Preparation and evaluation of tumour microenvironment response multistage nanoparticles for epirubicin delivery and deep tumour penetration

Abstract

Poor tumour penetration became a major challenge for the use of nanoparticles in anticancer therapy. To further enhance the tumour penetration efficiency, we developed a tumour-microenvironment-responsive multistage drug delivery system which was formed layer by layer via electrostatic interaction with cationic drug-loaded nanoparticles, hyaluronidase (HAase) and iRGD-modified gelatin (G-iRGD). The drug-loaded nanoparticles were formed by self-assembling mPEG-PDPA-PG and encapsulation with epirubicin (EPI). Due to the protonation of tertiary amine groups of PDPA segment in acid environment, mPEG-PDPA-PG could enhance the lysosomal escape and the intracellular release of EPI. This NPs/HAase/G-iRGD delivery system showed great biocompatibility in vitro, confirmed by MTT method. In vitro spherical tumour model penetration and in vivo tumour permeability investigation showed HAase coated NPs-EPI (NPs-EPI/HAase) could significantly enhance its penetrating efficiency. The NPs-EPI/HAase could assist in breaking down the hyaluronic acid (HA), which was a key component of extracellular matrix and thereby improving mass transport within the solid tumours. The flow cytometry studies showed that G-iRGD coated NPs-EPI (NPs-EPI/G-iRGD) was more easily taken up by HepG2 cells than gelatin coated NPs-EPI (NPs-EPI/G), which revealed the active targeting ability of iRGD. The results proved that this NPs/HAase/G-iRGD delivery system showed promising potential in enhancing tumour penetration efficiency.

Keywords:

• Drug delivery
• tumour penetration
• nanoparticles
• hyaluronidase

Introduction

Hepatocellular carcinoma (HCC) was one of the most invasively malignant human cancers [1]. With the increasing number of cancer patients, chemotherapy, which was widely used in cancer therapy, had attracted greater attention in recent years [2]. But traditional antitumour drugs showed a limited therapeutic index due to the toxic side effects combined with the poor tumour permeability, leading to tumour recurrence and metastasis [3]. Thus, it was important to deliver chemotherapy drugs directly to the tumour site, while maintaining high efficacy combined with low systemic exposure in terms of cancer treatment [4].

In the past decades, nanomedicines had been explored extensively as a promising strategy in the advancement of anticancer therapies owing to their significant advances of elongated drug circulation time, minimized systemic toxicity and enhanced therapeutic effectiveness [5]. Unfortunately, the therapeutic efficiency of nanoparticles was usually unsatisfied. Firstly, the targeting ability of conventional nanoparticles was too low to effectively accumulate around tumour sites, leading to a reduced the concentration of chemical drugs and compromised the therapeutic effect [6,7]. Secondly, after accumulating around tumours, nanoparticles faced vastly diffusional hindrance due to the presence of the compressed intratumoural blood as well as the rich extracellular matrix (ECM) and the elevation of the interstitial fluid pressure (IFP) [8]. Based on these considerations, the development of novel nanoparticles for HCC was still urgently needed and the above two issues should be addressed.

To improve the tumour targeting efficiency of nanoparticles, a variety of ligands were modified to the surface of nanoparticles [9,10]. Conventional active targeting ligands can compromise tumour penetration of nanoparticles, which was described as the “binding site barrier” effect [11–13]. Considering the overexpression of integrin receptors, such as α_vβ₃ on neovessel endothelial cells and HCC cells, the corresponding cell penetration peptide iRGD was employed to decorate nanoparticles to endow nanoparticles with active HCC targeting ability [14,15]. A large number of strategies had emerged in an attempt to overcame the limited tumour penetration inherent in nanoparticles systems, such as designing tumour-microenvironment-responsive size-shrinkable nanocarriers which could be broken down by tumour-associate proteases or enzymes into smaller nanoparticles following extravasation allowing them to effectively fully penetrate into the tumour [16,17]. Then the ECM degrading strategy had also been realized by designing nanoparticles functionalized with enzymes that degrade tumour ECM which was composed of collagen, fibronectin, hyaluronic acid. Hyaluronidase, degrading one of the major ECM components hyaluronic acid, was coupled to the surface of nanoparticles to facilitate efficient tumour penetration and significantly improved therapeutic efficiency [18–20]. Furthermore, normalizing tumour microvessel had also been a strategy to modulate the tumour environment for enhancing the nanoparticles penetration. The abnormal vascular network and impaired lymphatic function of tumours lead to increased IFP that hampers tumour penetration of intravenously injected therapeutics. Therefore, normalization of tumour vasculature was effective to improve the tumour penetration of drugs [21–23].

