A potent mannose-modified pillararene-BODIPY system for photodynamic therapy

Abstract

Photodynamic therapy (PDT) offers a promising, non-invasive approach to cancer treatment. However, its efficacy is often limited by the poor water solubility, low cellular uptake, and high dose requirements of traditional photosensitizers, which can lead to side effects like skin photosensitivity. This study presents a novel supramolecular photosensitizer, PS3⊂WP5, comprising a mannosylated pillar[5]arene (WP5) host and a near-infrared BODIPY photosensitizer (PS3). This host–guest complex exhibits a strong binding affinity (K_a = 5.10 × 10⁶ M⁻¹) and self-assembles into nanoparticles in water. PS3⊂WP5 demonstrates high singlet oxygen quantum yield (Φ_Δ = 0.95) upon irradiation at 633 nm, along with excellent photostability. In vitro experiments confirm that PS3⊂WP5 exhibits superior PDT efficacy, good biocompatibility, and low dark toxicity compared to the free PS3, which suffers from poor aqueous solubility, low stability, and limited cellular uptake. This supramolecular approach offers a promising strategy for the design of multifunctional nanomaterials for cancer phototherapy, potentially overcoming the limitations of conventional photosensitizers and paving the way for the development of more efficient PDT agents with enhanced clinical potential.

---

Introduction

Advances in optical and photonic technologies now play an important role in modern phototheranostics including photodynamic therapy (PDT).^1–3 PDT has emerged as particularly effective for skin, nasopharyngeal, and oral cancer treatments.^4–9 It is known for being minimally invasive, extremely precise, cost-efficient, and easily adjustable, making it an excellent therapeutic option.^6–9 The key factor in PDT is photosensitizers (PSs), which are molecules activated by certain light wavelengths.¹⁰ PSs work by transferring energy to nearby oxygen molecules upon activation, generating singlet oxygen (¹O₂) and reactive oxygen species (ROS). Designing these agents requires careful consideration of several factors, including low toxicity in the absence of light (dark toxicity), appropriate water solubility, strong absorption in the near-infrared (NIR) region for deep tissue penetration, photostability, little side-effects, target selectivity, minimal photobleaching, low-concentration treatment, and efficient singlet oxygen production.^11–15

The first generation of PSs was derived from porphyrin derivatives and presented several molecular and cellular challenges, such as low absorption band in the NIR region, high hydrophobicity, cytotoxicity, and phototoxicity.¹⁴ These issues arise from the inherent hydrophobicity and structural rigidity of porphyrins, limiting their pharmacokinetic properties. The second generation of PSs addressing these limitations by improving the photophysical and pharmacokinetic properties, are based on derivatives of porphyrin, phthalocyanine, cyanine, chlorin, and curcumin.^15,16 For instance, second-generation photosensitizers like benzoporphyrin derivatives,¹⁷ texaphyrins,¹⁸ and chlorin¹⁹ demonstrate enhanced targeted selectivity and deeper tissue penetration, but are limited by their solubility in water.²⁰ Therefore, the design and preparation of third generation PSs are dominated by modifying the PSs with carbohydrate²¹ and peptide²² for enhanced solubility and cellular uptake.²³ However, despite these advances, there is still a need for further research to optimize PSs in terms of targeting, stability, strong absorbance in the NIR region (600–850 nm) and toxicity profiles to expand their clinical applications.

Boron dipyrromethene (BODIPY) derivatives have gained attention as promising photosensitizers due to their high extinction coefficients, strong fluorescence, narrow spectral bands, and robust photostability.^24–26 Such characteristics make BODIPY derivatives valuable in various applications, including drug delivery^27,28 and cancer imaging,^29–31 and as a PS in PDT.^32–34 Despite their promise, BODIPY based PSs often suffer from low water solubility, poor biocompatibility, and limited selectivity, which restrict their biological applications. To overcome these limitations and enhance the solubility, selectivity, and biological compatibility, BODIPY derivatives have been modified with hydrophilic groups or biomolecules such as antibody,³³ peptide,^34,35 carbohydrate,^36,37 or folic acid,³⁸ each of which facilitates targeted delivery to cancer cells. Beyond these modifications, other strategies, such as the use of the heavy atom effect, have been explored to enhance the photodynamic efficacy of BODIPY-based photosensitizers.³⁹ For example, BDPI-lyso showed a singlet oxygen quantum yield (Φ_Δ) of 95% and revealed that the incorporation of heavy atoms at the 2,6-position of BODIPY could enhance the intersystem crossing and increase the singlet oxygen quantum yield.⁴⁰ However, excessive heavy atom incorporation increases dark toxicity, underscoring the need for a balanced approach.³⁰ BODIPY derivatives with orthogonal configurations and spatially separated donor–acceptor (D–A) systems can efficiently generate triplet states via the SOCT-ISC mechanism without heavy atoms, reducing dark toxicity. However, their Φ_Δ reaches a maximum of only 65%, indicating room for improvement in PDT efficacy.⁴¹

Supramolecular chemistry offers a promising approach for targeted delivery of PSs or drugs to cancer cells. For example, β-cyclodextrin has been utilized as a nanocarrier for the anticancer drug doxorubicin (DOX) to enhance delivery to target cells.³⁸ Pillar[5]arenes, due to their adaptable cavity sizes, have been extensively investigated as supramolecular hosts, demonstrating broad applicability.⁴² Functionalization of pillar[5]arenes with targeting moieties such as lactose, galactose, and mannose has shown significant potential for developing targeted anticancer drug delivery systems.^43,44 Specifically, mannosylated pillar[5]arene has been reported to effectively deliver tamoxifen, a poorly water-soluble anticancer drug, thereby improving its anticancer activity.^45,46 This success with mannosylated pillar[5]arene suggests its potential application in PDT.

In this study, we present a supramolecular photosensitizer system (PS3⊂WP5) formed by complexing a diiodo BODIPY-based photosensitizer (PS3) with a water-soluble, mannosylated pillar[5]arene (WP5) (Scheme 1). This design integrates heavy atom-enhanced singlet oxygen generation with improved solubility, stability, and cellular uptake via supramolecular chemistry. Mannose units were introduced to WP5 using a click reaction, forming stable, biocompatible triazole linkers.⁴⁸PS3 exhibits strong NIR absorption/emission for better tissue penetration in PDT, while the iodine atoms boost singlet oxygen yield for enhanced therapeutic efficacy. WP5 encapsulates PS3, protecting it from premature degradation and enhancing bioavailability. The resulting 215-nm spherical complex addresses common limitations of traditional photosensitizers and shows promise for effective anticancer PDT.

