Nanoencapsulation of Tomentosin-Rich Fraction in MIL-53(Fe) Improves the Release Profile and Anti-Colorectal Cancer Activity.

1. Introduction
Cancer remains a leading cause of morbidity and mortality worldwide, with 20 million new cases and nearly 10 million deaths reported in 2022 [1]. Its clinical management remains challenging due to tumor heterogeneity, chemoresistance, and treatment-related toxicities. These limitations restrict durable therapeutic responses and highlight the need for safer, affordable, and mechanistically informed interventions [2]. Among solid tumors, colorectal cancer (CRC) represents a major global burden. It is the third most commonly diagnosed cancer and the second leading cause of cancer-related deaths, accounting for 1.9 million new cases and 904,000 deaths in 2022 [1]. In Egypt, CRC incidence is rising and now ranks as the eighth most common cancer, with an incidence rate of 9.8 per 100,000 individuals. Most cases are diagnosed at advanced stages due to limited awareness and poor screening adherence [3]. Together, these trends underscore the need for accessible and regionally relevant therapeutic strategies.
Natural products have long played a central role in oncology drug discovery. They offer exceptional chemical diversity and engage multiple cancer-relevant pathways [4,5,6,7,8]. Reverse translation, in which traditional remedies with documented human use are interrogated using modern scientific tools, provides a powerful framework for identifying promising therapeutic leads.
In this context, Pulicaria crispa (Forssk.) Oliv. (Asteraceae), locally known as “Gethgath” in Egypt and Sudan, is of particular interest. It has been traditionally used to treat digestive disorders, inflammatory conditions, and fever [9,10,11]. This ethnopharmacological background provides a strong rationale for exploring its potential against CRC. Recent studies have begun to validate these uses, reporting antimicrobial, antioxidant, and anticancer activities, particularly in breast cancer models [9,12]. These effects are attributed to its chemically diverse profile, including flavonoids, phenolic acids, alkaloids, and terpenoids. Many of these classes are known modulators of apoptosis, cell-cycle progression, and inflammatory signaling [9,11,13,14].
Despite this promise, clinical translation of plant-derived agents remains limited. Common barriers include chemical instability, poor aqueous solubility, and non-selective biodistribution, all of which reduce bioavailability and therapeutic index [15]. Advanced nanocarrier systems offer an effective strategy to overcome these limitations.
Among them, metal–organic frameworks (MOFs) have emerged as versatile nanoporous drug carriers. Iron-based MOFs, as a class, are particularly attractive due to their high surface area, tunable pore chemistry, intrinsic biocompatibility, and capacity for controlled drug release, as highlighted in recent biomedical reviews [16,17,18]. Importantly, rational nanoformulation can do more than improve delivery. It can reshape the mechanistic behavior of phytochemicals. For example, nanoencapsulation has enhanced the anticancer efficacy and pathway selectivity of several natural compounds [19,20]. Silica nanoparticle-loaded quercetin was reported to preferentially disrupt the CDK4/6–cyclin D axis, inducing G1/S arrest in chemoresistant CRC cells [18]. Similarly, chitosan-based nanoencapsulation of curcumin shifted its dominant mode of cell death from apoptosis to necroptosis via RIPK1/MLKL activation in CRC models [18]. Free curcumin itself can upregulate necroptotic markers in colon cancer cells [21], an effect modulated by curcuminoid composition [22]. These examples illustrate how nanocarriers can fundamentally alter both pharmacokinetic and mechanistic profiles of plant-derived agents.
This perspective reflects a broader conceptual shift in nanomedicine. Nanoencapsulation is no longer viewed solely as a solution to solubility or stability issues. Instead, it is increasingly recognized as a tool to modulate intracellular trafficking, pathway engagement, and pharmacodynamic outcomes [23]. For bioactive flavonoids and terpenoids, advanced nanocarriers can enhance cellular uptake, protect labile compounds, and promote targeted release within the tumor microenvironment [24,25]. Together, these effects can markedly improve interaction with key molecular targets and amplify therapeutic efficacy [26,27].
Guided by this framework, we adopted a mechanism-guided, reverse-translation strategy to rationally formulate the most active fraction of P. crispa. Our prior screening identified the dichloromethane fraction (DCMF) as the most potent, albeit non-selective, fraction [28]. The present study therefore aimed to: (i) characterize the phytochemical profile of DCMF using GC–MS; (ii) design and physicochemically characterize a MIL-53(Fe)-based nanocomposite (DCMF@MIL-53(Fe)); (iii) assess the impact of nanoencapsulation on in vitro cytotoxicity, selectivity, and release kinetics against colorectal cancer cells; and (iv) elucidate the underlying cell-death mechanisms using cell cycle analysis, cell-death assays, and in silico molecular docking focused on the most abundant constituents.

2. Materials and Methods

2.1. Materials and Reagents
All chemicals and reagents were of analytical grade. For MOF synthesis, 1,4-benzenedicarboxylic acid (H2BDC, 99.7%), iron (III) nitrate nonahydrate (Fe(NO3)3·9H2O, 99.7%), and hydrochloric acid (HCl, 38%) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Potassium hydroxide (KOH, 99%) and sodium hydroxide (NaOH) were obtained from Egyptian Piochem (Giza, Egypt), while N,N-dimethylformamide (DMF, 99%) and methanol (99%) were sourced from Alpha Chemika (Mumbai, India). All solvents for plant extraction and fractionation (methanol, n-hexane, dichloromethane, and ethyl acetate) were procured from El Gomhouria Company for Pharmaceuticals, Chemicals, and Medical Supplies (Cairo, Egypt). Reagents and materials for biological assays, including cell culture media, supplements, assay kits, and cell lines, are specified in their respective methodology sections.