In this study, we proposed a tumour-microenvironment-responsive multistage drug delivery system as shown in Figure 1 which was designed by electrostatic interaction with small-sized drug-loaded nanoparticles formed by amphiphilic polymer mPEG-PDPA-PG encapsulated with the drug EPI and hyaluronidase as well as gelatin which was modified by cell penetration peptide iRGD. Amphiphilic triblock polymer mPEG-PDPA-PG with pH-sensitivity was designed, and their self-assembled polymeric nanoparticles were used as carriers for anticancer drug delivery and controlled release. Polymer mPEG-PDPA-PG was chosen as a pH-sensitive segment because of its beneficial superior pH-sensitivity [24,25]. The hydrophilic PEG was to maintain the stability of the system during long-time biological circulation [26]. Amphiphilic polymer mPEG-PDPA-PG with guanidyl which was the characteristics of arginine were prepared will effectively reduce cytotoxicity [27]. After systemic administration, the multistage nanoparticles with large size distribute to tumour tissue and cross the tumour blood vessel wall due to the function of iRGD [28,29]. Then the multistage nanoparticles enter into tumour interstitium after extravasation from leaky vasculature, following with the degradation of gelatin nanoparticles by matrix metalloproteinase-2 (MMP-2) [30]. The large-sized multistage nanoparticles turn to be the small-sized EPI-loaded nanoparticles which possesses better interstitial penetration and HAase which would deplete one of the major ECM components hyaluronic acid to further improve tumour penetration [31]. When EPI-loaded nanoparticles were under lysosome weak acidic conditions, the micellar structure was changed to swollen and loose because of protonation of the pH-sensitive segment PDPA, leading to accelerated drug release rate and accumulative.

Figure 1. The schematic diagram of tumour-microenvironment-responsive multistage drug delivery system.

Materials and methods

Materials

N-(tert-butoxycarbonyl)aminoethyl methacrylate (tBAM) was synthesized as reported; The chain transfer agent S-1-dodecyl-S-(α,α′-dimethyl-α″-acetic acid)trithiocarbonate conjugated mPEG (mPEG-RAFT, Mw= 5 kDa) was generously provided by Professor Dong Anjie from Tianjin University; near-infrared multiple methyl cyanine dye (Cypate) was synthesized by our lab (Ex = 768 nm, Em = 810 nm); azodiisobutyronitrile (AIBN), 2-(diisoprppylamino)rthly methacrylate (DPA), matrix metalloproteinase-2 (MMP-2), TRIS and 4-aminophenylmeruric acetate (AMPA) were purchased from Sigma-Aldrich (San Francisco, CA, USA); trifluoroacetic acid (TFA), tetrahydrofuran (THF) and N,N-dimethylformamide (DMF) were purchased from Yongda Chemical Reagent Crop (Tianjin, China); gelatin, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC), and N-hydroxy-succinimid (NHS), Brij-35, sodium chloride (NaCl) and calcium chloride(CaCl₂) were purchased from Macklin Biochemical Co. Ltd. (Shanghai, China). All other reagents were analytical grade and used without further purification. Cell penetration peptide iRGD was purchased from GL Biochem Crop (Shanghai, China). Dulbecc’s modified Eagle’s medium (DMEM), penicillin, streptomycin, Opti-MEM, trypsin and Fetal bovine serum (FBS) were purchased from Kangyuan Corporation (Tianjing, China). 3-(4,5-Dimethylthialzol-a-yl)-2,5-diphenyltetrazolium bromide (MTT) and Hoechst 33342 were purchased from Solarbio Corp (Beijing, China). LysoTracker Green DND-26 was purchased from Yusheng Bio Crop (Shanghai, China). Regular Agarose G-10 was purchased from Gene Corp (Hong Kong). Optimal Cutting Temperature (O.C.T.) was purchased from Sakura Finetek Corp (San Francisco, CA, USA). Human HA ELISA Kit (Shanghai, China) was purchased from XinYu Biotech Crop (Shanghai, China).

Synthesis of N-(tert-butoxycarbonyl)aminoethyl methacrylate (tBAM)

tBAM was synthesized according previous procedure [19]. Ethanolamine was dissolved in 100 ml, 1 M, NaOH, and stirred. Then, di-tert-butyl dicarbonate (Boc₂O) was dissolved in 50 ml 1,4-diethylene dioxide added into ethanolamine solution, reaction 48 h. Termination reaction with water addition and extracted with ethyl acetate three times. The organic phase was washed with saturated salt water and dried with anhydrous magnesium sulphate. After vacuum drying, a yellow oil product, Boc-ethanolamine, was obtained. Boc-ethanolamine and triethylamine were dissolved in anhydrous CH₂Cl₂, pre cooling in an ice water bath for 20 min and nitrogen protection, under magnetic stirring condition by dropwise addition of methacryloyl chloride and sealing. The reaction mixture was allowed to warm to room temperature and stirred overnight. The reaction mixture was washed with water, 10% citric acid and 10% K₂CO₃, sat.NaHCO₃ and brine. The organic layer was dried over Na₂SO₄, and the solvent was evaporated under reduced pressure. The product was purified by recrystallization from CH₂Cl₂/hexane to give the product.

Synthesis of amphiphilic triblock polymermPEG-PDPA-PG (PEDG)

mPEG-RAFT was reacted with 2-(diisopropylamino)ethyl-methacrylate (DPA) in the presence of AIBN in DMF under nitrogen armosphere. The solution was stirred at 40 °C in a water bath for 24 h. Then, mPEG-PDPA, tBAM, AIBN were dissolved in DMF under nitrogen atmosphere. The mixture was stirred at 40 °C for 24 h. After dialyzed in deionized water, mPEG-PDPA-PtBAM was freeze-dried and analyzed in a 400 MHz spectrometer. Polymer mPEG-PDPA-PtBAM was dissolved in TFA stirred at room temperature for 24 h, the reaction solvent was dialyzed in a dialysis bag (MW = 3500 Da). After the pH value was adjusted to 9.0, 1H-pyrazole-1-carboxamidine hydrochloride was added and stirred at room temperature for 24 h. The reaction solvent was dialyzed in a dialysis bag (MW = 3500 Da), then the solution in the bag was freeze-dried to obtain the mPEG-PDPA-PG. The synthesis route of amphiphilic triblock polymer mPEG-PDPA-PG is shown in Figure 2(A).

