Methoxypoly(ethylene glycol) (MPEG, Mn = 2000, Aldrich, United States), γ-benzyl-L-glutamate-N-carboxy anhydride (BLG-NCA), dichloromethane, pyridine, P-toluene sulfonyl chloride (TsCl), absolute ether, ammonium hydroxide, benzene, dimethylsulfoxide (DMSO) and 1,4-dioxane were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Petroleum ether was obtained from Fuyu Chemical Corp. (Shanghai, China). Simethicone was purchased from Yongda Chemical Corp. (Tianjin, China). Sodium bicarbonate was purchased from Dingshengxin Chemical Corp. (Tianjin, China). Tetrahydrofuran and sodium hydrate were purchased from Huirui Chemicals Corp. (Tianjin, China) and Jinshan Pharmaceutical Corp. (Sichuan, China), respectively. Spectrographic grade potassium bromide (KBr) was obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). HPLC-grade acetonitrile was purchased from Anaqua Chemical Supply (Houston, TX, United States). Cis-platinum was purchased from Meryer (Shanghai, China).

Dulbecco’s Modified Eagle’s Medium (DMEM) High Glucose and fetal bovine serum (FBS) were purchased from Gibco (Shanghai, China). Penicillin/streptomycin and 0.25% (w/v) trypsin-0.1% (w/v) Ethylenediaminetetraacetic acid (EDTA) were bought from Solarbio (Beijing, China). 3-(4,5-Dimethyl-thiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) was purchased from Sigma (Shanghai, China). Culture flasks and dishes were obtained from Corning (NY, United States).

Lewis lung carcinoma (LLC) cell line, derived from the C57BL/6 mouse, was purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). LLC cells were maintained in DMEM with 10% FBS in the humidified atmosphere containing 5% CO₂ at 37°C.

Healthy, six-to-eight-week-old male C57BL/6 mice were purchased from Qingdao Daren Fortune Animal Technology Co., Ltd. (Qingdao, China). Food and water were given to all mice ad libitum. Feeding temperature was controlled at 20–22°C, relative humidity 50–60%, light and dark cycles for 12 h. All experiments were conducted in accordance with the guidelines of the Institutional Animal Care and Use Committee, Ocean University of China.

MPEG-P(Glu) block copolymer was synthesized as previously reported (Cabral et al., 2007) (Figure 2). First, methoxypoly(ethylene glycol) amine (MPEG₂₀₀₀-NH₂) was prepared by the method of toluene sulfonate esterification. MPEG₂₀₀₀ (40 g, 20 mmol) and CH₂Cl₂ (200 ml) were added in a 3-neck round-bottomed flask (500 ml), stirred for 5 min. Subsequently, pyridine (150 ml, 1.89 mol) and TsCl (7.6 g, 40 mmol) were added into the solution above and stirred for 24 h under nitrogen protection at 30°C and then cooled to 0°C. The reaction mixture was poured into hydrochloric acid solution stirred for 15 min and the CH₂Cl₂ layer was collected. The aqueous phase was extracted from 200 ml dichloromethane (three times). The combined organic layer was washed with brine, aqueous sodium bicarbonate, and brine successively, dried over anhydrous sodium sulfate overnight. Next, tetrahydrofuran was added and stirred for 5 min. Then, the acquired white powder (MPEG₂₀₀₀-OTs) was washed 2–3 times with diethyl ether and dried under vacuum at 25°C for 4 h. After MPEG₂₀₀₀-OTs (7 g, 3.25 mmol) was dissolved in aqueous ammonia and stirred for 8.5 h at 135°C, MPEG₂₀₀₀-NH₂ was obtained by isolation and purification, dried for 4 h. The following steps were applied to generate methoxypoly(ethylene glycol)-poly (γ-benzyl-L-glutamate) (MPEG-PBLG) block copolymer. MPEG-PBLG was synthesized by ring-opening polymerization of BLG-NCA monomer using MPEG₂₀₀₀-NH₂ as macro initiator, and then the benzyl group was deprotected following the literature procedure (Song et al., 2012). BLG-NCA (4 g, 15.2 mmol) was suspended in a mixed solution of benzene (64 ml) and dioxane (16 ml) and then MPEG₂₀₀₀-NH₂ (1.2 g, 0.6 mmol) was added. The reaction mixture was reacted under nitrogen protection at ambient temperature for 4 days. After the reaction, the mixture was poured into a large volume of anhydrous ethanol and stirred for 1 h, rested for 30 min, suction filtration, and washed with little ethanol. The precipitated product was dried under vacuum at 30°C for 12 h. At last, the benzyl group was removed from protection by blending with 0.025 N NaOH at ambient temperature to produce MPEG-P(Glu).

