic MIA PaCa-2 cells in response to T. rubra inflorescence, bract, and flower extracts was analyzed by the Annexin V-FITC apoptosis detection assay according to the manufacturer’s protocol (Annexin V-FITC Reagent, Elabscience). 3 × 10⁵ cells/well seeded on 6-well plates were treated with 10 µM DOX as a positive control and T. rubra inflorescence, bract, and flower extracts for 16 h. After the completion of the incubation period, the media of the cells for each condition were collected, and the wells were washed with 1X PBS. The attached cells were then collected with 0.25% trypsin-EDTA, and the whole suspension of cells was centrifuged at 350xg for 5 min. 1 × 10⁶ cells from each sample were treated with Annexin Binding Buffer, Annexin V, and PI according to the manufacturer’s protocol. 20.000 events were analyzed immediately by Guava easyCyte Flow Cytometer (Merck Millipore, Germany).

Chemometrics analysis

17 different analyses were performed on 4 different plant species. An excel sheet were generated for chemometric analysis. An unsupervised (Principal Component Analysis) and a supervised (Partial Least Square- Discriminant Analysis) models were applied to data set. In order to perform this models, 4 groups were generated for Y-Block and 17 parameters were transferred into X-Block. Data were transferred into MATLAB 2023b software. Data cleaning, outlier detection and data normalization were applied. After all, data matrices were transferred into PLS Toolbox 9.5.0 for chemometric analysis. 17 × 36 data matrices were evaluated for classification and feature selection. PCA analysis were shown the discrimination between the species and PLS-DA analysis were performed to indicate the statistically significant parameters which are differentiated between the groups. A correlation map was also plotted in order to elucidate the relation between the parameters. A heatmap was also generated to monitor the whole differences and similarities of each sample regarding to different analysis.

Statistical analysis

All repeated experiments were conducted in triplicate. Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, Inc., San Diego, CA; Version 8.4.3) and Excel. Differences between groups were determined using one-way ANOVA followed by Tukey’s post-hoc tests. Correlations of the parameters were evaluated using Spearman’s correlation test. The differences were considered significant at p < 0.05.

Results

Phytochemical characterization and quantification of the marker components of Tilia species by HPTLC and HPLC

The chemical profiles of hydroethanolic extracts of different parts of four Tilia species were analyzed by HPTLC (Fig. 1). The flowers and bracts of Tilia species showed characteristic chemical fingerprints. Detailed information regarding the retardation factors (R_F) of standards can be found in Supplementary Table S4.

Fig. 1.

HPTLC chromatogram of flower, bract, and inflorescence extracts of T. cordata (TC), T. tomentosa (TT), T. rubra (TR), and T. platyphyllos (TP). F: Flowers; B: Bracts; I: Inflorescence.

According to the HPTLC profiles, most of the investigated standards as protocatechuic acid, tiliroside, quercitrin, astragalin, isoquercitrin, and hyperoside were identified in the samples. Apart from the investigated compounds as distinct yellow- and blue-colored zones, with R_F values of approximately 0.5 and 0.75, respectively, were observed in T. rubra bracts. These zones represent unknown compounds that require further investigation for identification.

In addition to the qualitative analysis of Tilia species, quantitative analysis was also performed on the investigated compounds. The calculated values are given in Table 1. The highest amounts of protocatechuic acid and hyperoside were found in the bracts of T. cordata and T. rubra at 30 mg/g and 3 mg/g, respectively. The highest amounts of isoquercitrin (15 mg/g) and astragalin (9 mg/g) were quantified in the flowers of T. cordata and T. tomentosa, respectively. All samples except the bracts of T. rubra contained tiliroside; however, the bracts of T. rubra were found to have the highest quercitrin content at 23 mg/g.

Table 1.

Quantitative results of compounds in Tilia species investigated by HPLC.