2.2. Plant Material
The aerial parts of P. crispa (Forssk.) Oliv. were collected in March 2021 from Saint Catherine, South Sinai, Egypt (28°33′42.88″ N, 33°56′57.62″ E) under official authorization. Botanical identification was performed by Mrs. Terase Labib, Taxonomist at Orman Garden, Giza, Egypt. A voucher specimen (Pc21) was deposited at the Herbarium of the Department of Pharmacognosy and Medicinal Plants, Faculty of Pharmacy (Girls), Al-Azhar University, Cairo, Egypt.

2.3. Extraction and Fractionation
The shade-dried, powdered aerial parts (300 g) were exhaustively macerated in methanol (3 × 2 L, 72 h each) at room temperature. The combined filtrates were concentrated under reduced pressure at 40 °C to yield the crude methanol extract (CME, 75 g, 25% w/w). The CME was subsequently suspended in distilled water and sequentially partitioned with solvents of increasing polarity: n-hexane, dichloromethane (DCM), and ethyl acetate. After solvent evaporation, the following fractions were obtained: the n-hexane (HF, 3 g, 1.0% w/w), dichloromethane (DCMF, 10 g, 3.3% w/w), ethyl acetate (EF, 13 g, 4.3% w/w), and the remaining aqueous (WF, 40 g, 13.3% w/w) fractions, as previously described [14]. All fractions were stored at −20 °C until further use.

2.4. Chemical Profiling of P. crispa DCMF
Chemical characterization of P. crispa DCMF was conducted using a Trace GC-TSQ mass spectrometer (Thermo Scientific, Austin, TX, USA) equipped with a TG-5MS capillary column (30 m × 0.25 mm × 0.25 µm film thickness). The column oven temperature was initially set to 60 °C, held for 2 min, then increased at a rate of 5 °C/min to 250 °C, followed by a 2 min hold. The temperature then increased at a rate of 30 °C/min to 300 °C. The injector temperature was set to 270 °C. Helium was used as the carrier gas at a constant flow rate of 1 mL/min. 1 µL of diluted sample was automatically injected using an Autosampler AS3000 in split mode with a 4 min solvent delay.
Electron ionization (EI) mass spectra were acquired at 70 eV over an m/z range of 50–650 in full-scan mode. The ion source and transfer line temperatures were set at 200 °C and 280 °C, respectively. Compound identification was achieved by comparing the acquired mass spectra with those in the WILEY 09 and NIST 14 spectral libraries. Retention indices (RI) were experimentally determined using a homologous series of n-alkanes (C10–C35) analyzed under identical chromatographic conditions. RI values were calculated according to the Van den Dool and Kratz equation and subsequently compared with reported literature values for non-polar stationary phases.

2.5. Synthesis of MIL-53(Fe) Metal-Organic Framework (MOF)
MIL-53(Fe) was synthesized using a hydrothermal approach, as previously described [29], with minor modifications. Briefly, 1.66 g of 1,4-benzenedicarboxylic acid (H2BDC) and 2.703 g of iron (III) chloride hexahydrate (FeCl3·6H2O) were dissolved in 50 mL of dimethylformamide (DMF) under continuous stirring for 10 min to ensure homogeneity. The resulting solution was transferred into a 160 mL Teflon-lined stainless-steel Parr autoclave, sealed, and heated at 150 °C for 65 h to promote MOF crystallization. After cooling to room temperature, the resulting solid was collected by centrifugation and washed three times with DMF, followed by three washes with methanol, to remove residual solvent and unreacted precursors. The final product was vacuum-dried overnight prior to characterization.

2.6. Loading DCMF onto MIL-53(Fe)
The P. crispa DCMF was loaded onto MIL-53(Fe) by controlled adsorption. Briefly, MIL-53(Fe) (100 mg) was dispersed in deionized water (20 mL) and added to a methanolic solution of DCMF (100 mg in 10 mL). The mixture was stirred continuously for 48 h at room temperature. Subsequently, the resulting DCMF@MIL-53(Fe) nanocomposite was collected by centrifugation (6000 rpm, 10 min, 4 °C), washed repeatedly with deionized water to remove any unbound DCMF, and vacuum-dried.

2.7. Characterization of MIL-53(Fe) and DCMF-Loaded Nanoformulation
The synthesized materials—MIL-53(Fe), DCMF (pure drug), and the DCMF-loaded nanocomposite (DCMF@MIL-53(Fe))—were comprehensively characterized to confirm successful nanocomposite formation and to evaluate their physicochemical properties. Powder X-ray diffraction (PXRD) patterns were recorded using a Bruker D8 ADVANCE diffractometer (Bruker AXS GmbH, Karlsruhe, Germany) with Cu Kα radiation (λ = 1.5406 Å) over a 2θ range of 5–70°, operating at 40 kV and 30 mA with a scan rate of 2° min−1, to assess crystallinity and phase purity. Functional groups were identified by Fourier-transform infrared (FTIR) spectroscopy (Bruker Vertex 70 spectrometer; Bruker Optics GmbH & Co., KG, Ettlingen, Germany) using KBr pellets over the range 4000 to 400 cm−1. The morphological features and elemental composition were examined using scanning electron microscopy (SEM, Quanta FEG250; Thermo Fisher Scientific, Hillsboro, Oregon, USA) coupled with energy-dispersive X-ray spectroscopy (EDX); SEM analyses were conducted on dry powder samples.

Sample Preparation for Colloidal Characterization (DLS and TEM)
For colloidal characterization, samples were analyzed as aqueous dispersions. Briefly, 5 mg of dried powder was dispersed in 10 mL of deionized water, stirred for 10 min, and subjected to bath sonication for 15 min to obtain a homogeneous suspension. For DLS measurements, the dispersion was diluted to a final concentration of 0.1–0.2 mg mL−1, allowed to stand for 5 min, and the supernatant was analyzed [30].
Transmission electron microscopy (TEM, JEOL JEM-2010; JEOL Ltd., Akishima, Tokyo, Japan) was performed on droplets of the prepared dispersions deposited onto carbon-coated copper grids. Colloidal stability was evaluated by dynamic light scattering (DLS) and zeta potential measurements using a Malvern Zetasizer Nano ZS90 (Malvern Panalytical Ltd., Malvern, Worcestershire, UK) at 25 °C and a backscattering angle of 173°. The intensity-weighted hydrodynamic diameter (Z-average), polydispersity index (PDI), and surface charge were reported. Thermogravimetric analysis (TGA) was conducted using a Netzsch STA 409 PC/PG instrument (NETZSCH Gerätebau GmbH, Selb, Germany) under a nitrogen atmosphere to assess thermal stability and composition.

