Enhancing the anti-glioma therapy of doxorubicin by honokiol with biodegradable self-assembling micelles through multiple evaluations

Combination chemotherapy is an important protocol in glioma therapy and honokiol shows synergistic anticancer effects with doxorubicin. In this paper, honokiol (HK) and doxorubicin (Dox) co-loaded Methoxy poly(ethylene glycol)-poly(ε-caprolactone) (MPEG-PCL) nanoparticles were prepared with a assembly method. The particle size (about 34 nm), morphology, X-ray Powder Diffraction (XRD), in vitro release profile, cytotoxicity and cell proliferation effects were studied in detail. The results indicated that honokiol and doxorubicin could be efficiently loaded into MPEG-PCL nanoparticles simultaneously, and could be released from the micelles in an extended period in vitro. In addition, honokiol and doxorubicin loaded in MPEG-PCL nanoparticles could efficiently suppress glioma cell proliferation and induce cell apoptosis in vitro. Furthermore, Dox-HK-MPEG-PCL micelles inhibited glioma growth more significantly than Dox-MPEG-PCL and HK-MPEG-PCL in both nude mice and zebrafish tumor models. Immunohistochemical analysis indicated that DOX-HK-MPEG-PCL micelles improved Dox’s anti-tumor effect by enhancing tumor cell apoptosis, suppressing tumor cell proliferation, and inhibiting angiogenesis. Our data suggest that Dox-HK-MPEG-PCL micelles have the potential to be applied clinically in glioma therapy.


Results
Preparation and characterization of DOX-HK-M nano-micelles. DOX-HK-M nano-micelles were prepared by a two-step self-assembly procedure, as shown schematically in Fig. 1C. First, MPEG-PCL copolymer and HK were co-dissolved in acetone. The organic phase was then evaporated in a rotary evaporator under reduced pressure, and subsequent addition of water formed the core-shell structured HK/MPEG-PCL nano-micelles with core-encapsulated HK. Afterward, phosphate-buffered saline (PBS) and doxorubicin solution were added into the HK/MPEG-PCL nano-micelles under continuous mechanical stirring, thus completing the DOX-HK-M nano-micelle preparation. This self-assembly procedure of amphiphilic MPEG-PCL, HK, and DOX created core-shell structured DOX-HK-M nano-micelles with core-encapsulated HK and DOX.
The DOX-HK-M nano-micelles were characterized in detail. The drug loading (DL) and Encapsulation efficiency (EE) of Dox and HK in DOX-HK-M nano-micelles were 5% and 93.4% (for Dox), and 5% and 99.8% (for HK). The particle size distribution spectrum of freshly-prepared DOX-HK-M nano-micelles is shown in Fig. 2A. The polydispersity index (PDI) of the prepared DOX-HK-M nano-micelles was 0.12, with a mean particle size of 34 nm, indicating the narrow particle size distribution of DOX-HK-M nano-micelles. The zeta potential of DOX-HK-M nano-micelles was − 2.3 mv, presented in Fig. 2B. Furthermore, the morphology of DOX-HK-M nano-micelles was determined by transmission electron microscopy (TEM), which is shown in Fig. 2C. The TEM images revealed that the DOX-HK-M nano-micelles were spherically-shaped in aqueous solution with a mean diameter of 29 nm, which was in good accordance with the results of dynamic light scattering (DLS). DLS allows The HK-Dox-M nano-micelles were prepared in twosteps by a self-assembly method. First the HK (A) and MPEG-PCL (B) were co-dissolved in acetone, and the mixture was then added into water with stirring. Secondly, 1 mL of 10x phosphate-buffered saline (PBS: 0.1 M, pH 7.4) was added to HK/MPEG-PCL nano-micelles (8 mL) and mixed completely. Next, 1 mL of doxorubicin solution (5 mg/mL) was dripped slowly into the mixture under continuous stirring by a mechanical stirrer.
for observation of the diameter of particles in aqueous solution, whereas TEM provides the diameter of particles in dry powder. The structure of amphiphilic block polymeric micelles was looser in aqueous solution, which could explain the slightly larger particle size determined by DLS than that by TEM. Combination of the above results showed that the prepared DOX-HK-M nano-micelles were stable and homogeneous in aqueous solution.
