Pure photosensitizer-driven nanoassembly with core-matched PEGylation for imaging-guided photodynamic therapy.

Pure drug-assembled nanomedicines (PDANs) are currently under intensive investigation as promising nanoplatforms for cancer therapy. However, poor colloidal stability and less tumor-homing ability remain critical unresolved problems that impede their clinical translation. Herein, we report a core-matched nanoassembly of pyropheophorbide a (PPa) for photodynamic therapy (PDT). Pure PPa molecules are found to self-assemble into nanoparticles (NPs), and an amphiphilic PEG polymer (PPa-PEG2K) is utilized to achieve core-matched PEGylating modification via the π‒π stacking effect and hydrophobic interaction between the PPa core and the PPa-PEG2K shell. Compared to PCL-PEG2K with similar molecular weight, PPa-PEG2K significantly increases the stability, prolongs the systemic circulation and improves the tumor-homing ability and ROS generation efficiency of PPa-nanoassembly. As a result, PPa/PPa-PEG2K NPs exert potent antitumor activity in a 4T1 breast tumor-bearing BALB/c mouse xenograft model. Together, such a core-matched nanoassembly of pure photosensitizer provides a new strategy for the development of imaging-guided theragnostic nanomedicines.

Introduction

Cancer is still considered as one of the most serious diseases threatening human health. Currently, surgery is the most common and effective therapeutic approach for cancer, especially for the early stage of solid tumors without metastasis. In addition to surgical removal of solid tumors, a variety of other treatment strategies have also been widely used in clinic, such as chemotherapy and phototherapy,. Among them, chemotherapy is the predominant therapeutic regimen for clinical treatment of cancer, especially for the patients with inoperable and massively metastatic tumors. However, serious toxicities are usually caused by systemic administration of chemotherapeutic agents, due to their narrow therapeutic windows and off-target distribution in the body. Therefore, it is desirable that local diseases could be controlled and treated with site-specific local treatment regimens.

Compared with systemic chemotherapy, photodynamic therapy (PDT) has been widely investigated as a non-invasive cancer therapeutic approach. Under tumor-localized laser irradiation, photosensitizers (PSs) induce the apoptosis and necrosis of tumor cells via producing large amounts of cytotoxic reactive oxygen species (ROS),. The cytotoxic ROS damages the normal physiological functions of tumor cells by damaging the cell membrane and oxidizing the intracellular macromolecules9, 10, 11. Notably, PSs almost show no cytotoxicity without laser treatment, making phototherapy as favorable non-invasive treatment modality for site-specific cancer therapy. Moreover, the clinical application of phototherapy has been extended to the treatment of tumors in the deep viscera, by virtue of the rapid development of novel PSs and fiber optic equipment. However, the therapeutic efficacy of phototherapy is still impeded by the inefficient accumulation of PSs in tumors after intravenous administration. Therefore, rational design of high-efficient drug delivery system (DDS) is of crucial importance for efficient phototherapy.

With the burgeoning biomedical nanotechnology, a variety of nanoparticulate drug delivery systems (nano-DDS) have been developed to improve the delivery efficacy of anticancer drugs, including PSs15, 16, 17, 18, 19, 20, 21. In particular, most PSs were non-covalently encapsulated in organic or inorganic nanocarriers for anticancer therapy22, 23, 24, 25. Nevertheless, non-covalent drug loading approaches have long been criticized for low drug loading efficiency, poor stability, premature drug leakage, and potential carrier material-related toxicity. Recently, pure drug-assembled nanomedicines (PDANs) of small-molecule drugs or prodrugs have emerged as a promising nanoplatform for efficient drug delivery27, 28, 29, 30. Particularly, some hydrophobic drugs were found to be able to self-assemble into nanoparticles (NPs) by themselves29, 30, 31, 32. However, PDANs usually show unsatisfactory colloidal stability, due to the relatively weak intermolecular interactions among small molecules.

Herein, we constructed a pure photosensitizer-driven nanoassembly with the feature of core-matched PEGylating stabilization for efficient PDT (Fig. 1). First, unique self-assembly phenomenon of a commonly-used photosensitizer (pyropheophorbide a, PPa) was observed. Moreover, an amphiphilic PEG polymer (PPa-PEG2K) was ingeniously utilized to achieve core-matched PEGylating modification on the PPa-nanoassembly via hydrophobic and π‒π stacking interaction between PPa and PPa-PEG2K. Meanwhile, PCL-PEG2K with a very similar molecular weight was utilized as a control to demonstrate the advantage of core-matched and π‒π stacking interaction between PPa and PPa-PEG2K. The self-assembly mechanisms and core-matched interaction were investigated by molecular docking simulation. Additionally, sodium chloride (NaCl), sodium dodecyl sulfate (SDS) and urea were utilized to investigate self-assembly mechanisms of PPa-nanoassembly. These results demonstrated that multiple interactions and forces involved in the self-assembly process of PPa, and PEGylation modification significantly improved the colloidal stability of PPa-nanoassembly. Such a core-matched nanoassembly demonstrated multiple drug delivery advantages, including ultrahigh drug-loading capacity (74.8%, w/w), core-matched stability, prolonged systemic circulation, accelerated tumor accumulation, as well as favorable cellular uptake. These advantages resulted in significantly enhanced antitumor efficacy against BALB/c mice bearing 4T1 xenograft tumors. This is the first time that pure PPa is found to self-assemble into NPs, and the nanostructure with core-matched PEGylating modification could significantly facilitate the efficient delivery and the potent antitumor activity of the PPa-nanoassembly. Such a pure photosensitizer-driven and core-matched nanoassembly offers a promising application prospect for efficient imaging-guided PDT against cancer.

