GSH-responsive SN38 dimer-loaded shape-transformable nanoparticles with iRGD for enhancing chemo-photodynamic therapy.

Accurate tumor targeting, deep penetration and superb retention are still the main pursuit of developing excellent nanomedicine. To achieve these requirements, a stepwise stimuli-responsive strategy was developed through co-administration tumor penetration peptide iRGD with shape-transformable and GSH-responsive SN38-dimer (d-SN38)-loaded nanoparticles (d-SN38@NPs/iRGD). Upon intravenous injection, d-SN38@NPs with high drug loading efficiency (33.92 ± 1.33%) could effectively accumulate and penetrate into the deep region of tumor sites with the assistance of iRGD. The gathered nanoparticles simultaneously transformed into nanofibers upon 650 nm laser irradiation at tumor sites so as to promote their retention in the tumor and burst release of reactive oxygen species for photodynamic therapy. The loaded d-SN38 with disulfide bond responded to the high level of GSH in tumor cytoplasm, which consequently resulted in SN38 release and excellent chemo-photodynamic effect on tumor. In vitro, co-administering iRGD with d-SN38@NPs+laser showed higher cellular uptake, apoptosis ratio and multicellular spheroid penetration. In vivo, d-SN38@NPs/iRGD+laser displayed advanced penetration and accumulation in tumor, leading to 60.89% of tumor suppression in 4T1 tumor-bearing mouse model with a favorable toxicity profile. Our new strategy combining iRGD with structural transformable nanoparticles greatly improves tumor targeting, penetrating and retention, and empowers anticancer efficacy.

Introduction

Progress in developing effective nanomedicines has been impeded by heterogeneity of the enhanced permeability and retention (EPR) effect in different tumor type, location and mass size1, 2, 3, 4. The tumor permeability and retention of nanocarriers are not only related to their size, but also closely associated with their shape and chemistry surface5, 6, 7, 8. However, it is worth noting that permeability and retention in “static” nanoparticles are often mutual contradictory,. To address the dilemma and compensate the poor EPR efficacy, shape transformable nanoparticles that can achieve in situ morphology change have attracted our attention11, 12, 13, 14, 15. Typically, the nanosphere offers the vehicle a proper stability to circulate and distribute to the tumor, while the nanofiber confers it a superb retention capability around and inside tumor,,. In light of this, convertible nanocarriers that can transform the morphology of nanospheres into nanofibers may fuel the nanotherapy outcomes.

Recently, transformable linear chimeric triblock molecules Ce6/BR-FFVLK-PEG were designed by our lab. The hydrophobic head [chlorin e6 (Ce6) or bilirubin (BR)], and hydrophilic tail [polyethylene glycol (PEG)] were tandem conjugated with Phe-Phe-Val-Leu-Lys (FFVLK) for forming spherical micelles. Notably, the formulation remained in spherical when circulated and distributed into the tumor. Upon the controllable 650 nm laser irradiation at tumor site, (1) the photosensitizer Ce6 was activated and reactive oxygen species (ROS) were generated for photodynamic therapy (PDT); (2) the strong and stable intermolecular forces of the micelles were broken,. Subsequently, spherical nanoparticles were transformed into nanofibers, which markedly facilitate their retention inside the tumor. However, to enlarge the therapeutic potency upon PDT, it is still necessary to combine with other strategies. On the other hand, the accumulation of this shape-transformable formulation in the tumor site mainly depended on the passive EPR effect, resulting in relatively low targeting efficacy. Active targeting provides a noteworthy additional benefit for high tumor location, which is the premise for nanomedicine to play the designed role,. In this context, constructing an actively targeted shape-transformable nanomedicine for combinational therapy is a feasible strategy to maximize the treatment outcome.

Mounting studies have shown that the combination of PDT and chemotherapy, namely chemo-photodynamic therapy, can increase susceptibility of tumor cells towards chemotherapeutics, thus producing better synergistic antitumor effect,24, 25, 26. To achieve the goal of chemo-photodynamic therapy, the chemotherapeutics should be released with locally confined treatment modalities. Additionally, stimuli-responsive drug release at the right place and right time plays a paramount role for reducing systemic toxicity of chemotherapeutics. Equipped with various responsive bonds to stimuli, including redox, pH, enzyme, light and temperature, dimeric prodrug has emerged as one of the most promising strategies to fulfill the requirement of on-demand drug release27, 28, 29, 30. Moreover, it has documented that dimeric prodrugs can also help to address the low drug loading efficiency in amphiphilic copolymers by increasing intermolecular hydrophobic interaction between molecules,. 7-Ethyl-10-hydroxy camptothecin (SN38), the biologically active metabolite of irinotecan (CPT-11), has proven to be approximately 1000-fold more cytotoxic than CPT-11 by inhibiting DNA topoisomerase,. The relatively high toxicology as well as the unfavorable pharmacology of SN38 limited its direct application,. In this perspective, SN38 dimer (d-SN38) linked via disulfide bond was synthesized and selected as the payload. The disulfide bond of d-SN38 was stable in bloodstream but can be efficiently cleaved by intracellular thiols, such as the high level of glutathione (GSH) in cytoplasm of tumor cells,. Therefore, loading d-SN38 in the above shape-transformable nanocarrier is capable to attain a better antitumor effect.