Figure 2. (A) The synthetic scheme of mPEG-PDPA-PG. (B) The synthetic scheme of gelatin-iRGD.

Nanoprecipitation of the blank nanoparticles (NPs) and EPI-loaded nanoparticles (NPs-EPI)

Blank nanoparticles were prepared by the nanoprecipitation method. The corresponding blank polymers were dissolved in THF at a final concentration of 10 mg/ml. Subsequently, THF polymer solution was added dropwise to the pure water under stirring. The THF/water (solvent/nonsolvent) ratio was kept constant at 0.1 for all suspensions. The THF was evaporated to yield aqueous suspensions with a final polymer concentration of 1 mg/ml.

Preparation of EPI-loaded nanoparticles was same to blank nanoparticles. Briefly, mPEG-PDPA-PG (10 mg) and EPI (1 mg) were dissolved in 1 ml THF. The solution was equilibrated for 30 min under stirring. Subsequently, the THF solution was added dropwise to 10 ml of pure water under continuous stirring. THF was evaporated and excess EPI was removed by dialysis (2 L, 24 h; MW = 3500 Da) against pure water.

Preparation of Cypate-loaded nanoparticles was same to blank nanoparticles. Briefly, mPEG-PDPA-PG (10 mg) and Cypate (1 mg) were dissolved in 1 ml THF. The solution was equilibrated for 30 min under stirring. Subsequently, the THF solution was added dropwise to 10 ml of pure water under continuous stirring. THF was evaporated and excess Cypate was removed by dialysis (2 L, 24 h; MW = 3500 Da) against pure water.

Characterization of nanoparticles (NPs) and EPI-load nanoparticles (NPs-EPI)

The morphological observation of NPs and NPs-EPI

NPs and NPs-EPI solution were dropped on a 300 mesh copper grid and kept for 3–5 min, the excess fluid was removed by absorbent paper. Negative staining was performed using 5% aqueous phosphotungstic acid for 30 min before observation with transmission electron microscope (TEM, JEM2010, JEOL).

Particle size and zeta potential of NPs and NPs-EPI

NPs and NPs-EPI solution were prepared to measure the particle size and zeta potential by using Malvern Zetasizer Nano ZS90 (Malvern, UK).

pH-response test of the NPs

For the pH-response tests of the NPs, 0.1 M acetate buffer saline (pH = 5.5) and 0.1 M phosphate buffer saline (pH = 7.4) were used. A 200 µL of suspension (1 mg/ml) was mixed with 800 µL acetate buffer saline and phosphate buffer saline in centrifuge tubes and stored at 37 °C while mixing at 200 rpm. Then, at predetermined intervals (2, 4, 12 and 24 h), particle size and zeta potential were performed.

Determination of drug loading efficiency (DLE) and drug loading content (DLC) of NPs-EPI

The NPs-EPI suspension was stored at 4 °C after filtration using a 0.45 µm syringe filter. A 1 ml of suspension was lyophilized to determine the suspension concentration. For drug loading quantification, fluorescence intensity calibration functions of EPI in DMSO were used as reported elsewhere. DLE and DLC were calculated according to the following equations:

$$\text{DLE=\hspace{0.17em}}\frac{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{ }\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{d}\mathrm{r}\mathrm{u}\mathrm{g}}{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{d}\mathrm{r}\mathrm{u}\mathrm{g}\mathrm{ }\mathrm{i}\mathrm{n}\mathrm{ }\mathrm{f}\mathrm{e}\mathrm{e}\mathrm{d}}\text{×100％}$$ $$\text{DLC=\hspace{0.17em}}\frac{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{ }\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{d}\mathrm{r}\mathrm{u}\mathrm{g}}{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{ }\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{ }\mathrm{o}\mathrm{f}\mathrm{ }\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{y}\mathrm{m}\mathrm{e}\mathrm{r}\mathrm{ }\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{ }\mathrm{l}\mathrm{o}\mathrm{a}\mathrm{d}\mathrm{e}\mathrm{d}\mathrm{ }\mathrm{d}\mathrm{r}\mathrm{u}\mathrm{g}}\text{×100％}$$

In vitro release

Drug release from EPI encapsulated nanoparticles was performed in two different buffers: (i) acetate buffer (100 mM, pH = 5.5), (ii) phosphate buffer (100 mM, pH = 7.4) using dialysis. Briefly, 3 ml of NPs-EPI suspensions in deionized water were added into dialysis bags (MWCO = 3500 Da) secured with cotton, and then placed in plastic tubes containing 30 ml acetate buffer saline (pH = 5.5) and phosphate buffer saline (pH = 7.4), respectively. The experiment was carried out at 37 °C and a horizontal shaking speed of 50 rpm. At appropriate time intervals, 1 ml media was taken away and the corresponding fresh buffer was supplemented. The amount of released DOX was detected by multifunctional enzyme standard instrument workstation (FlexStation-3).

Preparation of multistage nanoparticles (MGNPs)

The synthesis of gelatin-iRGD (G-iRGD)

Gelatin was dissolved in 10 ml phosphate buffer saline (pH = 8.0). Then 4-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)-benzoic acid and NHS were added to the mixture. After stirring for 20 min, EDC was added and stirred at room temperature for 24 h. The reaction solvent was purified by dialysis with a dialysis bag (MW = 8000–14000 Da). iRGD was added to the above-mentioned mixture and stirred at room temperature for 24 h. After dialyzed in deionized water, gelatin-iRGD was freeze-dried. The synthesis route of gelatin-iRGD is shown in Figure 2(B).

