Nanostructured lipid carrier co-delivering paclitaxel and doxorubicin restrains the proliferation and promotes apoptosis of glioma stem cells via regulating PI3K/Akt/mTOR signaling

The experiments involving animals were performed according to protocols approved by the Institutional Animal Care and Use Committee of North China University of Science and Technology Affiliated Hospital. All experiments were conducted in compliance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health.

The targets of PTX and DOX were retrieved from the drug-gene interaction database (DGIdb, www.dgidb.org). Glioma-related GSE79097 dataset (the platform annotation file: GPL15207) was downloaded from the Gene Expression Omnibus (GEO) database (https://ncbi.nlm.nih.gov/geo/). The dataset includes 3 normal samples and 11 tumor samples. Next, the R language 'limma' package (http://bioconductor.org/packages/release/bioc/html/limma.html) was used for differential analysis on the obtained GSE79097 dataset to screen the differentially expressed genes in glioma [20]. False discovery rates (FDRs) were calculated using the Benjamini and Hochberg methods, and the ∣log FoldChange∣ ≥ 1, FDR-corrected p value <0.05 was set as the screening criterion. Log FoldChange >1 suggests that the expression of this gene is significantly higher in glioma than that in normal control, while log FoldChange <−1 suggests that this gene expression is significantly lower in glioma than that in normal control. A heat map depicting the expression of the 50 top-most differentially expressed genes was plotted using the R 'pheatmap' package (https://cran.r-project.org/web/packages/pheatmap/). The overlap of PTX and DOX targets and significantly differentially expressed genes was then analyzed using the jvenn tool (http://jvenn.toulouse.inra.fr/app/example.html). The interaction network of drug-disease-genes was constructed by the Cytoscape 3.5.1 software [21]. To further study the signaling involved in the differentially expressed genes, the R language 'clusterProfiler' package (http:/bioconductor.org/packages/release/bioc/html/clusterProfiler.html) was applied for enrichment analysis via the Kyoto Encyclopedia of Genes and Genomes (KEGG) [22].

NLC loaded with PTX and DOX was prepared using the melt-emulsification technique [23]. A mixture containing 100 mg COMPRITOL^® 888 ATO, 100 mg soybean phosphatidylcholine, and 100 mg oleic acid were heated to prepare a thermal lipid phase. PTX (50 mg) and DOX (50 mg) were dissolved in 1 ml dimethyl sulphoxide (DMSO) and added with the thermal lipid phase. Then, 20 mg DOTMA was dissolved in 10 ml distilled water to obtain the aqueous phase. The aqueous and lipid phase were placed together in an oil bath at 80 °C and magnetically stirred for 15 min. The hot lipid phase was slowly added to the aqueous phase, followed by dispersing at high speed (10 000 rpm) for 5 min. The crude oil emulsion obtained above was treated for 5 min with an ultrasonic probe set at a power of 3 W. The hot emulsion was cooled to 4 °C in an ice bath and mechanically stirred for 30 min DMSO was removed by dialysis against deionized water. The hot emulsion was freeze-dried to obtain powdered PTX-DOX-NLC samples, which were stored at 4 °C for later use. PTX- and DOX-loaded NLC samples were prepared separately as described above.

The particle size of NLC was determined. In brief, NLCs were diluted 3000 times in mass ratio with ultrapure water. A nanometer particle size analyzer (Malvern Instruments Ltd, Malvern, Worcestershire, England) was applied to determine the particle size, particle size distribution, and surface charge of NLC.

Examination of drug loading capacity (DLC) and drug loading efficiency (DLE) of NLC: the content of DOX in NLC was analyzed by UV spectrophotometry at 480 nm. A total of 1 mg NLC was dissolved in 1 ml DMSO as the test product. DOX (0–100 μg ml⁻¹) was prepared with the same solvent and the absorption wavelength was measured at 480 nm to produce a standard curve. The content of PTX in microspheres was analyzed by high performance liquid chromatography (HPLC). In this procedure, the NLCs were dissolved in dichloromethane [21]. A Symmetry C18 chromatographic column was adopted as the stationary phase and acetonitrile/water (v/v = 4:1) as mobile phase, with 20 μl sample injection volume. Absorption wavelength was measured at 227 nm using an ultraviolet spectrophotometer for quantitative analysis. DLC and DLE were calculated respectively based on the following formula: DLC (%) = W_{loaded drug}/W_{drug loaded NCL} × 100%; DLE (%) = W_{loaded drug}/W_{free drug} × 100%.