2.8. Entrapment Efficiency (EE) and Loading Capacity (LC)
The entrapment efficiency (EE, %) and loading capacity (LC, %) of DCMF in MIL-53(Fe) were determined using a Thermo Fisher Evolution 350 UV-Vis spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). After centrifugation, as outlined in Section 2.6, 10 mL supernatant was collected, and the concentration of unloaded DCMF in the supernatant was determined from its UV absorbance at 230 nm against a standard calibration curve. To ensure accuracy, the absorbance contribution of a blank dispersion of pristine MIL-53(Fe) at an equivalent concentration was measured and subtracted from the supernatant reading prior to calculation [31]. The amount of DCMF loaded in MIL-53(Fe) was calculated by subtracting the amount of free DCMF in the supernatant from the initial DCMF amount added. EE and LC were calculated using the following equations:

2.9. Evaluation of Release Kinetics
The in vitro release profile of DCMF from the MIL-53(Fe) nanocomposite was compared with that of free DCMF in phosphate-buffered saline (PBS, pH 7.4). Dialysis bags (MWCO: 12,000–14,000 Da) were loaded with 5 mL of free DCMF or an equivalent DCMF-loaded nanocomposite suspension, matched for drug content. The bags were immersed in 100 mL PBS and incubated at 37 ± 0.5 °C under continuous shaking at 150 rpm.
Aliquots (1 mL) were collected at 0.5, 1, 2, 3, 4, 5, 6, 8, 12, 24, and 48 h. Each withdrawn volume was immediately replaced with fresh PBS to maintain sink conditions. Released DCMF was quantified by UV–Vis spectrophotometry at 230 nm, a wavelength corresponding to the α,β-unsaturated carbonyl chromophore characteristic of tomentosin, the predominant constituent of DCMF [32]. To account for potential optical interference from the MOF, blank MIL-53(Fe) dispersions were processed and analyzed under identical conditions. Their absorbance values were subtracted from all measurements to obtain background-corrected signals attributable solely to the released DCMF [31].
Method linearity, accuracy, and precision were verified before analysis. Release data were fitted to zero-order, first-order, Higuchi, Korsmeyer–Peppas, Hixson–Crowell, and Weibull models using appropriate software. The best-fitting model was identified based on the highest coefficient of determination (R2) and supporting selection criteria, indicating the predominant release mechanism [33].

2.10. Cell Lines and Culture Conditions
Colorectal carcinoma (HCT-116) cell line was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Normal human dermal fibroblasts (HFB4) were acquired from VACSERA (Dokki, Egypt). All cells were maintained at 37 °C/5% CO2 in: Dulbecco’s Modified Eagle Medium (DMEM; Lonza, Basel, Switzerland) for HFB4 cells and in McCoy’s 5A medium (Merck, Darmstadt, Germany) for HCT-116 cells. All media supplemented with 10% fetal bovine serum (FBS; Lonza, Switzerland), 2 mM L-glutamine (Merck, Germany), and during initial recovery only, penicillin–streptomycin (10,000 U/mL; Lonza, Switzerland) and amphotericin B (0.25 µg/mL). Media were replaced every 48–72 h or as needed, and cells were subcultured at ~75% confluence using 0.05% trypsin EDTA (Gibco, Waltham, MA, USA).

2.11. MTT Assay
Cytotoxicity was evaluated using the MTT assay ([3-(4,5-dimethylthiazol-2-yl) 2,5 diphenyl tetrazolium bromide], Merck, Germany) as previously described [7]. HCT-116 and HFB4 cells were seeded in 96-well plates (1 × 104 cells/well) in 100 μL complete medium and allowed to adhere for 24 h. Cells were then treated with serial dilutions of MIL-53(Fe), DCMF, DCMF@MIL-53(Fe) (3.1–200 µg/mL), or sorafenib (positive control; 3.1–50 µg/mL), and incubated for 48 h. To minimize optical interference from the nanomaterials, culture medium was removed and wells were gently washed once with PBS. Subsequently, 40 µL MTT solution (0.5 mg/mL) was added per well and incubated for 4 h at 37 °C. The resulting formazan crystals were dissolved in 180 µL acidified isopropanol, and absorbance was measured at 570 nm (FLUOstar OPTIMA, BMG LABTECH GmbH, Ortenberg, Germany). Data were presented as mean ± SD of three technical replicates (n = 3).
All treatments were applied based on apparent concentration, defined as the total mass of compound or nanocomposite per unit volume. For the nanocomposite, DCMF-equivalent concentrations were calculated post hoc using the experimentally determined loading capacity (LC = 20.6%). This normalization enables direct comparison of bioactive content between free and encapsulated DCMF.
Dose–response curves were generated from apparent concentration data. For comparative visualization, the fitted curve for DCMF@MIL-53(Fe) was mathematically re-parameterized and plotted on a DCMF-equivalent concentration axis.
Cell viability was calculated as:
IC50 values were determined by nonlinear regression (GraphPad Prism 9.0). For DCMF@MIL-53(Fe), corrected IC50 was calculated as:
where LC% is the loading capacity determined in Section 2.8.
The selectivity index (SI) was calculated as:

2.12. Cell Cycle Analysis
Cell cycle progression in HCT-116 cells treated with DCMF@MIL-53(Fe) was assessed using propidium iodide (PI) staining and flow cytometry, according to previously established methods [7]. Briefly, cells (1 × 105 cells/mL) were seeded overnight, then treated with DCMF@MIL-53(Fe) at its predetermined IC50 concentration for 24 h. Cells were harvested by trypsinization, washed, and fixed overnight at −20 °C in 70% ice-cold ethanol. After fixation, cells were stained with PI/RNase solution to label DNA content, enabling differentiation into G0/G1, S, and G2/M phases. Flow cytometric analysis was performed using a BD FACSCalibur flow cytometer (BD Biosciences, Milpitas, CA, USA).