Scheme 1 Schematic illustration of the formation of PS3⊂WP5 photosensitizer.

Results and discussion

Synthesis and characterization of PS3⊂WP5

The supramolecular photosensitizer PS3⊂WP5 was designed to enhance PDT efficacy through increased cellular uptake, water solubility, and singlet oxygen generation. PS3⊂WP5 was prepared via the host–guest interaction between the diiodo-substituted BODIPY PS3 with the mannosylated pillar[5]arene WP5. Diiodo groups on the BODIPY core increase singlet oxygen production, while mannose groups on the pillar[5]arene improve water solubility and enable receptor-mediated uptake by MCF-7 breast cancer cells.

Scheme S1 (ESI†) shows the synthesis of PS3. The starting BODIPY 1 was prepared by reaction between 4-hydroxybenzaldehyde and 2,4-dimethylpyrrole, followed by the addition of triethylamine and BF₃·OEt₂, according to the literature.^47,481 was then treated with 1,10-dibromodecane to afford a long-alkyl chain BODIPY 2 in 68% yield. After iodination of 2 using N-iodosuccinimide (NIS), a diiodo-BODIPY 3 was obtained in 97% yield. 3 underwent Knoevenagel condensation with anisaldehyde to afford the BODIPY 4 with extended π conjugation in 25% yield. In order to induce a strong affinity towards the pillar[5]arene, a trimethylamine moiety was introduced into the alkyl chain end of 4 to obtain PS3 in a quantitative yield. These compounds were characterized by ¹H-NMR, ¹³C-NMR spectroscopy, and mass spectrometry and their spectra are shown in the ESI.† In the ¹H-NMR spectrum of PS3 (Fig. S13, ESI†), the vinyl protons appear as two doublets at 8.06 ppm and 7.42 ppm with the average coupling constant of 16.6 Hz. Six equivalent protons of two methoxy groups resonate as a singlet peak at 3.82 ppm. The singlet resonance at 3.03 ppm is attributed to the three methyl protons attached to the ammonium ion. The triplet signal at 4.05 ppm is assigned to the methylene protons of the O–CH₂ group. The signal for the N–CH₂ methylene protons was not detected, likely obscured by the H₂O resonance in the 3.3–3.5 ppm region. Additionally, the signal at 1.46 ppm corresponds to the methyl groups attached to the BODIPY core. The ESI-MS result also confirms the formation of the BODIPY guest PS3 (Fig. S15, ESI†).

WP5 was synthesized via a CuAAC reaction between an alkyne-substituted pillar[5]arene (9) and azido-functionalized mannose (7) (Scheme S3, ESI†). Compound 9 was synthesized through a two-step process. Firstly, hydroquinone was reacted with propargyl bromide. Subsequently, the resulting intermediate 8 was cyclized using paraformaldehyde in the presence of BF₃·Et₂O as a catalyst. The formation of 9 was confirmed by the presence of a symmetrical aromatic proton signal at 6.82 ppm and a characteristic triplet signal at 2.28 ppm corresponding to the alkyne group in the ¹H-NMR spectrum (Fig. S26, ESI†). Compound 7 was synthesized by azidation of the bromo analog 6, following previously reported procedures (Scheme S2, ESI†).⁴⁹ Subsequently, a CuAAC reaction between 7 and 9 was performed, followed by deacetylation to yield the host WP5. The structure of WP5 was characterized by ¹H-NMR, ¹³C-NMR, and MALDI-TOF (Fig. S32–S34, ESI†). The formation of the triazole ring was confirmed by the presence of a chemical shift at 8.24 ppm in the ¹H-NMR spectrum and characteristic peaks at 124.34 and 143.25 ppm in the ¹³C-NMR spectrum. The successful removal of all acetyl groups was confirmed by the disappearance of the four singlet peaks around 2.04 ppm in the ¹H-NMR spectrum.

The formation of the PS3⊂WP5 host–guest complex was established using ¹H-NMR spectroscopy. Fig. 1 presents the ¹H-NMR spectra in 1 : 1 DMSO-d₆/D₂O for free PS3 (a), the PS3⊂WP5 complex (b), and free WP5 (c). Due to the limited solubility and self-aggregation of free PS3 in water, Fig. 1a shows only broad signals in the 1.06–1.58 ppm region, which are attributed to the decyl chain CH₂ groups (H-10 to H-17). Upon complexation with WP5, the ¹H-NMR spectrum displays new resonance peaks in addition to the characteristic signals of the mannosylated pillar[5]arene host (Fig. 1b). A significant upfield shift of the H-10 to H-17 signals, resulting in signals between 0.88 and −1.27 ppm, is observed. This result indicates inclusion-induced shielding by the WP5 cavity.⁵⁰ Additionally, the signals at 7.15–6.65 ppm are assigned to PS3 protons H-7, H-8, and H-4, along with WP5 proton H-a. The resonances at 8.22–8.02 ppm correspond to the vinylic protons H-1 of PS3 and the triazole protons H-b of WP5. The characteristic signals of WP5 are retained in the PS3⊂WP5 complex spectrum: 4.4–5.1 ppm (H-d, H-f, mannose protons H-g to H-j, overlapping with H₂O) and 3.0–4.0 ppm (H-c, H-e, mannose protons H-k, H-l, H-l′), as can be seen in Fig. 1c. These results confirm the successful formation of the stable PS3⊂WP5 host–guest complex and the disruption of PS3 self-aggregation.

Fig. 1 ¹H-NMR spectra (400 MHz, 1 : 1 DMSO-d₆/D₂O, 298 K): (a) 2.0 mM PS3, (b) 2.0 mM PS3⊂WP5, and (c) 2.0 mM WP5. The insets show the chemical shift region from 2.6 to −2.0 ppm for each spectrum.