Figure 1

Schematic representation of self-assembly of PPa and efficient PDT under laser irradiation. Core-matched PPa nano-assembly was formed by the modification of PPa-PEG2K. After PPa/PPa-PEG2K NPs are delivered to tumor tissues and endocytosed by tumor cells to exert effective photodynamic therapy under laser irradiation.

Materials and methods

Materials

PPa was purchased in Shanghai Dibo Biotechnology Co., Ltd. Cell culture medium, penicillin, streptomycin and fetal bovine serum (FBS) were all obtained from GIBCO, Invitrogen Corp. (Carlsbad, CA, USA). MTT, trypsin, Singlet Oxygen Sensor Green Reagent (SOSG) and Reactive Oxygen Species Assay Kit (DCFH-DA) were purchased from Dalian Meilun Biological Technology Co., Ltd. Hoechst 33342 was purchased from BD Biosciences. Cell culture consumables were purchased by Wuxi NEST Biotechnology Co., Ltd. PCL-PEG2K was obtained from Shanghai Ponsure Biological Technology Co., Ltd. PPa-PEG2K was obtained by our previous synthetic method. Other reagents were of analytical grade.

Preparation of PPa nanoassemblies

The non-PEGylated and PEGylated PPa NPs (PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs) were prepared using a one-step nanoprecipitation method. Briefly, PPa (1 mg) were dissolved in 200 μL of a mixed solution of tetrahydrofuran and ethanol (2:3, v/v). Then, the mixed solution was slowly added dropwise to 2 mL of water at a speed of 1200 rpm (Gongyi Yingyu Yuhua Instrument Factory, DF-101S, Shanghai, China). After stirring for 3 min, the organic reagents were removed by a vacuum rotary evaporator (YRE-5299, Yuhua, Gongyi, China) at room temperature. The final solution volume was made up to 2 mL. Similarly, the other two PEGylated nanosystems were prepared in the same procedure dissolving PPa with 10 mol% PCL-PEG2K or 10 mol% PPa-PEG2K in a mixed organic reagent. These prepared NPs were stored at 4 °C.

Characterization of PPa nanoassemblies

The loading rate of PPa was calculated by the following Eqs. (1), (2):where MPPa is average molecular weight of PPa; is average molecular weight of PCL-PEG2K; is average molecular weight of PPa-PEG2K.

Diameters and zeta potentials of the prepared NPs were measured by a Malvern Zetasizer (Nano ZS, UK). Morphological investigation was carried out using a transmission electron microscopy (Hitachi, HT7700, Japan). The samples were stained by phosphotungstic acid. In order to verify the stability of PEGylation on the NPs, non-PEGylated PPa NPs, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs were added dropwise to a phosphate buffer solution (PBS, pH 7.4), and the morphological changes of the NPs were observed. In order to further investigate the stability of the two PEGylated NPs, the changes of the NPs incubated in PBS containing 10% of FBS were determined at each time point in 37 °C. To analyze the effect of laser irradiation on the stability of PPa-nanoassembly, changes in particle size of PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs incubated in PBS (pH 7.4, 37 °C) were investigated under different laser irradiation time (660 nm, 200 mW/cm2) or without irradiation. The diameters of PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs after laser treatment were measured by a Malvern Zetasizer (Nano ZS).

Self-assembly mechanisms

Exploration of the self-assembly mechanisms of PPa molecules and analysis of intermolecular interactions were performed by computer simulations. Molecular docking calculations were conducted using the Vina protocol in Yinfo Cloud Platform. The 3D structure of PPa was constructed with energy minimization in MMFF94 force field. AutoDock Vina program was utilized to perform semi-flexible docking with maximum nine poses output after internal clustering.

To further explore self-assembly mechanisms of PPa-nanoassembly, non-PEGylated PPa NPs, PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs were added to the solutions containing NaCl (50 mmol/L), SDS (50 mmol/L) or urea (50 mmol/L), and were incubated in a shaker in 37 °C. The particle size changes of non-PEGylated PPa NPs, PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs were measured by a Malvern Zetasizer (Nano ZS).