Drug release characteristics of NLC in vitro: the drug release characteristics of PTX or DOX-loaded NLCs as well as PTX-DOX-NLC were determined as previously described [24, 25]. In brief, NLCs were suspend in 5 ml phosphate buffered saline (PBS) and transferred to a dialysis bag with a molecular weight cut-off of 3500 Da, and then placed in a 45 ml release medium and dialyzed with stirring at 100 rpm at 37 °C. A 4 ml portion of incubation solution was removed and replaced with the same volume of fresh PBS. The amount of DOX and PTX released was determined by ultraviolet spectrophotometry and HPLC, respectively.

The ratio of combined drug loading PTX and DOX was optimized, with the IC₅₀ of anti-proliferation as the parameter, where the PTX/DOX values of the drug loading ratio were as the following: 1:5, 1:2, 1:1, 2:1, 5:1 (w/w). Each ratio was diluted as the listed concentration ratio (table 1) to detect cell activity. Graphpad was employed to calculate the IC₅₀ of PTX and DOX at each ratio, and the group with the lowest IC₅₀ was selected as the optimal ratio.

The sorted CD133-positive (CD133⁺) U87 cells were seeded into 96-well plates with 5000 cells/well and cultured in 200 μl medium. Different drug concentration gradients (based on the determined drug combination ratio) (0, 0.3125, 0.625, 1.25, 2.5, and 5 μM) or time gradients (0, 12, 24, 36, 48, 60, and 72 h) were set, with three replicates for each concentration. After 72 h of culture, the cell viability was detected with CellTiter-Glo kit. Celltiter reagent (100 μl) was added to each well and mixed at room temperature on a shaker for 10 min. Cell viability was determined on a Glomax automatic fluorometer. The experiment was performed in the dark.

Cell suspension (1 × 10⁴–2 × 10⁴) was mixed with 0.4% agar in phenol red-free culture medium, and plated on top of a solidified layer of 0.8% agar in serum-free phenol red-free Roswell Park Memorial Institute (RPMI)-1640 medium in a 6-well plate. After the apical agar was solidified, normal growth medium supplemented with indicated treatments (Blank NLC, Free DOX group, Free PTX group, DOX-NLC group, PTX-NLC group, and PTX-DOX-NLC group), which was added on top of the apical layer and renewed every 3 d. After 2–3 weeks, the colonies were stained with 1 mg ml⁻¹ iodonitrotetrazolium violet (Shanghai ziyi Reagent Co., Ltd, Shanghai, China) overnight at 37 °C, and the colonies with diameter exceeding 2 mm were counted.

Cells in 6-well plates were collected and lysed in 1 ml cold TRIzol. Tumor tissues were grinded in the presence of liquid nitrogen, followed by dissolving in TRIzol (1 ml per 50–100 mg) with the assistance of a homogenizer. The concentration of total RNA was determined on aNanodrop2000.

Reverse transcription was performed according to the instructions of the reverse transcription kit (CAS:205311, QIAGEN, Valencia, CA, USA) to generate complementary DNA. Primers for RT-PCR (table 2) were designed using primer 5.0 software on the basis of sequences reported in the Gembank database, and were synthesized by Shanghai Sangon Company (Shanghai, China). RT-qPCR was performed using SYBR premix Ex Taq (RR420B, Takara Bio, Japan) and the relative expression of target genes was calculated by the 2^−ΔΔCt method with β-actin as the internal control gene.

Total protein of cells was extracted using radio-immunoprecipitation assay lysis buffer and the concentration was determined using bicinchoninic acid (BCA) protein assay kit (P0010, Beyotime Institute of Biotechnology, Shanghai, China). Samples of 30 μg of isolated proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). After separation, the protein was transferred to polyvinylidene fluoride (PVDF) membrane, and then blocked with 5% bovine serum albumin (BSA) for 1 h. The membrane was probed with diluted primary mouse anti-human antibodies to phosphorylated (p)-PI3K, p-Akt, mTOR, and actin (all from SANTA CRUZ, CA, USA) overnight at 4 °C. The membrane was then probed with goat anti-rabbit secondary antibody conjugated with horseradish peroxidase (HRP) for 1 h at room temperature. Electrogenerated chemiluminescence was applied for visualization in the dark. The Image J software was used for gray scale analysis to determine the relative expression of proteins.