To endow active targeting property to the formulation, targeting ligands were widely utilized due to their specific interaction with receptors that overexpressed in tumor. However, the modification of particles with ligand makes the preparation complex, and it may influence the responsiveness,. Therefore, co-injection particles with targeting enhancer are a promising strategy. iRGD (cRGDKGPDC), a widely used targeting enhancer with tumor-penetrating capacity, homes to tumors by initially binding to αvβ3 integrin receptors that are specifically expressed in tumor vessels and tumor cells. Subsequently, a C-end rule motif with specific affinity with neuropilin-1 (NRP-1) is produced through proteolytic cleavage,. When co-administered with drug or nanoparticles, iRGD can facilitate the penetration into extravascular tumor tissue. This co-administration strategy provides a versatile way to greatly simplify the path to clinical application. As such, we utilized the active tumor-specific tissue penetration feature conferred on iRGD to enhance cancer targeted drug delivery by co-injection.

Herein, to boost the tumor targeting, penetration and retention, we established a stepwise multiple stimuli-responsive strategy through co-administration iRGD with GSH-responsive SN38 dimer-loaded shape-transformable Ce6/BR-FFVLK-PEG micelles (d-SN38@NPs/iRGD) for chemo-photodynamic therapy of breast cancer (Scheme 1). The particle size, entrapment efficiency, GSH responsiveness, stability and drug release properties of d-SN38@NPs were characterized. Cellular uptake and cytotoxicity by the d-SN38@NPs/iRGD combined with PDT (d-SN38@NPs/iRGD+laser) were evaluated in mouse breast cancer 4T1 cells by both quantitative and qualitative methods. Their targeting and antitumor efficacies in tumor-bearing mice were determined by in vivo imaging and tumor growth curve. Overall, d-SN38@NPs/iRGD+laser showed superior tumor-targeting efficiency and excellent therapeutic effects.

Scheme 1

The schematic illustration of the preparation, drug delivery and tumor response of d-SN38@NPs/iRGD.

Materials and methods

Materials

iRGD and FFVLK-(PEG8)3 (MW = 1923.2 Da) were synthesized by Shanghai Dechi Biosciences Co., Ltd. (Shanghai, China). Ce6 was purchased from Dalian Meilun Biotech Co., Ltd. (Dalian, China). Bilirubin was gained from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). d-SN38 was supplied by Prof. Jun Cao (Sichuan University, China). 4′,6-Diamidino-2-phenylindole (DAPI) and Lysotracker green was purchased from Life Technologies (Grand Island, USA). FITC Annexin V Apoptosis Detection Kit I was obtained from BD Pharmingen™ (Franklin Lakes, NJ, USA). Calcein-AM/PI Double Stain Kit was supplied by Yeasen (Shanghai, China). 24-well hanging cell culture inserts (8.0 μm, PET) were obtained from Corning Inc. (Kennebunk, ME, USA). All other reagents and solvents were purchased from commercial resources and used without further purification unless otherwise noted. Mouse mammary breast tumor cell line 4T1 was obtained from Chinese Academy of Sciences Cells Bank (Shanghai, China). 4T1 cells were incubated in RPMI-1640 cell culture medium supplemented with 10% fetal bovine serum (FBS) and 100 U/mL streptomycin, and 100 U/mL penicillin at 37 °C in a humidified 5% CO2 atmosphere.

Animals

Female BALB/c mice (5–6 weeks, 18 ± 2 g) and nude mice (4–5 weeks, 18 ± 2 g) were supplied by Dashuo Biotechnology Co., Ltd. (Chengdu, China). All animal experiments were conducted according to the guidelines of the Ethics Committee of Sichuan University, Chengdu, China.

Preparation and characterization of d-SN38@NPs

Ce6-FFVLK-PEG and BR-FFVLK-PEG were synthesized and a molar ratio of 1:1 Ce6/BR was chosen for NPs preparation with notable and sensitive morphological change ability of nanofibers according to our previously report. The Ce6-FFVLK-PEG and BR-FFVLK-PEG were dissolved in chloroform (0.3 mL) and mixed with d-SN38 solution (4 mg in 0.1 mL of chloroform). Then the mixture was added into ultrapure water (7 mL) and subsequently emulsified using ultrasonic cell crusher (65 W, 5 s/5 s a cycle for 5 min). The suspension was transferred into rotary evaporation to remove chloroform, yielding uniform-sized d-SN38@NPs. The same procedure was followed to prepare blank nanoparticle (NPs) without d-SN38.