The stoichiometry of the PS3⊂WP5 complexation was determined to be 1 : 1 through both isothermal titration calorimetry (ITC) and fluorescence titration, as shown in Fig. 2 and Fig. S40 (ESI†), respectively. The ITC results exhibit a binding constant (K_a) of 5.10 × 10⁶ M⁻¹ which demonstrates a strong binding affinity between PS3 and WP5 (Table S2, ESI†). This may arise from a combination of intermolecular forces. Hydrophobic effects likely play a significant role,⁵¹ alongside London dispersion forces between the aromatic rings of the pillar[5]arene and the C–H units of the PS alkyl chain. Additionally, cation–π interactions between the trimethylammonium group of PS3 and the electron-rich aromatic cavity of the pillar[5]arene (WP5) may contribute to the observed association.⁵² The formation of the PS3⊂WP5 complex was further supported by a red shift of the maximum absorption and emission wavelengths observed in the UV-vis absorption and fluorescence spectra (Fig. 3a and b). Upon complexation, the absorption maximum of PS3 exhibited a noticeable red shift, indicating a change in the electronic environment of PS3.

Fig. 2 Isothermal titration calorimetry of WP5 and PS3 in DI water at 25 °C. (a) Raw ITC data for 19 sequential injections of 500 μM PS3 solution into 20 μM WP5 solution. (b) Plot of the enthalpy change (ΔH) versus molar ratio of guest to host molecules. The sharp change in ΔH around the inflection point indicates the binding event, with saturation occurring near the stoichiometric ratio, suggesting a 1 : 1 binding interaction.

Fig. 3 Normalized spectra of PS3 and PS3⊂WP5 in water (λ_ex = 680 nm): (a) absorption and (b) fluorescence. (c) DLS results of PS3⊂WP5 and PS3 in water. (d) The ζ potentials of PS3⊂WP5 and PS3 at 25 °C. Inset: Photograph showing the Tyndall effect of PS3⊂WP5.

Fig. S35 (ESI†) displays aqueous dispersions of PS3 and PS3⊂WP5 under ambient conditions, highlighting differences in solubility. PS3 at 50 μM appeared fully dissolved, whereas PS3 at 300 μM shows visible precipitation, indicating poor water solubility at higher concentrations. In contrast, the complex PS3⊂WP5 at 300 μM remained fully dispersed with no observable precipitate, demonstrating that complexation with WP5 significantly improves the aqueous solubility of PS3. To investigate the aggregation behavior of PS3⊂WP5, transmission electron microscopy (TEM) and dynamic light scattering (DLS) were employed. DLS was measured three times, and the results reveal the average hydrodynamic diameter of 228.5 nm in water indicating the formation of nanoscale self-assemblies (Fig. 3c). To assess colloidal stability, zeta potential (ζ) measurements were performed, yielding a value of 30.15 ± 4.06 mV, which suggests good electrostatic stabilization in aqueous solutions (Fig. 3d). Further stability evaluation in biologically relevant environments showed only a slight increase in particle size after 24 h incubation in phosphate-buffered saline (PBS, pH 7.4) and Eagle's minimal essential medium (EMEM) (Fig. S41, ESI†). These results confirm the good colloidal stability of PS3⊂WP5 nanoparticles under physiological conditions and highlight the potential of WP5-based vesicles as robust nanocarriers for photosensitizer delivery in cancer therapy.

In contrast, free PS3 exhibited a broad, bimodal size distribution, with peaks observed in both the low and high nanometer ranges and an overall average diameter of 771.5 nm, accompanied by a ζ of 5.35 mV (Fig. 3c and d). This behavior suggests poor dispersion and a pronounced tendency to form large aggregates in water, likely due to the hydrophobic nature of the PS3 core and the absence of sufficient amphiphilic balance. The appearance of multiple peaks further highlights the instability and heterogeneity of free PS3 particles in aqueous media. These findings support the hypothesis that WP5 encapsulation not only enhances the aqueous dispersibility of PS3 but also facilitates the formation of more homogeneous and colloidally stable nanoparticles, which is desirable for PDT.

The colloidal stability of the PS3⊂WP5 complex was evaluated over a 7-day period by monitoring both Z-average and ζ values (Fig. S42, ESI†). The hydrodynamic diameter remained stable throughout the measurement, with only minor fluctuations observed (from ∼229 by day 1 to ∼242 nm by day 7). These slight variations suggest that the supramolecular assembly remained intact and did not undergo significant aggregation or disassembly under aqueous conditions. Furthermore, the ζ values exhibited a slight initial decrease from +30.15 mV to +27.9 mV, followed by a gradual increase to approximately +29.5 mV by day 7. Despite these small changes, the values consistently remained above the threshold commonly associated with colloidal stability (i.e., ±25–30 mV), indicating sufficient electrostatic repulsion to prevent particle aggregation. Taken together, these results confirm that the WP5-mediated encapsulation not only improves the aqueous solubility and uniformity of PS3 but also ensures long-term colloidal stability. This sustained dispersion supports potential application of PS3⊂WP5 in PDT, where prolonged stability and storage are essential.

TEM analysis confirmed the formation of a spherical shape with an average size of 215 nm and a shell thickness of 7.7 nm (Fig. S36, ESI†). This shell thickness is consistent with the 7.5 nm thickness predicted by Chem3D calculations. These nanoparticles likely arise from the self-assembly of PS3⊂WP5 into a bilayer structure, featuring a hydrophilic mannose outer shell and a hydrophobic BODIPY core. Furthermore, the strong Tyndall effect observed for the PS3⊂WP5 complex further supports the formation of nanoparticles (Fig. 3d).

The photophysical properties of PS3 and PS3⊂WP5 were examined (Table S1 and Fig. S37–S39, ESI†). The extended π-conjugation introduced via the Knoevenagel reaction in PS3 resulted in strong absorption and fluorescence emission in the NIR region in both organic and aqueous solutions. In aqueous solution, the absorption maximum (λ_max, abs) of PS3 was 713 nm, whereas the absorption maximum of PS3⊂WP5 was bathochromically shifted to 730 nm due to self-assembly. Both photosensitizers exhibited red photoluminescence in water, as well as in water-DMSO mixtures, with emission maxima (λ_max, em) at 735 and 737 nm, respectively. The NIR emission bands of both compounds make them promising candidates for PDT due to the enhanced tissue penetration of NIR light. To evaluate the stability of the PS3⊂WP5 complex, UV-vis absorption spectra were recorded in various organic solvents. In dioxane, MeCN, and THF, the λ_max, abs appeared around 660 nm, similar to that of free PS3 (Fig. S39, ESI†). This result implies the complex dissociation under these conditions. In contrast, PS3⊂WP5 in DMSO showed the λ_max, abs at 724 nm, similar to that of PS3⊂WP5 in aqueous solutions. Notably, the spectrum in EtOH displayed two distinct peaks at ∼730 nm and ∼660 nm, suggesting partial disassembly of the complex and the coexistence of both the assembled and free forms of PS3. Overall, it can be concluded that the assembly of PS3⊂WP5 is solvent-dependent and stable in aqueous solutions.