U87 cells were incubated with ethylenediaminetetraacetic acid (EDTA) digestion solution (containing 2.5 mmol L⁻¹ EDTA and 1% FBS, pH7.4) for 30 min at 4 °C. The cells were suspended in 300 μl suspension buffer to a concentration of 1 × 10⁷ cell ml⁻¹. CD133⁺ and CD133 negative ⁽⁻⁾ U87 cells were sorted with the MiniMACS separation column according to the operating instructions and saved for future use.

Thirty-six healthy female mice (aged 5 weeks) were housed adaptively in the specific pathogen free animal laboratory cages with 60%–65% relative humidity, at 22 °C–25 °C with free access to food and water under 12 h light/dark cycle for one week. The sorted CD133⁺ U87 cells were subcutaneously inoculated into the underarms of the mice. Thereafter, the mice were randomly divided into 6 groups of six mice based on the treatment received. Tail vein injection was then adopted for administration of drugs or vehicles (DOX, PTX, DOX-loaded NLC, PTX-loaded NLC, or PTX and DOX-loaded NLC) for 21 consecutive days. During this period, tumor volume and body weight of the tumor-bearing mice were monitored and recorded. On the 28th day, mice were euthanized by intraperitoneal injection of excess barbital sodium. End-point tumor size and tumor weight were measured and the harvested tumors were fixed and prepared for histological analysis.

Paraffin sections of mouse tumor specimens were washed with tap water for 2 min after dewaxing and gradient alcohol rehydration. Endogenous peroxidase activity was quenched by incubating the sections with 3% hydrogen peroxide (H₂O₂) in methanol for 20 min. Sections were repaired in water bath with antigen repair solution (citrate buffer, pH 6.0) for antigen retrieval. Afterwards, the sections were blocked with goat serum blocking solution (C-0005, Shanghai Haoran Biotechnology Co., Ltd, Shanghai, China) at room temperature for 20 min. Sections were probed with rabbit anti-human antibodies to CD133 (1:200, Abcam, Cambridge, UK), Ki-67 proliferative index (ki-67; 1:200, Abcam), p-PI3K (1: 200, Abcam), p-Akt (1:200, Abcam), and mTOR (1: 200, Abcam) overnight at 4 °C, and then with the IgG (ab6785, 1:1000, Abcam) at 37 °C for 20 min. After that, HRP (Imunbio Biotechnology Co., Ltd, Beijing, China) was added and incubated at 37 °C for 20 min. Sections were visualized with 3,3'-diaminobenzidine (ST033; Guangzhou Weijia Technology Co., Ltd, Guangzhou, China) and the nuclei were counterstained with hematoxylin (PT001; Shanghai Biological Bogu Technology Co., Ltd, Shanghai, China) for 1 min and returned to blue by treatment with 1% ammonia. Then, sections were dehydrated with gradient alcohol, cleared with xylene, and sealed with neutral gum for microscopy. Five high-power microscopic fields were randomly selected from each section, with at least 100 cells in each field.

Healthy BALB/c mice (n = 12) were weighed and randomly divided into two groups (n = 6). PTX-DOX-NLC or PBS was injected intravenously via the tail vein. About 1 ml of blood was collected from the tail vein of the mouse within 24 h after injection, and immediately centrifuged at 5000 rpm for 3 min to collect serum for blood biochemical analysis to determine the level of liver (liver alkaline phosphatase (ALP), aspartate aminotransferase (AST), alanine aminotransferase (ALT)) and renal (blood urea nitrogen (BUN)) functional markers [24, 26–28].

The data were processed using SPSS 21.0 statistical software (IBM Corp. Armonk, NY, USA). Measurement data were expressed as mean ± standard deviation. Comparison among multiple groups was conducted by one-way analysis of variance (ANOVA) with Tukey's post hoc test. Cell proliferation data at different time points were compared by two-way ANOVA with Bonferroni's post hoc test. Repeated measurement ANOVA with Bonferroni's post hoc test was used to compare tumor data at different time points. p < 0.05 was considered as significantly different. All experiments were performed three times independently.