The size and morphology changes of d-SN38@NPs before and after receiving 650 nm laser irradiation (200 mW/cm2 for 5 min, the same below unless otherwise stated) were monitored by dynamic light scattering (DLS, Malvern Panalytical, Malvern Zetasizer Nano ZS, Malvern, UK) and measured by transmission electronic microscopy (TEM, Hitachi, H-600, Tokyo, Japan). For stability evaluation, d-SN38@NPs was incubated in medium containing 10% and 50% FBS at different temperatures (37, 25 and 4 °C) and pH values (7.4 and 6.5). The size changes were monitored for 24 h by DLS. Besides, to confirm the responsiveness of d-SN38 to GSH in tumor cytoplasm, d-SN38 was incubated with GSH at 10 × 10−3 mol/L, and the fluorescence spectrum change was detected by spectrofluorophotometer (Shimadzu, RF-5301PC, Tokyo, Japan) at 1, 2, 4 and 8 h.

The free drug was removed by centrifugal filters (10 K MWCO, Millipore Corp, MIlli-Q, Billerica, MA, USA), then concentration of d-SN38 in particles was determined with microplate reader (Thermo Scientific, Varioskan Flash, Waltham, MA, USA) at Ex/Em of 375 nm/427 nm. The encapsulation efficiency and loading capacity were obtained according to the following Eqs. (1), (2):

In vitro release of d-SN38@NPs

For in vitro release study of d-SN38@NPs, d-SN38@NPs (0.7 mL, 1 mg/mL) receiving irradiation or not were resuspended in PBS (7 mL) containing 0.5% Tween 80 at pH 7.4 or pH 6.5 respectively, then dialyzed for 48 h under 37 °C. 1 mL of release medium was taken out in each sample at 0.5, 1, 2, 4, 8, 12, 24 and 48 h, while an equal volume of fresh media was supplemented at certain time intervals. The fluorescent intensity of d-SN38 from different formulations in pH values 7.4 and 6.5 were determined at different time points.

ROS generation

4T1 cells were seeded at a density of 1 × 105 cells per well in 6-well plates. After grown to 80% confluent, NPs, d-SN38@NPs and d-SN38@NPs/iRGD were added into the wells at the same Ce6 concentration (5 μg/mL) for 2 h, respectively. The cells treated with PBS were set as negative control. After receiving 650 nm irradiation for 2 min, the ROS level was tested by ROS Assay Kit (10 μmol/L DCFH-DA, 35 min). Then the cells were washed, fixed and imaged by confocal laser scanning microscope (CLSM, Olympus, FV1000, Central Valley, PA, USA).

Cellular uptake

4T1 cells were inoculated at a density of 3 × 105 in a confocal dish (35 mm × 10 mm) and grown for 24 h. Thereafter, NPs and NPs/iRGD were added into the wells at the same Ce6 concentration (5 μg/mL), respectively. The concentration of iRGD was 43.3 μg/mL in NPs/iRGD group. One hour later, part of NPs- and NPs/iRGD-treated cells were carried out to receive 650 nm irradiation for 2 min. All treatment groups were maintained for another 1 h, then washed with cold PBS and fixed with 4% paraformaldehyde. To stain nuclei, the cells were incubated with 0.5 μg/mL of DAPI for 5 min. Finally, the cells were imaged by CLSM (Olympus).

For quantitative studies, 4T1 cells were seeded in 6-well plates at a density of 3 × 105 cells per well. After 24 h, different formulations were added to the plates and the cells were treated as described above. After 2 h incubation, the cells were harvested. The fluorescence intensity of Ce6 was detected by flow cytometry (Beckman Coulter, FC500, Brea, CA, USA).

Penetration and retention in multicellular spheroid

To prepare multicellular spheroids, 4T1 cells (1 × 103 cells per well) were seeded in 2% (w/v) low-melting point agarose coated 96-well plates. After seven days, the compacted and uniform spheroids were selected for penetration study. The tumor spheroids were transferred to a confocal dish and incubated with NPs and NPs/iRGD at an equivalent concentration of Ce6 (5 μg/mL), respectively. Four hours later, part of spheroids was taken out to receive 650 nm irradiation for 2 min. All spheroids were maintained for another 2 h, and then the spheroids were rinsed and fixed. Z-stack images of spheroids from the top to 100 μm were collected using CLSM (Olympus).

Cytotoxicity study

Firstly, MTT assay was carried out to investigate the cytotoxicity of different formulations. 5 × 103 per well of 4T1 cells were plated and cultured in 96-well plates for 24 h. Ce6, d-SN38, NPs, d-SN38@NPs and d-SN38@NPs/iRGD were added into each well. The final concentrations of d-SN38 and Ce6 ranged from 0 to 15 and 2.5 μg/mL, respectively. The d-SN38@NPs/iRGD group was coincubated with iRGD from 0 to 129.9 μg/mL. In the laser irradiation group, cells were stimulated at 12 h after administration and then cultured for another 12 h. The cell viability of these formulations was examined by MTT.