The appearance of DOX-HK-M nano-micelles in aqueous solution is shown in Fig. 2D. HK couldn't be dissolved in aqueous solution, instead forming a turbid white slurry, while the DOX-HK-M nano-micelles could be well-dissolved in aqueous solution as demonstrated by the clear solution. What is more, the dox can be filtered from ultrafiltration tube in the free DOX group and not be filtered in the DOX-HK-M group (shown in Fig. 2E). Those indicated that the HK and DOX were incorporated into the micelles. Scientific RepoRts | 7:43501 | DOI: 10.1038/srep43501 X-ray Powder Diffraction. X-ray Powder Diffraction (XRD) spectra of the HK powder, DOX powder, lyophilized blank MPEG-PCL nano-micelles, and DOX-HK-M nano-micelles are presented in Fig. 3. Compared with the XRD diagram of Dox and HK powder, the specific X-ray diffraction peaks of HK and Dox power disappeared in the diagram of DOX-HK-M nano-micelles.
In vitro drug release behavior. The release profiles of HK-M nano-micelles, DOX-M nano-micelles and DOX-HK-M nano-micelles in PBS (pH 7.4, 37 °C) with 10% fetal bovine serum are presented in Fig. 4. Dox and HK were able to be released from the DOX-HK-M nano-micelles, and the cumulative release rates of HK or DOX from the DOX-HK-M nano-micelles were slower when compared with the HK-M nano-micelles or DOX-M nano-micelles alone.
In vitro cytotoxicity evaluation. The in vitro cytotoxicity profiles of HK-M nano-micelles, DOX-M nano-micelles, and DOX-HK-M nano-micelles were determined by an MTT assay using C6 glioma cells for 24 (Fig. 5A) and 48 hours (Fig. 5B). As shown in Fig. 5, both DOX-M nano-micelles and DOX-HK-M nano-micelles-but not HK-M nano-micelles-inhibited the cell viability of C6 glioma cells in a dose-dependent manner. Compared with the HK-M nano-micelles and Dox-M nano-micelles, the DOX-HK-M nano-micelles conferred greater cytotoxicity to C6 glioma cells under the same dose of HK or DOX.

Induction of apoptosis by DOX-HK-M nano-micelles. Hoescht staining analysis was performed on
C6 glioma cells in order to investigate whether apoptosis contributed to the inhibition of cell growth. Stains were performed on C6 glioma cells in order to investigate the apoptotic potential of HK-M, DOX-M, and DOX-HK-M nano-micelles. As seen in Fig. 6, images showed more apoptotic cells with condensed and fragmented nuclei in DOX-HK-M nano-micelles group was than those of other groups.
Scientific RepoRts | 7:43501 | DOI: 10.1038/srep43501 Cellular uptake of DOX-HK-M nano-micelles. To reveal the mechanism of DOX-HK-M nano-micelles in enhancing cellular cytotoxicity and inducing apoptosis, a cellular uptake study of the nano-micelles was performed on C6 glioma cells. As shown in Fig. 8, no red fluorescence could be observed in the blank-M nano-micelles group, nor the HK-M nano-micelles group, at 4 hours. Compared with the DOX-M nano-micelles group, the DOX-HK-M nano-micelles group exhibited a brighter red fluorescence at 4 hours, indicating increased cellular uptake of DOX; the same result was also determined by FCM analysis. As seen in Fig. 9, the FCM histograms elucidated that the cellular uptake of DOX in DOX-HK-M nano-micelles group was greater than that of the other groups after 4 h incubation.  In vivo antitumor effects in tumor xenograft zebrafish models. Three days after injection of perivitelline into the zebrafish, U87 cells formed a solid tumor (green fluorescence, Fig. 11). Shown in Fig. 11, antitumor effects were observed in HK-M, DOX-M, and DOX-H-M nano-micelles groups when compared with the control and blank micelles group. As detected by confocal microscopy, the tumors in the DOX-H-M nano-micelles group were the smallest of all the groups, including the H-M and DOX-M nano-micelles groups; no antitumor effects was observed in blank micelles group. The average tumor volume in the DOX-HK-M nano-micelles group was significantly smaller than that of control, blank micelles, HK-M nano-micelles, and DOX-M nano-micelles groups. These data indicated that HK could enhance the anti-glioma cancer activity of Dox in vivo.