Ultraviolet and fluorescence spectra

The PPa ultraviolet (UV) absorbance spectra of PPa solution (PPa dissolved in DMSO), PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs (1 μmol/L, PPa equivalent) were measured by ultraviolet spectrophotometer (UV1102II, China). The PPa fluorescence spectra of PPa solution, PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs (1 mmol/L, PPa equivalent) were characterized using a microplate reader (Thermo Scientific, USA).

In vitro drug release

The in vitro drug release of PPa from the nanoassembly was explored using a common dialysis method. PBS (pH 7.4) containing 20% tetrahydrofuran was used as the release medium. First, a dialysis bag containing 1 mL of PPa solution, PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs were added to 30 mL release medium in a shaker (100 rpm) at 37 °C. PPa solution was prepared by dissolving PPa in a mixed solution of cremophor EL and absolute ethyl alcohol (50/50, v/v) and diluted by deionized water as dispersion medium. Then, 1 mL of the release medium was taken out at predetermined timepoints, and was diluted with DMSO and measured using a microplate reader with excitation at 415 nm and emission at 675 nm (Thermo Scientific). Meanwhile, 1 mL of fresh release medium was supplemented to the release medium.

In vitro singlet oxygen generation

The production of singlet oxygen under laser irradiation was detected by SOSG. Briefly, PPa solution, PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs (1 mmol/L, PPa equivalent) mixed with SOSG (1 mmol/L) were diluted in 1 mL of deionized water as dispersion medium. PPa solution was prepared by dissolving PPa in DMSO and diluted by deionized water as dispersion medium. The amount of singlet oxygen generated in these samples was detected under different laser irradiation time (660 nm, 200 mW/cm2) or without irradiation. The fluorescence signal intensity was measured using a microplate reader with excitation at 504 nm and emission at 525 nm (Thermo Scientific).

Cell culture

Mouse breast cancer 4T1 cells were cultured in Gibco cell culture medium containing 10% FBS, penicillin and streptomycin. After the cells reached 85%, they were digested with trypsin for 3 min, replaced with fresh culture medium, and further cultivated.

Cellular uptake and cellular ROS detection

4T1 cells (1 × 105 cells/well) were seeded in 12-well plates and cultured for 24 h. Then, the medium was replaced with PPa solution, PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs and further cultured for 0.5 h or 2 h at 37 °C. PPa solution was prepared by dissolving PPa in DMSO and diluted by fresh cell culture medium. All the formulations contained a total PPa concentration of 50 nmol/L (of which PPa/PPa-PEG2K NPs, free PPa/PPa-PEG2K = 10/1, mol/mol). After incubation, they were washed three times with cold PBS, fixed with 4% cell tissue fixation solution for 15 min, then washed three times with cold PBS, stained with Hoechst for 10 min, and washed three times with cold PBS. Finally, the fluorescence intensity of cells was observed using a laser confocal microscope (Nikon C2, Japan).

For further quantitative analysis, 1 × 105 cells were seeded in 12-well plates, and after 24 h of incubation, they were replaced with fresh culture medium containing drugs (PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs) for 0.5 or 2 h. After incubation, the cells were washed, digested, and after being dispersed in PBS. Then, their fluorescence signal intensity was detected by FACSCalibur flow cytometer. Moreover, in order to eliminate the fluorescent interference of aggregation caused quenching (ACQ) effect and to further accurately evaluate the cell uptake efficiency of PPa-nanoassembly, cells incubated with PPa solution or PPa-nanoassembly were digested and collected. Then, the cells were further destroyed by ultrasonic crushing, and methanol was utilized to destroy the nanostructure and dissolve the fluorescent agent (PPa). Finally, the real fluorescent intensity within cells were quantitatively measured using a microplate reader (Thermo Scientific).

4T1 cells were seeded in a 24-well plate at a density of 5 × 104 cells per well. After 24 h of incubation, the medium was discarded, and the cells were further placed in a 37 °C cell culture incubator with a fresh medium containing PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs (20 nmol/L, PPa equivalent) and continued to cultivate. After 4 h of incubation, the drug-containing medium was removed, and the serum-free 1640 medium containing DCFH-DA (20 mmol/L) was further incubated for 0.5 h. After the incubation, the cells were irradiated with laser (5 min, 660 nm, 60 mW/cm2), and the unirradiated cells were used as the control group. Cells were then washed three times with cold PBS, and the fluorescence signal of DCF in the cells was observed using an inverted fluorescence microscopy with an Ex/Em of 488/525 nm. After a similar procedure, the fluorescence intensity was quantified by flow cytometry.

Cytotoxicity evaluation

In order to test the anti-tumor activity of several preparations in vitro, 1 × 103 cells were seeded in 96-well plates and incubated for 24 h, then replaced with fresh culture medium containing drug (PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs) and incubated for 4 h. After that, the drug-containing medium was washed away with cold PBS and added fresh culture medium or continued the culture without removing the previous drugs, and then irradiated with laser (5 min, 660 nm, 60 mW/cm2), and continued to incubate for 44 h. Finally, MTT was used to evaluate cell viability.

