Supramolecular Self-assembly of Tanshinone IIA to Construct an Antitumor Drug Delivery Platform

Abstract

Salvia miltiorrhiza has been reported to exhibit significant antitumor effects, primarily due to its active ingredient, Tanshinone IIA (TSA). However, its low solubility and bioavailability pose challenges for its application. This study employed the antisolvent precipitation method to explore the self-assembly potential of TSA molecules for developing possible soluble nanomedicine. It was found that the morphology of self-assembled TSA aggregates was nanorods, and finally, numerous TSA nanorods (TSA NRs) were synthesized. Our findings indicate that TSA can self-assemble into nanorods through hydrophobic interactions and π-π stacking. Additionally, a polydopamine (PDA) coating, along with polyethylene glycol (PEG) and folic acid (FA) modifications, was applied to the TSA NRs to achieve high drug loading capacity, high stability, and a targeted delivery system. The resulting nanomedicine, PEG–PDA@TSA NRs, exhibits excellent dispersion and stability, remaining stable under physiological conditions. The antitumor efficacy of this system was evaluated using an H22 tumor mouse model, and the results demonstrated that FA-modified PEG–PDA@TSA NRs can significantly inhibit tumor growth in vivo. This study successfully developed a FA–PEG-PDA@TSA NR nanodrug delivery system, effectively enhancing the pharmacological activity of TSA and providing a novel formulation strategy for cancer treatment.

Introduction

As one of the diseases with the highest mortality rates, cancer has brought challenges to global medicine. Hepatocellular carcinoma (HCC), the most common primary liver cancer, remains a leading cause of cancer-related deaths globally. In the US, it is the ninth leading cause of cancer mortality. Despite advancements in prevention, screening, diagnosis, and treatment, HCC incidence and mortality rates continue to rise. Addressing HCC is critical, as it not only impacts patient survival but also places a significant burden on healthcare systems. Developing effective strategies to combat HCC is therefore an urgent priority for improving public health and societal well-being. It has been reported that approximately 60% of anticancer agents have emerged from nature bioactivity including microorganisms and plants. Traditional Chinese medicine (TCM), among these sources, is used in various cancer treatments because of its low side effects and favorable pharmacological profile.

Salvia miltiorrhiza Bunge, also known as Danshen (DS) in TCM, refers to the dried roots of the plant and exhibits various pharmacological effects. Diterpenoids are one group of the major bioactive components in DS, which possesses a variety of pharmacological activities, such as antibacterial, antioxidative, anti-inflammatory, and antineoplastic. Among all the diterpenoids, TSA is the major constituent of the processed herb, and many studies have mainly focused on its pharmacological activity. − Chen discovered TSA enhances cytotoxicity and pro-apoptotic effects against HepG2 cells, mediated by the upregulation of the intracellular ROS level, the increased cell cycle arrest at S phase, enhanced necrocytosis, and upregulation of caspase 3/7 and P38 protein expression. Ren et al. illustrated in their research that TSA can induce apoptosis in HCC through the miR30b-p53-PTPN11/SHP2 signaling pathway, suggesting its potential for clinical treatment and management of HCC. In vivo experiment confirmed that TSA can effectively inhibit tumor growth, no matter whether it is injected intravenously, subcutaneously, or intraperitoneally. However, TSA’s limited solubility and bioavailability pose significant challenges to its practical application.

Encouragingly, nanotechnology has been reported to be an effective solution to overcome the barriers to anticancer drug delivery. These nanoparticles are formed by the active constituents of TCM, which aggregate or self-assemble due to interactions that enhance the solubility of the drug. − Sheng et al. discovered a novel quercetin nanorod formulation prepared by the fluid bed coating crystallization technique. Due to the small particle size and large surface area, quercetin has a higher dissolution rate and higher antioxidant activity than the original quercetin. Fan et al. reported a simple and green approach to design a carrier-free, pure nanodrug (UA NP) by the self-assembly of UA molecules nanosized by pure UA self-assembly, which was without the help of a vehicle and water-soluble drugs. These nanoparticles exhibit enhanced solubility and stability relative to those of free UA.

Although carrier-free self-assembled nanomedicines offer a promising approach to enhancing the solubility of poorly soluble compounds, their limitations in targeting capability and functional modification pose significant challenges for overcoming physiological barriers. This results in low bioavailability and restricts their potential applications. The advancement of novel drug delivery systems presents a viable solution to these issues. A nanocarrier drug delivery system, effectively assisting to improve the solubility, bioavailability, and targeting of drugs in the human body, is a crucial means for treating cancer and other diseases. Fuster et al. constructed RA-loaded silk fibroin nanoparticles (RA-SFNs) to ameliorate the bioavailability of RA. Tang et al. lactoferrin (Lf) comodified NPs with borneol modification (Lf-BNPs) increased the cellular uptake of 16HBE cells in vitro and not only enhanced the penetration and transport of dopamine through the nasal mucosa, but also enhanced the drug delivery by opening the tight junctions of the BBB. For TSA, molecule is planar and belongs to the “I” linear structure, which is conducive to the formation of hydrophobic interactions between molecular frameworks. Meanwhile, the formation of π–π stacking also facilitates the self-assembly of molecules. However, as a lipid-soluble component, TSA suffers from poor water solubility and undergoes rapid metabolism in vivo, making it difficult to achieve effective concentrations at the tumor site; , the lack of targeted drug delivery systems presents significant challenges that limit their efficacy. This situation necessitates ongoing modifications of natural self-assembled nanomedicines to enhance both their circulation duration and targeting capabilities.