Singlet oxygen generation and photostability

¹O₂ plays a crucial role in PDT as the primary ROS responsible for inducing cytotoxicity in tumor cells during PDT. Therefore, assessment of singlet oxygen generation and photosensitizer stability is crucial for determining the therapeutic efficacy of a photosensitizer in PDT. To evaluate the efficiency of PS3 in producing ¹O₂, irradiation experiments were conducted. Initially, the singlet oxygen quantum yield of PS3 was determined in CH₂Cl₂ to establish its intrinsic photosensitizing capability under ideal conditions. Subsequently, the evaluation was extended to aqueous solutions to better mimic biological environments. The performance of PS3 alone was compared to that of PS3 encapsulated within the pillar[5]arene (PS3⊂WP5), providing a comprehensive understanding of the photosensitizer's behavior in both optimal and physiologically relevant conditions.

Prior to assessing singlet oxygen generation, the photostability of the experimental components was verified. As shown in Fig. 4a, the absorption spectrum of 1,3-diphenylisobenzofuran (DPBF), a ¹O₂ trapping molecule, remained unchanged in CH₂Cl₂ solution upon light irradiation for 90 s, confirming its photostability. Similarly, no change in DPBF absorbance at 415 nm was observed after incubation with PS3 in the dark for 20 minutes, confirming that PS3 does not generate singlet oxygen in the absence of light. Next, ¹O₂ generation by PS3 was investigated. Fig. 4b shows a decrease in the absorbance of DPBF at 415 nm upon irradiation with 633 nm light in the presence of PS3, indicating the consumption of DPBF by ¹O₂ generated by PS3. Methylene blue (MB), a known photosensitizer, was used as a reference (Fig. 4c). The decreasing absorbance of DPBF at 415 nm with increasing light exposure time in the presence of MB confirms its ability to generate ¹O₂.

Fig. 4 Absorption changes of (a) pure DPBF, (b) DPBF + PS3, and (c) DPBF + MB, in CH₂Cl₂ with increasing irradiation time. (d) Comparison of absorbance changes of pure DPBF, DPBF + PS3, and DPBF + MB, in CH₂Cl₂ under irradiation. Irradiation was carried out using a 633 nm LED lamp with a power density of 1 mW cm⁻².

The singlet oxygen quantum yield (Φ_Δ) of PS3 in CH₂Cl₂ was determined to be 0.95 relative to MB (Φ_Δ(MB) = 0.57 in CH₂Cl₂). The detailed methodology for calculating Φ_Δ is provided in the Experimental section and the ESI† (Fig. S43). Fig. 4d and Fig. S44 and Table S3 (ESI†) compare the decay rates of DPBF at 415 nm in the presence of PS3 and MB. The significantly faster rate of absorbance decay observed with PS3 (16.8 × 10⁻² s⁻¹) highlights its higher efficiency for singlet oxygen photogeneration compared to MB (9.7 × 10⁻² s⁻¹). This enhanced efficiency can be attributed to the heavy atom effect of the iodine atoms in PS, which promote intersystem crossing (ISC) from the singlet excited state to the triplet state, leading to increased singlet oxygen generation.^53,54 These results highlight the high efficiency of PS3 for singlet oxygen photogeneration. It should be noted that the absorption intensity of PS3 at 660 nm remained constant under light irradiation conditions demonstrating excellent photostability (Fig. 4b).

The irradiation of PS3⊂WP5 in aqueous solution was next investigated. Singlet oxygen generation in water was studied using 9,10-anthracenediyl-bis(methylene)dimalonic acid (ABDA) as a ¹O₂ trapping agent. As shown in Fig. 5a, ABDA exhibits characteristic absorbance peaks at 358, 378, and 400 nm, which remain unchanged upon irradiation. This result indicates that light alone does not cause ABDA degradation. In contrast, ABDA absorbance was found to be dependent on irradiation time in the presence of both PS3⊂WP5, PS3, and MB, with photobleaching of ABDA observed (Fig. 5a–c). These results confirm the photogeneration of ¹O₂ by both PS3⊂WP5 and PS3. Furthermore, PS3⊂WP5 and PS3 exhibited similar effects on the decrease in ABDA absorbance (Fig. 5d and Fig. S45 and S46, ESI†), with decay rates observed of around 1.4 × 10⁻⁴ s⁻¹ following first-order kinetics (Table S3, ESI†). The Φ_Δ values of PS3 and PS3⊂WP5 were determined by comparing the ABDA degradation rates to that of MB (Φ_Δ(MB) = 0.52 in water) under identical conditions. Based on the relative first-order decay slopes, the Φ_Δ values were calculated to be 0.917 for PS3 and 0.914 for PS3⊂WP5, indicating highly efficient singlet oxygen generation upon light irradiation. The results suggest that complexation with the mannosylated pillar[5]arene WP5 does not affect the efficacy of PS3 for singlet oxygen production. Notably, the PS3⊂WP5 complex retains these photodynamic properties while offering additional advantages such as improved solubility and stability, making it even more suitable for therapeutic applications.

Fig. 5 Absorption changes in the range of 300 to 500 nm of ABDA in the presence of (a) PS3⊂WP5, (b) PS3, and (c) MB, in aqueous solutions with increasing irradiation time. (d) Photobleaching comparison of ABDA + PS3⊂WP5, ABDA + PS3, and ABDA + MB, in aqueous solutions. Irradiation was carried out using a 633 nm LED lamp with a power density of 2 mW cm⁻².

To evaluate photostability, the UV-vis absorption spectra of free PS3 and PS3⊂WP5 were recorded every 2 min during continuous red light irradiation (45 mW cm⁻²) for 16 min. As shown in Fig. S47 (ESI†), both free PS3 and PS3⊂WP5 displayed remarkable photostability, with no significant change in its absorbance over the course of irradiation. These results indicate that both free PS3 and PS3⊂WP5 maintain reasonable photostability in aqueous solutions when exposed to continuous light.