The structure of MPEG-OH, MPEG–NH₂, and MPEG-PBLG was confirmed by Fourier-transform infrared (FT-IR) analyzer (Nexus 470 by Nicolet Instruments America). 2% (w/w) of the sample was dispersed in KBr and then compressed it into slices. Each slice was scanned between 400 and 4,000 cm⁻¹.

The structure and purity of MPEG₂₀₀₀-PBLG and MPEG₂₀₀₀-P(Glu) were determined by ¹H nuclear magnetic resonance spectroscopy (¹H-NMR) instrument (JNM-ECP600, JEOL TMS, Tokyo, Japan) at 25°C. MPEG₂₀₀₀-PBLG and MPEG₂₀₀₀-P(Glu) were dissolved in deuterated dimethyl sulfoxide (DMSO-d₆, 500 MHz), D₂O to a concentration of approximately 10 mg/ml, respectively.

Drug loading (DL) and encapsulation efficiency (EE) of the CISP-NPs were determined as follows: briefly, lyophilized CISP-NPS was dissolved in double distilled water and then hydrochloric acid (2 mol/L) was used to disrupt the structure of micelles. Afterwards, the solution was filtered by 0.22 μm syringe filter. The subsequent filtrate was measured by high performance liquid chromatography (HPLC) and inductively coupled plasma atomic emission spectroscopy (ICP-AES), respectively. According to the following formulas, DL and EE of CISP-NPs were calculated: DL= weight   of   CISP   in   micelle weight   of   the   polymer   and   CISP ×100% , (1) EE= weight   of   CISP   in   micelle weight   of   the   feeding   CISP ×100% . (2)

Determination of CISP concentration by HPLC (Shimadzu, Kyoto, Japan) is performed with the LC-20AT solvent delivery unit, the SPD-20A UV/VIS detector, the plus manual sampler, and the HT-230A column heater (Tianjin Hengao Technology, Tianjin, China). The chromatographic analysis was conducted on a Hypersil NH₂ (250 mm × 4.6 mm, 5μm, Elite, Dalian, China) with the detection wave length at 310 nm. The flow rate of mobile phase, ethyl acetate-methanol-dimethyl formamide-deionized water (25:16:5:5, v/v), was 0.8 ml/min whilst the column temperature was maintained at 25°C. The standard curve equation is A = 696.37*C-913.69 (A for the area of peak; C for the concentration of CISP-NPs) and the correlation coefficient is 0.9999.

Another test method, the metal ions standard solution, was the configuration at concentrations of 8 µg/L, 16 µg/L, 24 µg/L, 32 µg/L, and 48 µg/L to obtain standard curve. After the nanoparticles were nitrated with nitric acid (5 mg/ml), the manganese content of solution was measured with optima 8000 ICP-AES.

