The U87 MG American Type Culture Collection (ATCC) human glioblastoma cell line of unknown origin (cat no. CL-0238), SHG-44 human glioma cell line (cat no. CL-0207), 293T cell line (cat no. CL-0005) and human umbilical vein endothelial cell line (HUVEC; cat no. CL-0122) were purchased from Procell Life Science & Technology Co., Ltd. The SVG p12 human astroglia cell line (cat no. CRL-8621) was purchased from the ATCC. The HUVECs used in the present study were not primary cells. Suppliers have authenticated all the aforementioned cell lines by short tandem repeat profiling. ADSCs (cat no. HUXMD-01001) were purchased from Cyagen Biosciences, Inc. A total of 15 male, 6-week-old, specific pathogen-free, BALB/C nude mice, weighing 16–18 g, were purchased from Changzhou Cavens Laboratory Animal Co., Ltd., and were housed in a full-barrier rodent facility with a filtered air supply at constant temperature (25°C) and humidity (50%) under a 12-h light/dark cycle. Water and food for animals were provided ad libitum. The housing environment for animals, water and animal feeds were sterilized. All animal procedures were performed in accordance with the guidance of the Research Ethics Committee of The First Affiliated Hospital of Xi'an Jiaotong University (approval number: 2020-G-263).

The U87, SHG44, 293T and SVG p12 cells were cultured in Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.). HUVECs were cultured in F12K medium (Procell Life Science & Technology Co., Ltd.) supplemented with 10% fetal bovine serum. ADSCs were cultured in human adipose-derived stem cell growth medium (cat. no. HUXMD-90011; Cyagen Biosciences, Inc.). All cells were cultured in an atmosphere of 5% CO₂ at 37°C.

The lentiviral vector plasmid pLVX-mCMV-ZsGreen-IRES-Puro was purchased from Wuhan Viraltherapy Technologies Co., Ltd. The pUC57-NT4-alphastatin plasmid containing the sequence encoding the human neurotrophin-4 (NT4) signal peptide and alphastatin fusion gene fragment (NT4-Al) and FLAG tag was synthesized and purchased from GenScript. The NT4-Al-FLAG sequence was cloned into pLVX-mCMV-ZsGreen-IRES-Puro and identified by restriction enzyme digestion and sequence analysis. Lentiviral particles were produced by the transfection of 293T cells with a lentivirus packaging kit (cat. no. R003; Wuhan Viraltherapy Technologies Co., Ltd.) according to the manufacturer's instructions and lentiviral supernatants were collected after 48 h. The construction of negative control (NC) lentivirus was performed as described above. ADSCs were seeded in 24-well plates at a concentration of 5×10⁴ cells/well and infected with concentrated alphastatin lentiviral particles (Al-ADSCs) or NC lentiviral particles (NC-ADSCs; multiplicity of infection=20) in an atmosphere of 5% CO₂ at 37°C for 24 h. Successful transfection of cells was verified by expression of green fluorescent protein.

Flow cytometric detection of MSC surface markers was used to determine whether lentivirus transfection affected the biological characteristics of ADSCs. Al-ADSCs, NC-ADSCs and ADSCs were seeded in six-well plates at a concentration of 5×10⁵ cells/well and cultured in an atmosphere of 5% CO₂ at 37°C for 48 h. The cells were then quantified using a cell counting board and resuspended with 0.1 ml phosphate-buffered saline (PBS) containing 0.5% bovine serum albumin (BSA; Thermo Fisher Scientific, Inc.). Allophycocyanin (APC)-conjugated anti-human CD13 antibody (cat. no. 301705; BioLegend, Inc.), APC-conjugated anti-mouse/human CD44 antibody (cat. no. 103011; BioLegend, Inc.), phycoerythrin (PE)/Cyanine7-conjugated anti-human CD90 (Thy1) antibody (cat. no. 328123; BioLegend, Inc.) and PE-conjugated anti-human CD105 (endoglin) antibody (cat. no. 12-1057-41; eBioscience; Thermo Fisher Scientific, Inc.) were added according to the manufacturer's instructions. Following incubation at 4°C for 30 min, the cells were washed twice with PBS containing 0.5% BSA. A single cell suspension was prepared by the addition of 0.2 ml PBS and stem cell surface markers were then detected using flow cytometry (CytoFLEX; Beckman Coulter, Inc.) and CytExpert 2.4 software (Beckman Coulter, Inc.).