Samples		Investigated compounds (mg/g extract)
		Protocatechuic acid	Hyperoside	Isoquercitrin	Astragalin	Quercitrin	Tiliroside
T. cordata	Flowers	13.35 ± 0.41	n.d.	15.06 ± 0.34	7.76 ± 0.12	1.71 ± 0.04	4.78 ± 0.05
	Bracts	30.26 ± 0.82	1.74 ± 0.03	n.d.	2.12 ± 0.05	17.87 ± 0.21	0.24 ± 0.02
	Inflorescence	28.85 ± 0.27	1.47 ± 0.01	14.82 ± 0.09	6.94 ± 0.08	11.68 ± 0.10	3.1 ± 0.06
T. platyphyllos	Flowers	3.12 ± 0.10	n.d.	7.24 ± 0.15	6.40 ± 0.05	3.43 ± 0.03	5.57 ± 0.06
	Bracts	2.03 ± 0.06	n.d.	2.12 ± 0.03	1.37 ± 0.004	1.07 ± 0.01	3.12 ± 0.05
	Inflorescence	2.68 ± 0.09	n.d.	5.41 ± 0.10	3.74 ± 0.03	2.32 ± 0.02	4.36 ± 0.06
T. rubra	Flowers	6.19 ± 0.01	n.d.	13.43 ± 0.03	4.85 ± 0.02	2.51 ± 0.02	0.90 ± 0.03
	Bracts	6.68 ± 0.17*	3.01 ± 0.04	4.44 ± 0.11	1.73 ± 0.04	23.13 ± 0.33	n.d.
	Inflorescence	7.46 ± 0.51*	2.28 ± 0.02	10.16 ± 0.17	3.64 ± 0.05	13.67 ± 0.19	0.59 ± 0.01
T. tomentosa	Flowers	1.18 ± 0.03	n.d.	11.02 ± 0.22	8.89 ± 0.09	3.43 ± 0.02	4.30 ± 0.65
	Bracts	1.15 ± 0.01*	n.d.	2.20 ± 0.05	1.77 ± 0.02	0.83 ± 0.01	3.23 ± 0.02
	Inflorescence	0.99 ± 0.04*	n.d.	7.24 ± 0.15	5.81 ± 0.06	2.27 ± 0.03	4.18 ± 0.03

The quantified investigated standards were expressed as mg in hydroethanolic extract (g) of Tilia species; n = 3.

n.d.: not detected.

*Co-eluted with an unknown compound.

TPC, TFC, and antioxidant activity of the extracts

The highest phenolic content was determined in T. rubra flowers, followed by T. rubra bracts and T. rubra inflorescence (Table 2). Additionally, the highest flavonoid content was determined in T. rubra bracts.

Table 2.

TPC, TFC, and antioxidant activity results of the Tilia samples.

Samples		TPC (mg GAE/g)	TFC (mg QE/g)	DPPH (mg TE/g)	FRAP (mg TE/g)	CUPRAC (mg TE/g)
T. cordata	Flowers	237.93 ± 9.03	21.26 ± 0.60	556.27 ± 22.44	302.66 ± 8.57	787.79 ± 78.08
	Bracts	223.03 ± 3.18	45.67 ± 1.53	518.59 ± 14.55	263.72 ± 20.62	768.38 ± 26.23
	Inflorescence	245.32 ± 13.20	34.33 ± 3.60	558.34 ± 12.67	277.13 ± 8.84	806.72 ± 36.66
T. platyphyllos	Flowers	128.48 ± 8.14	15.38 ± 0.26	165.38 ± 8.41	127.03 ± 17.18	425.55 ± 12.64
	Bracts	57.92 ± 9.12	27.17 ± 1.92	144.90 ± 4.55	70.74 ± 2.84	197.89 ± 21.79
	Inflorescence	153.18 ± 2.79	23.70 ± 0.71	312.35 ± 6.37	183.29 ± 5.12	558.24 ± 60.96
T. rubra	Flowers	336.31 ± 19.89	19.64 ± 1.10	561.45 ± 19.88	446.05 ± 5.88	693.15 ± 22.59
	Bracts	307.50 ± 11.64	58.61 ± 2.85	727.17 ± 2.11	374.20 ± 7.64	203.36 ± 18.84
	Inflorescence	295.43 ± 8.87	36.67 ± 2.22	638.44 ± 14.39	340.85 ± 17.77	487.81 ± 47.37
T. tomentosa	Flowers	232.68 ± 11.29	18.93 ± 0.79	495.27 ± 22.69	86.54 ± 14.74	693.15 ± 22.59
	Bracts	56.55 ± 3.67	24.31 ± 2.80	53.92 ± 10.01	32.75 ± 4.12	203.36 ± 18.84
	Inflorescence	165.63 ± 12.12	22.32 ± 0.82	270.88 ± 16.20	90.08 ± 7.16	487.81 ± 47.37

TPC Total phenolic content; TFC Total flavonoid content; DPPH DPPH radical scavenging activity; FRAP Ferric reducing antioxidant power; CUPRAC Copper reducing antioxidant capacity.

The hydroethanolic extracts of bracts belonging to T. rubra, which had the highest TFC values, also showed potent antioxidant activity with DPPH, FRAP, and CUPRAC assays.