Cytotoxicity of free CISP, CISP-NPs, and MPEG-P(Glu) polymer was determined by the MTT assay on LLC cells. LLC cells were seeded in 96-well culture plates (Jet Biofil) at a density of 6 × 10³ cells/well and incubated for 24 h. Cells were then treated with a series of drugs at different concentrations of free CISP (DMSO <0.5%), CISP-NPs, and MPEG-P(Glu) polymers for 24, 48, and 72 h, respectively. Tests were repeated for 6 times. Briefly, at the determined incubation-time, 20 μl of 5 mg/ml MTT was added per well of the plate. After incubation for another 4 h at 37°C, purple formazan crystals were observed. The supernatant was aspirated off gently, and 150 μl/well of DMSO was added into each well and the plate was gently shaken for 10 min at room temperature to dissolve the formazan crystals, and the absorbance of the resulting DMSO solution at 490 nm was measured with a SPECTRA-max M5 microplate reader (Molecular Devices, United States). Cells without drug treatment were served as a negative control with a 100% survival rate, while cells without MTT were used as a blank control to calibrate the spectrometer meter. Relative cell viability (%) in comparison to control cells was calculated as (Abs_sample/Abs_control) × 100%, expressed as the mean ± SD of six measures.

Cell cycle analysis was evaluated as described below (Wang et al., 2013). In brief, LLC cells were seeded at a density of 5 × 10⁵ cells in each well in a six-well plate (NEST Biotechnology) followed by incubation for 12 h. Then, culture media was substituted with fresh medium containing free CISP or CISP-NPs formulations at dosage of 12.55 μg/ml of CISP and cultivated for 48 h, with untreated cells as the negative control. At the end of the incubation, the collected cells (1 × 10⁶) were centrifuged at 1,000 rpm for 5 min, resuspended and washed twice with ice-cold PBS, and then fixed with 75% cold ethanol. After storage for 24 h at 4°C, the cell suspension was centrifuged again and washed twice with cold PBS and then incubated with the freshly prepared staining solution containing propidium iodide (PI) and RNase (Beyotime, Haimen, China) at 37°C under light-proof conditions for 30 min. Subsequently, the DNA content was determined by Cytomics FC500 MPL flow cytometry system (Beckman Coulter, United States). The percentage of cell populations in each stage of the cell cycle, including G0/G1, S, and G2/M phases, was assessed using MultiCycle AV software (San Diego, CA).

Antineoplastic activity of CISP-NPs was elucidated in the mouse xenograft model. 1 × 10⁶ LLC cells were inoculated subcutaneously into the right axillary space of the C57BL/6 mice. Briefly, the treatment was initiated when the tumor size reached an average of 7–9 mm and randomly divided into four groups (n = 6 for per group). Mice were separately injected with saline and MPEG-P (Glu) polymer, free CISP (6 mg/kg), CISP-NPs (6 mg/kg CISP or 12 mg/kg CISP) via the tail vein on days 7, 10, 13, 16, and 19, respectively. On day 21, three mice in each group were executed by cervical dislocation, followed by the tumor, heart, liver, spleen, and kidney which were excised with histopathological examination by Hematoxylin-Eosin (H&E) staining. The remaining three mice in each group were further used for survival experiments.

Histopathological impairment evaluated by hematoxylin and eosin (H&E) assay according to manufacturer’ s instruction. Liver, heart, spleen, and kidney from treated mice were stained with hematoxylin-eosin for assessment of toxicity. Briefly, tumor, heart, liver and spleen and kidney tissues in different groups were harvested and then fixed in 10% neutral buffered formalin. The fixed tissues were subsequently embedded in paraffin and sectioned in 3-μm increments. The sections were dewaxed by dipping them in xylene and then placed on glass slides and stained with H&E. Observations were made by optical microscopy (OLYMPUS BX60, Japan) with corresponding image analysis software (OLYMPUS DP25, Japan).

All statistical data were expressed as means ± standard deviation (SD) and total experiments were performed at least three replicates. The significant difference between the groups were examined by Student’s t-test. *p < 0.05, **p < 0.01, and ***p < 0.001 were deemed statistically significant.