2.13. Apoptosis Detection
Apoptosis in HCT-116 cells was evaluated using the Annexin V-FITC Apoptosis Detection Kit (BioVision Inc., Milpitas, CA, USA, K101-25), following the manufacturer’s protocol. This assay detects phosphatidylserine translocation (early apoptosis) via Annexin V-FITC staining, while propidium iodide (PI) stains the DNA of late apoptotic and necrotic cells. HCT-116 cells (1 × 105 cells/mL) were cultured for 24 h and treated with DCMF@MIL-53(Fe) at its IC50 concentration for 48 h. After treatment, cells were trypsinized, washed with serum-containing media, and centrifuged at 300× g for 10 min. The cell pellet was resuspended in 500 µL of 1X binding buffer and stained with 5 µL of Annexin V-FITC and 5 µL of PI for 5 min in the dark. Flow cytometry analysis was performed using a BD FACSCalibur flow cytometer (BD Biosciences, CA, USA).
Apoptotic populations were classified by flow cytometry as follows: live cells (Q4: Annexin V−/PI−), early apoptotic cells (Q3: Annexin V+/PI−), late apoptotic/necrotic cells (Q2: Annexin V+/PI+), and dead/necrotic cells (Q1: Annexin V−/PI+). The percentages of cells in each quadrant were calculated to determine the extent of apoptosis and necrosis.

2.14. In Silico Predictions and Molecular Modeling

2.14.1. ADME and Pharmacokinetics
The physicochemical and ADME properties of tomentosin, selected based on its high abundance and reported anticancer activity [34,35], were predicted using the SwissADME [36] and PreADME [37] servers. These tools were used to evaluate key pharmacokinetic parameters, including absorption, distribution, metabolism, excretion, and toxicity.

2.14.2. Molecular Docking Study
A molecular docking of tomentosin was performed against selected protein targets relevant to the observed cell-cycle arrest and cell death responses. The investigated targets included CDK4/Cyclin D3 (PDB ID: 7sj3) [38], MDM2-P53 (PDB ID: 7bmg) [39], RIPK1 (PDB ID: 6nw2) [40], and RIPK3-MLKL (PDB ID: 7mon) [41]. Docking simulations were conducted using AutoDock 4.2 [42].
Protein and ligand structures were prepared following standard molecular modeling protocols, and ligand files were converted to PDBQT format. Grid boxes were set to 60 × 60 × 60 Å with target-specific coordinates: CDK4/Cyclin D3 (12.93, −38.61, 10.71 Å), MDM2-P53 (−22.03, −9.21, 13.16 Å), RIPK1 (8.13, −14.55, −26.47 Å), and RIPK3-MLKL (−21.05, −20.28, −3.14 Å).
Docking poses were visualized and analyzed using Discovery Studio Visualizer (DSV) [43]. Binding affinity scores and RMSD values were used to assess interaction strength and complex stability.

3. Results

3.1. GC-MS Chemical Profiling of P. Crispa DCMF
GC-MS analysis of P. crispa DCMF identified 24 phytoconstituents (Figure 1 and Table 1). The chromatogram was dominated by the sesquiterpene lactone tomentosin (28.82%). Its EI mass spectrum showed a molecular ion [M]•+ at m/z 248, with sequential fragment losses at m/z 230 (−H2O) and 215 (−CH3). A characteristic side-chain cleavage fragment was observed at m/z 190, with a base peak at m/z 91 corresponding to the tropylium ion (Figure 2).
Oxygenated terpenoids (marked by * in Table 1), including tomentosin, β-caryophyllene oxide, α-Cadinol, a cycloheptafuranone-type lactone derivative, and stigmasterol, collectively accounted for approximately 39.83% of the total detected constituents. In addition, the fraction was characterized by a substantial lipid component. This included several unsaturated fatty acids, such as palmitoleic, linoleic, α-linolenic, and eicosatetraenoic acids, together with saturated fatty acids, mainly palmitic and docosanoic acids. Straight-chain alkanes were also detected as minor constituents of the fraction.
Given its clear quantitative predominance within the fraction and its well-documented anticancer activity [34,35], tomentosin was selected as a representative molecular probe for subsequent in silico investigations aimed at contextualizing the observed biological responses. While the biological effects reported in this study arise from the whole DCMF, this approach enables mechanistic hypothesis generation grounded in the dominant bioactive constituent, without excluding potential synergistic or contributory roles of minor components.

3.2. Characterization of the MIL-53(Fe) Before Loading

3.2.1. FTIR Spectroscopy for MIL-53(Fe)
The FTIR spectrum of the synthesized MIL-53(Fe) (Figure 3a) showed a broad absorption band in the region of 3195–3752 cm−1, corresponding to O-H stretching vibrations. Characteristic symmetric and asymmetric vibrations of the carboxylate group from the BDC linker were observed at 1388 cm−1 and 1594 cm−1 respectively. Additional peaks were identified at 1659 cm−1 (C=O stretch), 1015 cm−1 (benzene ring vibration), and 750 cm−1 (C-H bonding of benzene rings). Faint bands at 546 cm−1 and 462 cm−1 were attributed to Fe-O stretching vibrations.

3.2.2. Powder X-Ray Diffraction (PXRD) of the Prepared MIL-53(Fe)
The PXRD pattern of the synthesized MIL-53(Fe) (Figure 3b) exhibited characteristic peaks at 2θ = 9.47°, 11.01°, 17.43°, 19.6°, 21.16°, and 25.02° which pointed using star shape, which matched the simulated pattern from CIF files 690,314–690,316 [44,45], confirming successful formation of the MIL-53(Fe) framework.