DOX-HK-M nano-micelles inhibited embryonic angiogenesis in a Tg
In vivo antitumor effects in a subcutaneous tumor model. Subcutaneous injection of C6 glioma cells was used to evaluate the antitumor effects of DOX-HK-M nano-micelles. Figure 12A and B show the tumor growth curves and weight in each group, respectively; Fig. 12C shows representative tumors from each group. According to the results, DOX-HK-M nano-micelles exhibited stronger antitumor effects than that of the other groups, while the blank micelles exhibited no antitumor effect at all. Figure 12D shows the body weight curves of different groups, demonstrating that there was no change among the groups. Determination of tumor cell apoptosis. An immunofluorescent TUNEL assay was used to evaluate the degree of apoptosis in each group; only in the regions of intact tumor cells were the TUNEL positive cells counted, while the cells in the central necrotic region were excluded. As shown in Fig. 13A-E, there were more apoptotic tumor cells in the DOX-HK-M nano-micelles group than in other groups. In Fig. 13F, the apoptotic index in the DOX-HK-M nano-micelles group (128 per field) was significantly higher than that in control (2 per field, p < 0.05), blank micelles (2 per field, p < 0.05), HK-M nano-micelles (41 per field, p < 0.05) and DOX-M nano-micelles (88 per field, p < 0.05) groups, while there is no difference between the control and blank micelles groups.  Alginate-encapsulated tumor cell assay. An alginate-encapsulated tumor cell assay was used to further investigate the anti-angiogenic effects of DOX-HK-M nano-micelles in vivo. As seen in Fig. 15A-D, sparser blood vessels in alginate beads were observed in the DOX-HK-M nano-micelles group than that of other groups. Uptake of FITC-dextran was also measured in order to quantify angiogenesis in the alginate implant. As shown in Fig. 15E, uptake of FITC-dextran in the DOX-HK-M nano-micelles group (0.53 μ g per bead) was significantly lower compared to that in the control (3.05 μ g per bead, p < 0.05), blank micelles (3.12 μ g per bead, p < 0.05), HK-M nano-micelles (1.53 μ g per bead, p < 0.05), and DOX-M nano-micelles (1.77 μ g per bead, p < 0.05) groups. These results indicated that DOX-HK-M nano-micelles significantly suppressed tumor angiogenesis in tumor-bearing mice, which might participate in the inhibition of tumor growth and metastasis. Immunohistochemical

Discussion
Chemotherapy is one of the most basic treatment strategies for cancer. Unfortunately, the administration of anti-cancer drugs often causes serious side effects such as nausea, vomiting, diarrhea, and hair loss, which limits the dose and frequency of chemotherapy. In recent years, nanomedicine, with fewer side effects and well-controlled release of drugs, has shown promising applications in the field of drug delivery 27,28 . DOX is a hydrophobic drug used mainly for the treatment of lung, breast, liver, prostate, and brain cancers, with the cardiac toxicity limiting the dose that can be used 29,30 . HK, on the other hand, was demonstrated to be a potential chemosensitizer and have therapeutic effect in various cancers 13,31,32 , but the therapeutic effect of HK is restricted by its poor water solubility.
In this study, we used MPEG-PCL micelles to co-deliver HK and DOX, and successfully developed an intravenously-injectable formulation. Furthermore, our formulation was characterized in detail, and its anti-angiogenic and anti-tumor effects were evaluated both in vitro and in vivo. In this protocol, the DOX-HK-M nano-micelles were prepared by a two-step nano-precipitation method, which was easy to scale up. The MPEG-PCL copolymer self-assembled into core-shell structures, encapsulating HK, DOX, or DOX-HK during the process. The DL and EL of the DOX-HK-M nano-micelles were 5% and 93.4%, respectively. The DOX-HK-M nano-micelles were monodisperse, with an average particle size of about 34 nm. As confirmed by the transparent appearance of DOX-HK-M nano-micelles, encapsulation in the nano-micelles enabled both DOX and HK to completely disperse in aqueous solution (Fig. 2D).
Several investigations about the therapeutic effects of the DOX-HK-M nano-micelles were made in this work. An in vitro release study showed that DOX-HK-M nano-micelles released DOX and HK more slowly than that of HK-M nano-micelles or DOX-M nano-micelles individually. Combining their small size (34 nm) and slow release of DOX and HK, the DOX-HK-M nano-micelles could potentially circulate in vivo for a long time after systemic application, which may help to maintain plasma drug concentrations.