Secondly, the cytotoxicity was evaluated by Calcein-AM/PI double stain method. In brief, 4T1 cells (1 × 105 per well) were grown in 24-well plates as above for 24 h. Free Ce6, free d-SN38, NPs, d-SN38@NPs and d-SN38@NPs/iRGD were added to each well with 0.1 μg/mL of Ce6, 0.5 μg/mL of d-SN38 and 4.33 μg/mL of iRGD correspondingly. After 4 h of culture, half of the cells received 650 nm irradiation for 2 min. Cells incubated with fresh culture medium was set as control. After another 4 h of incubation, the cells were washed and stained by Calcein-AM (5 μg/mL) and PI (10 μg/mL) for 30 min. Finally, the stained cells were imaged by fluorescent microscope (Shimadzu).

To assess the cell apoptosis induced by the architected formulations, 4T1 cells (5 × 104 per well) were plated in 12-well plates. The cells were treated with Ce6, d-SN38, NPs, d-SN38@NPs and d-SN38@NPs/iRGD followed by the same irritation schedule as MTT assay. The concentrations of Ce6, d-SN38 and iRGD were 0.05, 0.25 and 2.165 μg/mL, respectively. After 24 h incubation, the cells were harvested and stained with Annexin V-FITC and PI. Finally, the cell apoptosis in different groups was analyzed by flow cytometer (Beckman Coulter).

Wound healing assay

For wound healing assay, 5 × 105 4T1 cells were seeded in 12-well plates and incubated until 90% confluence. The confluent cell monolayers were wounded with a 10-μL pipette tip to get two cell islands, then washed and incubated with d-SN38@NPs or d-SN38@NPs/iRGD at a concentration of 0.25 μg/mL Ce6 for another 24 h. In the laser irradiation group, laser irradiation was given after 12 h. The wound healing area was photographed at the beginning (0 h) and 24 h with a microscope. The cell motility was calculated according to the following Eq. (3) :

Pharmacokinetics study

Female BALB/c mice (n = 3) were intravenously injected with d-SN38@NPs and d-SN38@NPs/iRGD at a dosage of 5 mg/kg of Ce6, respectively. 40 μL of blood sample was collected at 0.25, 0.5, 1, 2, 4, 8, 12, 24 and 48 h post injection. The Ce6 fluorescence signal (Ex 640 nm, Em 710 nm) of each sample was analyzed by Lumina III Imaging System (PerkinElmer, IVIS Lumina III, Waltham, MA, USA).

In vivo targeting efficiency

100 μL of 4T1 cells (2 × 105) were subcutaneously injected to the left mammary fat pads of female BALB/c nude mouse. 7 days later, mice with tumor around 100 mm3 were randomly divided into three groups (n = 3). Two groups of mice were given d-SN38@NPs and d-SN38@NPs/iRGD, respectively. The last group of mice was intravenously administrated with d-SN38@NPs/iRGD, while the tumor sites of the mice were irradiated with 650 nm laser at 1 h before every imaging point. The fluorescence signal (Ex 640 nm, Em 710 nm) was determined at 4 and 24 h post-intravenous administration using Lumina III Imaging System (PerkinElmer). The mice were sacrificed by cervical dislocation after last imaging. Tumors and major tissues were collected, fixed, dehydrated and sliced for fluorescence study. After stained the nuclei with DAPI (0.5% μg/mL), images of the slices were captured by a confocal microscope (Olympus).

In vivo antitumor efficacy

Implantation of 4T1 cells was performed in BALB/c mice as described above. Mice with tumor around 100 mm3 were randomized into five treatment groups and the control group that received saline (n = 5). The treatment groups were intravenously given NPs, d-SN38@NPs, d-SN38@NPs/iRGD, d-SN38@NPs+laser and d-SN38@NPs/iRGD+laser at a dose of d-SN38 10 mg/kg once every 3 days for a total of 4 times, while the dose of iRGD was 86.6 mg/kg in iRGD co-injection groups. Twenty four hours post administration, the laser groups received irradiation at tumor site (200 mW/cm2 for 5 min). Tumor volume and body weight were recorded every 2 days. Mice were sacrificed one day after the last irradiation. Tumors and major organs were collected and sampled for hematoxylin and eosin (H&E) staining and TUNEL staining.

Statistical analysis

Mean ± standard deviation (SD) values were used for data expression, unless specified otherwise. Statistical analyses were performed using Student's t-test and P < 0.05 was considered statistically significant.