Preparation of multistage nanoparticles (MGNPs)

A 0.5 ml of HAase-B aqueous solution (isoelectric point = 5.55, 1 mg/ml) was added dropwise to 1 ml of NPs-EPI aqueous solution (mPEG-PDPA-PG 1 mg/ml), the nanocomposite solution (NPs-EPI/HAase) was formed by the interaction of the positive charge carried by mPEG-PDPA-PG with the negative charge carried by HAase-B. Then, 0.5 ml G-iRGD aqueous solution (gelatine 1 mg/ml) was added dropwise to the nanocomposite solution. The multistage nanoparticles (NPs-EPI/HAase/G-iRGD) were prepared by the electrostatic force between the excess positive charge of the nanocomposite surface and the negative charge carried by gelatine.

The nanocomposite NPs-EPI/gelatin-iRGD (NPs/G-iRGD) was formed by the interaction of the positive charge carried by polymer mPEG-PDPA-PG with the negative charge carried by gelatin. The preparation of nanocomposite NPs-EPI/gelatin (NPs/G) was same to NPs-EPI/gelatin-iRGD (NPs/G-iRGD).

Degradation of MGNPs triggered by MMP-2 in vitro

The expression level of matrix metalloproteinase (MMP), especially MMP-2 and MMP-9, was found to be relatively high in liver cancer, lung cancer and so on [16]. To investigate the enzyme-sensitivity of NPs-EPI/HAase/G-iRGD, 1 ml NPs-EPI/HAase/G-iRGD was incubated with 0.4 µg MMP-2 in TCNB buffer (4 µl). At prearranged time intervals, the particle size of NPs-EPI, NPs-EPI/G-iRGD, NPs-EPI/G-iRGD incubated with MMP-2 was measured by Malvern Zetasizer Nano ZS90 (Malvern, UK).

Degradation of HA by HAase in vitro

Due to the loading procedure which HAase loading into the nanoparticles may lead to denature of HAase, the activity of HAase should be evaluated by human hyaluronic acid (HA) ELISA Kit (Shanghai, China). Set standard group, sample group and blank group. Standard groups were added with different concentration of standard products (HA) 50 µl. Sample groups first added with the standard solution (HA, 80 pg/ml) 50 µl and then added with sample solution HAase (100 µg/ml) and NPs/HAase (HAase, 100 µg/ml) 50 µl, respectively. Blank groups were first added with the standard solution (HA, 80 pg/ml) 50 µl and then added with distilled water 50 µl. Each hole was added with Human HA HRP-Conjugate Reagent 100 µL and incubated at 37 °C for 4 h. Then discard the liquid of each hole, and wash each hole with wash solution five times. Each hole was added with Chromogen Solution A and Chromogen Solution B 50 µL, respectively, and dark incubated at 37 °C for 15 min. Then each hole was added with stop solution 50 µl and the absorbance of each hole was measured at 450 nm wavelength by Microplate reader in 15 min. The concentration of HA in the blank group and the sample group was measured by the standard curve.

In vitro cytotoxicity assays

MTT assay was used to test the cytotoxicity of EPI, NPs, NPs/HAase, NPs/HAase/gelatin, NPs-EPI on HepG₂ cells. HepG₂ cells were planted in 96-well plates at a density of 2 × 10³ per well and cultured in 5% CO₂ humidified incubator at 37 °C for 24 h. Then cells were treated with a series of different concentration (from 0.5 mM to 32 mM of EPI) of EPI, NPs, NPs/HAase, NPs/HAase/gelatin, NPs-EPI. After incubation for 24 h, 20 μl of MTT (5 mg/ml) was added to each cell and incubated for another 4 h. Then the medium was removed, the cells were washed by PBS once and replaced by 150 μl of dimethyl sulfoxide (DMSO). Finally, the absorbance was measured by a microplate reader (Thermo Scientific, MA, USA) at 490 nm. The HepG₂ cells treated with medium were evaluated as control. Cell viability was calculated by the following formula: $$\mathrm{C}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{ }\mathrm{v}\mathrm{i}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}（\%）=\frac{A_{\mathrm{treated}}}{A_{\mathrm{control}}}\times 100$$ in which A_treated and A_control represented the absorbance of treated cells and control cells, respectively.

Tumour spheroid penetration assay

To prepare the three-dimensional tumour spheroids, HepG₂ cells were seeded at a density of 8 × 10³ cells/100 μl per well in 96-well plates coated by 100 μl of 2% low-melting-temperature agarose. Several days later, the uniform spheroids were selected and treated with NPs-EPI, NPs-EPI/HAase at an equivalent concentration of EPI(8 μg/ml). After 12 h of incubation, the spheroids were rinsed with ice-cold PBS for three times and were transferred to glass bottom petri-dish. Then the fluorescence distribution was observed via confocal microscopy (Nikon, Japan).

In vitro cellular uptake study

For flow cytometry experiment, HepG₂ was seeded on 6-well plates at 2 × 10⁵ cells/well and incubated for 24 h. Then, the cells were treated with various samples (8 μg/ml EPI equal) under 37 °C for 4 h. After incubation, the cells were washed twice with cold PBS, trypsinized and resuspended in proper volume of PBS. To determine the relative uptake of EPI, 10,000 cells were measured by flow cytometry (Cytomics™ FC 500, Beckman Coulter, Miami, FL). The amount of EPI which were taken up by the cells was measured by the mean fluorescence intensity (MFI) of all cells. The experiments were performed at least three times independently.