Our DOX-HK-M nano-micelles showed stronger anti-cancer effects than DOX-M nano-micelles or HK-M nano-micelles both in vitro and in vivo. In vitro, DOX-HK-M nano-micelles caused increased cytotoxicity and induction of apoptosis in C6 glioma cells when compared with DOX-M or HK-M nano-micelles, and no difference was observed between the blank micelles-treated and control group, as confirmed by an MTT assay, Hoescht staining, and FCM analysis. Moreover, in vivo studies showed that systemic application of DOX-HK-M nano-micelles was more efficient in inhibiting tumor growth than DOX-M or HK-M nano-micelles in a subcutaneous C6 glioma model, which was further confirmed by the staining of TUNEL, CD31, and Ki67 in tumor tissues. In addition, the anti-tumor effects were also investigated in a tumor xenograft zebrafish model, which indicated stronger inhibition of tumor growth in the DOX-HK-M nano-micelles treated group than that of other groups. Finally, a cellular uptake Scientific RepoRts | 7:43501 | DOI: 10.1038/srep43501 study was implemented by using a fluorescence microscope, with its results suggesting that the increased cytotoxicity of DOX-HK-M nano-micelles was related to their enhanced cellular uptake.
Zebrafish are widely used as an ideal model to screen small molecules that affect vascular formation due to their well-described and highly conserved vascular system 33,34 . In this work, Tg (FLK-1: EGFP) zebrafish were used to investigate the inhibitory effects of DOX-HK-M nano-micelles on embryonic angiogenesis, as their endothelial cells exhibit green fluorescence. In the embryonic anti-angiogenesis assay, DOX-HK-M nano-micelles caused stronger inhibitory effects on angiogenesis than that of DOX-M nano-micelles or HK-M nano-micelles, while the blank micelles showed no inhibitory effects at all. In addition, the alginate-encapsulated tumor assay, a specific tumor angiogenesis model in vivo, demonstrated that DOX-HK-M nano-micelles could more efficiently inhibit tumor angiogenesis than that of DOX-M or HK-M nano-micelles. Therefore, the enhanced anti-tumor effects of DOX-HK-M nano-micelles may be due to direct cytotoxicity of cancer cells and inhibitory effects on tumor angiogenesis [35][36][37][38][39] . Alternatively, angiogenic blood vessels in tumor tissues, which are different from those in normal tissues, have gaps between adjacent endothelial cells 40 . This defective structure, coupled with poorly developed lymphatic drainage, creates the enhanced permeability and retention (EPR) effect, which permits the nanoparticles to extravasate through these gaps into extravascular spaces and then accumulate inside tumor tissues 41 . For such a passive targeting mechanism (i.e., the EPR effect) to work, the circulation time of nanoparticles must be long enough. To maximize the targeting effect and circulation time, the optimal nanoparticle diameter must be less than 100 nm, and a hydrophilic surface is needed to avoid clearance by macrophages 40 . The monodisperse DOX-HK-M nano-micelles are small enough < 100 nm) for this purpose. Indeed, the prepared DOX-HK-M nano-micelles also have a core-shell structure, allowing the hydrophilic PEG to act as a brush-like protective shell. As such, the particles' small size and hydrophilic shell may contribute to increasing the accumulation of DOX and HK in tumor tissues due to the EPR effect, thus enhancing the anti-cancer effects of DOX and HK.  In this work, we successfully prepared intravenously-injectable DOX-HK-M nano-micelles, and their enhanced anti-cancer effects when compared to DOX-M and HK-M nano-micelles was due to the direct cellular cytotoxicity, stronger inhibitory effects on angiogenesis, and slower release, causing the EPR effect. In summary, our findings indicate that the DOX-HK-M nano-micelles have potential application in the treatment of glioma.

Conclusion
Monodisperse DOX-HK-M nano-micelles were prepared by a self-assembly method, which rendered DOX and HK completely dispersible in aqueous solution. They had promising effects on inhibiting the growth of C6 glioma cells, both in vitro and in vivo, by directly killing cancer cells, inducing apoptosis, decreasing the density of tumor microvasculature, inhibiting angiogenesis, and EPR effects. In conclusion, DOX-HK-M nano-micelles, an excellent intravenously-injectable formulation of DOX and HK, hold promising clinical application in cancer therapy in the future.  Characterization of DOX-HK-M nano-micelles. The particle size distribution and zeta potential of prepared DOX-HK-M nano-micelles were measured by dynamic light scattering (DLS) (Malver Nano-ZS 90; Malvern Instruments, Malvern, UK). The temperature was kept at 25 °C during the measuring process. All results are the mean of three independent test runs.

Materials
The morphological characteristics of the DOX-HK-M nano-micelles were observed under a transmission electron microscope (TEM) (H-6009IV; Hitachi, Tokyo, Japan). Micelles were diluted with distilled water and placed on a copper grid covered with nitrocellulose. Samples were negatively stained with phosphotungstic acid and dried at room temperature before observation.