Cell culture studies

Evaluation of inflammatory response

The inflammation parameters were evaluated for the concentrations at which the extracts yielded a viability above 70% in the RAW264.7 cell viability assay (Table 3).

Table 3.

The effects of Tilia sp. on LPS-induced NO, IL-6, and PGE₂ production in RAW264.7 cells.

Groups	Concentration (mg/mL)	Cell viability (%)	Nitrite (µM)	Nitrite inhibition (%)	IL-6A (pg/mL)	PGE2A (pg/mL)
Control	–	124.99 ± 1.94	1.31 ± 0.52	–	313.38 ± 0.42	13.55 ± 2.78
Control + LPS	–	100.00 ± 0.52	33.04 ± 1.12	–	5817.03 ± 460.80	269.10 ± 18.94
Indomethacin	100 µM	98.57 ± 1.46	16.40 ± 1.52*	50.36 ± 4.42	2499.47 ± 338.26*	25.09 ± 3.64*
L-name	100 µM	98.05 ± 1.42	19.67 ± 0.79*	40.43 ± 2.76	3899.16 ± 574.38*	27.79 ± 8.15*
T. cordata flowers	0.125	98.51 ± 0.67	14.56 ± 0.53*	55.80 ± 2.23
	0.25	98.77 ± 1.75	14.43 ± 1.33*	56.23 ± 4.82
	0.5	90.39 ± 0.24	10.95 ± 1.46*	66.77 ± 5.90
	1	84.67 ± 2.59	4.67 ± 0.74*	85.87 ± 2.94	582.98 ± 108.23*	73.71 ± 19.84*
T. cordata bracts	0.125	97.39 ± 3.20	12.20 ± 1.66*	62.96 ± 0.50
	0.25	93.95 ± 0.81	11.60 ± 1.22*	64.78 ± 0.48
	0.5	81.32 ± 0.46	8.55 ± 1.20*	74.08 ± 3.56
	1	74.43 ± 2.26	5.13 ± 0.88*	84.50 ± 4.31	516.15 ± 32.58*	236.35 ± 32.11
T. cordata inflorescence	0.125	94.41 ± 0.10	13.96 ± 1.43*	57.62 ± 1.25
	0.25	87.68 ± 0.92	13.64 ± 0.78*	58.60 ± 0.66
	0.5	84.63 ± 0.10	11.65 ± 1.53*	64.66 ± 2.28
	1	80.07 ± 2.96	4.52 ± 0.87*	86.33 ± 1.47	522.60 ± 10.54*	136.98 ± 10.89*
T. platyphyllos flowers	0.125	95.85 ± 0.33	16.28 ± 0.95*	50.59 ± 0.40
	0.25	89.54 ± 1.12	13.78 ± 1.42*	58.20 ± 2.44
	0.5	87.19 ± 0.39	8.82 ± 1.63*	73.23 ± 4.76
	1	82.48 ± 3.05	4.76 ± 2.02*	85.76 ± 2.15	597.3 ± 6 32.90*	52.92 ± 0.08*
T. platyphyllos bracts	0.125	89.24 ± 3.11	11.46 ± 1.34*	67.27 ± 1.12
	0.25	83.91 ± 1.48	9.24 ± 2.03*	71.79 ± 1.98	695.41 ± 36.44*	33.91 ± 0.31*
	0.5	64.90 ± 2.17	–	–
	1	62.45 ± 0.37	–	–
T. platyphyllos inflorescence	0.125	92.55 ± 0.33	14.10 ± 1.50*	57.20 ± 0.14
	0.25	81.65 ± 4.23	10.17 ± 1.76*	58.88 ± 0.92
	0.5	74.12 ± 0.14	9.06 ± 1.27*	72.54 ± 2.73	467.24 ± 12.70*	40.41 ± 12.01*
	1	65.07 ± 1.16	–	–
T. rubra flowers	0.125	89.43 ± 3.42	11.56 ± 0.37*	64.94 ± 2.89
	0.25	89.22 ± 0.72	10.86 ± 2.93*	67.06 ± 3.69
	0.5	73.74 ± 0.57	8.82 ± 1.54*	73.23 ± 9.96	310.53 ± 12.10*	58.82 ± 5.37*
	1	67.19 ± 3.28	–	–
T. rubra bracts	0.125	96.22 ± 1.02	14.06 ± 0.46*	57.36 ± 4.21
	0.25	90.63 ± 1.01	12.20 ± 1.80*	62.96 ± 0.90
	0.5	89.40 ± 2.63	11.28 ± 0.25*	65.78 ± 2.51	2885.66 ± 101.15*	45.49 ± 7.42*
	1	58.60 ± 4.11	–	–
T. rubra inflorescence	0.125	98.79 ± 1.61	12.48 ± 2.33*	62.13 ± 2.87
	0.25	96.45 ± 2.35	11.83 ± 2.45*	64.10 ± 4.88
	0.5	82.60 ± 1.49	6.69 ± 1.67*	79.69 ± 3.22	583.18 ± 131.34*	30.99 ± 6.36*
	1	64.69 ± 3.91	–	–
T. tomentosa flowers	0.125	101.39 ± 2.74	13.36 ± 1.32*	59.51 ± 1.54
	0.25	84.29 ± 3.10	11.23 ± 0.79*	65.95 ± 1.26
	0.5	81.27 ± 1.28	10.56 ± 0.66*	68.94 ± 2.84
	1	71.35 ± 3.60	3.87 ± 1.22*	88.29 ± 1.88	500.33 ± 83.00*	84.56 ± 12.26*
T. tomentosa bracts	0.125	100.81 ± 2.01	7.67 ± 1.02*	84.03 ± 0.52
	0.25	85.99 ± 2.46	4.24 ± 1.77*	88.95 ± 2.27
	0.5	73.50 ± 4.507	3.96 ± 0.65*	89.10 ± 0.71
	1	71.13 ± 1.81	3.29 ± 0.87*	90.11 ± 1.65	452.69 ± 1.22*	71.16 ± 9.99*
T. tomentosa inflorescence	0.125	98.85 ± 5.47	12.94 ± 1.65*	60.72 ± 1.68
	0.25	92.32 ± 0.28	12.20 ± 1.80*	62.97 ± 2.09
	0.5	90.33 ± 0.96	11.46 ± 2.01*	65.22 ± 1.49
	1	84.62 ± 1.99	3.55 ± 0.43*	89.26 ± 2.10	1806.48 ± 277.63*	143.39 ± 12.28