Intracellular distribution study

Due to the excitation and emission wavelengths of EPI and Lysotracker Green overlapped, we choose near-infrared dye Cypate instead of drug EPI. HepG₂ (10⁵ cell/ml) was seeded on two glass bottomed dishes and cultured for 24 h in a humidified 5% (v/v) CO2 atmosphere. Cypate-loaded nanoparticles (NPs-Cypate) were added to the cells and incubated for 2h and 4 h, respectively. After incubation, the glass bottomed dishes were washed with cold PBS for three times. Then, LysoTracker Green DND-26 (50 nM) was added to stain lysosomes for 30 min, and washed with cold PBS three times later. Then, the intracellular distribution was observed by confocal microscope (Nikon, Japan).

In vivo imaging and tissue distribution

Due to the excitation and emission wavelengths of EPI were not in the NIR band of the in vivo imaging system, we choose near-infrared dye Cypate instead of EPI. All of the animal experiments were conducted in accordance with guidelines that were evaluated and approved by the Ethics Committee of China Pharmaceutical University and the humane care of the animals were carried out for these animal studies. Male nude mice (18–20 g) were obtained from Model Animal Research Center of Nanjing University and the tumour model was established according to a method that was previously reported. Briefly, HepG₂ cells were incubated into the left armpit of each nude mouse. When the tumour volume reached 300–400 mm³, the tumour-bearing mice were randomly divided into four groups (n = 3) and intravenously injected treated with PBS (mock), NPs-Cypate, NPs-Cypate/G, NPs-Cypate/G-iRGD. At 2, 8, 24 h, the mice were anesthetized by an intraperitoneal injection of pentobarbital (50 mg/kg) and the whole body fluorescent distribution was observed through an in vivo imaging system (IVIS Spectrum, Caliper, Shelton, USA). After being imaged, the mice were sacrificed. Tumour and all major tissues were collected and imaged as well.

In vivo tumour permeability investigation-frozen section

The HepG₂ subcutaneous tumour model was established according the experiment above. When the tumour volume reached 300–400 mm³, the tumour-bearing mice were treated NPs-EPI, NPs-EPI/HAase, NPs-EPI/HAase/G-iRGD through peritumoural injection. Four hours later, the mice were sacrificed and the tumour tissues were removed. Then the tumours were embedded and frozen in OCT embedding medium, frozen slices of 15 μm thickness were prepared with Cryotome Cryostat (Leica, Germany). At the same time, the slices were coverslipped with the anti-fluorescence decay mount medium (including DAPI with stain the nuclei). Fluorescent imaging was performed with confocal microscope (Nikon, Japan).

Results and discussions

Synthesis and characterization of amphiphilic triblock polymer mPEG-PDPA-tBAM

The structure of the mPEG-PDPA-tBAM was confirmed by Gel Permeation Chromatography (GPC, PL1110-6504, Agilent Technologies, Palo Alto, USA) and ¹H NMR (JNM-ECP600, JEOL, Tokyo, Japan). As shown in Figure 3(B), it could be seen that the increase in molecular weight of PEG-PDPA-PtBAM after the reaction.

Figure 3. (A) ¹H NMR spectrum of mPEG-PDPA-tBAM. (B) The contrast of GPC between mPEG-PDPA and mPEG-PDPA-tBAM. (C) ¹H NMR spectrum of gelatin-iRGD.

Figure 4. TEM images of blank nanoparticles (NPs) (A) and EPI-loaded nanoparticles (NPs-EPI) (B). Particle size of NPs (C) and NPs-EPI (E). Zeta potential of NPs (D) and NPs-EPI (F).

Meanwhile the ¹H NMR spectrum is shown in Figure 3(A). ¹H NMR (600 MHz, CDCl₃) δ: 3.35 (2H, CH₂NH), 4.03 (2H, CH₂CH₂NH), 3.61 (2H, COCH₂) [19]. The representative functional groups could be found in the spectra, suggesting the products were synthesized successfully. The ¹H NMR spectrum of gelatin-iRGD is shown in Figure 3(C), the peak of iRGD was found as 7.198 ppm.

Characterization of blank nanoparticles (NPs) andEPI-loaded nanoparticles (NPs-EPI)

As shown in Figure 4(A) and (B), the morphology of NPs and NPs-EPI was investigated using transmission electron microscopy (TEM), the NPs and NPs-EPI were regular spherical particles, and their surfaces were smooth and not conglutinated. The NPs and NPs-EPI had a unimodal size distribution and the hydrodynamic diameter was 54.8 ± 0.3 nm (C) and 85.1 ± 0.2 nm (E), respectively. And as well as zeta potential was 31.2 ± 0.5 mV (D) and 39.9 ± 0.6 mV (F), respectively.