Results and discussion

Characterization of d-SN38@NPs

The nanoparticles were formed by self-assembly of the two linear chimeric triblock molecules (Supporting Information Fig. S1), and d-SN38 was incorporated in the hydrophobic core through hydrophobic interactions and the π–π stacking,. The diameter of d-SN38@NPs was 46.62 ± 2.15 nm with a uniform dispersity. However, the size of d-SN38@NPs after receiving 650 nm laser irradiation apparently increased to 207.0 ± 4.3 nm (Fig. 1A). Furthermore, an extensive morphological change from sphere to nanofiber was observed by TEM (Fig. 1B). The notable size and morphology changes collectively suggested that the morphology transformable system was successfully prepared (Supporting Information Fig. S2); this structural transformable ability could endow d-SN38@NPs a superior retention property.

Figure 1

Characterization of d-SN38@NPs. (A) Size changes of d-SN38@NPs post 650 nm irradiation measured by DLS; (B) TEM image of d-SN38@NPs and d-SN38@NPs post 650 nm irradiation; (C) Stability of d-SN38@NPs in different pH conditions for 24 h (n = 3); (D) Fluorescence emission spectra changes of d-SN38 after incubating with GSH; (E) Drug release from d-SN38@NPs upon or not upon 650 nm irradiation in different pH medium (n = 3).

The hydrodynamic diameter of d-SN38@NPs showed no perceivable changes in different pH values (6.5 and 7.4) and different temperatures (4, 25 and 37 °C) with a stable size around 46 nm (Fig. 1C). Furthermore, the turbidity changes of d-SN38@NPs in 10% and 50% FBS were quantified with UV–Vis spectra by detecting the light transmittance and no obvious variations were observed after 24 h incubation (Supporting Information Fig. S3). Both results verified that d-SN38@NPs possessed excellent stability.

Due to the abnormal metabolism of tumor tissue, there is a strong reduction environment in tumor cells. The concentration of GSH in tumor cells is more than 1000 times higher than that in normal cells. In order to maximize the accuracy of tumor targeted therapy, d-SN38 was designed to generate monometric drug SN38 activity in tumor cells when released from nanoparticles. As shown in Fig. 1D, the fluorescence intensity of d-SN38 under the GSH-abundant circumstance extensively reduced with time, which verified the GSH responsiveness of the dimers. And as shown in Supporting Information Figs. S4 and S5, the chemical structures of SN38 and d-SN38 were further confirmed by NMR and IR. Besides, the reduced molecular rigidity of drug dimers makes it easier to incorporate in nanoparticles. In this composition, the drug loading content was 33.92 ± 1.33% with a proper encapsulation efficiency (82.91 ± 3.24%), which is more than 6-fold higher than that of most reported micellar systems (normally lower than 5%).

The release behaviors of d-SN38 from the nanoparticles were monitored using a dialysis method. As shown in Fig. 1E, post irradiation, approximately 30% of d-SN38 was burst released from d-SN38@NPs in pH 6.5 medium at the first 0.5 h, whereas that without irradiation was less than 10%. When the incubation medium came to pH 7.4, the release ratio slightly decreased in d-SN38@NPs upon irradiation group (13.47 ± 1.54%), which was also higher than that of laser off (8.22 ± 1.45%). After 48 h, the cumulative release of d-SN38@NPs upon irradiation at pH 6.5 and 7.4 were 65.14 ± 1.67% and 56.05 ± 2.00%, respectively, and those without irradiation was 36.85 ± 3.26% and 35.59 ± 1.65%, respectively. The highest cumulative release of d-SN38@NPs upon irradiation at pH 6.5 suggested that irradiation induced structural transformation as well as the acidic circumstance promoted the drug release of d-SN38@NPs, which is also favorable for the efficient delivery of drug into tumor cells.

The ROS generation ability of different formulations on 4T1 cells was determined. As shown in Fig. 2A, without irradiation, almost no ROS was generated in all the formulations. However, upon irradiation, ROS signal significantly increased, demonstrating that NPs in tumor area can provide abundant ROS level. Interestingly, the ROS generated by d-SN38@NPs and d-SN38@NPs/iRGD was slightly higher than that of NPs. This result may attribute to the down-regulation of GSH by d-SN38, which reduced the consumption of ROS.

Figure 2

ROS generation and cellular uptake of various formulations. (A) ROS generation of PBS, NPs, d-SN38@NPs and d-SN38@NPs/iRGD upon 650 nm irradiation or not at cell level, scale bar = 10 μm; (B) Cellular uptake of NPs and NPs/iRGD upon 650 nm irradiation or not using confocal microscope, scale bar = 10 μm; (C) Flow cytometry analysis, error bars indicate SD (n = 3), ∗P < 0.05.