In this study, we focused on preparing and optimizing TSA nanodrugs through the antisolvent precipitation technique first. Subsequently, we investigated the feasibility of TSA self-assembly, specifically discovering that TSA can coassemble to form nanorod TSA NRs, driven by π-π stacking and the hydrophobic effect. Next, we coated the TSA NRs with PDA to reduce the level of nanorod agglomeration and achieve a more uniform size. We then functionalized the TSA NRs with PEG-NH₂ to enhance stability and modified them with FA to improve drug accumulation at the tumor site. Ultimately, we prepared FA–PEG-PDA@TSA NRs. Physicochemical characterization revealed that FA–PEG-PDA@TSA NRs exhibit enhanced solubility and stability compared with TSA alone. Additionally, we evaluated the antitumor potential of FA–PEG-PDA@NRs. In vitro experiments demonstrate that FA–PEG-PDA@TSA NRs possess excellent biosafety and exhibit no toxicity to tissues. In an in vivo H22 tumor mouse model, FA–PEG-PDA@TSA NRs displayed a significant inhibitory effect on tumor growth. In summary, our work proposes a self-assembly nanotechnology based on TCM, modified by the incorporation of a variety of materials. This approach highlights the advantages in physicochemical properties and pharmacological activity, thereby broadening the scope of development of nanotechnology.

Materials and Methods

Materials

Tanshinone IIA (>98%) was purchased from Shanghai Yuanye Biotechnology Co. THF (99.5%) was purchased from Anhui Zaisheng Technology Co. Dopamine was purchased from Sigma-Aldrich, USA. mPEG-NH₂ was purchased from Shanghai McLean Biochemical Technology Co.

SPF-grade KM mice aged 6–8 weeks, with a body weight of 18–22 g, were purchased from Jiangsu Hua Chuang Xin Nuo Biology Science and Technology Co., Ltd., with license SCXK (Su) 2020-0009. All experiments were conducted in accordance with the guidelines for animal research of the Ethics Committee of China and received approval from the Animal Ethics Committee of Nanjing University of Chinese Medicine (Approval ID: 202404A066)

Preparation of Tanshinone IIA Nanorods

The TSA NRs were prepared using the antisolvent precipitation method. First, a TSA stock solution was prepared by dissolving 0.1 mg of TSA in 0.1 mL of THF, resulting in a concentration of 1 mg/mL. This TSA solution was then injected into 80 mL of water under sonication. Sonication was continued for 5 min after complete injection to produce carrier-free TSA NRs.

To prepare stable and homogeneous TSA NRs, we first investigated the effects of various conditions on their formation: the types of organic phases (THF, DMSO, and acetone), the ratios of the organic phase to the aqueous phase (1:10, 1:25, 1:50, and 1:100), and the initial concentration of TSA (10, 20, and 30 μg/mL) on the TSA NRs. We optimized the latest preparation process using particle size, polydispersity index (PDI), and zeta potential as indicators.

Characterization of Tanshinone IIA Nanorods

The morphology of TSA NRs was examined by transmission electron microscopy (TEM, Hitachi HT-7800, Japan). Particle size distribution and PDI were determined by using dynamic light scattering (DLS, Brookhaven BI9000AT system, Brookhaven Instruments Corporation). Spectral properties were analyzed with UV–visible spectroscopy (Shimadzu UV-2700, Japan), fluorescence spectroscopy (Hitachi F-7000, Japan), and Fourier transform infrared spectroscopy (FT-IR, Shimadzu IR Tracer 100, Japan). Crystallographic analysis of raw TSA and lyophilized TSA NRs was performed by X-ray diffraction (XRD, Bruker D8 QUEST, Germany) with Cu Kα radiation, scanning from 5° to 55° (2θ) at 10°/min. Thermal behavior was evaluated using differential scanning calorimetry (DSC, Shimadzu DSC-60, Japan) under nitrogen purge (10 mL/min), with temperature ramps from 25 to 500 °C at 10 °C/min using sealed aluminum pans.

HPLC Quantification of Tanshinone IIA Content

TSA analysis was performed using a Shimadzu HPLC system (Shimadzu Corporation, Kyoto, Japan) equipped with a Hydrosphere C18 analytical column (250 × 4.6 mm, 5 μm) maintained at 30 °C. An isocratic mobile phase of acetonitrile/0.1% phosphoric acid aqueous solution (76:24, v/v) was delivered at 1.0 mL/min, with 10 μL injections and detection at 270 nm. For quantification, the TSA reference standard was precisely weighed and dissolved in methanol by sonication to prepare a 1 mg/mL stock solution, which was serially diluted with methanol to standard concentrations of 5, 20, 50, 80, 120, 160, and 180 μg/mL. After filtration through 0.22 μm membranes, a linear calibration curve (5–180 μg/mL) was established by plotting peak area (A) versus concentration (C, μg/mL), yielding the regression equation: A = 74794C + 74255 (R ² = 0.9996, Figure S1), confirming excellent method linearity.

Determination of Saturated Solubility

Saturation solubility testing was performed by placing the TSA or TSA NRs in centrifuge tubes. The tubes were weighed and then shaken at 150 rpm for 48 h in a constant-temperature water-bath shaker at 37 ± 0.5 °C. After the system attained equilibrium, the samples were centrifuged, and the supernatant was analyzed by HPLC under predefined chromatographic conditions.

Examination of the Assembly Mechanism of Tanshinone IIA Nanorods

AutodockTools v1.5.7 was used to perform semiflexible molecular docking simulations of a possible TSA with the Lamarckian genetic algorithm and calculated binding energy. In addition, the NRs solution was supplemented with various blocking agents with different strengths (0.5% sodium dodecyl sulfate (SDS), 100 mM NaCl, 100 mM urea, 100 mM NaOH, 100 mM HCl, DMSO, THF, or DMF), and the absorbance was then measured at 270 nm and UV spectral data was recorded.