The hydrodynamic diameter and zeta potential of blank NPs at different pH values were determined to investigate the pH-sensitivity of amphiphilic triblock polymer, as shown in Figure 5 [31]. With regard to the hydrodynamic diameter as shown in Figure 5(A), DLS measurements revealed that as time went on NPs showed stable size distribution at neutral pH values (pH = 7.4), a sharply increase (from 89.2 to 402.3 nm) of the pH value from 7.4 to 5.5. This was due to the PDPA segment was protonated in an acidic environment, which enlarged the hydrophilic chain and enhanced the repulsive force, resulting in more difficulty for micellization [31,32]. As shown in Figure 5(B), as pH value decreased (from 7.4 to 5.5), the zeta potential of nanoparticles increased rapidly (from 2.1 to 27.5 mV). We observed that upon decreasing the pH value from neutral to acidic, a continuous increase in the particle diameter together with an increase in the zeta potential [33–35]. Figure 5(C) presents the TEM images of the NPs in the buffer solution with pH = 7.4 and pH = 5.5. It was found that the NPs exhibited a clear spherical morphology at a pH = 7.4 but the NPs were swelling under weakly acidic conditions (pH = 5.5). In conclusion, all the results above demonstrated the sharp pH-sensitivity of the amphiphilic triblock polymer mPEG-PDPA-PG.

Figure 5. Particle size (A) and zeta potential (B) of blank nanoparticles (NPs) dependent on pH values. Typical TEM images of NPs in different solutions with pH 7.4 (C) and 5.5 (D).

In vitro release of EPI from EPI loaded nanoparticles (NPs-EPI)

The anticancer drug EPI was entrapped into nanoparticles using the membrane dialysis method for further study. A drug loading efficiency of 31.7% was obtained at a drug loading content of 4.4%.

The release of EPI was investigated under two different pH buffer solutions: pH = 7.4 (simulate normal physiological conditions) and pH = 5.5 (simulate the pH of the lysosome). The plots of EPI cumulative release as a function of time in different buffer solutions are shown in Figure 6. The 72 h cumulative release ratio from NPs-EPI was 50% under pH = 7.4, whereas that under pH = 5.5 was 82%. The key reason for this phenomenon could be that the structure of NPs-EPI (at pH = 5.5) became porous, swollen and loose because of stronger protonation of the amine groups, which led to a water-soluble PDPA segment under weakly acidic conditions. The release results as well as the change of particle size and zeta potential of NPs under different pH conditions showed the sharp pH sensitivity of the amphiphilic triblock polymer mPEG-PDPA-PG. Meanwhile the pH of lysosome was nearly 5.5, the pH-responsive property of PDPA segment enabled EPI was specifically and almost completely released in tumour cells after the internalization of the NPs-EPI [36,37].

Figure 6. Cumulative release of NPs-EPI at pH 7.4 and pH 5.5.

Preparation of multistage nanoparticles (MNPs)

The particle size and zeta potential change of NPs-EPI, NPs-EPI/HAase, NPs-EPI/HAase/G-iRGD are shown in Table 1. The size of NPs-EPI, NPs-EPI/HAase and NPs-EPI/HAase/G-iRGD increased in turn, while the zeta potential decreased accordingly, indicating that NPs-EPI could form multistage nanoparticles by electrostatic interaction with HAase and G-iRGD.

Table 1. The size and zeta potential of NPs-EPI, NPs-EPI/HAase, NPs-EPI/HAase/G-iRGD.

	Size (nm)	Zeta (mV)
NPs-EPI	85.1 ± 0.2	39.9 ± 0.6
NPs-EPI/HAase	113.4 ± 0.6	19.2 ± 0.8
NPs-EPI/HAase/G-iRGD	186.9 ± 0.5	−6.9 ± 0.2

Degradation of MGNPs triggered by MMP-2

As shown in Figure 7(B,C), the particle size of NPs-EPI, NPS-EPI/G-iRGD was 85.4 and 189.4 nm at the beginning. The particle size of NPS-EPI/G-iRGD shrank to 100.2 nm after incubation with MMP-2, which was demonstrated by DLS (Figure 7(D)). With the prolongation of the incubation time, the particle size of NPs-EPI/G-iRGD notably decreased from 189.4 to 100.2 nm after incubation with MMP-2, which was attributed to the degradation of gelatin of NPs-EPI/G-iRGD (Figure 7(A)). However, the diameter of NPs-EPI and NPS-EPI/G-iRGD remained stable (Figure 7(A)). Above results suggested that NPS-EPI/G-iRGD could accomplish large-to-small shrinkage in response to MMP-2.

Figure 7. (A) Hydrated diameter of NPs-EPI, NPs-EPI/G-iRGD and NPs-EPI/G-iRGD incubated with MMP-2 during 24 h. DLS data of NPs-EPI (B), NPs-EPI/G-iRGD (C), NPs-EPI/G-iRGD incubated with MMP-2 (D).

Degradation of HA by HAase in vitro

HA (80 pg/ml, 50 µl) was first added to each group and then the blank group, the HAase group and the NPs/HAase group were added distilled water, HAase, NPs/HAase 50 µl, respectively, so as to compare the ability of HAase and NPs/HAase to degrade HA. As shown in Figure 8, the blank group did not degrade HA, and the degradation rate of HA in the HAase group and the NPs/HAase group were 81.9% and 50.2%, respectively. The rate of HA degradation in the NPs/HAase group was lower than that in the HAase group. It may be that HAase combined with nanoparticles resulted in the denaturation of enzymes, and the ability of NPs/HAase to degrade HA was weaker than that of free HAase.

Figure 8. The results of HA degradation rate of HA, HAase, NPs/HAase.

In vitro cytotoxicity

The cytotoxicity of NPs, NPs/HAase and NPs/HAase/gelatin against HepG2 cells was performed by MTT assay, as shown in Figure 9(A). The cytotoxicity of NPs, NPs/HAase, NPs/HAase/gelatin increased slightly with the increase in concentration of NPs, and the cell viability was higher than 90% even at the highest concentration of polymer (400 μg/ml) for 24 h, the drug delivery system (NPs/HAase/gelatin) showed bare and negligible toxicity to HepG₂ cells [38,39].