To investigate the cell penetration behavior of the prepared nanoparticle when co-administrated with iRGD, qualitative and quantitative cellular uptake studies were carried out in αvβ3 and NRP-1 overexpressed 4T1 cells,. The results of laser confocal fluorescence microscopy and flow cytometry showed that laser irradiation can effectively promote the uptake of d-SN38@NPs in 4T1 cells (Fig. 2B and C), suggesting that shape transformation to nanofibers could elevate cellular uptake. It is reported that some slender materials, such as nanofibers, nanorods, and nanoworms, etc., can increase cell adhesion and other functions through their large specific surface area, thus promoting cell uptake. When combined with iRGD, the amount of NPs swallowed by 4T1 cells after irradiation was 2.08 times as much as that by NPs alone, and was much higher than that of non-irradiation group. Introducing of iRGD indeed enhanced the cellular uptake of the transformable nanoparticles, which could be credited to the homing of iRGD to 4T1 cells through the αvβ3 receptor. iRGD could trigger the endocytic bulk transport pathway by NRP-1-mediated “bystander effect”. Thus, the difference of the signal distribution indicated that the uptake of these nanoparticles was mainly related to endocytic pathway and iRGD could enhance the uptake. These results suggested that the combination strategy of PDT and iRGD was able to synergistically facilitate the penetration of the shape-transformable formulations.

To confirm whether the shape change nanoparticles co-administrated with iRGD further enhanced penetration and retention in tumors, multicellular spheroids were prepared to simulate the tumor tissue. After 4 h incubation, due to the small particles size, all tested formulations distributed more than 40 μm of Z-axis to the tumor spheroid (Fig. 3A). Among them, NPs/iRGD+laser-treated spheroid maintained the strongest signal, which is about 1.69-fold of that without iRGD, even 2.93-fold of that of simple NPs (Supporting Information Fig. S6). Although the signal weakened as the depths deepened, enhanced signal could be clearly observed at 100 μm in NPs/iRGD+laser treated spheroid, while the signal of NPs tended to vanish. Corresponding quantify analysis of tumor spheroid sections at 100 μm was also consistent with this result (Fig. 3B and Fig. S6). Collectively, these results showed that the combination of PDT with iRGD could indeed accelerate the accumulation of shape changeable nanoparticles into deep region of solid tumor. The study well demonstrated that with the help of iRGD, spherical NPs/iRGD could first penetrate in tumor by initially binding with αvβ3 integrin receptors. And upon irradiation, the spherical NPs transformed into nanofibers, which were hard to be pumped back and thereby improved retention of NPs in the spheroids.

Figure 3

Penetration and retention in multicellular spheroid. (A) Penetration of NPs and NPs/iRGD upon or not upon 650 nm irradiation for 4 h by confocal microscopy, scale bar = 100 μm; (B) Quantitative analyses of tumor spheroid sections at 100 μm.

The enhanced cellular uptake prompted by combining PDT with iRGD paved the way for further systematically cytotoxicity evaluation of d-SN38@NPs/iRGD+laser. First, the relationship between in vitro cytotoxicity and incubation concentration of the prepared formulations was measured by MTT assay. The cytotoxicity of iRGD was evaluated, and it showed that iRGD almost has no cytotoxicity to cells at the tested concentrations of 0–100 μg/mL (Supporting Information Fig. S7). Then, we measured the cytotoxicity of NPs to determine if the nanoparticle themselves caused any cytotoxicity. The results indicated that NPs almost had no toxic effect on 4T1 cells even at the highest concentration (Fig. 4A). Whereas, when the concentration of Ce6 was more than 0.02 μg/mL, the irradiated NPs showed obvious cytotoxicity. When come to the irradiated free Ce6 group, the working concentration was much lower than that in irradiated NPs with an IC50 of 0.03 μg/mL. This result can be attributed to the generation of ROS, resulting in BR transformation. Compared to d-SN38, the nanoparticle formulations were more toxic to 4T1 cells. With the assistance of iRGD, d-SN38@NPs/iRGD (IC50 of 0.48 μg/mL) displayed an enhanced cytotoxic activity compared with d-SN38@NPs (IC50 of 1.33 μg/mL). Notably, post irradiation, when the concentration of d-SN38 was less than 0.6 μg/mL, d-SN38@NPs/iRGD+laser showed the strongest cytotoxicity compared with other preparations; while at higher concentrations, the cell viability of d-SN38@NPs or d-SN38@NPs/iRGD upon irradiation became comparable less than 10% owing to the synergy of chemotherapy and PDT.

Figure 4

In vitro cytotoxicity studies of 4T1 cells incubated with different formulations. (A) Cell viability by MTT assay (n = 3); (B) Fluorescence images of calcein AM/PI double stain assay, scale bar = 50 μm; (C) Apoptosis analysis by flow cytometry (n = 3).

Subsequently, dead-live staining assay was employed to directly observe the antitumor effect at cellular level. With or without irradiation, almost all the cells were alive in PBS group (Fig. 4B). In parallel, no obvious difference was observed in d-SN38 group. Whilst upon irradiation, Ce6 and NPs induced substantial cell death and displayed comparable vitalities at the detection concentration in line with the results of MTT assay. When treated with d-SN38@NPs and d-SN38@NPs/iRGD, nearly all the cells were died after irradiation, which further confirmed the synergistic effect of chemotherapy and PDT.