Animal studies

All experimental procedures were executed according to the protocols approved by Shenyang Pharmaceutical University Animal Care and Use Committee.

Pharmacokinetics

Male Sprague–Dawley rats (200–220 g) were utilized to explore the pharmacokinetics of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs. PPa solution was prepared by dissolving PPa in a mixed solution of cremophor EL and absolute ethyl alcohol (50/50, v/v) and diluted by PBS. The animals were fasted for 12 h before administration. PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs were respectively injected into the tail vein of rats (1 mg/kg, PPa equivalent) At predetermined intervals, blood samples were collected and immediately centrifuged at 10,000 rpm (Shanghai Anting Scientific Instrument Factory, TGL-16B, Shanghai, China) for 10 min to obtain plasma. Plasma samples were stored at −20 °C for the following analysis. PPa in plasma was extracted by a methanol protein precipitation method. Briefly, 50 μL plasma sample was added to 250 μL methanol and vortexed for 3 min. After vortex, the solution was centrifuged at 5000 rpm (Shanghai Anting Scientific Instrument Factory) for 5 min to obtain free PPa. Then, the concentrations of PPa in various formulations were measured using a microplate reader with excitation at 415 nm and emission at 675 nm (Thermo Scientific).

In vivo real-time and ex vivo biodistribution

A 4T1 tumor-bearing BALB/c mouse model was established by subcutaneously injecting 100 μL of tumor cells (1 × 107 cells/mL) into the right back side of the mice (20–22 g). When the tumor volume reached 300 mm3, the mice were randomly divided into three groups (n = 3). PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs (1 mg/kg, PPa equivalent) were injected intravenously into mice. At predetermined times (2, 4, 6 and 12 h), real-time fluorescent images of the living were taken using the IVIS spectroscopy small animal imaging system. Additionally, for the quantitative analysis of major organs and tumor tissues, a new batch of 4T1 tumor-bearing BABL/c mice were established and treated with the same dose of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs (1 mg/kg, PPa equivalent). The mice were sacrificed at predetermined time intervals (2, 4, 6 and 12 h) after intravenous injection. The hearts, livers, spleens, lungs, kidneys and tumors were collected and analyzed using the IVIS spectroscopy small animal imaging system. Moreover, high intensity ultrasound was used to destroy the cellular structures of main organs and tumors, and methanol was utilized to destroy the nanostructures of nanoassemblies and to dissolve PPa for the precise quantitative analysis using a microplate reader with excitation at 415 nm and emission at 675 nm (Thermo Scientific).

In vivo antitumor activity

The BALB/c mice bearing 4T1 xenograft tumors were used for the evaluation of antitumor in vivo. When the tumor volume grew to about 150 mm3, the established tumor-bearing mice were randomly divided into 4 groups (n = 5). PPa solution, PPa/PCL-PEG2K NPs or PPa/PPa-PEG2K NPs were injected via the tail vein of mice at a dose of 4 mg/kg (PPa equivalent) and taken once every other day. After 4 h of administration, the set light group was given a laser (5 min, 660 nm, 200 mW/cm2). The non-light group was placed in a dark place to avoid being activated. Tumor volumes and body weights of mice were measured and recorded daily.

After the in vivo anti-tumor experiment was completed, blood was collected from the tumor-bearing mice. The collected blood was centrifuged to take the upper serum and sent to the blood analysis center to measure liver and kidney function indicators. The main indicators included: alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urine nitrogen (BUN) and creatinine (CRE).

After the tumor-bearing mice were sacrificed, mice were dissected to obtain the main organs and tumors. After being washed three times with PBS, 4% paraformaldehyde solution was used to treat the organ and tumor tissues. Then, the samples were sent to the center of tissue section, embedded in paraffin, frozen section, and finally stained. The sections were observed using an inverted biological microscopy (Nikon, Japan).

Statistical analysis

All the data were displayed as mean value ± standard deviation. Student's t-test and one-way analysis of variance were applied to statistical difference, and P < 0.05 was regarded as significant difference.

Results and discussion

Self-assembly of PPa

Although a variety of PPa-encapsulated nano-DDS have been widely investigated for PDT, the self-assembly capacity of PPa hasn't been found yet. In the present study, we found that pure PPa could self-assemble into NPs by using a facile nano-precipitation method. When PPa dissolved in a mixed solution of tetrahydrofuran and ethanol was added dropwise to water, nanoassembly occurred spontaneously even without the help of any surfactant (Supporting Information Fig. S1). Tetrahydrofuran and ethanol were removed by a vacuum rotary evaporator at room temperature. The particle size of PPa-nanoassembly was about 100 nm, with a zeta potential of around −20 mV (Fig. 2B and Supporting Information Table S1). By contrast, PPa powder couldn't be dissolved in water and precipitated immediately due to its poor water solubility (Supporting Information Fig. S1).