The FTIR spectra of MPEG₂₀₀₀-OH, MPEG₂₀₀₀-NH₂, and MPEG₂₀₀₀-PBLG were shown in Figure 4. The characteristic peak of amine bond appearing at 1,648 and 1,548 cm⁻¹ indicated the N-H stretch of MPEG₂₀₀₀-NH₂ [Figure 4(2)]. The characteristic IR absorption frequencies for MPEG₂₀₀₀-PBLG [Figure 4(3)] were 3,300 (OH, NH Str), 2,885 (CH₂ Str), 1,734 (C = O Str), 1,654 and 1,542 (CONH Str), and 778 (CH bend aromatic). The ¹H-NMR spectra (Figure 5) also showed that MPEG₂₀₀₀-PBLG and MPEG₂₀₀₀-P(Glu) block copolymers were successfully synthesized. In the ¹H NMR spectrum (Figure 5A), five major signals were observed at 3.50 (Figure 5A-a), 3.90–4.4 (Figure 5A-b), 1.791–2.180 (Figure 5A-c), 2.364 (Figure 5A-d), 5.034 (Figure 5A-e), and 7.256 (Figure 5A-f) ppm, which represented the carbons of (CH₃O−, 3H), (−CH₂CH₂O−, 4H), (−CH₂CH₂COO⁻Na⁺, 2H), (−CH₂CH₂COO⁻Na⁺, 2H), (C₆H₅CH₂−, 2H), and (C₆H₅−, 5H), respectively. Taking above mentioned results into consideration, its proved MPEG₂₀₀₀-PBLG was successfully synthesized. The deprotection of the benzyl group was performed by mixing with 0.025 N NaOH at ambient temperature to obtain MPEG₂₀₀₀-P(Glu), the peak of which at δ 5.034 and 7.256 ppm disappeared, confirming the complete deprotection (Figure 5B). The peak of 3.630 (Figure 5B-a), 4.24 (Figure 5B-b), 2.17 (Figure 5B-c), and 1.85–1.95 (Figure 5B-d) ppm represented the carbons of (CH₃O−, 3H), (−CH₂CH₂O−, 4H), (−CH₂CH₂COO−Na⁺, 2H), and (−CH₂CH₂COO⁻Na⁺, 2H), respectively. The whole average molecular weight of MPEG₂₀₀₀-P(Glu)₁₉₇₉ was 3,979 Da, while the number of units of the P(Glu) was 13.

As CISP is widely used for chemotherapy drug in clinical trials, it still has a lot of side effects, especially nephrotoxicity, which make many patients suffered. In order to resolve this problem, nanotechnology-based delivery systems were devised to be capable of loading all sorts of drugs in recent decades (Xu et al., 2017). In this study, the biocompatible methoxypoly(ethylene glycol)-b-poly(L-glutamic acid) (MPEG-P(Glu)) was used to encapsulate CISP. The typical XRD patterns of pure MPEG₂₀₀₀-P(Glu)₁₉₇₉ material (1), CISP crystalline powder (2), CISP-NPs lyophilized powder (DL = 31.7%) (3), and the mixture of CISP and MPEG₂₀₀₀-P(Glu)₁₉₇₉ (4) were illustrated in Figure 6A. The XRD spectrum of free CISP showed many sharp peaks confirming that CISP was a crystalline compound, where narrow peaks represented some paracrystalline phases. While blank carrier was amorphous, the mixture of CISP and MPEG₂₀₀₀-P(Glu)₁₉₇₉ presented some peaks (in the 2θ range of 10–30°) suggesting crystallizability of CISP. However, CISP-NPs lyophilized powder was not detected with crystalline peaks, and the crystalline diffraction peaks of CISP disappeared or were masked since the CISP was encapsulated into CISP-NPs in an amorphous form. Furthermore, it was reported that the change of XRD patterns was due to the possible inhibition of CISP crystallization by polymer matrices. Compared with free CISP, CISP-NPs showed superior dissolution rate and water solubility through transforming a crystalline drug into its amorphous form (Hancock and Zografi, 1997; Lobmann et al., 2011). The amorphous crystalline form of CISP inside the MPEG₂₀₀₀-P(Glu)₁₇₉₇ contributes to its sustained release (Ji et al., 2016). The drug-polymer amorphous system could reach the goal of improving solubility and bioavailability of poorly soluble drugs and the stability of the amorphous could be improved by distributing the drug in an amorphous state in highly dispersed carrier materials (Li and Huang, 2008).