In vitro cellular uptake assay of PS3⊂WP5

To investigate the intracellular distribution of PS3⊂WP5, MCF-7 breast cancer cells were incubated with 250 and 500 nM of the complex for 24 h and imaged using confocal laser scanning microscopy. As shown in Fig. 6, cells incubated with 250 nM PS3⊂WP5 exhibited red fluorescence signals (white arrows), indicating cellular uptake. Increasing the concentration of PS3⊂WP5 to 500 nM resulted in slightly higher accumulation within the cells. Conversely, MCF-7 cells treated with PS3 alone showed a very weak fluorescence signal (Fig. S50, ESI†). These results highlight that the mannosylated pillar[5]arene facilitates the uptake of the PS3⊂WP5 complex into MCF-7 breast cancer cells, suggesting its potential as a photosensitizer for PDT in cancer treatment.

Fig. 6 CLSM images of MCF-7 cells treated with 250 nM and 500 nM of PS3⊂WP5 for 24 h. Each panel shows the cell nuclei stained by Hoechst 33342 (blue), PS3⊂WP5 (red), phalloidin (green), and an overlay of three channels.

Cytotoxicity and phototoxicity of PS3⊂WP5 and PS3

The biocompatibility and PDT efficacy of PS3⊂WP5 and PS3 were assessed using an MTT assay to evaluate the viability of two different cell lines: MCF-7 breast cancer cells and MCF-10A normal breast cells. The cells were incubated with either PS3⊂WP5 or PS3 at different concentrations ranging from 0–500 nM (Fig. 7). The results showed that PS3⊂WP5 exhibited low to moderate cytotoxicity in the absence of light; the viability of MCF-10A cells did not significantly change from the control group, while the viability of MCF-7 cells decreases to approximately 60% at a concentration of up to 500 nM (Fig. 7a and b). Remarkably, upon 666 nm red light exposure at 100 nM PS3⊂WP5, the MCF-7 cell viability dropped dramatically to around 20%, while MCF-10A cells maintained high viability at 92%. These results underscore the potent and selective phototoxic activity of PS3⊂WP5 towards cancer cells.

Fig. 7 Photodynamic cytotoxicity of PS3⊂WP5 and PS3 against MCF-7 and MCF-10A cells. (a) and (b) MCF-7 breast cancer and MCF-10 normal breast epithelial cells were treated with PS3⊂WP5 (0–500 nM) for 2 h, followed by red LED irradiation (45 mW cm⁻², 30 min) and incubated for 24 h. (c) and (d) MCF-7 breast cancer and MCF-10A normal breast epithelial cells were treated under the same conditions with free PS3. 1% Triton X-100 was used as a positive control for complete cell death. Data are shown as means ± SEM from three independent experiments (n = 3). Statistical significance was analyzed using two-way ANOVA; **p < 0.01, ****p < 0.0001, ns = not significant.

In contrast, treatment with free PS3 resulted in over 80% viability for both MCF-7 and MCF-10A in both dark and PDT conditions, as shown in Fig. 7c and d. Similarly, MB, a well-known commercial photosensitizer, also demonstrated minimal phototoxicity under the same conditions in both MCF-7 and MCF-10A cells (Fig. S51a and b, ESI†), indicating that PS3⊂WP5 provides superior PDT efficacy compared to both MB and free PS3. This observation aligns with cellular uptake findings, suggesting that PS3⊂WP5 effectively penetrates cells better than PS3, resulting in induced photodynamic toxicity in the cells. These results highlight its potential as a photosensitizer for generating singlet oxygen species, triggering apoptosis, and serving as a promising agent for PDT cancer treatments.

ROS production in MCF-7 breast cancer cells

Intracellular reactive oxygen species (ROS) production by PS3⊂WP5 was further analyzed using 2′,7′-dichlorofluorescein diacetate (DCFDA), which is oxidized into the green-fluorescent compound dichlorofluorescein (DCF).^30,55 The fluorescence signal was visualized using the Operetta High Content Imaging System. As illustrated in Fig. S52a and b (ESI†), intracellular ROS levels significantly increased when MCF-7 cells were treated with PS3⊂WP5 at concentrations ranging from 100 to 500 nM, followed by 666 nm red light irradiation. In contrast, minimal changes were observed without light exposure (Fig. S52b, ESI†). These results suggest that PS3⊂WP5 generates singlet oxygen species, inducing photodynamic cellular damage in MCF-7 breast cancer cells. It should be noted that the basal ROS signal observed in untreated MCF-7 cells (PS3⊂WP5 0 nM) can be attributed to endogenous ROS production from mitochondria and NADPH oxidases. Cancer cells are known to maintain elevated ROS levels due to oncogenic signaling, metabolic alterations, and hypoxic microenvironments, even in the absence of external stimuli.⁵⁶

Conclusions

This study demonstrates the successful synthesis and characterization of a diiodo-substituted BODIPY photosensitizer (PS3) incorporated into a mannosylated pillar[5]arene (WP5) host, forming the PS3⊂WP5 complex. The complex exhibits near-infrared absorption and efficient singlet oxygen generation, highlighting its potential as a photosensitizer for PDT. Owing to its ability to form nanoparticles in water, the photosensitizer complex PS3⊂WP5 has high water solubility, photostability, and enhanced cellular uptake in MCF-7 breast cancer cells, compared to its molecular photosensitizer analog PS3. Evaluation of the PS3⊂WP5 complex shows that it retains the same singlet oxygen generation capacity in water as PS3, indicating that complexation with WP5 does not diminish its PDT efficacy. In vitro experiments further confirm that the complex induces significant cytotoxicity in MCF-7 cells upon light activation, demonstrating its potential as an anticancer agent.