Figure 9. (A) Cytotoxicity of test of NPs, NPs/HAase and NPs/HAase/geltain in HepG2 cells after 24 h incubation. (B) Cytotoxicity of test of NPs-EPI and free EPI in HepG2 cells after 24 h incubation.

Additionally, the cytotoxicity of the NPs-EPI on the HepG₂ cells after 24 h of incubation was investigated and compared with the cytotoxicity of the free EPI (Figure 9(B)) to evaluate the anticancer efficacy. The NPs-EPI could decrease the cell viability gradually with increasing concentration of EPI. For 24 h of incubation, NPs-EPI exhibited slightly lower toxicities than the free EPI. Specifically, 60% of the cells died after 24 h of exposure to NPs-EPI with 32 μg/ml of EPI. When the concentration of EPI in the NPs-EPI was 32 μg/ml, the concentration of polymer was 400 μg/ml, and the remaining concentrations remained the same. Since, NPs were not display cytotoxicity, it was reasonable to posit that cytotoxicity of the NPs-EPI comes from release of EPI in the nanoparticles [40,41].

Penetration efficiency of NPs-EPI and NPs-EPI/HAase

Three dimensional tumour spheroids were usually used to mimic a number of native tumour environments including extracellular matrix, tumour macrostructure and diffusion gradients. Thus, tumour spheroid models were established herein to evaluate tumour penetrating ability of NPs-EPI and NPs-EPI/HAase. HepG₂ tumour spheroids were incubated with 400–500 μm in diameter and exposed to NPs-EPI and NPs-EPI/HAase for 4 h. Then, we assess the ability of EPI to penetrate into the three dimensional tumour spheroids via confocal microscopy. As shown in Figure 10(A), tumour spheroids treated with NPs-EPI/HAase exhibited higher intensity in all slices than that of NPs-EPI, demonstrating HAase, degrading HA which was one of the major ECM components, facilitating small sized NPs-EPI penetrate into the tumour spheroid. By comparison, NPs-EPI was located at the edge of tumour spheroid, while the fluorescence of NPs-EPI/HAase distributed more extensively and penetrated much deeper in the distance of 90 μm [4]. In order to better quantitatively compare the penetration efficiency of NPs-EPI and NPs-EPI/HAase, as shown in Figure 10(B), the core/surface coefficient was introduced [17]. At 90 μm depth, the core/surface coefficient of NPs-EPI/HAase was higher than NPs-EPI. Besides, as shown in Figure 10(C), the fluorescence intensity profiles of NP at 90 μm depth also indicated that the penetration ability of NPs-EPI/HAase was stronger than NPs-EPI [42]. All the results confirmed that NPs-EPI functionalized with HAase would facilitate penetration in tumour.

Figure 10. (A) Fluorescent images of HepG2 tumour spheroids upon incubation with NPs-EPI, NPs-EPI/HAase. (B) The ratio of fluorescence intensity of core and surface at 90 μm. (C) The fluorescence intensity profile from surface to core of NPs-EPI, NPs-EPI/HAase at 90 μm. Scale bars represent 100 μm.

Tumour tissues were characterized by a high density of ECM and leaky vasculature and impaired lymphatic circulation which overall result in a high IFP. These factors together form a biological barrier that prevents nanomedicines from penetrating into tumour tissues, as it was frequently observed that nanomedicines hardly diffuse deeper into tissues. In our study, targeting the dense ECM associated with solid tumours by co-delivery of HAase, which can assist in breaking down the ECM thereby improving mass transport within the solid tumours [43].

In vitro cellular uptake study

Flow cytometry was used for quantitative evaluation of the drug delivery capacity of the NPs-EPI/G-iRGD into human HepG₂ cells in vitro. As shown in Figure 11(A) and (B), NPs-EPI/G-iRGD, the fluorescence intensity of EPI was higher than that of NPs-EPI/G and NPs-EPI, indicating that cellular uptake of NPs-EPI/G-iRGD was extremely enhanced. This was owing to αvβ₃ and NRP-1 receptors being overexpressed in HepG₂ cells: iRGD could home to tumour cells through the αvβ₃ receptor; then, a motif with C-terminal was shown to NRP-1 receptor after protease hydrolysis on cell membrane surface. Thus, the cellular uptake of iRGD modified nanocarriers was enhanced [44].

Figure 11. (A) Intracellular fluorescence intensity of different groups measured by flow cytometry. (B) Mean fluorescence intensity of different groups measured by flow cytometry.

Meanwhile, flow cytometry results indicated that the intracellular EPI content released from the nanoparticles was higher than that of the free EPI after 4 h. Positive charge on the surface of cationic NPs was strongly affinity with the negative charge of phospholipid bilayer of cell membranes. Therefore, the enhanced cellular uptake of the NPs-EPI was attributed to their cationic characteristic [30].

Intracellular distribution study

The intracellular distribution of the Cypate loaded nanoparticles (NPs-Cypate) were qualitatively monitored in HepG₂ cells at different time points (2–4 h) by confocal laser scanning microscopy. By observing co localization of Cypate fluorescence (red) and Lysotracker Green (green), Figure 12(A) shows that the NPs-Cypate was incubated 2 h, the superimposed signal of these two signals (yellow) higher than that of incubated 4 h (Figure 12(B)), indicating that the NPs-Cypate escape from the lysosomal more as the incubated time went on. By observing the distribution of Cypate fluorescence, we could see that the red fluorescence signal of NPs-Cypate that was incubated 2 h was the distribution of cluster of points, yet the red fluorescence signal of NPs-Cypate that was incubated 4 h were dispersed distribution indicating that after NPs-Cypate incubated 4 h, Cypate release from the NPs and in the cytoplasm in the form of free Cypate.