Preparation of PDA@TSA NRs

DA (0.1 mg/mL) was added to the TSA NRs solution, suspension, and polymerized at pH 8.5. The resulting precipitate was collected by centrifugation at 13,000 rpm for 15 min. In order to prepare TSA NRs with high effectiveness, we first examined the effects of different condition forces on TSA NRs: the concentration of DA (50, 100, and 200 mg/mL) on the PDA@TSA NRs and the coating time (2, 3, and 4 h) of DA on the TSA NRs and screened the latest preparation process using drug loading content (DLC) and entrapment efficiency (EE) as indicators.

Determination of Encapsulation Efficiency and Drug Loading Capacity

PDA@TSA NRs were dispersed into methanol, followed by vortexing and sonication to ensure complete drug release. The TSA content was quantified by HPLC under the chromatographic conditions described previously.

Subsequently, the DLC and EE were calculated by using the appropriate formulas:

$$\mathrm{EE}\%=\frac{\mathrm{Amount}\mathrm{of}\mathrm{drug}\mathrm{in}\mathrm{nanoparticles}}{\mathrm{Total}\mathrm{amount}\mathrm{of}\mathrm{deeding}\mathrm{drug}}\times 100\%$$

$$\mathrm{DLC}\%=\frac{\mathrm{Amount}\mathrm{of}\mathrm{drug}\mathrm{in}\mathrm{nanoparticles}}{\mathrm{Total}\mathrm{amount}\mathrm{of}\mathrm{drug}\mathrm{loaded}\mathrm{micelles}}\times 100\%$$

Preparation of PEG–PDA@TSA NRs

The PDA@TSA NR aqueous dispersion was mixed with mPEG-NH₂ (2 mg/mL). PEG–PDA@TSA NRs were subsequently synthesized by adjusting the solution to pH 9 using a Tris buffer solution, followed by overnight stirring and centrifugal collection of the precipitate.

Stability Evaluation

PDA@TSA NRs and PEG–PDA@TSA NRs were submerged underwater for storage, with size and PDI being measured over 7 consecutive days to assess the nanoparticles’ stability. Additionally, the PEG–PDA@TSA NRs were dispersed in either PBS or DMEM, and the stability in different media was evaluated by measuring size within 48 h.

Preparation of FA–PEG-PDA@TSA

The PDA@TSA NRs were dispersed in water, followed by the addition of FA–PEG-NH₂ (1 mg/mL). FA–PEG-PDA@TSA NRs were then obtained by adjusting the solution to pH 9 using a Tris buffer solution, stirring overnight, and finally collecting the precipitate through centrifugation.

In Vitro Release

PDA@TSA NRs and FA–PEG-PDA@TSA NRs were encapsulated in a dialysis bag, which was immersed in a centrifuge tube filled with PBS (with 1% (w/v) SDS for solubilization) and shaken horizontally at 37 °C and 100 r/min. Samples were collected at various time points (1, 2, 3, 4, 5, 6, 8, 10, 12, 24, 36, and 48 h) and supplemented with an equal amount of PBS containing 1% SDS. The amount of released TSA was quantified using HPLC. Cumulative drug release (Q%) was calculated using the established formula:

$$Q\%=\frac{V_0{C}_i+V_t\sum C_{i-1}}{M}\times 100\%$$

where V ₀ is total release medium volume (mL), C_i is TSA concentration at time i (μg/mL), V_t is sampling volume (mL), and m is total drug mass (μg).

Biosafety Evaluation

In order to verify the safety of the material, a hemolysis test was performed first, and blood was extracted from the rat abdominal aorta and treated to form a red blood cell suspension. The solution of each group (negative control group (plus PBS), positive control group (plus distilled water), and drug treatment group (concentration 80, 160, 200, 320, and 400 mg/mL FA–PEG-PDA@TSA NRs)) was added to red blood cell suspension, gently shaken, and incubated at 37 °C for 3 h; the supernatant was centrifuged, and the color of the supernatant and its absorbance was determined at 540 nm:

$$\mathrm{Hemolysis}\%=\frac{A_{\mathrm{sample}}-A_{\mathrm{neg}}}{A_{\mathrm{pos}}-A_{\mathrm{neg}}}\times 100\%$$

To assess the potential toxicity of nano preparations in vivo, healthy mice were injected with saline and FA–PEG-PDA@TSA NRs (TSA, 5 mg/kg) via the tail vein every 2 days. After 7 days, the mice were euthanized. The major organs, namely, the heart, liver, spleen, lungs, kidneys, and brain, were excised and fixed in 4% paraformaldehyde. The tissues were then embedded in resin, sectioned, and subjected to H&E staining for observation.

In vivo biodistribution

First, we utilized H22 cells to establish a tumor-bearing mouse model through the injection of 100 μL of H22 cell suspension containing 10⁷ cells. Ten tumor-bearing mice were randomly assigned to two groups (n = 5). Group 1 received an intravenous injection of PEG–PDA@TSA NRs via the tail vein, while group 2 received FA–PEG-PDA@TSA NRs. Both groups were administered a TSA-equivalent dose of 2.5 mg/kg. At 4 h postinjection, the mice were euthanized by cervical dislocation. The heart, liver, spleen, lungs, kidneys, and tumor tissues were immediately harvested. The collected tissues were perfused with normal saline to remove residual blood, blotted dry on filter paper, weighed, and stored at −80 °C for analysis.