In addition, similar results were obtained with regard to apoptosis. After introducing a lower concentration of d-SN38 (0.25 μg/mL), the early and late apoptosis percentage of 4T1 cells treated with d-SN38@NPs/iRGD were 12.93% and 2.76%, which were higher than those of d-SN38 (9.66%, 1.82%) and d-SN38@NPs (10.71%, 3.23%; Fig. 4C). On the other hand, all the irradiation groups showed better apoptosis-inducing ability than non-irradiation groups. d-SN38@NPs/iRGD+laser induced 28.4% of cells in apoptotic stage, which was 1.15-fold higher than that of d-SN38@NPs+laser (24.77%). These data further proved that combination of PDT and iRGD could give assistance to the uptake of d-SN38@NPs/iRGD+laser, which lead to more apoptosis compared with control nanoparticles.

To explore the inhibition effect of d-SN38@NPs/iRGD+laser for tumor cell motility, wound healing assay was performed. The results showed that laser irradiation can effectively promote inhibitory capability of the formulations in 4T1 cells (Fig. 5A and B). Remarkably, d-SN38@NPs/iRGD+laser-treated cells displayed the weakest healing ability at 24 h after scratching. Upon irradiation, the d-SN38@NPs/iRGD treatment reduced the wound healing rate to 2.51 ± 1.44%, which was 3.25-fold lower than that of the d-SN38@NPs treated group, indicating the inhibitory effect of d-SN38@NPs was further strengthen by co-administration of iRGD. This enhanced inhibition capability may originate from (1) the in situ transformation of the nanoparticles into fibrils to construct artificial extracellular matrix, which was critical to the motility of metastatic cells; and (2) the blockage of NPR receptors of cancer cells by iRGD, which induced partial cell detachment and collapse of cellular processes.

Figure 5

Wound healing assay of 4T cells incubated with different formulations. (A) Microscopy images and (B) quantitative analysis of wound healing, scale bar = 200 μm (n = 3).

Ideally, the co-injection of d-SN38@NPs and iRGD is expected to improve the tumor targeting and therapeutic effect. Before the chemo-photodynamic therapy, pharmacokinetic profiles of d-SN38@NPs and d-SN38@NPs/iRGD were tested after a single intravenous injection in mice. As shown in Supporting Information Fig. S8 and Table S1, both d-SN38@NPs and d-SN38@NPs/iRGD have a long blood circulation half-life (t1/2z), which was longer than 15 h. The mean residence time (MRT0‒∞) for d-SN38@NPs/iRGD was 25.97 ± 8.75 h, which was 1.21-fold higher than that of d-SN38@NPs group. The result suggested that the prepared vehicles could circulate for a long time in circulatory system and co-administration iRGD with d-SN38@NPs could prolong the blood retention time to a certain extent, which were beneficial to tumor targeting.

Then, the tumor targeting and penetration efficiency of d-SN38@NPs/iRGD+laser in vivo were validated in 4T1 mice model. At 24 h post injection, all the tested formulations showed stronger distribution at the tumor site than 4 h (Fig. 6A). d-SN38@NPs/iRGD irrespective of irradiation showed extensive signal in the whole body, suggesting that the enhanced circulating comparing with d-SN38@NPs, which was consistent with the results of pharmacokinetic experiments. On the other hand, d-SN38@NPs/iRGD-treated group demonstrated that the iRGD endorsed more nanoparticles accumulation than d-SN38@NPs, evidenced by stronger fluorescent signal in tumor. Importantly, upon irradiation, d-SN38@NPs/iRGD displayed the strongest florescent intensity than laser off nanoparticles and d-SN38@NPs, supporting the notions that both iRGD and in situ nanoparticle-to-nanofiber transformation elevated tumor retention ability. The results were further confirmed by ex vivo imaging and the confocal imaging of sectioned organs and tumors (Fig. 6B‒D). In consistent with the results of in vivo imaging at 24 h, the semi-quantification of ex vivo imaging showed d-SN38@NPs/iRGD with irradiation significantly accumulated 1.79-fold (P < 0.001) and 1.52-fold (P < 0.01) more in tumor when compared with d-SN38@NPs and d-SN38@NPs/iRGD without irradiation, respectively. The heavily accumulation at the liver of all formulations indicated they were cleared faster in liver than other organs. Similarly, confocal imaging of sectioned organs and tumors showed that the fluorescent intensity of d-SN38@NPs/iRGD without irradiation was 1.46-fold higher than that of d-SN38@NPs, while the fluorescent intensity of d-SN38@NPs/iRGD with irradiation was the strongest, which was 1.94- and 1.33-fold higher than that of d-SN38@NPs and d-SN38@NPs/iRGD, respectively (Supporting Information Figs. S9 and S10). These results collectively suggested that presence of iRGD and the laser triggered shape transformation could effectively improve the targeting and penetration of nanoparticles to tumor parenchyma.