3.2.3. Surface Morphology and Elemental Composition
SEM imaging revealed that the MIL-53(Fe) particles exhibited a bipyramidal polyhedral prismatic morphology with smooth surfaces (Figure 3c). The average particle length was approximately 2.2 μm. EDX analysis confirmed the presence of carbon, oxygen, and iron as the primary elemental constituents (Figure 3d).

3.2.4. Thermogravimetric Analysis (TGA)
The TGA profile of MIL-53(Fe) (Figure 3e) showed an initial weight loss between 60 and 160 °C, attributed to removal of adsorbed water molecules. A second weight loss occurred between 250 and 350 °C, corresponding to decomposition of the organic linker, followed by a further loss between 400 and 620 °C due to framework collapse and iron ox-ide formation.

3.3. Characterization of the DCMF-Loaded MIL-53(Fe)

3.3.1. FTIR Spectroscopy
The FTIR spectrum of the composite material after loading with DCMF (DCMF@MIL-53(Fe)) is presented in Figure 4a, alongside the spectra of the pristine MOF and free DCMF. Critically, the fundamental fingerprint of the MIL-53(Fe) framework remains intact upon DCMF loading. The characteristic carboxylate vibrations (at 1579 cm−1 and 1388 cm−1) and the Fe–O band (at 535 cm−1) persist, demonstrating the structural stability of the MOF following the loading process. The absence of significant new absorption bands in the DCMF@MIL-53(Fe) spectrum, when compared to the distinct peaks of the free DCMF extract, suggests that the encapsulation is primarily a physical phenomenon within the pores, without the formation of new chemical bonds.

3.3.2. Powder X-Ray Diffraction (PXRD)
The MIL 53(Fe) pattern shows the expected reflections near 9.4°, 12.6°, 17.8°, and 25.4–25.7° 2θ (Cu Kα), consistent with the narrow pore/large pore phases reported for MIL 53(Fe) and with COD CIF 690,314–690,316 (Figure 4b). After DCMF loading, all MIL 53(Fe) Bragg peaks are retained without new reflections, indicating preservation of the framework; minor intensity changes are attributable to guest occupancy and scattering contrast (Figure 3b).

3.3.3. Colloidal Stability and Morphological Analysis
The colloidal properties of the DCMF@MIL 53(Fe) nanocomposite were assessed in aqueous dispersion (Section Sample Preparation for Colloidal Characterization (DLS and TEM)). Dynamic light scattering (DLS) measurements yielded an intensity weighted hydrodynamic diameter (Z average) of 218.3 ± 9.6 nm with a polydispersity index (PDI) of 0.41 ± 0.03 (Figure 5a). This nanoscale dimension contrasts with the micrometer scale prisms observed by SEM (Section 3.2.3), a difference attributable to sample state: SEM imaged dry, intact crystals, while DLS analyzed sonicated dispersions where fragile crystals fragment into nanoscale colloids.
Zeta potential measurements indicated good colloidal stability. The pristine MIL-53(Fe) carrier had a zeta potential of +29 ± 2 mV, which decreased to +18 ± 2.6 mV after DCMF loading (Figure 5b). This reduction in positive surface charge suggests partial shielding by the loaded phytochemical fraction.
Transmission electron microscopy (TEM) visualized particle morphology. Unloaded MIL 53(Fe) fragments appeared as discrete nanoplates (Figure 6a), whereas the DCMF loaded composite showed an amorphous, electron dense coating on the particle surfaces (Figure 6b), providing direct evidence of successful encapsulation. The observed surface coating and a modest degree of particle aggregation in the loaded sample correlate with the measured decrease in zeta potential.

3.4. Entrapment Efficiency (EE) and Loading Capacity (LC)
The entrapment efficiency (EE, %) and loading capacity (LC, %) of DCMF in MIL-53(Fe) were assessed as critical indicators for evaluating the performance of nanocarriers in delivering bioactive compounds. DCMF in MIL-53(Fe) exhibited an entrapment efficiency of 83.6 ± 1.5% and a loading capacity of 20.6 ± 2.2%. These results demonstrate the high encapsulation potential of the MIL-53(Fe) framework in efficiently retaining the bioactive fraction, DCMF.

3.5. In Vitro Release Profile of DCMF from the MIL-53(Fe)
The release kinetics of DCMF from the MIL-53(Fe) framework were evaluated over 48 h and compared to the free extract (Figure 7). Free DCMF exhibited rapid release, reaching completion within 12 h. In contrast, the DCMF@MIL-53(Fe) composite demonstrated a sustained and prolonged release profile, with only 62.5% released at 12 h and a gradual increase to 100% by 48 h.
To investigate the release mechanism, the data were fitted to standard kinetic models (Table 2). The DCMF@MIL-53(Fe) release profile correlated strongly with the Korsmeyer-Peppas (R2 = 0.99) and Higuchi (R2 = 0.98) models. The superior fit to these models, compared to the free DCMF (R2 = 0.97 and 0.95, respectively), indicates that encapsulation shifts the release mechanism to one that is predominantly diffusion-controlled. The high correlation with the Korsmeyer-Peppas model specifically suggests that the release is governed by the diffusion through the porous framework of the MOF.