^AThe parameters IL-6 and PGE₂ were tested at the highest non-cytotoxic dose of the extracts.

Ctrl Control group treated with DMEM; Ctrl + LPS Control group only stimulated with LPS; IND Indomethacin (100 µM); L-NAME N G-nitro-L-arginine methyl ester hydrochloride (100 µM). Results were expressed as mean ± SD compared with control. The significant differences between groups and Ctrl + LPS were denoted with *p < 0.05. LPS Lipopolysaccharides from E. coli; L-NAME Nv-Nitro-L-arginine methyl ester hydrochloride; IND Indomethacin. IL-6 interleukin-6.

• Nitrite inhibition

According to the results, 2 h of pre-treatment with Tilia samples showed significant reducing effects on nitrite levels compared with 100 µM indomethacin. According to the comparative evaluation of different Tilia species extracts, all extracts effectively reduced inflammation by preventing nitric oxide (NO) generation dose-dependently (Table 3). The percentage of nitrite inhibition was over 50% for all non-cytotoxic concentrations (0.125–1 mg/mL).

• IL-6 and PGE ₂ releasing ınhibition

According to the findings, T. rubra flowers at a 0.5 mg/mL concentration exerted the highest inhibitory activity against LPS-activated release of IL-6 (Table 3).

As per our results, T. rubra inflorescence at a concentration of 0.5 mg/mL exerted the highest inhibitory activity against PGE₂ production, followed by T. platyphyllos bracts at the same concentration. The minimum inhibitory effect was obtained for T. cordata bracts at 1 mg/mL concentration, followed by T. tomentosa inflorescence at the same concentration (Table 3).

Anticancer activity

Two-dimensional cell cytotoxicity

All extracts exerted selective cytotoxic activity against MIA PaCa-2 cells, while at least 70% cell viability was obtained with normal human dermal cells (Fig. 2). All extracts showed potent anticancer activity in a dose-dependent manner against pancreatic cells, decreasing cell viability by at least 50% at a concentration of 1 mg/mL. T. rubra bracts exerted the highest cytotoxic activity on MIA PaCa-2 cells at the highest concentration with an IC₅₀ of 0.16 ± 0.01 mg/mL followed by T. rubra flowers (IC₅₀ = 0.21 ± 0.01 mg/mL) and inflorescence (IC₅₀ = 0.24 ± 0.05 mg/mL) (Supplementary Table S5).

Fig. 2.