A TSA stock solution (1 mg/mL) was prepared by dissolving 1 mg of TSA reference standard in methanol via sonication. Similarly, an internal standard (IS) solution (1 mg/mL cryptotanshinone, CT) was prepared in methanol. Both solutions were stored light-protected at 4 °C. For analysis, 100 μL of tissue homogenate (0.5 g of tissue/1 mL of saline) was mixed with 10 μL of IS solution. Proteins were precipitated with 300 μL of methanol. After vortexing and centrifugation, the supernatant was lyophilized. The residue was reconstituted in 100 μL of methanol–water (9:1, v/v).

Regarding the LC-MS/MS (6500⁺, American AB) quantification conditions: chromatographic separation used an XBridge BEH C18 column (1.7 μm, 2.1 × 100 mm; 40 °C) with a mobile phase of 0.1% formic acid in water (A) and acetonitrile (B). The gradient program was as follows: 5% → 95% B (0–5.0 min), 95% B (5.0–6.0 min), 95% → 5% B (6.0–7.0 min), and 5% B (7.0–8.0 min) at 0.30 mL/min. Injection volume was 3.00 μL (autosampler: 8 °C). Mass detection employed positive ESI-MRM with ion spray voltage 5500 V, source temperature 550 °C, and nebulizer gas 3.0 L/min. Analyte-specific transitions and collision energies were monitored as follows: m/z 295 → 277 for TSA with a collision energy of −26.62 V and m/z 296 → 254 for cryptotanshinone (CT) with a collision energy of −35 V.

TSA concentration was determined from the calibration curves:

$$Y=4.4976X+0.6813$$

Y represents peak area ratio, and X denotes concentration in ng/mL, R ² = 0.9997.

Tissue concentration was computed by the following equation:

Tissue concentration (ng/g) = Homogenate concentration (ng/mL) × 2 (mL/g)

The factor 2 represents the dilution factor due to homogenization of 0.5 g tissue in 1 mL of saline.

Antitumor Efficacy Evaluation

To investigate the antitumor effects of FA–PEG-PDA@TSA NRs, this study utilized H22 cells to establish a tumor-bearing mouse model through the injection of 100 μL of H22 cell suspension containing 10⁷ cells. Once the tumors reached a size of 100 mm³, KM mice were randomly assigned to four groups. They were then treated with intravenous injections of PBS, TSA, PEG–PDA@TSA NRs, and FA–PEG-PDA@TSA NRs (TSA, 12 mg/kg) every other day for 15 days. Tumor size, animal survival, and body weight were monitored every day. After consecutive administration seven times every other day, the mice were put to execution at 24 h after the last administration to collect tumor. Tumor volume (V) was calculated using the formula: V (mm³) = (d ² × D)/2, where d is shortest diameter (mm) and D is longest diameter (mm). −

Tumor Tissue H&E Staining

After treatment, the mice were killed and the tumor tissue was removed and preserved in 4% paraformaldehyde at room temperature for a duration of 48 h. Subsequently, the tumor tissues underwent dehydration and were embedded in paraffin and frozen sections were prepared, followed by H&E staining for observation.

K _i-67 Immunohistochemistry

The tumor tissue samples were fixed, paraffin embedded, sliced, and then made into thin slices. After K _i-67 immunohistochemical staining, observation and counting were performed to calculate the percentage of positive cells.

Data Processing and Statistical Analysis

Data analysis was conducted using one-way analysis of variance (ANOVA) and nonparametric Tukey’s test. A P value less than 0.05 was considered indicative of a significant difference. Image processing was carried out using GraphPad Prism 9.5 and ImageJ software.

Result

To investigate the potential of TSA, the active ingredient of Salvia miltiorrhiza, for self-assembly into nanomedicines, we initially prepared and optimized TSA nanomedicines using antisolvent precipitation technology. TSA, being a hydrophobic drug with limited solubility in water, was dissolved in an organic solvent and subsequently injected slowly into an excess volume of water. We assessed the particle size and PDI of the self-assembled nanomedicines across various solvents (DMSO, THF, and acetone), solvent volume ratios, and drug concentrations. The results presented in Table S1 (in the Supporting Information) indicate that utilizing THF as a solvent yields TSA nanorods exhibiting favorable hydrodynamic diameter and low PDI, suggesting that THF is effective for dissolving TSA. This study further examined the impact of varying the ratio of good to poor solvents on the size of TSA nanorods. The findings revealed that increasing the proportion of the good solvent resulted in larger nanomedicine sizes, while a ratio of 1:100 led to smaller particle sizes. Additionally, it was observed that the concentration of TSA also influences the size of the nanomedicines; higher concentrations correlate with larger sizes and decreased stability. To achieve consistent morphology and optimal particle size, the recommended approach is to utilize THF as the solvent, maintain a 1:100 solvent ratio, and employ a TSA concentration of 10 μg/mL.

TSA can self-assemble to form nanomedicine using the antisolvent precipitation method (Figure A). Upon irradiation with a laser pen under light-shielding conditions, TSA NRs exhibited a significant Tyndall effect compared to the TSA aqueous solution, with a clear visible light path (Figure B). The TEM image in Figure C confirms this observation, indicating the interaction among TSA molecules and the formation of nanoscale nanorods; however, the nanorods displayed irregular sizes and poor dispersion. The average particle size is measured by DLS exhibiting a bimodal size distribution with peaks at 65.76 and 249.59 nm and a PDI of 0.233 (Figure D). At the same time, quantitative size analysis of TEM images using Nano Measurer 1.2 software revealed nanorods with mean dimensions of 283.17 ± 10.55 nm (length) × 57.46 ± 15.12 nm (width), resulting in an average aspect ratio of 5.07.