3.6. Cytotoxicity and Selectivity
The cytotoxic effects of free DCMF, pristine MIL-53(Fe), the DCMF@MIL-53(Fe) nanocomposite, and the reference control Sorafenib were evaluated against HCT-116 and HFB4 cell lines (Figure 8, Table 3).
The DCMF@MIL-53(Fe) nanocomposite demonstrated a significantly enhanced cytotoxicity against HCT-116 cells, with an apparent IC50 of 2.05 ± 0.15 μg/mL. Correction for the drug loading capacity (20.6%) yielded a DCMF-equivalent IC50 of 0.40 ± 0.05 μg/mL, representing a 53.8-fold increase in potency over free DCMF (IC50 = 21.51 ± 1.2 μg/mL). The MIL-53(Fe) carrier alone showed no cytotoxicity (IC50 > 200 μg/mL). The potency of the nanocomposite was also compared to Sorafenib, a standard chemotherapeutic. The corrected IC50 of the nanocomposite (0.40 μg/mL) was substantially lower than that of the standard chemotherapeutic Sorafenib (12.53 ± 0.28 μg/mL), indicating superior in vitro potency against HCT-116 cells.
Selectivity was quantified using the Selectivity Index (SI = IC50(HFB4)/IC50(HCT-116)). The DCMF@MIL-53(Fe) nanocomposite (SI = 1.62) demonstrated a clear preferential cytotoxicity for cancer cells over normal cells. This contrasts with both free DCMF (SI = 0.87) and Sorafenib (SI = 0.47), which showed higher toxicity to normal cells.

3.7. Cell Cycle Analysis via Flow Cytometry
Flow cytometry analysis revealed that DCMF@MIL-53(Fe) treatment induced significant cell cycle dysregulation in HCT-116 colorectal carcinoma cells compared to untreated controls (Figure 9, Table 4). The nanocomposite triggered pronounced G0/G1 phase accumulation (80.89% ± 4.85 vs. 64.00% ± 5.12 in controls; p < 0.001), representing a 1.6-fold increase consistent with G1/S transition blockade. Concurrently, S-phase populations decreased dramatically (17.25% ± 1.03 vs. 36.32% ± 2.6; p < 0.001), indicating impaired DNA synthesis. Minimal change occurred in G2/M phase distribution (1.18% ± 0.06 vs. 0.55% ± 0.03), confirming the primary impact at the G1 restriction point.

3.8. Analysis of Cell Death Mode (Apoptosis/Necrosis)
Flow cytometry analysis revealed that DCMF@MIL-53(Fe) treatment predominantly induced necrotic cell death, as indicated by an increased Annexin V-/PI+ population (Figure 10, Table 5). Untreated controls showed high viability (95.89% ± 1.8), while nanocomposite-treated cells exhibited significantly reduced viability (85.96% ± 2.1; p < 0.001) with a 4.7-fold increase in necrosis (12.85% ± 0.9 vs. 2.71% ± 0.2 in controls; p < 0.001). Apoptosis induction was statistically insignificant, with early apoptosis decreasing to 0.96% ± 0.05% from 1.27% ± 0.09% and late apoptosis remaining negligible (0.23% ± 0.01% vs. 0.13% ± 0.01%). These data indicate that DCMF@MIL-53(Fe) primarily compromises plasma membrane integrity under tested conditions, resulting in a necrotic, non-apoptotic cell death modality.

3.9. In Silico Adme Profiling of Tomentosin
SwissADME and preADMET analyses predicted generally favorable oral drug-likeness for tomentosin (Supplementary Table S1). The compound complied with Lipinski’s Rule of Five (MW = 248.32 g/mol, MlogP = 2.38, HBD = 0, HBA = 3, zero violations) and Veber’s Rule (rotatable bonds = 3, TPSA = 43.37 Å2). Key absorption parameters indicated high intestinal permeability (Caco-2 = 31.63 nm/s; moderate) and near-complete human intestinal absorption (HIA = 98.2%). Distribution profiling revealed high plasma protein binding (PPB = 94.8%) and predicted blood–brain barrier penetration. Crucially, tomentosin was predicted not to be a P-glycoprotein substrate. The bioavailability score was 0.55, indicating high oral bioavailability potential, and synthetic accessibility (SA) score was 4.14, suggesting moderate synthetic tractability.

3.10. Molecular Docking Study of Tomentosin
Tomentosin adopted well-defined poses in the active sites of CDK4/Cyclin D3, MDM2–p53, RIPK1, and RIPK3–MLKL (Table 6; Figure 11, Figure 12, Figure 13 and Figure 14; Supplementary Table S2). Docking returned a competitive score for CDK4/Cyclin D3 (−8.6 kcal/mol) relative to the co-crystallized ligand (−8.5 kcal/mol), and a comparable score at MDM2–p53 (−8.5 kcal/mol vs. −8.5 kcal/mol). By contrast, tomentosin scored weaker than the reference ligands at RIPK1 (−8.0 kcal/mol vs. −9.0 kcal/mol) and RIPK3–MLKL (−6.5 kcal/mol vs. −8.0 kcal/mol). Re-docking of the native ligands reproduced their crystallographic poses with RMSD = 0.62 Å (CDK4/Cyclin D3), 0.21 Å (MDM2–p53), 0.53 Å (RIPK1), and 0.71 Å (RIPK3–MLKL) (Figures S1–S4), satisfying the <2 Å validation criterion and supporting protocol reliability.