Figure 2

Self-assembly of PPa and in vitro singlet oxygen production efficiency of formulations. (A) Self-assembly simulation of PPa in water. (B) TEM images of non-PEGylated PPa NPs, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs (scale bar represents 200 nm). (C) UV absorption spectra at 250–800 nm wavelength of PPa solution, non-PEGylated PPa NPs, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs. (D) Changes in particle size of PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs with different laser time. (E) In vitro accumulative drug release curves of PPa from PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs in PBS containing 20% tetrahydrofuran within 24 h. (F) Fluorescence intensity of singlet oxygen produced by formulations in vitro (parameters of microplate reader: excitation at 415 nm and emission at 675 nm). Data are shown as the mean ± SD, n = 3; ∗P < 0.05, ∗∗P < 0.01.

In order to further strengthen the nanostructure of PPa-nanoassembly, an amphiphilic polymeric PEG2K-PPa was synthesized according to our previous method. PEG2K-PPa was utilized to prepare a uniquely core-matched nanostructure (PPa/PPa-PEG2K NPs). Another amphiphilic polymer PCL-PEG2K with similar molecular weight was used as control. Finally, PPa-PEG2K (10%, mol/mol) and PCL-PEG2K (10%, mol/mol) were used to form a hydrophilic shell in the fabrication of PPa/PPa-PEG2K NPs and PPa/PCL-PEG2K NPs, respectively. As shown in Fig. 2B and Supporting Information Table S1, the drug loading rates of PPa in PPa/PPa-PEG2K NPs and PPa/PCL-PEG2K NPs were 74.8% and 68.3%, respectively. The slightly higher drug loading capacity of PPa/PPa-PEG2K NPs should be attributed to the contribution of the PPa in PPa-PEG2K polymer.

The self-assembly mechanisms of PPa were explored through molecular docking simulation. As shown in Fig. 2A, PPa molecules could form nano-aggregates in a stable thermodynamic state. Notably, the multiple pyrrole hybrid structures of PPa endowed strong π‒π stacking effect and hydrophobic force among PPa molecules, which made a significant contribution to the formation of self-assembled NPs (Fig. 2A). Additionally, as shown in Fig. 2C and Supporting Information Fig. S4, obvious red shift was observed in the UV absorbance spectra and fluorescence spectrum of non-PEGylated PPa NPs, PPa/PPa-PEG2K NPs and PPa/PCL-PEG2K NPs when compared with PPa solution. However, the fluorescence intensity of PPa-nanoassembly decreased when compared with PPa solution, due to the ACQ effect and the exciton migration during π‒π stacking interaction between PPa molecules and the molecular conformation change in process of assembly,,. Notably, when compared with non-PEGylated PPa NPs, the fluorescence intensity of PPa/PCL-PEG2K NPs further decreased, but no fluorescent reduce was observed in PPa/PPa-PEG2K NPs group. These results indicated that PPa/PPa-PEG2K NPs showed advantages in relieving the ACQ effect.

To further explore the self-assembly mechanisms of PPa-nanoassembly, SDS, urea and NaCl were utilized to verify the existence of hydrophobic interaction, hydrogen-bond interaction and electrostatic interaction in the PPa-nanoassembly. As shown in Supporting Information Fig. S5, non-PEGylated PPa NPs, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs showed no significant change in particle size after incubation with urea, but their particle size obviously increased after incubation with SDS. Notably, the particle size of non-PEGylated PPa NPs treated with sodium chloride significantly increased when compared with PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs. These results demonstrated that multiple interactions and forces were involved in the self-assembly process of PPa. Moreover, PEGylation modification significantly improved the colloidal stability of PPa-nanoassembly.

Stability of PPa-nanoassembly

The colloidal stability of NPs is extremely important for efficient drug delivery. As shown in Supporting Information Fig. S2, the non-PEGylated PPa NPs quickly precipitated in PBS due to the salting-out effect. In contrast, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs could remain good colloidal stability in PBS even for one week. Notably, PPa/PPa-PEG2K NPs showed much better colloidal stability in PBS containing 10% FBS than PPa/PCL-PEG2K NPs (Supporting Information Fig. S3). The excellent colloidal stability of PPa/PPa-PEG2K NPs should be attributed to the core-matched stabilizing effect of PPa-PEG2K, which could form a more stable core-shell nanostructure due to the π‒π stacking effect and hydrophobic interaction between the PPa core and the PPa-PEG2K shell. These results suggest that core-matched PEGylating modification could be utilized as an efficient strategy to significantly increase the colloidal stability of pure drug-driven nanoassembly. Which would benefit the efficient drug delivery of PPa-nanoassembly in vivo.