Experimental section

Materials

All the chemicals were purchased from commercial sources and used without further purification. 2,4-Dimethylpyrrole, 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), triethylamine, trimethylamine, boron trifluoride diethyl etherate (BF₃·Et₂O), N-iodosuccinimide, p-anisaldehyde, 1,10-dibromodecane, D-mannose, trifluoroacetic acid (TFA), 2-bromoethanol, and sodium azide were obtained from Tokyo chemical industry (TCI). 4-Hydroxybenzaldehyde, 9,10-anthracenediyl-bis(methylene)dimalonic acid (ABDA), sodium methoxide, piperidine, and amberlite IRC120H hydrogen form were obtained from Merck. 1,3-Diphenylisobenzofuran (DPBF) was obtained from Thermo Fisher. Tetrahydrofuran (THF) was distilled over Na and benzophenone. Dichloromethane (CH₂Cl₂) was distilled from calcium hydride (CaH₂) and stored under 4 Å molecular sieves. Water utilized in the analysis processes was obtained from the Millipore system and deionized before use. Thin layer chromatography (TLC) was performed using aluminum-backed silica gel 60 F254 analytical plates from Merck. Column chromatography was performed using Merck silica gel 60 (70–230 mesh) and aluminium oxide 90 active neutral (70–230 mesh).

Instruments

¹H-NMR spectra were obtained using either a Bruker Ascend 400 MHz or a JEOL JNM-ECZR 400 MHz spectrometer, and ¹³C-NMR spectra were obtained at 100 MHz. Chemical shifts (δ) and coupling constants (J) are reported in parts per million (ppm) and Hertz (Hz), respectively. Tetramethylsilane (TMS) or residual non-deuterated solvent signals were employed as internal standards for all NMR spectral analyses. HRMS spectra were recorded using an HR-TOF-MS Micromass model VQ-TOF2, direct analysis in real time (DART) mass spectrometer (JEOL), and liquid chromatography mass spectrometer (Q-trap). UV-vis and fluorescence experiments were carried out using a Shimadzu UV-2600 UV-vis spectrophotometer and a HORIBA Fluoro Max Plus spectrofluorometer, and transmission electron microscopy (TEM) images were obtained using a JEOL JEM-2100 transmission electron microscope. DLS measurements were carried out on Malvern Zetasizers Pro. ITC experiments were performed using an Isothermal Titration Calorimeter (ITC)-PEAQ.

Evaluation of singlet oxygen quantum yield (Φ_Δ)

The singlet oxygen quantum yield (Φ_Δ) of PS3 was determined indirectly using 1,3-diphenylisobenzofuran (DPBF) as the ¹O₂ trapping agent in dichloromethane (CH₂Cl₂), with MB as the reference standard (Φ_Δ = 0.57 in CH₂Cl₂). The generation of singlet oxygen was monitored by observing the photobleaching of DPBF, quantified by the decrease in absorbance, when the sample was irradiated with a 633 nm LED light source operating at 1 mW cm⁻². The light source was positioned approximately 1 cm from the sample cuvette (Fig. S49, ESI†).where “x” and “ref” designate “PS3” and “MB”, respectively. “m” is the slope of the absorbance attenuation curve of DPBF at 415 nm, and “F” is the absorbance correction factor, which is obtained by F = 1–10^−O.D. (O.D. is the absorbance of the solution at 666 nm).

Singlet oxygen generation by PS3 and PS3⊂WP5 in aqueous solutions was demonstrated using 9,10-anthracenediyl-bis(methylene)dimalonic acid (ADBA), a water-soluble singlet oxygen-selective trapping molecule, under irradiation with a red LED light source at 2 mW cm⁻².

Photostability measurement

Photostability was evaluated using 105 μM solutions of PS3 and PS3⊂WP5 in deionized water. For each compound, 1 mL of the solution was transferred to a quartz cuvette with a 1 cm path length and irradiated using a red LED light source (666 nm, 45 mW cm⁻²). The absorbance spectra were recorded every 2 min over a period of 16 min using a UV-vis spectrophotometer. Photostability was assessed by monitoring the decrease in the absorbance maximum of the photosensitizers over time.

Cell culture

Human breast cancer cells (MCF-7) were purchased from the American Type Culture Collection (ATCC®) and cultured in minimum essential medium (MEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin. Cells were maintained at 37 °C under 5% CO₂ in a humidified incubator to reach ∼80% confluence. Subculture was done using 0.25% trypsin-EDTA. The media were changed every 2 days. All cell culture supplies were purchased from Gibco, UK.

Cellular uptake of PS3⊂WP5

MCF-7 cells were seeded onto coverslips in a 24-well plate (2 × 10⁵ cells per mL, 0.5 mL) and incubated in a humidified incubator at 37 °C in the presence of 5% CO₂ overnight to allow the cells’ attachment. The media were removed and replaced with fresh media containing either 250 nM or 500 nM of PS3⊂WP5, then incubated in a humidified incubator for 24 h. After incubation time, cells were washed three times with PBS. Then, the cells were fixed with 3.6% formaldehyde in PBS, pH 7.4 for 15 min at RT, and washed three times with PBS. The cells were stained with Phalloidin-Fluor 488 conjugate (1 : 1000, AB176753, Abcam, United Kingdom) at RT for 1 h under dark conditions and washed three times with PBS. The cells were incubated with 1 μg mL⁻¹ of Hoechst 33342 in PBST for nucleus staining and rinsed three times for 5 min, each with PBST. Finally, the coverslips of the cells were mounted with a drop of anti-fade mounting medium onto a glass slide, sealed with nail polish to prevent drying and movement, and stored in the dark at 4 °C. An inverted confocal laser scanning microscope was employed to visualize the cellular uptake of PS3⊂WP5. The following excitation wavelengths were used: 350 nm for Hoechst 33342, 495 nm for Phalloidin-Fluor 488, and 647 nm for PS3⊂WP5.