We first retrieved the drug target database DGIdb and obtained the corresponding gene interaction targets of PTX and DOX. Then we performed a differential gene expression analysis on a glioma-related expression dataset GSE79097, which revealed 3011 differentially expressed genes, including 1372 highly-expressed genes and 1639 lowly-expressed genes (figure 1(A)). A heat map illustrating the top 50 differentially expressed genes is shown in figure 1(B). Following Venn diagram analysis of the aforementioned genes, 37 genes were found at the intersection (figure 1(C)). An interaction network regarding the drug-disease-gene is shown in figure 1(D). In addition, KEGG enrichment analysis showed that 37 genes were mainly enriched in microRNAs in cancer and PI3K-Akt signaling (figure 1(E)). Moreover, we found links between the PTX/DOX target genes and PI3K/Akt/mTOR signaling (map04151) by retrieving the KEGG database (figure 1(E)). The aforementioned bioinformatics results suggested that PTX and DOX may mediate the development of glioma by regulating the PI3K/Akt/mTOR signaling, which inspired us to investigate the regulation of PI3K/Akt/mTOR signaling by PTX-DOX-NLC in glioma.

Zoom In Zoom Out Reset image size

PTX and DOX-loaded NLC was prepared by the fusion emulsification technique (figure 2(A)). The results showed that the mean diameter of NLC ranged from 115 and 125 nm, with a uniform particle size distribution of each type of nanoliposome. The median particle size of drug-free NLCs was 119.72 nm, and that of DOX-loaded NLCs was 122.83 nm. The detection of Zeta potential, a key factor to predict and evaluate the colloidal dispersion stability, displayed, that compared with drug-free NLCs, PTX- and DOX-loaded NLCs exhibited lower Zeta potential. Moreover, PTX-DOX-NLC had the lowest Zeta point. Taken together, the particle size, particle size distribution, surface charge number, and polydispersity index of the prepared liposomes revealed their favorable dispersibility, which would promote their applicability for delivery to cancer cells (figure 2(B)). Although PTX-DOX-NLC had lower DLC and DLE compared with PTX-loaded or DOX-loaded NLCs, their stability was not affected, as reflected by the particle size and Zeta potential (figure 2(C)). We further evaluated drug release behaviors of PTX-DOX-NLC, DOX-loaded NLC, and PTX-loaded NLC in vitro. The results revealed that the PTX and DOX were gradually released during the test, which conformed to the Higuchi equation. The release rate of PTX and DOX reached more than 70% within 24 h, indicating that PTX-DOX-NLC had the capacity to gradually and steadily release PTX and DOX (figure 2(D)).

Zoom In Zoom Out Reset image size

We initially observed that CD133⁺ cells could grow in clumps in stem cell medium to form stem cell globules (figure 3(A)), confirming that the obtained cells were GSCs. Next, the expression of Nestin, GFAP, and β-tubulin was determined in isolated CD133⁺ U87 cells by immunofluorescence staining. The result showed that Nestin was strongly positive, and GFAP and β-tubulin were negative (figure 3(B)). Meanwhile, the optimized combination ratio of PTX and DOX for NLC loading was also determined by evaluating their effect on cell viability, as reflected by the IC₅₀. The results showed that IC₅₀ was the lowest when PTX and DOX were loaded in a 1:1 ratio, rather than combinations. Therefore, this PTX/DOX ratio was selected for subsequent studies (figure 3(C)). Celltiter-Glo was applied to detect the cell activity of different groups (Blank NLC, Free DOX group, Free PTX group, DOX-NLC group, PTX-NLC group, and PTX-DOX-NLC group) after treatment with different concentrations (0, 0.3125 μM, 0.625, 1.25, 2.5, and 5 μM) (figure 3(D)). We found that all interventions suppressed the cell viability of GSCs in a dose-dependent manner compared to the control group, in which cells were treated with un-loaded NLC. Moreover, PTX-DOX loaded-NLC achieved stronger inhibitory effect on GSCs' viability than either PTX or DOX loaded-NLC groups, as well as the free drugs.