The drug loading (DL) and encapsulation efficiencies (EE) of DOX-HK-M, HK-M, and DOX-M nano-micelles were determined using HPLC. Briefly, 10 mg of lyophilized DOX-HK-M, HK-M, and DOX-M nano-micelles were dissolved in 0.1 mL of methanol. The concentrations of HK and DOX were determined by high-performance liquid chromatography (HPLC, Waters Alliance 2695; Waters, Milford, MA). The solvent delivery system was equipped with a plus auto-sampler and a column heater. Detection was conducted on a Waters 2966 detector. Chromatographic separations were performed on a reversed-phase C 18 column (4.6 × 150 nm, 5 μ m, Sunfire Analysis column), and the column temperature was kept at 28 °C. Methanol-water (70/30, v/v) was used at a flow rate of 1 mL/min as eluent. The EE and DL of each micelle variant was calculated according to equations (1) and (2): Experimental drug loading Theoretical drug loading 100% (2) Crystallographic studies. Crystallographic assays were performed on doxorubicin powder, honokiol powder, lyophilized blank MPEG-PCL nano-micelles, and DOX-HK-M nano-micelles using an X-ray diffractometer (XRD) (X' Pert Pro, Philips, Netherlands) with Mo Kα radiation. microtiter terazolium (MTT) assay. Cells in the exponential phase were seeded at densities of 5 × 10 3 and 3 × 10 3 cells per well in 96-well culture plates and grown for 24 h at 37 °C in the presence of 5% CO 2 . The cells were then exposed to each formulation at different concentrations ranging from 20 to 320 ng/mL for 24 or 48 hours, respectively. After these treatments, each well was added to 20 μ L of 5 mg/mL MTT solution and then incubated for 3 hours. The supernatants were aspirated and 150 μ L of dimethyl sulfoxide were added in each well to dissolve the formazan crystals for 10 minutes with shaking. Finally, the absorbance at 570 nm was read on a microplate reader.
The mean percentage of cell survival relative to that of untreated cells was estimated using data from six individual experiments and all data are presented as the means ± SD. (The relative cell viability was estimated/determined by comparing the absorbance at 570 nm with the control wells which contained only cell culture media).
Apoptosis assays. Hoescht staining assay. C6 glioma cells were treated by NS, blank micelles (0.08 μ g/ml), HK-M nano-micelles (HK: 0.08 μ g/mL), DOX-M nano-micelles (Dox: 0.08 μ g/mL) and DOX-HK-M nano-micelles (HK and Dox: 0.08 and 0.08 μ g/mL) for 48 hours in 24-well plates and washed with cold PBS twice. Hoescht 33342 was then added to the culture media at a final concentration of 5 μ g/mL and the changes of nuclear morphology in treated cells were observed immediately under fluorescence microscopy with a filter for Hoescht 33342. Embryos were maintained at 28 °C in Holtfreter's solution after transplantation procedures. Embryos at 14 hours post-fertilization (hpf ) were exposed to Holtfreter's solution (control), blank micelles, H-M nano-micelles, DOX-M nano-micelles, and DOX-H-M nano-micelles at a final DOX and/or HK concentration of 1 μ g/mL. At 24 hpf, embryos were stripped off the egg sheath and anaesthetized with 0.01% tricaine. Finally, the images of zebrafish blood vessels were taken under a Zeiss M2Bio fluorescence microscope (Carl Zeiss Microimaging Inc.). Ten embryos per group were used for these experiments, and each experiment was performed in three independent replicates.
Anti-tumor activities in a transgenic zebrafish model. Our previous work showed that implanting U87 tumor cells into the perivitelline space of zebrafish allowed for tumor growth in zebrafish embryos 43 . Therefore, these zebrafish with tumor xenografts could serve as an ideal model to investigate the anti-tumor effect of our nano-micelles. Indeed, the U87 tumor cells were also transfected with EGFP by lentivirus in order to aid in analysis. Zebrafish embryos were stripped off the egg sheath and anesthetized with 0.01% tricaine at 48 hpf. Using a Zeiss Stemi 2000-C dissecting microscope (Carl Zeiss Microimaging Inc., Thornwood, NY), zebrafish were then injected with 300 tumor cells into the perivitelline space using a Cell Tram Vario injector (Eppendorf, USA) with a glass micropipette (50 mm length, diameter of the needle opening about 25 μ g/mL). At 24 h post-implantation, Holtfreter's solution (control), blank micelles, H-M nano-micelles, DOX-M nano-micelles, and DOX-H-M nano-micelles were added into the incubating Holtfreter's solution at a final DOX and/or HK concentration of 1 μ g/mL. Finally, images of the tumors were taken using a confocal microscope (DM6000 CS, Leica, Germany) at 5 days post-implantation.