Effects of Tilia species on the viability of HDF (A) and MIA PaCa-2 (B) cells.

A strong negative correlation was detected between hyperoside and quercitrin with IC₅₀ level (mg/mL) for Mia PACA-2 cell line (r= – 0.914, p < 0.0001, r= – 0.479, p = 0.0031 respectively). Moreover, a significant negative correlation was found between TPC and both MIA PaCa-2 cell viability (r = – 0.516, p = 0.01) and the IC₅₀ value for cell viability (r = – 0.593, p = 0.0001).

Three-dimensional spheroid formation/growth

T. rubra extracts from different parts caused a significant reduction in the growth rate of the spheroids obtained, while the control spheroids-maintained growth for up to 9 days. Even at the lowest dose (0.125 mg/mL), all investigated parts of T. rubra reduced the growth rate of spheroids compared with the control. At the highest dose (1 mg/mL), bracts of T. rubra showed the most substantial spheroid size change (50% reduction). The changes in spheroid diameters of MIA-PaCa-2 cells treated with extracts from flowers, bracts, and inflorescences of T. rubra are provided in Supplementary Table S6. DOX as a reference molecule slowed spheroid growth by 70% compared with the control after 9 days of treatment at a 10 µM concentration. Figure 3 shows spheroid images taken every 3 days.

Fig. 3.

Images of Mia PACA-2 tumor spheroids at the beginning (Day 0) and the end (Day 9) of the experiment (after incubation with 1 mg/mL of the different parts of T. rubra extract). Magnification rate: 10X.

As inflorescences of Tilia species are traditionally used, extracts at the highest doses (1 mg/mL) were further investigated with a 3D spheroid growth assay. The variations in spheroid diameters of cells are outlined in Supplementary Table S7. All species reduced the growth rate of spheroids compared with the control. Similar to our results obtained by the 2D cell cytotoxicity assay, at a 1 mg/mL dose, T. rubra showed the highest anticancer activity, with a 50% reduction in size, followed by T. cordata (32%) and T. tomentosa (24%).

According to the 2D cytotoxicity assay and 3D spheroid growth assay results, all parts of T. rubra (1 mg/mL) were further investigated for apoptosis and necrosis profiles.

The effect of plant derivatives on cell cycle progression of MIA PaCa-2 cells

In order to demonstrate the effect of T. rubra inflorescence, bract, and flower extracts on cell cycle arrest, MIA PaCa-2 cells were treated with 1 mg/mL of extracts from each part for 24 h. (Fig. 4). The average G0/G1 cell population among non-treated, T. rubra inflorescence-treated, bract-treated, and flower-treated cells was 68.45%, 86.94%, 92.25%, and 79.38%, respectively. The average percentage of cells in the S phase for non-treated MIA PaCa-2 cells was 6.13%, while T. rubra inflorescence, bract, and flower extract-treated cell percentages were found to be 6.13%, 5.91%, and 9.93%, respectively. The population of cells in G2/M phase was 7.32%, 1.98%, and 11.0% after treatments with T. rubra inflorescence, bract, and flower in a respective order. For the control group, the frequency of the G2/M cell cycle phase was found to be 15.77%.

Fig. 4.

Cell cycle profiles of MIA PaCa-2 cells examined by flow cytometry. (A) Control MIA PaCa-1 cells, (B) 1 mg/mL of T. rubra inflorescence treated MIA PaCa-2 cells, (C) 1 mg/mL of T. rubra bract treated MIA PaCa-2 cells (D) 1 mg/mL of T. rubra flower-treated MIA PaCa-2 cells and (E) overall cell cycle profiles after treatment with plant derivatives. (****p ≤ 0.001).

In vitro anti-cancer activity of plant derivatives on MIA PaCa-2 cells

MIA PaCa-2 cells were treated with 1 mg/mL of T. rubra inflorescence, bract, and flower extracts, along with 10 µM of DOX as a positive control, for 24 h. Cell death was measured using a flow cytometry method employing an Annexin V-FITC/PI apoptosis assay detection kit. The results revealed minimal apoptotic and necrotic cells in the control group (Fig. 5). In contrast, cells treated with 10 µM DOX, a cell death inducer, exhibited 65% necrosis. Similarly, T. rubra inflorescence, bract, and flower extract treatments led to an increase in necrotic cell percentages. Treatment with 1 mg/mL of T. rubra inflorescence resulted in 60.84% necrotic cells stained with PI and 4.16% Annexin-V-stained apoptotic cells. Likewise, bract and flower treatments showed 54.39% and 66.53% PI-stained necrotic cells and 3.80% and 5.21% Annexin-V-stained apoptotic cells, respectively, after 24 h. Overall, T. rubra inflorescence, bracts and flowers induced statistically similar levels of necrotic cell death at 24 h.