Phototoxicity of PS3⊂WP5

To measure the cellular toxicity of PS3⊂WP5 and PS3, cells were seeded onto a 96-well plate (1 × 10⁴ cells per well) and incubated in a humidified incubator at 37 °C in the presence of 5% CO₂ overnight to allow the cells’ attachment. The media were removed and replaced with fresh media containing different concentrations of PS3⊂WP5 or PS3 (0–500 nM, 100 μL per well) compared with 1% Triton-X100 in media solution. The cells were incubated in a humidified incubator for 2 h. The cells were then exposed to the 666 nm red LED lamp (Fig. S48, ESI†) at a power of 45 mW cm⁻² for 30 min and further incubated in a humidified incubator for 24 h. The nonirradiated cells were incubated as the control condition. The treatment solutions were removed, and the cells were washed with complete media and 1× PBS solution. Following this, 100 μL of methyl-thiazolyl diphenyl-tetrazolium bromide (MTT) solution (0.5 mg per mL in serum-free media) was added, and the mixture was incubated at 37 °C for 3 h in a humidified incubator with a 5% CO₂ atmosphere. After incubation, the MTT was removed, and 200 μL of DMSO was added to each well to dissolve the formazan pellets. A microplate reader (Tecan Spark 10M) was used to measure the absorbance of the dissolved formazan at 540 nm and calculate the viability of the cells by the following equation:

Reactive oxygen species production assay

The intracellular ROS, produced by PS3⊂WP5, were measured by DCFDA/H₂DCFDA Cellular ROS Assay kit (ab113851, Abcam, UK). MCF-7 cells were seeded into a 96-well black plate (2.5 × 10⁴ cells per well) and incubated in a humidified incubator at 37 °C in the presence of 5% CO₂ overnight to allow the cells’ attachment. Cells were incubated with 25 μM DCFDA solution for 45 minutes at 37 °C in the dark, after which the media were removed. Following washing with 1× buffer, the cells were incubated with PS3⊂WP5 at concentrations ranging from 100 to 500 nM for 24 h. tert-Butyl hydroperoxide (TBHP) was used as a positive control for ROS induction. ROS levels were quantified by measuring fluorescence intensity (Ex/Em = 485/535 nm) using a fluorescence plate reader (Tecan, Infinite 200 PRO) and visualized with a fluorescence imaging microscope (IX83, Olympus).

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by Mahidol University (Fundamental Fund: fiscal year 2025 by National Science Research and Innovation Fund (NSRF)). We acknowledge the CIF-CNI grant, Faculty of Science, Mahidol University and the Center of Excellence for Innovation in Chemistry (PERCH-CIC), Ministry of Higher Education, Science, Research, and Innovation. Created with BioRender.com./Mahidol University.