Subcutaneous injection of C6 glioma cells. A brain tumor model was established by subcutaneous injection of 1 × 10 7 C6 glioma cells in exponential phase into the right flank of BALB/c nude mice (6-8 weeks, 20 ± 2 g). Mice bearing tumors around 100 mm 3 were selected and randomly assigned to 5 groups (5 mice per group). The five groups were intravenously injected with NS (control), blank micelles (3 mg/kg), HK-M nano-micelles (3 mg/kg), DOX-M nano-micelles (3 mg/kg), or DOX-HK-M nano-micelles (3 mg/kg) four times, once every five days. The tumor-bearing mice in each group were weighed every three days. The tumor size was also measured externally every three days using calipers during the experimental period. The tumor volume was approximated according to the following equation: tumor volume = 0.52 × length × width 2 . The mice were sacrificed at the end of the experiment. Solid tumors from each group were then harvested and processed for subsequent immunohistochemical analysis and terminal deoxynuleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assay, described below.
Scientific RepoRts | 7:43501 | DOI: 10.1038/srep43501 TUNEL assay. Cellular apoptosis in C6 gliomas was determined using a TUNEL assay. First, the subcutaneous tumor tissues described above from each group were harvested, fixed in 4% (w/v) paraformaldehyde, embedded in paraffin, and sectioned according to conventional methods. TUNEL staining was performed with an in situ cell death detection kit (DeadEnd TM Fluorometric TUNEL System, Promega, Madison, USA) according to the manufacturer's protocol. Five high-power fields of sections from each group were randomly selected and analyzed. The apoptotic index was calculated as the number of apoptotic cells in each high-power field.
Alginate-encapsulated tumor cell assay. An alginate-encapsulated tumor cell assay was carried out as previously described to evaluate the anti-angiogenic activity of DOX-HK-M nano-micelles 44 . Briefly, C6 glioma cells were re-suspended in an alginate solution (1.5% w/v) and then dropped into a calcium chloride solution (250 mM) to form alginate beads containing 1 × 10 5 tumor cells per bead. At day 0, the prepared alginate beads were subcutaneously implanted into an incision made on the dorsal sides of nude BALB/c mice. Afterwards, fifteen bead-bearing mice were randomly assigned to five groups (3 mice per group), and were intravenously injected with NS (control), blank micelles (3 mg/kg), HK-M nano-micelles (3 mg/kg), DOX-M nano-micelles (3 mg/kg), or DOX-HK-M nano-micelles (3 mg/kg) once a day for 3 days. Finally, mice were intravenously injected with 0.1 mL of a 100 mg kg-1 FITC-dextran solution at day 12. Alginate beads were imaged and rapidly removed within 20 minutes following the injection of FITC-dextran solution. The uptake of FITC-dextran was measured as described previously 44 .
Immunohistochemical analysis. The angiogenic activity and cell proliferative potential of each tumor was detected by CD31 or Ki-67 immunofluorescence staining of paraffin-embedded tumor tissue sections, described above. Briefly, frozen sections of tumor tissue from each group were fixed in acetone, washed with PBS, and then stained with a rat anti-mouse CD31 polyclonal antibody (BD Pharmingen TM , USA) or rabbit anti-mouse Ki67 monoclonal antibody (Abcam, USA). Subsequently, the sections were washed with PBS twice, and were then incubated with a rhodamine-conjugated secondary antibody (Abcam, USA). The microvessel density (MVD) was determined by counting the number of microvessels per high-power field (400x magnification) in each section using a fluorescence microscope. Five high-power fields of tumor section were randomly selected and analyzed in a blind fashion by two independent investigators for this procedure. The Ki-67 labeling index (Ki-67 LI) was used to quantify the expression of Ki-67 by calculating the number of KI-67-positive cells per total number of cells.
Statistical analysis. All data were expressed as the means ± SD. Statistical analysis was performed by one-way analysis of variance (ANOVA) (Tukey test) using SPSS software, version 11.5. Differences were considered to be statistically significant if p < 0.05.