We then investigated the impact of laser irradiation on the colloidal stability of PPa-nanoassembly by determining the changes in particle size changes with/without laser irradiation. As shown in Fig. 2D, the particle size of PPa/PPa-PEG2K NPs significantly increased within 4 min under laser irradiation. By contrast, there was almost no significant increase in the particle size of PPa/PCL-PEG2K NPs exposed to the laser even for a longer irradiation time (8 min). Obviously, PEGylation modification on PPa-nanoassembly could significantly increase the colloidal stability. But only PPa/PPa-PEG2K NPs showed laser-triggered disassembly phenomenon, owing to the photobleaching of the PPa fraction in PPa-PEG2K polymer under laser irradiation. As a result, the amphipathic structure of PPa-PEG2K polymer was broken, resulting in the decrease of PEGylation stabilization on PPa-nanoassembly. In contrast, laser exerted marginal influence on PCL-PEG2K. As a result, PCL-PEG2K could maintain the good stability of PPa/PCL-PEG2K NPs with or without laser treatment. The slight change (∼30 nm) in particle size of PPa/PCL-PEG2K NPs exposed to laser treatment should be contributed to the gradual photobleaching effects on the PPa core. These results illustrate that core-matched PPa/PPa-PEG2K NPs could not only significantly improve the colloidal stability of PPa-nanoassembly, but also demonstrated laser-triggered destabilization of the nanoassembly. Such a laser-triggered destabilization is expected to facilitate its photoconversion and ROS generation efficiency.

As shown in Fig. 2E, the release rate of PPa from PPa solution was much faster than that from both PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs. More than 60% of PPa was released from PPa solution within 4 h, while only about 20% of PPa was released from PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs under the same conditions. These results indicate that PPa solution showed an initial burst release phenomenon, but PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs revealed sustained drug release features.

The in vitro singlet oxygen generation efficiencies of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs were investigated using SOSG. PPa solution was prepared by dissolving PPa in DMSO and diluted by deionized water. As shown in Fig. 2F, the amount of singlet oxygen produced by the three formulations was extremely low without laser irradiation, but the fluorescence intensity of SOSG significantly increased in an irradiation time-dependent manner. Under laser irradiation, the amount of singlet oxygen produced by PPa solution was somewhat higher than that of PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs. Which should be attributed to the aggregation-caused reduction of photo-induced ROS generation in PPa-nanoassembly. Notably, although there was no significant difference in the fluorescence intensity of PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs exposed to the laser within 2 min, the amount of singlet oxygen produced from PPa/PPa-PEG2K NPs was much more than that from PPa/PCL-PEG2K NPs with the laser exposure time extended to 4 or 8 min. These results reveal that short time laser treatment triggered the rapid increase of particle size and the formation of the fluffy state of PPa/PPa-PEG2K NPs, significantly relieving the aggregation-caused reduction of photo-induced ROS generation (Fig. 2D and F). As a result, singlet oxygen generation from PPa/PPa-PEG2K NPs was much higher than that from PPa/PCL-PEG2K NPs when exposed to a longer laser irradiation time. Moreover, when compared with non-PEGylated PPa NPs, the fluorescence intensity of PPa/PCL-PEG2K NPs decreased, but there's no fluorescent reduce observed in PPa/PPa-PEG2K NPs group (Supporting Information Fig. S4). These results indicate that PPa/PPa-PEG2K NPs showed advantages in relieving the aggregation-caused reduction of photo-induced ROS generation and ACQ effect of fluorescence. These results are in well consistence with the fluorescence spectra and the colloidal stability results (Fig. 2D and Supporting Information Fig. S4).

Cellular uptake and intracellular ROS generation

The inherent fluorescence of PPa makes it convenient to monitor the cellular uptake of the nanoassembly. Moreover, the fluorescence spectra of PPa-PEG2K had no significant change compared with PPa, making it convenient to accurately calculate the molar dose of PPa. The cellular uptake was explored using 4T1 breast cancer cells treated with PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs (50 nmol/L, PPa equivalent), As shown in Fig. 3A‒D and Supporting Information Fig. S6, the cellular uptake of all formulations demonstrated a time-dependent manner, but PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs showed much higher cellular uptake efficiency than PPa solution at both 0.5 and 2 h. In addition, there's no obvious difference in the fluorescence intensity between PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs, due to their very similar nanostructures. These results suggest that PPa-nanoassembly could significantly improve the cellular internalization of PPa, probably due to nanoparticle-mediated endocytosis,.

Figure 3

Cellular uptake of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs. (A) Confocal laser scanning microscopy (CLSM) images at 0.5 h or (B) 2 h. (C) Flow analysis diagram by flow cytometry (a: PPa/PPa-PEG2K NPs 2 h, b: PPa/PCL-PEG2K NPs 2 h, c: PPa solution 2 h, d: PPa/PPa-PEG2K NPs 0.5 h, e: PPa/PCL-PEG2K NPs 0.5 h, f: PPa solution 0.5 h, g: Blank). (D) Quantitative analysis for the fluorescent intensity by flow cytometry. Data are shown as the mean ± SD, n = 3; ∗P < 0.05, ∗∗P < 0.01. Scale bar = 10 μm.