1.

Preparation and properties of the TSA NRs. (A) Chemical structure and self-assembly of TSA; (B) Tyndall effect of TSA and TSA NRs; (C) TEM images of TSA NRs; and (D) particle size and PDI of TSA NRs.

To further investigate the formation mechanism of the nanorods, we conducted spectral analysis. The results from UV spectral analysis indicate that the absorption peak of the assembled nanorods exhibits a blue shift, specifically a 15 nm shift toward shorter wavelengths. As illustrated in Figure A, the nanorods also display a significant hypochromic effect compared to unassembled TSA molecules, likely due to interactions among the TSA molecules. Furthermore, the fluorescence emission intensity of the nanorods is markedly lower than that of free TSA, as shown in Figure B, which further supports the existence of energy migration within the stacked TSA molecules. Conversely, the results from infrared spectrum analysis revealed only slight changes, suggesting that the structure of the TSA molecules remains stable, as depicted in Figure C. We hypothesize that the formation of TSA NRs primarily depends on the spatial arrangement of the molecules. This finding provides a crucial foundation for a more in-depth exploration of the assembly mechanism. XRD analysis (Figure D) revealed sharp diffraction peaks for raw TSA at 2θ = 7.13°, 9.53°, 10.1°, 11.95°, 14.4°, 17.8°, and 19.24°, confirming its highly crystalline nature. In contrast, TSA nanorods (NRs) exhibited broadened and attenuated peaks with burr-like profiles, indicating reduced crystallinity and possible amorphous or small-crystalline domain formation. To quantitatively compare crystallinity differences, the average crystallite sizes of TSA and TSA NRs were systematically calculated using the Debye–Scherrer equation with detailed parameters provided in Supplementary Table S2 (in the Supporting Information). Although the difference in XRD peaks of TSA and TSA NRs is not prominent, the discernible reduction in crystallite size and increased full width at half maximum (FWHM) provide conclusive evidence of modified crystallinity. In Figure E, the DSC curve of the TSA original drug exhibits a distinct endothermic peak at approximately 208.1 °C. In contrast, the intensity of the endothermic peak for TSA NRs is diminished and shifted to a lower temperature of 192.8 °C. This observation suggests that TSA molecules self-assemble to form TSA NRs, leading to a reduction in the crystallinity of TSA within the NRs. Collectively, these results confirm the successful preparation of the TSA NRs.

2.

Spectra characterization of the TSA NRs. (A) UV spectra, (B) fluorescence spectra, (C) FT-IR, (D) XRD, (E) DSC, and (F) saturated solubility of TSA and TSA NRs.

Finally, we conducted a comparative analysis of the saturated solubility of TSA before and after self-assembly. The initial measurement of TSA’s solubility in water was recorded at 0.46 μg/mL; however, following its conversion to carrier-free nanorods, the saturated solubility increased significantly to 40.7 μg/mL, representing an 88-fold enhancement over the original solubility of TSA (Figure F). This study demonstrates that the self-assembly process of TCM can significantly improve the solubility of its active ingredients, which is essential for advancing the development and utilization of low-solubility TCM components.

In this study, we aimed to conduct a comprehensive analysis of the molecular mechanisms underlying the self-assembly behavior of TSA molecules. Utilizing molecular docking technology, we accurately determined the spatial arrangement patterns of these molecules. The results revealed that TSA molecules exhibit a planar configuration and belong to the I type linear structure; this characteristic facilitates effective hydrophobic interactions between molecules, thereby promoting the self-assembly of the molecular framework. The molecular docking results obtained from Autodock Tools 1.5.6 software indicated a favorable binding activity of TSA molecules, with a binding energy of −3.7 kcal/mol. This finding suggests a significant binding affinity between TSA and its molecular targets, consistent with previous studies that indicate a good affinity when binding energy is <−1.2 kcal/mol. Furthermore, the coplanar and orderly arrangement of the molecules highlights the critical role of π-π conjugation forces in the self-assembly process of TSA NRs (Figure A). These forces not only facilitate the self-assembly of TSA molecules but also promote their aggregated growth in coplanar or unidirectional configurations, , ultimately resulting in the formation of rod-like nanostructures.

3.

Assembly mechanism investigation of TSA NRs. (A) Molecular docking of TSA molecules; (B) appearance and (C) maximum absorption intensity of TSA NRs solutions with addition of different force blockers; (D) UV spectra of TSA NR solutions in SDS; (E) maximum absorption intensity (n = 3); and (F–H) UV spectra of TSA NR solutions with different proportions of organic solvents.

According to the literature, SDS, NaCl, and urea act as blockers for hydrophobic, electrostatic, and hydrogen bonding interactions, respectively, while THF, DMSO, and DMF serve as inhibitors of π-π stacking. This study systematically investigates the behavior of nanorods in various chemical environments. Experimental observations indicate that the appearance of the nanorods differs significantly across different solutions. Specifically, when TSA NRs are exposed to SDS and solvents such as THF, DMSO, and DMF, the solution’s color deepens, suggesting structural destruction of the nanorods. Moreover, as the concentrations of THF, DMSO, and DMF increase, the color of the solution becomes increasingly intense. In contrast, in other solutions, the nanorod solution exhibited a light orange hue and remained highly transparent (Figure B). These observations provide an experimental basis for understanding the stability and self-assembly behavior of nanorods in various chemical environments.