4. Discussion
Building on the traditional gastrointestinal use of P. crispa for treating digestive ailments, our prior research identified the dichloromethane fraction (DCMF) as the most potent anti-proliferative agent against CRC cells from its crude extract and other fractions [28]. However, its therapeutic potential was limited by a low selectivity index. To address this limitation and bridge ethnopharmacological knowledge with modern therapeutic strategies, this study employed nanoencapsulation to enhance the bioavailability and selective delivery of DCMF’s bioactive compounds. This approach aimed not only to improve cytotoxic efficacy but also to elucidate the mechanistic basis for its traditional therapeutic potential against CRC.
Initially, the chemical composition of the active DCMF was investigated using GC-MS analysis to identify the phytoconstituents responsible for its antiproliferative activity. Tomentosin, a sesquiterpene lactone and major constituent of the Asteraceae family [46], was identified as the most abundant phytoconstituent (28.82%), alongside fatty acids. Tomentosin has demonstrated anticancer activity in various models, primarily through induction of cell-cycle arrest (SubG1 and G2/M), apoptosis (↑Bax, Bim, p53; ↓Bcl-2), and ROS generation [34,35,47,48]. Furthermore, our in silico ADME predictions of tomentosin suggest it as a drug-like compound within the DCMF, with favorable properties such as Rule-of-Five compliance, high predicted human intestinal absorption (~98%), non-substrate behavior for P-glycoprotein (P-gp), and a bioavailability score of 0.55 (Supplementary Table S1). These findings position tomentosin as a promising candidate for further development, supporting its potential as a key bioactive compound for CRC therapy.
To address the inherent challenges of selective delivery, the tomentosin rich DCMF was encapsulated into MIL-53(Fe), a metal-organic framework known for its potential in targeted drug delivery. Comprehensive characterization confirmed the successful formation of the DCMF@MIL-53(Fe) composite. PXRD analyses confirmed the retention of the MIL-53(Fe) crystalline structure (e.g., Bragg reflections at ~9.4°, 12.6°, 17.8°, 25.4–25.7° 2θ) without the formation of new phases after loading. Similarly, FTIR spectra preserved the characteristic linker vibrations ν_as/ν_s(OCO) bands (~1590/1390 cm−1) and the Fe–O bond (~535–550 cm−1), collectively indicating the framework’s integrity was maintained during the encapsulation process. The composite achieved a high entrapment efficiency of 83.6 ± 1.5% with a loading capacity of 20.6 ± 2.2%, and formed a monodisperse, nanoscale system with a hydrodynamic diameter of 218 nm, as determined by DLS. The shift in zeta potential from +29 ± 2 mV (pristine MIL-53(Fe)) to +18 ± 2.6 mV after loading indicates partial surface masking by DCMF, while maintaining a net positive charge. This positive charge is believed to facilitate electrostatic interactions with the typically anionic membranes of cancer cells, a mechanism that has been shown to enhance the cellular uptake of cationic nanocarriers [49].
Moreover, the encapsulation fundamentally altered the drug release profile from a burst release in the free DCMF (~100% release by ~12 h) to a sustained (~62.5% at 12 h; near-complete by 48 h), diffusion-driven release from the nanocomposite, as confirmed by high fit to the Korsmeyer-Peppas and Higuchi models (with high coefficients of determination: R2 = 0.99/0.98 for DCMF@MIL-53(Fe); 0.97/0.95 for free). This sustained release behavior is a known characteristic of MIL-type MOFs and is critical for maintaining therapeutic drug levels over extended periods, improving efficacy, and potentially reducing side effects [50]. This pH-responsive and sustained release is particularly beneficial for targeting the acidic tumor microenvironment, enhancing the potential for localized treatment of CRC.
The primary objective of nano-encapsulation was to enhance the therapeutic index of DCMF, and the cytotoxicity results support this outcome. The DCMF@MIL-53(Fe) nanocomposite exhibited a significant increase in potency, with a DCMF-equivalent IC50 of 0.40 ± 0.05 μg/mL in HCT-116 cells. This represents a ~53-fold enhancement over the free drug (21.51 ± 1.2 μg/mL) and, notably, a ~31-fold superiority over the standard chemotherapeutic agent, Sorafenib (12.53 ± 0.28 μg/mL). Furthermore, the strategy conferred a crucial shift in selectivity. While both free DCMF (SI = 0.87) and Sorafenib (SI = 0.47) displayed greater toxicity to normal cells, the nanocomposite demonstrated a clear selectivity for cancer cells (SI = 1.62). While this value represents a modest improvement in absolute terms, it signifies a critical transition from a non-selective (SI < 1) to a selective (SI > 1) profile, which is a primary goal in nano-formulating phytochemicals to overcome their inherent lack of specificity. The achieved selectivity is consistent with reports for other nano-formulated plant extracts in CRC models (e.g., SI of 2.17 for a chitosan-based system) [51]. Coupled with the major gain in potency, this shift successfully rebalances the therapeutic window of the crude extract, enhancing its translational potential.
This significant enhancement is attributed to a combination of factors. The nanoscale size and positive surface charge promote cellular internalization, as evidenced by the increased cytotoxicity, while potentially also contributing to passive tumor targeting (e.g., the EPR effect) in vivo [52,53]. Furthermore, the Fe-based MOF carrier may contribute to a synergistic chemodynamic activity by catalyzing Fenton-like reactions in the tumor microenvironment, generating reactive oxygen species that exacerbate cellular stress [54]. Although the final biological interactions are influenced by the serum protein corona, the data clearly indicate that the net result is a highly effective and more selective anti-CRC agent [52,53].
The mechanistic basis for the enhanced antiproliferative activity of DCMF@MIL-53(Fe) was elucidated by profiling cell-cycle dynamics and cell-death modalities in HCT-116 cells. Flow cytometry revealed a significant G0/G1 accumulation (80.89 ± 4.85% vs. 64.00 ± 5.12% in controls) with concomitant S-phase depletion (17.25 ± 1.03% vs. 36.32 ± 2.60%), indicating a robust disruption of G1/S checkpoint commitment. Mechanistically, this pattern is consistent with inhibition of the CDK4–Cyclin D complex, preventing retinoblastoma (Rb) phosphorylation and E2F-driven S-phase entry [55], coupled with stabilization and activation of p53, potentially by disrupting the MDM2-p53 axis, and induction of p21, which directly suppress CDK activity [56].
Given the p53-wild-type background of HCT-116 and our G1/S arrest, CDK4/Cyclin D3 (to probe direct checkpoint control) and MDM2–p53 (to probe p53 stabilization) were prioritized as the most biologically congruent targets for docking. While the biological effects arise from the integrated activity of the whole fraction, the predominance of tomentosin provides a rational molecular anchor for mechanistic interpretation without excluding supportive or modulatory contributions from co-occurring constituents. Accordingly, tomentosin (the most abundant bioactive compound in DCMF) adopted well-defined poses at both targets and returned competitive binding at CDK4/Cyclin D3 (–8.6 kcal/mol) versus its co-crystallized ligand (−8.5 kcal/mol) and a comparable score at MDM2–p53 (−8.5 vs. −8.5 kcal/mol). Re-docking of native ligands reproduced crystallographic poses (RMSD 0.62 Å for CDK4/Cyclin D3; 0.21 Å for MDM2–p53), supporting protocol reliability. These in silico data are consistent with direct interference at CDK4–Cyclin D and engagement of the MDM2–p53 interface, providing a structural rationale for the observed G1/S arrest.
Previous published work in hepatocellular models shows tomentosin ↑p53 (Ser15/20/46), ↑p21, ↑p27 with ↓CDK4/6 and ↓Cyclins D/E, alongside apoptosis markers, precisely the molecular constellation that enforces G1/S restriction [47]. While some studies report G2/M or Sub-G1 enrichment [34,47], differences in cell-line genetics (HCT-116 p53-WT) and exposure kinetics/formulation (our sustained, mildly cationic MIL-53(Fe) delivery, which augments ROS via Fenton-like activity) plausibly shift the dominant checkpoint toward G1/S in our system. Thus, nano-encapsulation appears to amplify P. crispa’s innate bioactivity to reactivate intrinsic tumor-suppressive pathways in CRC cells.
Complementary to the observed cell cycle arrest, the mode of cell death induced by DCMF@MIL-53(Fe) was investigated using Annexin V/PI flow cytometry. The analysis revealed a near-absence of apoptotic populations, with early and late apoptosis accounting for only 0.96% ± 0.05% and 0.23% ± 0.01% of cells, respectively. In contrast, a marked increase in the Annexin V−/PI+ population was detected (12.85% ± 0.9% versus 2.71% ± 0.2% in control cells; p < 0.001). This staining pattern reflects a primary loss of plasma membrane integrity without phosphatidylserine externalization and confirms the predominance of necrotic, non-apoptotic cell death under these conditions.
While Annexin V/PI staining cannot distinguish regulated from accidental necrosis, the context strongly suggests a regulated mechanism. The treatment employs an Fe-based MIL-53(Fe) framework, a material documented to catalyze Fenton-like reactions and generate reactive oxygen species (ROS) [54]. Excessive ROS is a recognized upstream trigger for regulated necrotic pathways, including necroptosis, by facilitating the activation of the RIPK1–RIPK3–MLKL axis [57].
To explore potential direct interactions, in silico docking of tomentosin—the major bioactive constituent—was performed against RIPK1 and RIPK3–MLKL. Tomentosin adopted defined poses with docking scores of −8.0 kcal/mol (RIPK1) and −6.5 kcal/mol (RIPK3–MLKL), compared with −9.0 kcal/mol and −8.0 kcal/mol for the respective co-crystallized ligands. Protocol validation via native ligand re-docking yielded low RMSD values (0.53 Å and 0.71 Å, respectively). However, the modest binding affinities, particularly for RIPK3–MLKL, suggest direct protein modulation is unlikely the primary driver.
Therefore, the dominant necrotic outcome is best attributed to profound ROS generation from both the tomentosin-rich fraction [58] and, more significantly, the Fe-based MIL-53(Fe) carrier [54]. This oxidative burst is hypothesized to be the principal activator of the necroptotic cascade, potentially bypassing the need for strong ligand-receptor binding. Concurrently, a highly oxidizing microenvironment can destabilize key apoptotic regulators (e.g., the MDM2–p53 axis), further shifting the death modality from apoptosis to necrosis [56]. Finally, the data confirm DCMF@MIL-53(Fe) induces necrotic cell death. Based on the established properties of the components, a ROS-driven regulated pathway such as necroptosis is proposed as a plausible, contributing mechanism. Definitive validation requires future analysis of pathway-specific effectors, such as phosphorylated MLKL (p-MLKL) oligomers [59]. This mechanistic shift is therapeutically promising, as it may circumvent common apoptotic resistance in cancer cells [54,60].
This study provides foundational in vitro and in silico insights, though several aspects require further development. The findings, primarily established in p53-wild-type HCT-116 cells, warrant validation across genetically diverse CRC models. Mechanistically, while docking supports CDK4/Cyclin D and MDM2–p53 engagement, protein-level confirmation of this axis (e.g., p53, p21, phospho-Rb) is required to substantiate the G1/S arrest, just as pathway-specific inhibition is needed to definitively establish necroptosis over alternative ROS-driven pathways. Furthermore, the tomentosin-rich fraction—consistent with ethnopharmacological use—requires robust standardization for batch-to-batch reproducibility. Future work will therefore focus on this experimental validation and advancing the lead formulation into comprehensive in vivo pharmacokinetic and efficacy studies.