The expression level of alphastatin was detected indirectly by detecting the expression of the FLAG tag. Extraction of total cellular proteins and western blot analysis were performed as previously described (13). Briefly, cells were washed twice with PBS and lysed on ice following the addition of 300 µl RIPA buffer (Beyotime Institute of Biotechnology) with 1 mmol/l phenylmethylsulfonyl fluoride. Protein concentration was determined using a bicinchoninic acid protein assay kit (Beyotime Institute of Biotechnology). Protein samples (40 µg) were boiled in 1X SDS-PAGE sample loading buffer, resolved using 10% SDS-PAGE and transferred onto polyvinylidene fluoride membranes (MilliporeSigma). The membranes were then blocked with TBS-0.1% Tween-20 (TBST) containing 5% non-fat dry milk at room temperature for 2 h and probed with primary antibodies overnight at 4°C, followed by incubation with secondary antibodies conjugated to horseradish peroxidase at 37°C for 2 h. Membranes were developed using SuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo Fisher Scientific, Inc.). The primary antibodies employed in western blotting were as follows: Goat anti-FLAG tag antibody (1:1,000; cat. no. ab1257; Abcam) and rabbit anti-GAPDH antibody (1:5,000; cat. no. ap0063, Bioworld Technology, Inc.). The secondary antibodies employed in western blotting were as follows: HRP-conjugated goat anti-rabbit immunoglobulin G (IgG) antibody (1:5,000; cat. no. BA1054) for GAPDH detection and HRP-conjugated rabbit anti-goat IgG antibody (1:5,000; cat. no. BA1060) for FLAG tag detection (both from Wuhan Boster Biological Technology, Ltd.). ImageJ 1.8 software (National Institutes of Health) was used for densitometry.

The secretion levels of alphastatin from alphastatin lentivirus-transfected ADSCs (Al-ADSCs) were assessed using a FLAG tag protein ELISA kit (cat. no. E4700-100; BioVision, Inc.). Briefly, the supernatant of ADSCs in each group was collected at 48 h following infection with lentivirus. The samples were centrifuged at room temperature for 20 min at 850 × g to remove particulates and were immediately assayed using the aforementioned kit according to the manufacturer's instructions.

U87 and SHG44 cells were resuspended in 1 ml PBS containing 0.5% BSA in an Eppendorf tube containing with 1×10⁶ cells per tube. Following centrifugation (200 × g for 3 min at 4°C), the cells were washed with PBS containing 0.5% BSA once and then incubated with 5 µl anti-CD133 antibody (cat. no. 17-1338-42; Invitrogen; Thermo Fisher Scientific, Inc.) at 4°C for 30 min. Subsequently, the cells were washed twice with PBS containing 0.5% BSA and resuspended for cell sorting using a flow cytometer (FACSAria III; BD Biosciences).

Cell migration assay was used to analyze the tropism of Al-ADSCs to GSCs in vitro. Transwell chambers (cat. no. 353097; BD Biosciences) with 8-µm pore size polyethylene terephthalate (PET) membranes were placed in 24-well plates (Corning, Inc.). GSCs, CD133⁻ glioma cells or SVG p12 cells were seeded in the lower chambers (2.5×10⁵ cell/well) and cultured overnight in an atmosphere of 5% CO₂ at 37°C. Cell suspension of Al-ADSCs, NC-ADSCs or ADSCs (3×10⁵ cells/ml) was then added to the upper chambers (200 µl/well) and co-cultured with cells in the lower chamber in an atmosphere of 5% CO₂ at 37°C overnight. The migrated cells adhering to the bottom surface of the membranes were fixed with a 70% ethanol solution at 4°C for 1 h and stained with 0.5% crystal violet (cat. no. C0121; Beyotime Institute of Biotechnology) at room temperature for 10 min. The migrated cells were then counted under an optical microscope (IX51; Olympus Corporation; magnification, ×200).

The effect of Al-ADSCs on the angiogenesis of ECs was analyzed using a tube formation assay in vitro. Matrigel (cat. no. 356234; Corning, Inc.) was incubated at 4°C for 24 h and then 0.1 ml Matrigel was evenly plated onto 24-well plates (Corning, Inc.), precooled at −20°C for ≥10 min and then incubated for 30 min at 37°C. Subsequently, HUVECs (1.5×10⁵ cells/well) were seeded onto the Matrigel-coated plates and incubated in an atmosphere of 5% CO₂ at 37°C for 2.5 h. Transwell chambers (cat. no. 353095; BD Biosciences) with 0.4-µm pore size PET membranes were then placed in the 24-well plates and a cell suspension of Al-ADSCs, NC-ADSCs or ADSCs (3×10⁵ cells/ml) was added to the Transwell chambers (200 µl/well). The cells in the upper and lower chambers were co-cultured in an atmosphere of 5% CO₂ at 37°C for 8 h. The EC-derived tube-like structures in each well were visualized and analyzed directly under an optical microscope (IX51; Olympus Corporation; magnification, ×100).