Fig. 5.

Effect of T. rubra inflorescence, bract, and flower extracts on the pancreatic carcinoma cell line, MIA PaCa-2, following 24-h treatment. (A) Non-treated MIA PaCa-2 cells, (B) MIA PaCa-2 cells treated with 1 mg/mL T. rubra inflorescence, (C) 1 mg/mL T. rubra bract, and (D) 1 mg/mL T. rubra flower were subjected to both Annexin V-FITC and propidium iodide prior to analysis using flow cytometry. (E) Doxorubicin (10 µM) was used as a positive control. Dot plot representing control and plant-derivative treated cells. (F) The scatter plot represents the percentages of necrosis (upper left), late apoptosis (upper right), viable cells (lower left), and early apoptosis (lower right) populations of MIA PaCa-2 cells in respective order. (****P ≤ 0.001).

Chemometric analysis

Data matrices were generated in order to monitor the alteration of different species in accordance with our experimental findings. A multivariate data analysis was performed via 18 × 36 data matrices including flowers, bracts and inflorescence of 4 different plat species (cordata, rubra etc.) (Fig. 6). Principal component analysis was carried out and specificity and sensitivity were found to be better than 1.00 (for each group) and 0.889 (the worse value for cordata), respectively. Root Mean Square of Calibration (RMSEC) value were calculated to be 0.53 while Root Mean Square Error of Cross Validation (RMSECV) values were 0.69. Three components were selected to explain model 54.64% of variance for X block and 83.99% of variance for Y block were captured by the PCA model. A PLS-DA analysis was also performed for feature selection and VIP scores were monitored in Fig. 7 which elucidated the significant alterations between the groups. PLS-DA analysis was also indicated that groups were significantly separated. ROC curves for each species were plotted and provided in Supplementary Fig. 1. The correlation between the parameters were emphasized via correlation map by PLS Toolbox for MATLAB software (Supplementary Fig. 2). A heat map was also plotted in order to summarize all findings. In order to perform such analysis, all parameters were scaled by calculating z-score and new data matrices were generated via MATLAB 2023b software (Fig. 8).

Fig. 6.

Score plot for 4 different species (T. cordata, T. platyphyllos, T. rubra, T. tomentosa).

Fig. 7.

PLS-DA Variable in Projection scores for each parameter for T. rubra. Parameters has greater value than the threshold (1.0) accepted to be significantly altered.

Fig. 8.

Heatmap for 4 different species T. cordata (A), T. platyphyllos (B), T. rubra (C), T. tomentosa (D).

According to VIP scores and Heatmap illustrations; Tiliroside, TPC, TFC, DPPH, FRAP, cytotoxicity against Mia PACA-2 cell line and IC₅₀ values were passed the significant threshold. These parameters were found to be statistically significant for T. rubra species against the others.

Discussion

This study investigated the anti-cancer potentials of hydroethanolic Tilia species extracts from different plant parts against MIA PaCa-2 pancreatic cancer cells, as well as the underlying cancer-triggering mechanisms.

Oxidative stress and chronic inflammation are involved in all stages of cancer. ROS may directly damage DNA by introducing structural alterations and gene mutations into the DNA in the initiation stage; block cell-to-cell communication, induce abnormal gene expression, and modify secondary messengers in the promotion stage; and eventually cause an increase in cell proliferation and a decrease in apoptosis. While ROS produced as a part of the inflammatory response facilitates the clearance of pathogens, when produced in excessive amounts and for prolonged periods, it leads to oxidative stress and diseases associated with chronic inflammation^1,2. Therefore, we analyzed the chemical components indicating antioxidant and anti-inflammatory characteristics.