5. Conclusions
This study demonstrates that nanoformulation of a bioactive Pulicaria crispa dichloromethane fraction within an MIL-53(Fe) metal–organic framework yields a sustained-release system with enhanced in vitro activity against colorectal cancer cells. The resulting DCMF@MIL-53(Fe) nanocomposite showed good physicochemical stability and controlled release, quantified using a background-corrected UV–Vis method. These properties were associated with a 53-fold increase in cytotoxic potency, calculated on a DCMF-equivalent basis, and a modest but measurable improvement in selectivity toward HCT-116 cells relative to the non-formulated fraction.
Mechanistically, the enhanced anticancer activity was accompanied by pronounced G0/G1 cell-cycle arrest and a shift toward necrotic, non-apoptotic cell death. Docking of the predominant constituent, tomentosin, to CDK4/Cyclin D3 and MDM2–p53 complexes supports interference with G1/S checkpoint regulation. Given the strong ROS-generating capacity of both the Fe-based carrier and the tomentosin-rich fraction, involvement of a ROS-driven regulated necrotic pathway, such as necroptosis, is proposed as a plausible contributing mechanism, pending direct biochemical confirmation.
Although further molecular validation is required, these findings provide clear proof-of-concept that rational nano-delivery can mitigate key pharmaceutical limitations of plant-derived fractions, including uncontrolled release and limited selectivity. By integrating ethnopharmacological knowledge with formulation-guided design, this work offers a reproducible strategy to enhance the anticancer potential of traditional medicinal resources.