A scratch wound healing assay was used to examine the effects of Al-ADSCs on the migration of ECs. The HUVECs were seeded at a density of 5×10⁵ cells/well in six-well plates. After 24 h, the cell monolayer was scraped in a straight line using a 20-µl pipette tip and the cells were washed three times with PBS. Transwell chambers (cat. no. 353090; BD Biosciences) with 0.4-µm pore size PET membranes were then placed in the six-well plates and a cell suspension of Al-ADSCs, NC-ADSCs or ADSCs (5×10⁵ cells/well) was added to the Transwell chambers. The cells in the upper and low chambers were co-cultured in serum-free medium in an atmosphere of 5% CO₂ at 37°C for 24 h. Images of the plates were captured under an optical microscope (IX51; Olympus Corporation; magnification, ×100) at 0 and 24 h, respectively. ImageJ 1.8 software (National Institutes of Health) was used to measure the migration area and the percentage of wound healing was calculated. The calculation method was (Area _{0 h}-Area _{24 h)}/Area _{0 h}.

A CCK-8 kit (cat. no. E-CK-A362; Elabscience Biotechnology, Inc.) was used to examine the effect of Al-ADSCs on the proliferation of ECs. HUVECs were seeded at a density of 5×10³ cells/well in 24-well plates. After 24 h, Transwell chambers (cat. no. 353095; BD Biosciences) with 0.4-µm pore size PET membranes were placed in the 24-well plates and a cell suspension of Al-ADSCs, NC-ADSCs or ADSCs (5×10³ cells/well) was added to the Transwell chambers. The cells in the upper and low chambers were co-cultured in an atmosphere of 5% CO₂ at 37°C. The HUVECs in low chambers were then analyzed by CCK-8 assay at 24 and 48 h following co-culture. Briefly, 50 µl CCK-8 solution was added to each well. Following incubation at 37°C for 4 h, the absorbance at 450 nm was measured with a microplate reader (Flexstation 3; Molecular Devices, LLC).

The establishment of intracranial xenograft models was conducted as previously described (14). The 15 BALB/C nude mice were randomly divided into three groups. To prevent accidental animal death, one additional mouse was added to each group. Each group of six mice was randomly numbered 1 to 6. The animals were anaesthetized with an intraperitoneal injection of 30 mg/kg pentobarbital sodium. U87 glioma cells transfected with LV-mCherry-Puromycin Lentivirus [cat. no. FLV050; Fubio (Suzhou) Biomedical Technology Co., Ltd.] and able to stably express a red fluorescent protein mCherry were injected intracranially using a stereotaxic apparatus (2.5 µl/mouse; cell concentration, 2×10⁵ cells/10 µl). At day 3 after model establishment, a suspension of NC-ADSCs or Al-ADSCs (2.5 µl/mouse; cell concentration, 2×10⁵ cells/10 µl) was injected into the periphery of the tumor modeling area. The number of intracranial injection cells followed the guidelines for the welfare and use of animals in cancer research (15). Animal health and behavior were monitored twice a week. After 28 days of tumor cell implantation, due to no accidental animal death, 5 mice numbered 1 to 5 in each group underwent subsequent experiments, while mice numbered 6 continued to be raised until the humane endpoint. The 15 tumor-bearing mice (5 mice/group) were sacrificed after observing intracranial tumorigenesis using an in vivo imaging instrument (IVIS Lumina III; PerkinElmer, Inc.) at 28 days following tumor cell implantation by an intraperitoneal injection of 200 mg/kg pentobarbital sodium solution. Cardiac arrest for 2 min was used to identify mortality. The humane endpoints included labored breathing, inability to remain upright, impaired mobility, a hunched posture for >48 h and no response to external stimuli. The mice were perfused with 4% paraformaldehyde for 1 h at 4°C immediately following sacrifice, before the brain and tumor samples were obtained and fixed overnight with 4% paraformaldehyde at 4°C. Paraffin-embedded sections were prepared of a thickness of 4 µm. Briefly, the tissues were dehydrated by serial incubations in 75% alcohol (4 h), 85% alcohol (2 h), 90% alcohol (1.5 h), 95% alcohol and 100% alcohol (2 times, 0.5 h each). The tissues were then washed in a 1:1 mixture of 100% alcohol and xylene (10 min) and xylene (2 times, 10 min each). After washing, the tissues were immersed in paraffin (3 times, 1 h each) and then embedded in paraffin blocks. The distribution of NC-ADSCs or Al-ADSCs was observed under a fluorescence microscope (BX53; Olympus Corporation).