Phenolic compounds are the biggest phytochemical group and have drawn increasing interest due to their ability to prevent and treat oxidative stress-related diseases by inhibiting oxidative damage to proteins, DNA, and lipids, which may lead to mutation, apoptosis, and carcinogenesis³⁰. Toker et al. (2001) found quercetin 3,7-dirhamnoside, quercitrin, astragalin, isoquercitrin, and rutin to be the main and common compounds among the Tilia species T. platyphyllos, T. rubra, and T. tomentosa³¹. Ziaja et al. (2020) investigated the flavonoids in flowers of T. cordata, T. platyphyllos, T. x vulgaris, T. tomentosa, and T. americana, revealing isoquercitrin, astragalin, tiliroside, and protocatechuic acid as the common compounds among them³². These compounds of Tilia species can exhibit a wide range of biological activities, including antioxidant, anti-inflammatory, and anticancer properties^6,11,33–35. In this study, we characterized Tilia species in terms of the amounts of protocatechuic acid, hyperoside, isoquercitrin, astragalin, quercitrin, and tiliroside and investigated their contributions to the pharmacological activity. We found that hyperoside and quercitrin levels as well as TPC were strongly correlated with reduced IC₅₀ values in Mia PACA-2 cells. However, since hyperoside was not detected in all extracts, we attributed the observed effects to quercitrin. It should be noted that other compounds detected in Tilia species might have contributed to the pharmacological activity since the extract constituents might exhibit additive or synergistic effects.

Free radical scavenging activities of flavonoids have been attributed to their ability to donate hydrogen from their phenolic hydroxyl groups, stabilizing free radicals³⁶. Accordingly, the hydroethanolic T. rubra bract and flower extracts, which had the highest TFC and TPC values, respectively, showed potent antioxidant activity.

Macrophages play a crucial role in the immune system by mediating inflammation and initiating defense mechanisms. When stimulated, RAW 264.7 macrophages produce a variety of pro-inflammatory mediators, such as NO and interleukins, making them an ideal model for assessing the anti-inflammatory and immunomodulatory properties of any compound³⁷. Our results showed that all extracts effectively and dose-dependently reduced LPS-induced inflammation in RAW 264.7 cells by preventing NO generation. Toker et al. (2004) reported that 50 mg/kg of T. tomentosa leaves inhibited inflammation in a paw edema model stimulated with carrageenan³⁸. Similarly, T. tomentosa extracts inhibited nitrite production by 89–90%, providing potent protection against inflammation. Moreover, Fawzy et al. (2018) reported that a 70% methanolic extract of T. cordata aerial parts exerted anti-inflammatory activity against carrageenan-induced rat paw edema test equal to indomethacin at a 300 mg/kg dose³⁹. Jabeur et al. (2017) from Portugal reported that 0.2 mg/mL of hydroethanolic T. platyphyllos aerial part extract inhibits NO production by 50% in RAW264.7 cells¹³. In line with these findings, in our study, all extracts showed more than 50% inhibition even at 125 µg/mL, the lowest dose tested.

Supporting the anti-inflammatory activities of Tilia species, LPS-activated release of IL-6, an important pro-inflammatory cytokine, was inhibited in this study, with T. rubra flowers exerting the highest inhibitory activity. Moreover, the inflorescence extract was the most potent in reducing the release of PGE₂, which occurs following a harmful stimulus and results in localized edema⁴⁰.

The anticancer activities of Tilia extracts were assessed by HDF and MIA PaCa-2 cell monolayers prior to the 3D spheroid model. The cytotoxic effects of Tilia extracts on normal and cancer cells vastly differed in that potent anticancer activity was observed against MIA PaCa-2 pancreatic cancer cells, while no toxicity was exerted on HDF cells. Further, T. rubra extracts exerted the highest cytotoxic activity. In line with our findings, Barreiro et al. (2006) demonstrated that exposure to T. cordata flower extracts resulted in a selective and dose-dependent antiproliferative effect on tumoral BW 5147 lymphocytes, leading to apoptosis¹⁵. In normal lymphocytes, all extracts suppressed mitogen-induced proliferation. Jabeur et al. (2017) demonstrated the antitumor activity of a 70% ethanolic extract of T. platyphyllos aerial parts on HeLa cervix cancer, HepG2 hepatocellular carcinoma, NCI-H460 lung carcinoma, and MCF-7 breast adenocarcinoma cells¹³. T. tomentosa extracts prepared using an ethanol/ethyl acetate/water mixture also showed a selective effect on A549 lung adenocarcinoma cells compared with nonmalignant HaCaT cells¹⁶.

This is the first study to demonstrate the anticancer effect of T. rubra. Moreover, the anticancer activity of Tilia species on 3D spheroid growth was evaluated for the first time, which is crucial since 3D models more closely resemble the conditions in the microenvironment than 2D models. While we assessed the anti-cancer activities of different Tilia species, since all parts of T. rubra showed more potent anticancer activity than other extracts in the 2D cytotoxicity assay, only T. rubra extracts were included in the spheroid growth assay.