Zoom In Zoom Out Reset image size

Cell activity of GSCs was also measured after exposing to different NLC groups for different periods (0, 12, 36, 48, 60, and 72 h) (figure 3(E)). The results revealed that the cell viability of GSCs was suppressed in a time-dependent manner while the PTX-DOX-NLC group showed a significantly stronger inhibitory effect in comparison to all other groups (Free DOX, Free PTX, DOX-NLC, PTX-NLC). The impact of PTX-DOX-NLC on GSC apoptosis was also assessed by treating GSCs with different NLCs for 24 h followed by apoptosis assay. Results from flow cytometry-based apoptosis assay suggested that, compared with the NLC group where cells were treated with un-loaded NLCs, all drug-loaded NLCs (as well as NLC-free drugs) resulted in increased apoptosis in GSCs (figure 3(F)). Furthermore, the effect of PTX-DOX-NLC on the colony formation potency of GSCs were also evaluated via soft agar colony formation assay. We observed that all drug-presenting interventions resulted in reduced colony formation potency in GSCs compared with that in the control group where cells were incubated with NLC alone. More importantly, the inhibitory effect of colony formation was most significant in PTX-DOX-NLC treated group (figure 3(G)). Taken together, these data suggested that NLC co-delivering PTX and DOX achieved better anti-proliferative and apoptosis induction effects in GSCs.

The mechanism of the inhibitory role of PTX-DOX-NLC in GSCs was further explored. RT-qPCR and Western blot analysis were employed to measure the mRNA and protein expressions of PI3K, Akt, and mTOR in GSCs treated with NLC, DOX-NLC, PTX-NLC, or PTX-DOX-NLC. The results revealed that the mRNA (figure 4(A)) and protein levels (figure 4(B)) of PI3K, Akt, and mTOR in GSCs treated with PTX-DOX-NLC was significantly downregulated relative to that treated with NLC. Hence, both PTX and DOX suppress proliferation and induce apoptosis in GSCs via regulating PI3K/Akt/mTOR signaling.

Zoom In Zoom Out Reset image size

To further evaluate our speculation that PTX-DOX-NLC inhibits glioma growth via suppressing proliferation and inducing apoptosis in GSCs by regulating PI3K/Akt/mTOR signaling, a human xenograft model of glioma was established in mice. For this purpose, nude mice transplanted with CD133⁺ cells were randomly divided into 6 groups, and received different treatments for 21 consecutive days, and the tumor growth curve were plotted 30 days post drug administration. The result showed that the growth of xenograft tumors was significantly inhibited by free DOX, free PTX, DOX-NLC, PTX-NLC, and PTX-DOX-NLC, compared with growth in the mice treated with NLC alone. More importantly, the inhibitory effects were most significant in the PTX-DOX-NLC treated group (figure 5(A)). Meanwhile, the body weight of nude mice was also monitored, which showed no difference among the groups, suggesting that the drug injection had no conspicuous toxic side effects for nude mice (figure 5(B)). In addition, the liver (ALP, AST, and ALT) and renal (BUN) functional markers were also monitored after injection of PTX-DOX-NLC or PBS, and no significant difference was observed, suggesting no conspicuous metabolic toxic side effects to nude mice (figure S1 (available online at stacks.iop.org/NANO/32/225101/mmedia)). After the mice were euthanized, the tumors were removed and weighed, which showed reduced weight in the mice treated with free DOX, free PTX, DOX-NLC, PTX-NLC, and PTX-DOX-NLC, compared with those treated with NLC alone, with the inhibitory effect being the most significant in the PTX-DOX-NLC treated mice (figure 5(C)).

Zoom In Zoom Out Reset image size

We further analyzed the expression of a GSC marker (CD133), proliferation marker (ki67), key proteins of PI3K/AKT/mTOR signaling (p-PI3K, p-AKT, and mTOR), and caspase3 in the xenograft tumors (figure S2). The result showed that, compared with control group in which mice were treated with NLC alone, the expression of CD133, ki67, p-PI3K, p-AKT, and mTOR was decreased and caspase3 was increased by the treatments of free DOX, free PTX, DOX-NLC, PTX-NLC, as well as PTX-DOX-NLC. Moreover, the treatment with PTX-DOX-NLC resulted in the most significant effects (figure 5(D)). Thus, we concluded that PTX-DOX-NLC could suppress the proliferation and accelerate apoptosis of GSCs by inhibiting the PI3K/Akt/mTOR signaling in vivo.