Figure 6

In vivo targeting efficiency of different formulations in 4T1 tumor-bearing mice. (A) Representative in vivo fluorescent imaging of mice from different groups at 4 and 24 h post injection; (B) Ex vivo fluorescence distribution imaging of major organs and tumors at 24 h after tail injection; (C) The semi-quantification of major organs and tumors at 24 h post injection (n = 3); (D) Fluorescence distribution in frozen sections of tumor tissue at 24 h after tail injection. The unit of color scales and radiant efficiency is [p s cm2/sr]/[μW/cm2]; In (B), a, b, c denotes d-SN38@NPs, d-SN38@NPs/iRGD, and d-SN38@NPs/iRGD+laser, respectively; ∗∗P < 0.01 and ∗∗∗P < 0.001; The bar in (D) represents 50 μm.

Encouraged by the superior tumor penetration and distribution effect of chemotherapy/PDT combination strategy of d-SN38@NPs/iRGD, in vivo antitumor study was carried out using 4T1 tumor-bearing mice. Compared to the rapid tumor growth curves of PBS, the groups receiving treatments all displayed a relatively slower growth rate. As shown in Fig. 7A, d-SN38@NPs group exhibited a moderate antitumor efficiency down to the monotherapy without laser. However, co-administration of iRGD with d-SN38@NPs apparently enhanced the inhibition efficiency with an inhibition rate of 30.55% on Day 18. When further combined with PDT, d-SN38@NPs/iRGD exhibited the strongest antitumor responsiveness with a tumor inhibition rate of 60.89%, which was 7.42-fold higher than simple d-SN38@NPs. In corresponding, the tumor weights of d-SN38@NPs/iRGD +laser were the lightest (Fig. 7B). In addition, the slightly body weight growth curve (Supporting Information Fig. S11) and histological images without obvious changes in the major organs (Supporting Information Fig. S12) collectively revealed that all treatments had satisfactory biocompatibility. This best performance of d-SN38@NPs/iRGD +laser could be put down to (1) the active targeting capacity of iRGD, which led to more nanoparticles homed to the tumor site; (2) the structural transformation of the formulation, resulting in an enhanced retention; (3) the synergistic effects by a combination of chemotherapy and PDT.

Figure 7

In vivo antitumor efficacy. (A) Tumor volume and (B) tumor weight of 4T1 tumor-bearing mice administered with PBS, NPs, d-SN38@NPs, d-SN38@NPs/iRGD, d-SN38@NPs+laser and d-SN38@NPs/iRGD+laser by intravenous injection, ∗P < 0.05, ∗∗P < 0.01 and ∗∗∗P < 0.001; Representative photographic image of (C) H&E staining and (D) TUNEL staining images of tumor tissues, scale bar = 20 μm.

Immunohistochemistry analysis

To explore the underlying mechanism that enabled enhanced therapeutic efficacy, H&E staining and TUNEL staining assay were employed (Fig. 7C and D). In PBS and NPs groups, typical features of tumor cells with large and irregular nuclei were observed. On the contrary, d-SN38-loaded nanoparticle formulations presented fewer complete tumor shape and more apoptotic bodies. Strikingly, d-SN38@NPs/iRGD +laser showed the lowest density of live tumor cells, and the most apoptotic and necrotic cells than all the other examined formulations, confirming the improved tumor inhibition efficiency with the help of PDT and iRGD.

Conclusions

In this study, we set forth a stepwise stimuli-responsive strategy in which we co-administrated tumor penetration peptide iRGD with structural transformable nanoparticles to enhance tumor targeting, penetration and retention. d-SN38@NPs combined with iRGD could home to tumors after intravenous administration and penetrate into the deep region of tumor sites. Then laser responsiveness of the Ce6 component could not only release ROS upon laser irradiation, but also stimulate the gathered nanoparticles transformed into nanofibers at the tumor sites. Subsequently, the released d-SN38 inside the tumor cell responded to the high level of GSH, resulting in a high antitumor effect in 4T1 tumor-bearing mice due to the enhanced chemo-photodynamic therapy effect. Collectively, the results of this study demonstrate the potential of combination iRGD with transformable nanomedicines for enhancing chemo-photodynamic therapy against cancer.

Acknowledgments

The authors gratefully acknowledge the financial support from (Nos. 81961138009; 82071915), (No.19YYJC2250, China), (No. B18035, China), , and (No. YQ2019H004). We would be grateful to Dr. Chenghui Li (Analytical & Testing Center, Sichuan University, China) for her technical help.

Author contributions

Huile Gao and Xiangrong Yu designed the research. Congcong Lin and Fan Tong carried out the experiments and performed data analysis. Rui Liu, Rou Xie, Ting Lei, Yuxiu Chen and Zhihang Yang participated part of the experiments. Congcong Lin wrote and revised the manuscript. All of the authors have read and approved the final manuscript.

Conflicts of interest

The authors have no conflict of interest to declare.