Extracts from various parts of T. rubra led to a significant reduction in spheroid growth even at the lowest dose tested. This is in line with a previous finding, where plant-based compounds demonstrated an ability to inhibit cancer cell proliferation in both 2D and 3D models⁴¹. Other studies have reported that natural compounds induce cell cycle arrest, consequently inhibiting cancer progression^42,43. Here, we report a similar mechanism, where T. rubra inflorescence, bract, and flower extracts trigger the activation of different checkpoints in the cell cycle, further supporting the notion that plant-derived compounds can interfere with critical regulatory points in cancer cell proliferation⁴⁴.

Given that both cytotoxicity and spheroid assays confirmed the notable effects of T. rubra extracts, their apoptotic and necrotic potential was further investigated. Consistent with findings from earlier research, where necrosis was identified as a key pathway in plant extract-induced cytotoxicity⁴⁵, cells treated with the T. rubra flower extract exhibited the highest necrotic population. Furthermore, both T. rubra inflorescence and bract extracts induced similar levels of necrotic cell death within 24 h of treatment, suggesting that plant extracts from various parts contribute similarly to the induction of necrosis. The ability of T. rubra extracts to arrest the cell cycle and induce necrosis presents a compelling case for its potential as an anticancer agent⁴⁶. Taken together, results from cytotoxicity assays, spheroid growth inhibition, and cell death analyses provide a robust foundation for further investigation into the molecular mechanisms underlying the anticancer properties of T. rubra. In addition, according chemometric analysis, tiliroside, TPC, TFC, DPPH, FRAP, cytotoxicity against Mia PACA-2 and IC₅₀ parameters were passed the significant threshold. These parameters were found to be statistically significant for T. rubra species against the others. These findings suggest that antioxidant capacity and cytotoxicity of T. rubra species were significantly altered compared to others.

We acknowledge that, despite the promising biological activities of the Tilia species, the lack of data on combination treatments with known anticancer agents such as doxorubicin might be considered a limitation of this study.

Conclusion

In conclusion, this study demonstrated potent and promising anti-inflammatory and anti-cancer activities of different hydroethanolic extracts of chemically characterized Tilia species from Türkiye. The anti-inflammatory effects of Tilia species rich in phenolic compounds were supported by their strong capacity to reduce nitrite and inhibit IL-6 and PGE₂ release, while selective and dose-dependent cytotoxicity on pancreatic cancer was demonstrated by a significant reduction in viability. Of note, T. rubra exerted the highest anticancer effect, with extracts from all parts leading to arrest at the G0/G1 phase. Taken together, this study highlights a potential use for Tilia extracts, particularly T. rubra with the major quercitrin content, in the prevention and treatment of pancreatic cancer.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

CUPRAC

Cupric reducing antioxidant capacity

DAD

Diode array detector

DMEM

Dulbecco’s modified eagle’s medium

DNA

Deoxyribonucleic acid

DOX

Doxorubicin

DPPH

2,2-diphenyl-1-picrylhydrazyl

ELISA

Enzyme-linked immunosorbent assay

FBS

Fetal bovine serum

FRAP

Ferric ion reducing antioxidant power

GAE

Gallic acid equivalent

HPLC

High-performance liquid chromatography

HPTLC

High-performance thin layer chromatography

IC50

The half maximal inhibitory concentration

IL-6

Interleukin-6

LPS

Lipopolysaccharide

2D

Two-dimensional

3D

Three-dimensional

MeOH

Methanol

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

PBS

Phosphate buffer saline

PEG 400

Polyethylene glycol 400

PGE₂

Prostaglandin E2

PI

Propidium Iodide

QE

Quercetin equivalent

RNase A

Pancreatic ribonuclease

ROS

Reactive oxygen species

TE

Trolox equivalent

TFC

Total flavonoid content

TLC

Thin-layer chromatography

TPC

Total phenolic content

TPTZ

2,4,6-tris(2-pyridyl)-s-triazine

UV-Vis

Ultraviolet-visible

Author contributions

GY contributed to the conceptualization, data curation, formal analysis, validation, investigation, methodology, and writing of the manuscript. YO, DG, NBŞ, İB and OS contributed to the data curation, formal analysis, validation, investigation, and methodology of the manuscript. GE contributed to investigation. DT and VP contributed to the critical review and editing of the manuscript. EG and HS contributed to the conceptualization, supervision, funding acquisition, data curation, formal analysis, and critical review and editing of the manuscript.

Data availability

Corresponding author, Prof. Dr. Hande Sipahi will share the research data supporting the results of our article via the URL link upon request.

Declarations

Competing interests

The authors declare no competing interests.