By observing the subtractive color effect produced by TSA molecules upon the formation of nanorods, we gained insights into the dynamics of their assembly and disassembly processes through quantitative analysis of the absorbance values of these nanorods. As illustrated in Figure C, the introduction of SDS into the TSA NRs solution led to a significant increase in absorbance. Specifically, the addition of 0.5% (w/v) SDS resulted in an enhancement of the ultraviolet–visible (UV–vis) absorption spectrum of TSA nanorods, as shown in Figure D. This enhancement is attributed to the disassembly process of TSA NRs, which exposes TSA chromophores and consequently increases UV absorption. The changes in the nanorod size distribution before and after SDS treatment further support this conclusion (Figure S2). These findings indicate that hydrophobic interactions play a crucial role in the formation of nanorods.

In contrast, the absorbance of TSA NRs did not change significantly when exposed to NaCl and urea solutions, indicating that electrostatic and hydrogen bonding interactions did not affect the stability of the nanorods under these conditions. Additionally, the presence of solvents such as THF, DMSO, and DMF significantly enhanced the UV absorption capacity of TSA NRs, underscoring the importance of π-π conjugation in the self-assembly process of the nanorods (Figure C). Furthermore, we examined the effects of varying proportions of THF, DMSO, and DMF on the UV absorbance of the TSA NRs (Figure E). The results demonstrated a progressive increase in UV absorbance with rising organic solvent ratio, particularly evident beyond 50% concentration. Specifically, as shown in Figure F–H, the maximum absorption peak of the UV absorption spectrum at a concentration of 100% was enhanced compared to that at 50%. This enhancement may be attributed to the π-π stacking inhibitors, which prevents the formation of an effective π-π stacking structure between the nanorods, resulting in a reduction in the light absorption cross-section of the nanorods.

This change in absorbance is believed to be associated with the synergistic effects of hydrophobicity and π-π stacking interactions. At an organic solvent ratio of 25%, although the π-π stacking forces decreased, this did not result in the disintegration of the nanorods; DLS characterization (Figure S3) confirms that exposure to 25% DMSO, THF, or DMF did not induce significant disintegration of the nanorods or notable alterations in their size distribution, indicating that hydrophobic interactions play a predominant role in the drug self-assembly process. However, when the organic solvent ratio increases to 50 and 100%, the nanorods disintegrate due to the complete disruption of π-π stacking interactions. In summary, the self-assembly mechanism of TSA primarily involves hydrophobic interactions and π-π stacking, with hydrophobic interactions being more significant and crucial for the formation and stability of the nanorods.

Dopamine can self-polymerize at an alkaline pH, resulting in the formation of PDA that can be deposited almost spontaneously on various organic or inorganic materials. This process leads to the creation of a surface-adhered PDA film that is stable and biocompatible within biological systems. , Compared to other surface-functionalized materials, PDA presents distinct advantages and significant effects in the modification of nanoparticles, demonstrating substantial application potential. To enhance the dispersibility of TSA NRs and prevent drug adherence to their surfaces, a PDA coating was applied to the surface of the nanorods.

After PDA coating, TEM revealed that the nanorods exhibit a core–shell structure, with TSA nanorods functioning as the core and the PDA coating serving as the shell. In comparison to TSA, the PDA@TSA nanorod structure is more uniform and the surface coating is smoother and more complete (Figure A,B). To optimize the coating conditions for PDA, a single-factor analysis was performed. Table S3 (in the Supporting Information) demonstrates that the concentration of dopamine (DA) does not significantly influence the particle size of PDA@TSA NRs; however, it has a notable impact on the drug encapsulation efficiency and drug loading capacity. At a DA concentration of 0.2 mg/mL, TSA exhibited the highest encapsulation efficiency. However, increasing the DA concentration could result in a thicker coating, potentially hindering the release of TSA in subsequent experiments; therefore, the 0.2 mg/mL concentration was excluded from further consideration. In contrast, a DA concentration of 0.05 mg/mL yielded the lowest encapsulation efficiency and yield for the PDA@TSA NRs formulation, prompting its exclusion. After a thorough evaluation, the final DA concentration was established at 0.1 mg/mL (Figure C). The coating time had minimal impact on particle size; however, as the coating duration increased, drug loading gradually decreased. At a coating duration of 2 h, drug loading reached its maximum value of 23%; thus, 2 h was selected as the optimal coating time for subsequent experiments (Figure D).

4.

Characterization of polydopamine-coated Tanshinone IIA nanorods (PDA@TSA NRs). TEM images of TSA NRs (A) before and (B) after polydopamine coating; drug loading content (DLC) and encapsulation efficiency (EE) effect of PDA@TSA NRs with (C) different dopamine concentrations and (D) different coating time; (E) zeta potential and (F) size of TSA NRs (A) before and (B) after polydopamine coating.

After coating with PDA, the zeta potential of TSA NRs changed from −22.6 ± 1.4 to −30.2 ± 1.9 mV, indicating an improvement in the dispersion and stability of the final PDA@TSA NRs (Figure E). The bimodal hydrodynamic size distribution of TSA NRs migrated to larger dimensions with peaks at 76.6 and 266.65 nm, while the PDI decreased to 0.175, suggesting a more uniform distribution (Figure F).