The drug loading (DL) and encapsulated efficiency (EE) of CISP-NPs were (28.54, 27.82%) and (85.63, 83.74%), which were measured by HPLC and ICP-AES, respectively. Optimizing nanocarrier MPEG2000-P(Glu)1979 and the particle size of nanodrug, the average size of CISP-NPs obtained in this study was (18 ± 1.33) nm (Figure 6C), with PDI of 0.185 and zeta potential of −21 mV, which could enhance tumor accumulation and aggregation via the EPR effect and achieve longer blood circulation. The TEM (Figure 6B) and AFM (Figure 6D) image revealed that CISP-NPs was spherical in shape and uniformly distributed without significant agglomeration. The particle size of CISP-NPs ranged from 8 to −40 nm with an average size of 18 ± 1.33 nm, which not only could evade renal clearance but also advance the penetration of nanodrugs into tumor tissues.

To evaluate whether free CISP and CISP-NPs bring into play antineoplastic effect against lung carcinoma, MTT assay was performed to evaluate the cell viability in response to drug treatment in LLC cell lines. Accordingly, both the free CISP and CISP-NPs displayed time-dependent and dose-dependent manner against LLC cells (Figure 7A). Fifty percent growth inhibitory concentration (IC50) was used as the measure of relative cytotoxicity. The IC50 (μg/ml) values of the free CISP group vs. CISP-NPs group were (17.9 vs. 908), (6.789 vs. 33.2), and (3.538 vs. 4.55), at 24, 48, and 72 h, respectively. CISP-NPs showed remarkably lower cytotoxicity than free CISP at 24 and 48 h. CISP-NPs could slightly inhibite the proliferation of LLC cells with cells viability greater than 80%, which were incubated in CISP at a concentration of 15 μg/ml for 24 h. This indicated that most of the Pt complexes came out of the micelle through a ligand exchange reaction and kept active. Then, the conjugation of CISP to the CISP-NPs polymer reduced its in vitro cytotoxic activity, but it retained its antitumor activity. However, there was no significant difference between the CISP-NPs group and free CISP group according to the IC50 values at 72 h.

The viabilities of LLC cells treated with MPEG₂₀₀₀-P(Glu)₁₉₇₉ polymer with concentrations from 0.001 to 1,000 µg/ml after 24, 48, and 72 h incubation were also evaluated (Figure 7B). The blank copolymer micelles were found to be almost nondeleterious, and even if the concentration of the copolymer was up to 1,000 µg/ml, the cell viabilities were still higher than 85.82%. These results indicated that MPEG₂₀₀₀-P(Glu)₁₉₇₉ polymer was biocompatible with no cytotoxicity and suitably served as a drug delivery vehicle.

As shown in Figure 8, after 48 h incubation with different formulations, the results of distribution of LLC cells in the five phases of the cell cycle clearly illustrated that both the group of free CISP and the group of CISP-NPs induced a typical G₂/M phase arrest in LLC cells. Specifically, the G₂/M proportions, in quantitative analysis of the cell cycle distribution, were 36.23 and 70.94% for the free CISP group and CISP-NPs group, while the G₂/M proportion of control group (NS) was only 9.28%. Obviously, LLC tumor cells after CISP-NPs treating displayed a better arrest than the free-CISP-treated group, in terms of the G₂/M phase. Moreover, cells cycle arrest in the G₀/G₁ phase decreased significantly and this would result in enhanced cancer cells death.