We further detected the intracellular ROS generation capacity of PPa-nanoassembly. An ROS probe (2′,7′-dichlorofluorescein diacetate, DCFH-DA) can be hydrolyzed by the intracellular esterase to generate DCFH, and the fluorescent DCF could be generated after the oxidization of DCFH by ROS. The fluorescence intensity of DCF was utilized for intracellular ROS detection. As shown in Fig. 4A and B, the fluorescence intensity of PPa/PPa-PEG2K NPs and PPa/PCL-PEG2K NPs were significantly higher than that of PPa solution, owing to the higher cellular uptake of PPa-nanoassembly. Notably, despite of their similar cellular uptake efficiency, PPa/PPa-PEG2K NPs exhibited higher intracellular ROS generation capacity than that of PPa/PCL-PEG2K NPs, which should be attributed to the relieved aggregation-caused reduction of ROS generation in PPa/PPa-PEG2K NPs under laser irradiation.

Figure 4

In vitro cellular ROS production efficiency and photodynamic cytotoxicity. (A) Fluorescence image of ROS produced in cells with PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs using inverted fluorescence microscope. (B) Quantitative analysis of fluorescence intensity using flow cytometry (a: PPa solution (+); b: PPa/PCL-PEG2K NPs (+); c: PPa/PPa-PEG2K NPs (+), d: Blank). (C) Cell viability treated with PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs. (D) Cell viability with removing the drug-containing culture medium before laser irradiation. Data are shown as the mean ± SD, n = 3. Scale bar = 10 μm.

Cytotoxicity

Higher cellular uptake efficiency of PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs might result in more potent cytotoxicity. Distinguished from other cytotoxic drugs, the cytotoxicity of PSs depends on the generation of ROS under laser irradiation. Notably, both the extracellular and intracellular PSs could produce ROS and damage the structures and functions of cells under laser treatment. As shown in Fig. 4C and Supporting Information Table S2, there's no significant difference in the cytotoxicity of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs without removing the drug-containing culture medium before laser irradiation. By contrast, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs showed more potent cytotoxicity against 4T1 cells than PPa solution when the drug-containing cell culture medium was replaced by fresh medium before laser irradiation (Fig. 4D and Supporting Information Table S2). Moreover, PPa/PPa-PEG2K NPs had a higher cytotoxicity than PPa/PCL-PEG2K NPs, which should be attributed to the higher ROS generation capacity of PPa/PPa-PEG2K NPs (Fig. 4A and B). As expected, all the three formulations showed negligible cytotoxicity without laser irradiation. These results suggested that both the extracellular and intracellular PPa contributed to the ROS-induced cytotoxicity against tumor cells. When the drug-containing cell culture medium was replaced by fresh medium before laser irradiation, the cellular uptake efficiency and ROS production capacity of PPa-nanoassembly exerted obvious impact on the cytotoxicity. As a result, PPa/PPa-PEG2K NPs, with good colloidal stability, favorable cellular uptake and efficient ROS production, showed more potent phototoxicity against 4T1 cells than PPa/PCL-PEG2K NPs and PPa solution.

The pharmacokinetic profiles of PPa solution, PPa/PPa-PEG2K NPs and PPa/PCL-PEG2K NPs were investigated in rats. The concentration–time curves and pharmacokinetics parameters were shown and summarized in Fig. 5 and Supporting Information Table S3, respectively. As expected, PPa/PPa-PEG2K NPs and PPa/PCL-PEG2K NPs significantly prolonged the systemic circulation time of PPa when compared with PPa solution. Notably, PPa/PPa-PEG2K NPs exhibited longer circulation time in blood than PPa/PCL-PEG2K NPs, which should be attributed to the core-matched PEGylating stabilization of PPa-PEG2K. These results illustrate that PEGylating modification could be of benefit to the in vivo drug delivery efficiency of PPa-nanoassembly. More importantly, such a core-matched PEGylated strategy could further extend the systemic circulation of PPa, thus facilitating preferable drug accumulation in tumors via the enhanced permeability and retention (EPR) effect.

Figure 5

Pharmacokinetics. Plasma concentration-time profiles of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs after a single intravenous administration. Data are shown as the mean ± SD, n = 5.

In vivo imaging and ex vivo biodistribution

As PPa is a near-infrared dye, its nanoassembly is inherently equipped with self-tracking capacity for imaging-guided drug delivery. The in vivo real-time imaging of PPa-nanoassembly was investigated in 4T1 tumor-bearing mice. As shown in Supporting Information Fig. S7, PPa solution was quickly cleared from the body with little tumor accumulation. By contrast, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs exhibited stronger fluorescence intensity in tumors than PPa solution. As expected, the highest fluorescence in tumor was observed at 4 h in mice treated with PPa/PPa-PEG2K NPs due to the better stability and longer circulation time in blood. Moreover, in order to more visually observe the fluorescence intensity of specific organs and tumors, the ex vivo biodistribution of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs was also explored in 4T1 tumor-bearing mice. As shown in Fig. 6A and B, and Supporting Information Figs. S8 and S9, the ex vivo biodistribution is consistent with the result of in vivo imaging. Therefore, the optimal laser irradiation time for in vivo antitumor activity assay should be 4 h after intravenous injection.