References

1. E. Ö. Gündüz, M. E. Gedik, G. Günaydın and E. Okutan, ChemMedChem, 2022, 17, e202100693.
2. M. Guan, J. Ge, J. Wu, G. Zhang, D. Chen, W. Zhang, Y. Zhang, T. Zou, M. Zhen, C. Wang, T. Chu, X. Hao and C. Shu, Biomaterials, 2016, 103, 75–85.
3. X. Yan, G. Niu, J. Lin, A. J. Jin, H. Hu, Y. Tang, Y. Zhang, A. Wu, J. Lu, S. Zhang, P. Huang, B. Shen and X. Chen, Biomaterials, 2015, 42, 94–102.
4. R. Laptev, M. Nisnevitch, G. Siboni, Z. Malik and M. A. Firer, Br. J. Cancer, 2006, 95, 189–196.
5. M. Iranifam, Trends Anal. Chem., 2014, 59, 156–183.
6. Q. Zheng, X. Liu, Y. Zheng, K. W. K. Yeung, Z. Cui, Y. Liang, Z. Li, S. Zhu, X. Wang and S. Wu, Chem. Soc. Rev., 2021, 50, 5086–5125.
7. B. Kim, B. Sui, X. Yue, S. Tang, M. G. Tichy and K. D. Belfield, Eur. J. Org. Chem., 2017, 25–28.
8. T. J. Dougherty, G. B. Grindey, R. Fiel, K. R. Weishaupt and D. G. Boyle, JNCI, J. Natl. Cancer Inst., 1975, 55, 115–121.
9. B. M. Luby, C. D. Walsh and G. Zheng, Angew. Chem., Int. Ed., 2019, 58, 2558–2569.
10. P. Agostinis, K. Berg, K. A. Cengel, T. H. Foster, A. W. Girotti, S. O. Gollnick, S. M. Hahn, M. R. Hamblin, A. Juzeniene, D. Kessel, M. Korbelik, J. Moan, P. Mroz, D. Nowis, J. Piette, B. C. Wilson and J. Golab, Ca-Cancer J. Clin., 2011, 61, 250–281.
11. R. R. Allison and C. H. Sibata, Photodiagn. Photodyn. Ther., 2010, 7, 61–75.
12. C. H. Sibata, V. C. Colussi, N. L. Oleinick and T. J. Kinsella, Braz. J. Med. Biol. Res., 2000, 33, 869–880.
13. M. J. Garland, C. M. Cassidy, D. Woolfson and R. F. Donnelly, Future Med. Chem., 2009, 1, 667–691.
14. M. Triesscheijn, P. Baas, J. H. M. Schellens and F. A. Stewart, Oncologist, 2006, 11, 1034–1044.
15. H. Wall Mark, P. Basu, T. Buranda, B. S. Wicks, E. W. Findsen, M. Ondrias, J. H. Enemark and M. L. Kirk, Inorg. Chem., 1997, 36, 5676–5677.
16. J. Kou, D. Dou and L. Yang, Oncotarget, 2017, 8, 81591–81603.
17. M. Kramer, J. W. Miller, N. Michaud, R. S. Moulton, T. Hasan, T. J. Flotte and E. S. Gragoudas, Ophthalmology, 1996, 103, 427–438.
18. J. L. Sessler and R. A. Miller, Biochem. Pharmacol., 2000, 59(7), 733–739.
19. J. Shi, Q. Zeng, P. Wang, Q. Chang, J. Huang, M. Wu, X. Wang and H. Wang, Photodiagn. Photodyn. Ther., 2023, 41, 103321.
20. S. Kwiatkowski, B. Knap, D. Przystupski, J. Saczko, E. Kędzierska, K. Knap-Czop, J. Kotlińska, O. Michel, K. Kotowski and J. Kulbacka, Biomed. Pharmacother., 2018, 106, 1098–1107.
21. P. M. R. Pereira, W. Rizvi, N. V. S. D. K. Bhupathiraju, N. Berisha, R. Fernandes, J. P. C. Tomé and C. M. Drain, Bioconjugate Chem., 2018, 29, 306–315.
22. J. C. H. Chu, C. Yang, W.-P. Fong, C. T. T. Wong and D. K. P. Ng, Chem. Commun., 2020, 56, 11941–11944.
23. I. S. Mfouo-Tynga, L. D. Dias, N. M. Inada and C. Kurachi, Photodiagn. Photodyn. Ther., 2021, 34, 102091.
24. L.-Y. Niu, Y.-S. Guan, Y.-Z. Chen, L.-Z. Wu, C.-H. Tung and Q.-Z. Yang, J. Am. Chem. Soc., 2012, 134, 18928–18931.
25. S. Ye, J. Rao, S. Qiu, J. Zhao, H. He, Z. Yan, T. Yang, Y. Deng, H. Ke, H. Yang, Y. Zhao, Z. Guo and H. Chen, Adv. Mater., 2018, 30, 1801216.
26. A. Turksoy, D. Yildiz and E. U. Akkaya, Coord. Chem. Rev., 2019, 379, 47–64.
27. M. Porubský, S. Gurská, J. Stanková, M. Hajdúch, P. Džubák and J. Hlaváč, Mol. Pharmaceutics, 2021, 18, 2385–2396.
28. W. Saiyasombat, S. Nuchpun, K. P. Katewongsa, S. Pornsuwan, J. J. Weigand and S. Kiatisevi, New J. Chem., 2022, 46, 22525–22532.
29. M. Li, R. Tian, J. Fan, J. Du, S. Long and X. Peng, Dyes Pigm., 2017, 147, 99–105.
30. J. Zou, Z. Yin, K. Ding, Q. Tang, J. Li, W. Si, J. Shao, Q. Zhang, W. Huang and X. Dong, ACS Appl. Mater. Interfaces, 2017, 9, 32475–32481.
31. Y. Zhang, Z. Yang, X. Zheng, L. Yang, N. Song, L. Zhang, L. Chen and Z. Xie, Dyes Pigm., 2020, 178, 108348.
32. J. Zhu, J. Zou, J. Zhang, Y. Sun, X. Dong and Q. Zhang, J. Mater. Chem. B, 2019, 7, 3303–3309.
33. N. Lu, Y. Tian, W. Tian, P. Huang, Y. Liu, Y. Tang, C. Wang, S. Wang, Y. Su, Y. Zhang, J. Pan, Z. Teng and G. Lu, ACS Appl. Mater. Interfaces, 2016, 8, 2985–2993.
34. Q. Hu, W. Sun, Y. Lu, H. N. Bomba, Y. Ye, T. Jiang, A. J. Isaacson and Z. Gu, Nano Lett., 2016, 16, 1118–1126.
35. T. Ji, S. Li, Y. Zhang, J. Lang, Y. Ding, X. Zhao, R. Zhao, Y. Li, J. Shi, J. Hao, Y. Zhao and G. Nie, ACS Appl. Mater. Interfaces, 2016, 8, 3438–3445.
36. Y. Lim and M. Lee, Org. Biomol. Chem., 2007, 5, 401–405.
37. M. Delbianco, P. Bharate, S. Varela-Aramburu and P. H. Seeberger, Chem. Rev., 2016, 116, 1693–1752.
38. Z. Ye, Q. Zhang, S. Wang, P. Bharate, S. Varela-Aramburu, M. Lu, P. H. Seeberger and J. Yin, Chem. – Eur. J., 2016, 22, 15216–15221.
39. Y. Cai, P. Liang, Q. Tang, X. Yang, W. Si, W. Huang, Q. Zhang and X. Dong, ACS Nano, 2017, 11, 1054–1063.
40. C. Wang and Y. Qian, Org. Biomol. Chem., 2019, 17, 8001–8007.
41. P. Zhao, Z. Wang, Y. Wang, Z. Wu, Y. Guo, C. Wang, X. Fang, Z. Qu, H. Wang and G. Zhao, Dyes Pigm., 2023, 214, 111214.
42. K. Khamphaijun, P. Namnouad, A. Docker, A. Ruengsuk, J. Tantirungrotechai, R. Díaz-Torres, D. J. Harding and T. Bunchuay, Chem. Commun., 2022, 58, 7253–7256.
43. T. Ogoshi, S. Kanai, S. Fujinami, T. Yamagishi and Y. Nakamoto, J. Am. Chem. Soc., 2008, 130, 5022–5023.
44. C. Sathiyajith, R. R. Shaikh, Q. Han, Y. Zhang, K. Meguellati and Y.-W. Yang, Chem. Commun., 2017, 53, 677–696.
45. L. Shangguan, Q. Chen, B. Shi and F. Huang, Chem. Commun., 2017, 53, 9749–9752.
46. H. Peng, B. Xie, X. Cen, J. Dai, Y. Dai, X. Yang and Y. He, Mater. Chem. Front., 2022, 6, 360–367.
47. L.-B. Meng, W. Zhang, D. Li, Y. Li, X.-Y. Hu, L. Wang and G. Li, Chem. Commun., 2015, 51, 14381–14384.
48. J. Zou, Z. Yin, K. Ding, Q. Tang, J. Li, W. Si, J. Shao, Q. Zhang, W. Huang and X. Dong, ACS Appl. Mater. Interfaces, 2017, 9, 3247–32481.
49. I. Nierengarten, K. Buffet, M. Holler, S. P. Vincent and J.-F. Nierengarten, Tetrahedron Lett., 2013, 54, 2398–2402.
50. Q. Duan, Y. Cao, Y. Li, X. Hu, T. Xiao, C. Lin, Y. Pan and L. Wang, J. Am. Chem. Soc., 2013, 135, 10542–10549.
51. A. J. De Jesus and H. Yin, in Comprehensive Supramolecular Chemistry II, Elsevier, 2017, pp. 311–328.
52. T. Ogoshi and T. Yamagishi, Chem. Commun., 2014, 50, 4776–4787.
53. K. Chen, Y. Dong, X. Zhao, M. Imran, G. Tang, J. Zhao and Q. Liu, Front. Chem., 2019, 7, 821.
54. T. Yogo, Y. Urano, Y. Ishitsuka, F. Maniwa and T. Nagano, J. Am. Chem. Soc., 2005, 127, 12162–12163.
55. Q. Zhang, Y. Cai, Q. Li, L. Hao, Z. Ma, X. Wang and J. Yin, Chem. – Eur. J., 2017, 23, 14307–14315.
56. L. B. Sullivan and N. S. Chandel, Cancer Metab., 2014, 2, 17.

---

Footnote

† Electronic supplementary information (ESI) available: Additional references. See DOI: https://doi.org/10.1039/d5tb00670h

---