As illustrated in Figure A, PDA@TSA NRs readily aggregate and precipitate in PBS and DMEM, exhibiting poor stability that results in their rapid clearance from the body’s circulatory system during in vivo applications. To address this issue, we reviewed relevant literature and identified that surface modification of PDA with PEG can enhance its stability. Accordingly, we selected mPEG-NH₂ as the binding agent for the PDA coating surface and employed a Schiff base ligation reaction. TEM results indicated that the morphology of the PEG–PDA@TSA NRs remained unchanged compared to PDA@TSA NRs (Figure A,C); however, their stability in PBS and DMEM improved, as evidenced by the absence of precipitates (Figure C). Subsequently, we evaluated the stability of the nanoparticles under simulated in vivo conditions, and the particle size and PDI of the nanorods were characterized by DLS. The results demonstrated that both PDA@TSA NRs and PEG–PDA@TSA NRs exhibited satisfactory storage stability, with size and PDI remaining relatively constant over a period of 7 days in the aqueous solution (Figure B,D). Moreover, PEG–PDA@TSA NRs maintained stability for 48 h in PBS and DMEM, showing relatively constant size and PDI (Figure E). These findings effectively underscore the beneficial role of PEGylation modification in enhancing the stability of PDA@TSA NRs.

5.

Characterization of surface-modified Tanshinone IIA nanorods (FA–PEG-PDA@TSA NRs). (A) TEM image, appearance in different solvents, and (B) storage stability of PDA@TSA NRs; (C) TEM image, appearance in different solvents, and (D) storage stability of PEG–PDA@TSA NRs; (E) storage stability of PEG–PDA@TSA NRs in different solvents; (F) particle size; (G) zeta potential; (H) infrared spectroscopy; (I) in vitro release.

For antitumor drugs, it is essential to enhance drug accumulation at the tumor site. Literature reports indicate that FA is utilized as a targeting moiety in tumor-targeted delivery systems due to its high affinity for folate receptors, which are overexpressed in various tumors. We modified FA onto the surface of PEG–PDA@TSA NRs to leverage the tumor-targeting properties of FA and enhance the therapeutic effect. The particle size and zeta potential of PDA@TSA NRs, PEG–PDA@TSA NRs, and FA–PEG-PDA@TSA NRs were evaluated by using DLS technology. The results indicated that following modification with mPEG-NH₂ and FA–PEG-NH₂ induced bimodal peak migration in NR hydrodynamic size, exhibiting peaks at 118.56 and 407.37 nm compared to PDA@TSA NRs (Figure F). Additionally, as illustrated in Figure G, the modification also resulted in alterations to the zeta potential. These findings confirm the successful modification of FA–PEG-NH₂. Figure H presents the FT-IR spectra of TSA, PDA@TSANRs, and FA–PEG-PDA@TSA NRs. In comparison to the infrared spectrum of the TSA group, no distinct absorption peak unique to TSA was observed at 1641 cm^–1. Conversely, a broad absorption peak associated with −OH was detected in the range 3100–3400 cm^–1, indicating the successful encapsulation of the PDA shell on the nanorods. In the infrared spectra of PEG–PDA@TSA NRs and FA–PEG-PDA@TSA NRs, a characteristic absorption peak at 1062 cm^–1 was noted, signifying the presence of C–O–C, which demonstrates the successful attachment of PEG. Furthermore, the spectrum of FA–PEG-PDA@TSA NRs exhibits a prominent CN absorption peak that is absent in the spectrum of PEG–PDA@TSA NRs, confirming the successful incorporation of FA.

Finally, we assessed the release rates of FA–PEG-PDA@TSA NRs. Compared with PDA@TSA NRs, PEG modification resulted in a slight decrease in the release rate. After 48 h, the cumulative release amounts for PDA@TSA NRs and FA–PEG-PDA@TSA NRs were 72.4 and 62.1%, respectively (Figure I). The curve of the TSA concentration over time is shown in Figure S4. This is because PEG has a long hydrophilic chain, which can form a hydration layer on the surface of nanoparticles to effectively slow down the release of drugs.

Following the completion of material preparation and modification, a red blood cell hemolysis experiment was conducted to verify the biological safety of FA–PEG-PDA@TSA NRs. The results, illustrated in Figure A, revealed that no hemolysis occurred when using PBS as a negative control. In contrast, distilled water, employed as a positive control, resulted in significant hemolysis, indicating its potential to damage cell membranes. Notably, the hemolysis rate of FA–PEG-PDA@TSA NRs at varying concentrations remained below 5%, with no statistically significant differences observed between the groups, suggesting that the FA–PEG-PDA@TSA NRs possess good blood safety.

6.

Safety assessment of the formulations. (A) Hemolytic assay of FA–PEG-PDA@TSA NRs; (B) H&E staining of different tissues (scale bar = 50 μm).

Subsequently, H&E staining was performed to conduct preliminary in vivo toxicity studies. This staining allowed for the examination of the morphology of animal tissues and organs following treatment with different nanopreparations, facilitating an evaluation of their biological toxicity. After administration, the primary organs of the mice were dissected and visually inspected. The findings indicated that there were no apparent lesions in the heart, liver, spleen, lungs, kidneys, brain, and other organs across all experimental groups. Furthermore, histopathological examination of the aforementioned organs under a microscope, as shown in Figure B, revealed that the tissue structure remained intact, with no significant lesions or necrosis, and the functions of the organs were well-preserved at the experimental dosage.

Following comprehensive safety validation of FA–PEG-PDA@NRs, we evaluated their in vivo biodistribution in tumor-bearing mice (Figure ). At 4 h postintravenous administration, the TSA was detected in the heart, liver, spleen, lungs, kidneys, and tumor tissues. Significant accumulation was observed in the liver, spleen, lungs, and kidneys, with the highest TSA concentration found in the liver. Following FA modification, TSA concentrations demonstrated a marked increase across all examined tissues. Notably, the TSA concentration detected in tumor tissue was approximately 1.6-fold higher than that observed in mice administered the non-FA-modified formulation. This enhancement is attributed to folate receptor-mediated active targeting, which promotes tumor-specific accumulation and provides a pharmacokinetic basis for the subsequent antitumor efficacy.

7.