Figure 6

Ex vivo biodistribution. (A) Ex vivo fluorescent distribution images of PPa in rats after injection with various formulation at 4 h. (B) Quantitative analysis of the percent injected dose. Data are shown as the mean ± SD, n = 3; ∗P < 0.05, ∗∗P < 0.01. Scale bar = 50 mm.

Multiple drug delivery advantages make PPa/PPa-PEG2K NPs as a promising nanomedicine for antitumor treatment. The antitumor activity of PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs was evaluated in BALB/c mice bearing 4T1 tumor. When the tumor volume grew to about 150 mm3, PBS, PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs (2 mg/kg, PPa equivalent) were intravenously administrated once every other day for five injections, respectively. According to the in vivo and ex vivo biodistribution results, the laser treated groups received a tumor-localized light exposure (200 mW/cm2) for 5 min at 4 h post administration. As shown in Fig. 7A, C and Supporting Information Fig. S10, PPa solution showed a moderate inhibitory effect on tumor growth when compared with PBS. PPa/PCL-PEG2K NPs showed a better inhibition effect on tumor growth than that of PPa solution. Notably, PPa/PPa-PEG2K NPs exhibited much more potent anti-tumor activity than both PPa/PCL-PEG2K NPs and PPa solution, which could be attributed to multiply advantages of PPa/PPa-PEG2K NPs: (i) stable nanostructure by virtue of core-matched PEGylating stabilization; (ii) long systemic circulation and high tumor accumulation, (iii) efficient cellular uptake; and (iv) abundant generation of ROS under laser irradiation.

Figure 7

Evaluation on the in vivo antitumor activity in BALB/c mice bearing 4T1 tumors. (A) Changes of the tumor volume after PPa solution, PPa/PCL-PEG2K NPs and PPa/PPa-PEG2K NPs treatments. (B) Changes in body weight of mice during treatment. (C) Tumor burden (tumor weight/average body weight) after the last treatment. (D) Liver and kidney functional parameters after the last treatment. Data are shown as the mean ± SD, n = 5; ∗P < 0.05, ∗∗P < 0.01.

Moreover, there was no significant change in body weights of mice in each group throughout the treatment period (Fig. 7B). Moreover, there was no significant change found in the liver and kidney function parameters based on the serum analysis results (Fig. 7D). In the histological analysis of H&E section staining, there was no significant damage found in the sections of the main organs of each group (Supporting Information Fig. S11). In tumor tissue sections, the largest apoptosis and necrosis spread area was observed in the PPa/PPa-PEG2K NPs group. Therefore, PPa/PPa-PEG2K NPs hold great potential to be used as a promising nanoplatform with high efficiency and low toxicity.

Conclusions

In the present study, we reported a unique self-assembly phenomenon of PPa. To enhance the stability of PPa-nanoassembly, core-matched PEGylating modification was developed using a photodynamic PEG polymer (PPa-PEG2K) in the nano-formulation. The core-matched PEGylated nanoassembly (PPa/PPa-PEG2K NPs), with high drug loading capacity and colloidal stability, demonstrated long circulation time in blood, high tumor accumulation, efficient cellular uptake and potent cytotoxicity. As expected, PPa/PPa-PEG2K NPs performed much more potent antitumor activity in 4T1 tumor-bearing BALB/c mice than PPa/PCL-PEG2K NPs and PPa solution. This novel nanosystem could be utilized as a promising nanomedicine for high-efficient imaging-guided cancer therapy by integrating pure drug self-assembly technique and core-matched PEGylation stabilization into one platform.

Acknowledgments

This work was supported by Science and Technology Major Project of Liaoning (No. 2019JH1/10300004, China), the (No. 81773656 and 81703451), the Excellent Youth Science Foundation of Liaoning Province (No. 2020-YQ-06, China), the (No. 2020M670794) and the Liaoning Revitalization Talents Program (No. XLYC1907129 and XLYC1808017, China).

Author contributions

Shenwu Zhang, Cong Luo, Jin Sun and Zhonggui He conceived the work. Shenwu Zhang, Yuequan Wang, Zhiqiang Kong participated in the design and implementation of the experiment. Shenwu Zhang, Xuanbo Zhang, Bingjun Sun, Han Yu and Qin Chen carried out data processing and analysis. Shenwu Zhang and Cong Luo wrote and modified paper. All authors discussed and approved the manuscript.

Conflicts of interest

There are no conflicts to declare.