Figure 12. (A) The intracellular distribution of the Cypate-loaded nanoparticles (NPs-Cypate) that was incubated 2 h. (B) The intracellular distribution of the Cypate-loaded nanoparticles (NPs-Cypate) that was incubated 4 h. Red represents Cypate, Lysotracker Green stained Lysosomes (Green), yellow represents Cypate was located in Lysosomes. Scale bars represent 10 μm. Refer online version for the color representation of the figure.

In vivo imaging and tissue distribution

Figure 13(A) and (B) displays the in vivo images and ex vivo tumours, normal organs of HepG₂-tumour bearing after systemic administration of different nanoparticles. As shown in Figure 13(A), the fluorescence signals of Cypate were more concentrated in the abdominal cavity after injection of 2–8 h. This may be due to the fact that the nanocomposite was more easily captured by the liver and kidney in the blood, and the Cypate fluorescence signal increased in tumour as time went on. As shown in Figure 13(B), after 48 h the Cypate fluorescence signal in the anatomy organ of the mice could also be seen in NPs-Cypate, NPs-Cypate/G, NPs-Cypate/G-iRGD that the signal of the liver was strongest, the kidney was second, and the fluorescence intensity of the NPs-Cypate/G-iRGD group was the strongest, demonstrating that NPs-Cypate/G-iRGD was easily captured by the reticuloendothelium. In addition, a certain amount of Cypate fluorescence signal was detected in the tumour blocks after 48 h in all groups but the amount of enrichment was not high meanwhile there was no significant difference in fluorescence intensity between the three groups. This was due to the particle size of nanocomposite as around 200 nm, usually scavenge through entero-hepatic circulation in vivo, meanwhile NPs-Cypate/G-iRGD was formed by physical electrostatic interaction between NPs-Cypate and G-iRGD demonstrating NPs-Cypate/G-iRGD was not very stable in vivo so may be the active targeting ability of iRGD was not so strong.

Figure 13. (A) In vivo images of HepG_2-tumour bearing mice at 2 h, 8 h and 24 h post-injection of NPs-Cypate, NPs-Cypate/G and NPs-Cypate/G-iRGD. (B) Ex vivo images of heart, liver, spleen, kidney and tumour of HepG₂-tumour bearing mice at 24 h post-injection of NPs-Cypate, NPs-Cypate/G and NPs-Cypate/G-iRGD.

In vivo tumour permeability investigation

As shown in Figure 14, the nuclei were stained to blue and the distribution of EPI (red) in the tumour slices was determined. The fluorescent intensity of NPs-EPI was weaker than of NPs-EPI/HAase which was consistent with tumour spheroid penetration efficiency analysis indicating hyaluronidase, degrading one of the major ECM components HA, facilitating small sized NPs-EPI penetrated into the tumour spheroid. The fluorescent intensity of NPs-EPI/HAase/G-iRGD was weaker than NPs-EPI/HAase but higher than NPs-EPI. The release of NPs-EPI/HAase from NPs-EPI/HAase/G-iRGD needs to break through the barriers of G-iRGD, so it needs a certain time for NPs-EPI/HAase/G-iRGD to release NPs-EPI/HAase after the degradation of the gelatin. HA, which was constituted of repeating units of N-acetylglucosamine and glucuronic acid disaccharide, was a key component of ECM [40]. HAase, which functions as an enzyme to break down HA at specific sites, had been used as an adjuvant for chemotherapy to enhance the penetration of drugs [40]. So HAase coated NPs-EPI could enhance penetration by surface coating to transport across the ECM [41].

Figure 14. Fluorescent distribution in slice of HepG₂ tumours with NPs-EPI (A), NPs-EPI/HAase (B) and NPs-EPI/HAase/G-iRGD (C). The upper left image of Figure (A–C) represents DAPI stained nuclei and the lower left image of Figure (A–C) represents EPI.

Conclusion

In this study, we had designed the multistage drug delivery system which was non toxicity in vitro was formed by electrostatic interaction with small-sized EPI-loaded nanoparticles, hyaluronidase as well as iRGD modified gelatin. The amphiphilic triblock polymer mPEG-PDPA-PG was synthesized successfully and its self-assembly polymeric nanoparticles were used in hydrophobic anticancer drug delivery with sustained controlled release. Polymer mPEG-PDPA-PG had sharp pH-sensitivity owing to the protonation of tertiary amine groups of PDPA segment in acid environment and thereby enhanced the release of EPI and the endosomal/lysosomal escape. Owing to the HAase would degrade one of the major ECM components HA, the HAase coated NPs-EPI delivered EPI to the least accessible area of solid tumour, the core in tumour tissues, where tumour stem cells usually stayed, thus enhancing therapeutic efficacy. The active targeting ability of iRGD had some effect in vitro, but the effect was not obvious in vivo and the improvement was still under the study. We believe that this study provide facile strategy towards the design of more intelligent nanocarriers for deep tumour penetration in future.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [81673360, 81601591], and Natural Science Foundation of Shandong Province [ZR2016HM45, ZR2016HB15].