To further evaluate the safety of CISP-NPs, the analyses of pathology were performed. Cell nucleus was dyed blue by staining with hematoxylin, and, meanwhile, both cytoplasm and extracellular matrix were dyed red by staining with eosin in H&E staining. The cell morphology was not clear, and the chromatin deepened or diffused out of the cell if cell necrosis happened. It was seen that the cells had a large nucleus with a spherical or spindle shape in the tumor tissue of the NS group. At the same time, high nucleocytoplasmic ratio and the soakage of inflammatory cells were seen, which illustrated a fast-growing tumor (Figure 9A). By comparison, in the groups of both CISP-NPs, the tissue is in a state of necrosis to varying degrees, concentrated chromatin was distributed around the edge, and the nuclei of the tumor cells of CISP-NPs (12 mg/kg) group (Figure 9D) were extremely shrinking, fragmented or absent. Of all the tested groups, the CISP-NPs (12 mg/kg) group has the largest necrosis area, whereas CISP-NPs (6 mg/kg) (Figure 9C) administration groups at the same dose levels as free CISP (6 mg/kg) (Figure 9B) showed no significant difference in the degree of necrotic tissue. In addition, PEG could improve the circulation time of the nanoparticles in the bloodstream (Roberts et al., 2012), and CISP-NPs usually had remarkably prolonged blood circulation time in comparison with free CISP.

As shown in Figure 10, free CISP groups showed obvious damage to kidney, while the groups treated with CISP-NPs were normal. In addition, it was seen that these organs, like heart, liver, spleen, and kidney, had no damage in both CISP-NPs groups. Free CISP disappeared rapidly from circulation and was distributed to each organ. Especially, rapid and high accumulation was observed in the kidney, which was illustrated by the yellow arrows in the figure. NS groups showed a small amount of inflammatory cell infiltration. As for free CISP group (6 mg/kg), tubular epithelial cells in the convoluted proximal tubule became swollen and enlarged. The disappearance of the tubular lumen, cell edema, glomerular wall hyperplasia, nuclear atrophy, and glomerular inflammatory cells infiltration occurred, while CISP-NPs groups (6 mg/kg) were normal. For the CISP-NPs groups (12 mg/kg), due to the larger dose, a small amount of inflammatory cells infiltration was observed, indicating that the CISP-NPs could significantly reduce the nephrotoxicity caused by CISP. Therefore, our results showed that the efficacy of CISP-NPs was statistically equivalent to that of free CISP at the same dosage. However, CISP-NPs have successfully reduced the nephrotoxicity that hampers the clinical application of CISP. Furthermore, the higher concentration of CISP-NPs not only reduced the side effects but also had a better antitumor activity.

The key factors influencing the effect of cellular internalization were the size and composition of particles. PEGylation could enhance the stability and solubility of the drug in plasma and reduce the toxicity of the drug (Saravanan et al., 2017). The renal filtration threshold was about 4–5 nm, and particles with size more than 50 nm were mostly sequestrated by the reticuloendothelial system (RES) of the liver, lungs, kidney, and spleen (Li and Huang, 2008; Dai et al., 2017). The average size of CISP-NPs was (18 ± 1.33) nm, which could efficiently avoid renal clearance for reducing nephrotoxicity and nonselective RES scavenge as well as showing enhanced permeability and retention (EPR) effects of passive targeted drug at solid tumor sites (Figure 3B) (Yang et al., 2014). There are two main factors which contributed to the hypoxia in the tumor microenvironment. The entire tissue was incapable of getting sufficient oxygen due to the abnormal vascular network in the solid tumor, especially for the poorly vascularized internal area. At the same time, the overproliferation of tumor cells further accelerated the consumption of oxygen. Therefore, in the tumor microenvironment, the regions far from the blood vessels were typically hypoxic, which was not conducive to the large size, low penetration of nanoparticles in deep tissues. Jain and coworkers proved that nanoparticles with a 12-nm size had a better tumor penetration than larger sized nanoparticles (Cabral et al., 2011; Chen et al., 2017). To this end, CISP-NPs exhibited significantly superior efficacy in impeding tumor growth with no obvious side effects than that of free CISP in a mouse model of LLC cell line.