In vivo biodistribution (n = 5, *p < 0.05, ***p < 0.001 compared with PEG–PDA@TSA NRs).

To evaluate the antitumor effect of the nanodelivery system in vivo, a tumor mouse model was established by administering H22 tumor cells to the right axilla. According to the dosing schedule depicted in Figure A, various formulations (PBS, free TSA, PEG–PDA@TSA NRs, and FA–PEG-PDA@TSA NRs) were injected through the tail vein every other day. The mice were monitored for changes in tumor volume and body weight. The results indicated that the weight of mice in each group increased following treatment (Figure B), suggesting that the nanodrug delivery system exhibits good safety and does not produce significant toxicity. Furthermore, during the treatment period, the tumor volume in the model group increased significantly; however, each drug group effectively inhibited tumor growth compared to the model group. Notably, at the same dose, the antitumor effect of the FA–PEG-PDA@TSA NRs group was superior to that of the free TSA and PEG–PDA@TSA NR groups (Figure C). This enhanced efficacy is attributed to surface modification by FA, which improves drug distribution and accumulation in tumor tissue. Following the treatment, the mice were euthanized and the tumor tissue was excised. The tumor tissue image is presented in Figure D, the group administered FA–PEG-PDA@TSA NRs exhibited the smallest tumor volume, and the tumor weight results were consistent with the tumor volume findings (Figure E). These results demonstrate that the nanodelivery system can effectively sustain the antitumor effect of TSA in vivo, with FA–PEG-PDA@TSA NRs displaying the most pronounced antitumor activity.

8.

In vivo antitumor effect. (A) Schematic illustration of the experimental process; change in (B) body weight, (C) tumor volume in each group of mice during the treatment period; (D) isolated tumor images, (E) tumor weight, (F) H&E and immunohistochemical staining of tumor section in each group after the end of the treatment. (G) Percentage of Ki67 positive cells (n = 6, *p < 0.05, **p < 0.01 compared with model group or indicated group; ###p < 0.001 vs TSA group).

Next, H&E staining was employed to further analyze the tumor tissue, as illustrated by the blue-purple nuclei indicated by red arrows in Figure F. The tumor cells in the model group exhibited vigorous growth, high density, and compact arrangement, accompanied by strong nuclear staining and evident mitosis. In contrast, the distribution of tumor cells in each treatment group appeared more scattered, with shrunken and fragmented nuclei. Notably, the extent of cellular damage and necrosis in the FA–PEG-PDA@TSA NRs group was the most pronounced, as evidenced by a reduction in the blue-purple cell nuclei, indicating that its anticancer effect is the most significant. The K _i-67 immune response further corroborated this conclusion (Figure G), the model group had the highest K _i-67 expression level (92.16%); conversely, the TSA group (68.23%) and the PEG–PDA@TSA NRs group (31.93%) exhibited significantly reduced numbers of K _i-67 positive cells. Importantly, the FA–PEG-PDA@TSA NRs group demonstrated the lowest expression level (17.76%). These results indicate that FA–PEG-PDA@TSA NR can inhibit tumor growth by killing tumor cells and destroying cell nuclei, further demonstrating its excellent antitumor efficacy.

Conclusions

We developed a novel method employing antisolvent precipitation and self-assembly technology to address the high lipophilicity and low solubility of the natural active ingredient TSA. TSA self-assembles into nanorods, driven by π-π stacking and hydrophobic interactions. This carrier-free nanodrug technology effectively overcomes TSA’s low solubility, achieving an 88-fold solubility increase in TSA NRs compared to free TSA. Subsequently, using these self-assembled TSA NRs as a core, we successfully prepared FA–PEG-PDA@TSA NRs. The PDA and PEG-coated nanorods demonstrated enhanced uniformity, improved water dispersibility, and increased stability. The H22 in vivo tumor mouse model confirmed that modification with FA enhanced tumor targeting and significantly inhibited tumor growth. This study introduces a novel drug delivery system that provides a direct approach to expanding the application of TSA. Furthermore, this nanotechnology and drug delivery strategy holds significant potential for clinical applications, particularly in enhancing drug solubility and pharmacological activity at equivalent concentrations.

Glossary

Abbreviations

SSA

Supramolecular self-assembly

TCM

Traditional Chinese medicine

Danshen, DS

Salvia miltiorrhiza

TSA

Tanshinone IIA

PDA

Polydopamine

FA

Folic acid

FR

Folate receptor

HCC

Hepatocellular carcinoma cells

CT

Cryptotanshinone

TEM

Transmission electron microscopy

DLS

Dynamic light scattering

UV–vis

Ultraviolet–visible spectroscopy

Flu

Fluorescence spectroscopy

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c03030.

• Effect of different solvents, solvent ratio, drug concentration on particle size of TSA NRs; average crystallite sizes of TSA and TSA NRs by Debye-Scherrer equation; effect of different PDA concentration, coating time on particle size of PDA@TSA NRs; standard curve for determination of TSA; particle size of TSA NRs before and after the addition of SDS (PDF)

Yiman Zou: data curation, investigation, formal analysis, validation, writing (original draft); Yu Zheng: formal analysis; Yue Liu: investigation, validation; HaoYu Zhao: visualization; Yunyu Zhang: methodology; Qun Zhao: methodology; Linwei Chen: supervision; Gang Li: supervision; Rui Chen: conceptualization, supervision, funding acquisition, writing (review and editing).

Supported by National Natural Science Foundation (No. 82474195), Youth medical Innovation research project of China (P24021887623), Nanjing Medical University (TZKY20230104, 2024KF0292), and Science and Technology Support Project of Taizhou (TS202420).

The authors declare no competing financial interest.