CD133 immunofluorescence staining was used to detected CD133⁺ GSCs in the xenograft model. Sections were incubated overnight with an anti-CD133 antibody (cat. no. 66666-1-IG; Proteintech Group, Inc.) diluted 1:100 at 4°C. The NC sections were incubated with PBS instead of the antibody. After washing, the sections were incubated with cyanine 3-conjugated goat anti-mouse IgG (cat. no. BA1031; Wuhan Boster Biological Technology, Ltd.) diluted 1:100 at 37°C for 1 h. The CD133⁺ GSCs were observed under a fluorescence microscope (BX53; Olympus Corporation).

CD34 staining was performed to detect MVD. All procedures were performed according to the manufacturer's protocols. The sections were incubated overnight at 4°C with a rabbit anti-CD34 antibody (1:200; cat. no. ab81289; Abcam). The NC sections were incubated with PBS instead of the antibody. Antibody localization was determined using a 3,3′-diaminobenzidine substrate kit (Wuhan Boster Biological Technology, Ltd.). MVD was evaluated by detecting the cluster of CD34⁺ cells. Briefly, the tumor sections were scanned at low magnifications (×40 or ×100) to determine the areas of most intense tumor angiogenesis, termed ‘hot spots’. Following ‘hot spot’ identification, the MVD was calculated by averaging the number of individual microvessels in five fields under an optical microscope (IX51; Olympus Corporation) at high magnification (×400).

Data are presented as the mean ± standard deviation of three replicates and analyzed using SPSS 17.0 software (SPSS, Inc.). One-way analysis of variance was used to compare the groups and the least significant difference post hoc test (≤3 groups) or Tukey's test (>3 groups) were performed to further determine inter-group comparisons. P<0.05 was considered to indicate a statistically significant difference.

As presented in Fig. 1, following transfection with alphastatin lentivirus (Fig. 1A), the FLAG tag protein could be detected in Al-ADSCs, which indicated that the Al-ADSCs expressed alphastatin. No alphastatin expression was detected in the control or NC groups (Fig. 1C). In addition, FLAG tag protein could be detected in the supernatant of Al-ADSCs using ELISA, which demonstrated that Al-ADSCs had the function of exocrine alphastatin secretion (Fig. 1D). Compared with the ADSCs, there was no significant difference in the expression of MSC surface markers, including CD13, CD44, CD90 and CD105, in Al-ADSCs, which indicated that alphastatin lentiviral transfection had no significant effect on the stem cell characteristics of ADSCs (Fig. 1B).

As presented in Fig. 2, GSCs were isolated from U87 and SHG44 glioma cell lines using flow cytometry (Fig. 2B). Cell migration assay revealed that ADSCs exhibited more obvious tropism to GSCs than to CD133⁻ glioma or SVG p12 cells (F=708.439; P=0.001; Fig. 2A and C). There was no significant difference between ADSCs, NC-ADSCs or Al-ADSCs regarding tropism to GSCs (F=0.484; P=0.639 for GSCs-SHG44; and F=1.122; P=0.386 for GSCs-U87; Fig. 2D), which indicated that alphastatin lentiviral transfection had no significant effect on the tropism of ADSCs to GSCs.

As presented in Fig. 3, tube formation assay revealed that Al-ADSCs significantly inhibited the angiogenesis of HUVECs in vitro (F=98.262; P=0.001; Fig. 3C and D). Scratch wound healing assay revealed that the wound healing ability of the HUVECs decreased following coculture with Al-ADSCs, suggesting that the Al-ADSCs inhibited HUVEC migration (F=554.852; P=0.001; Fig. 3A and B). In addition, cell proliferation assay revealed that the proliferation of HUVECs was inhibited by Al-ADSCs. At 48 h following co-culture with Al-ADSCs, the absorbance of HUVECs at 450 nm was lower than that of the control group, indicating that Al-ADSCs significantly inhibited the proliferation of HUVECs (F=310.614; P=0.001; Fig. 3E and F). Notably, the ADSCs and NC-ADSCs exerted promoting effects on the proliferation, migration and tube formation abilities of HUVECs in vitro.

As presented in Fig. 4, both NC-ADSCs and Al-ADSCs were detected in glioma xenograft tissue using a fluorescence microscope (Fig. 4B), which indicated that Al-ADSCs could migrate into tumor tissue. In vivo imaging revealed that the Al-ADSCs significantly inhibited tumor growth in the glioma xenograft model. Furthermore, the number of CD133⁺ cells in the Al-ADSC group was significantly reduced (Fig. 4A). Immunohistochemical assay revealed that Al-ADSCs significantly reduced the MVD in glioma xenograft tissue (F=16.236; P=0.001; Fig. 4C).