GA was purchased from Sigma-Aldrich (St. Louis, MO, USA). The MET selective inhibitor, PHA665752, was purchased from Selleck Chemicals (Houston, TX, USA). All drugs used in the present study were dissolved in sterile dimethylsulfoxide (DMSO; Sigma-Aldrich); a 10 mM working solution was prepared and stored in aliquots at −22°C. Rabbit polyclonal IgG antibodies against human phosphorylated (p-) MET (#sc-101736), p-AKT (#sc-101629), p-extracellular-signal-regulated kinase (ERK; #sc-101760), MET (#sc-10), AKT (#sc-8312) and ERK (#sc-292838) were purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Rabbit anti-Ki-67 monoclonal IgG antibodies for immunohistochemistry (IHC) were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA; #9129). All the chemicals used in the present study were of analytical reagent grade.

The human NSCLC cell line, NCI-H1993, which harbors a MET gene amplification (12), was obtained from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in DMEM supplemented with 10% fetal bovine serum, 10⁵ U/l penicillin and 100 mg/l streptomycin (GE Healthcare Life Sciences, Logan, UT, USA) at 37°C in an atmosphere containing 5% CO₂.

BALB/c female nude mice, obtained from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China), were used when they were 7–9 weeks old. The health of all animals was monitored daily by gross observation, and the experimental animals were housed in the laminar airflow cabinet. All the animals were allowed to acclimatize and recover from any stress associated with shipping for at least three days prior to experimental manipulation. Autoclaved water and irradiated food (Vital River Laboratory Animal Technology Co., Ltd.) were provided ad libitum, and the animals were maintained in a 12 h light and dark cycle. Cages, bedding and water bottles were autoclaved prior to use and were changed twice weekly. All the animal experiments were performed in accordance with protocols approved by the Experimental Animal Center of the Second Military Medical University Animal Care and Use Committee (Zhuozhou, China).

NCI-H1993 cells were harvested, pelleted by centrifugation at 600 × g for 10 min, and resuspended in sterile serum-free medium supplemented with 50% Matrigel (BD Biosciences, Franklin Lakes, NJ, USA). The cells (5×10⁶ in 100 µl) were then subcutaneously implanted into the hind-flank region of each mouse and allowed to grow to a volume of 150–200 mm³ prior to the administration of GA.

Nude mice bearing NCI-H1993 tumors (150–200 mm³) received vehicle [10% DMSO, 15% ethanol and 75% phosphate-buffered saline (PBS) (Sigma-Aldrich)] or 10, 20 or 30 mg/kg GA (10% DMSO, 15% ethanol and 75% PBS via intraperitoneal (i.p.) injection; or 10 mg/kg PHA665752 (L-lactate and 10% polyethylene glycol; Selleck Chemicals, Houston, TX, USA) via tail intravenous (i.v.) injection for 21 consecutive days. On day 21 of the efficacy study, at 2 h following the final treatment with GA, the mice were humanely sacrificed by CO₂ overexposure, and the tumors were resected. The tumor volume (TV) was determined by measurement with electronic vernier calipers, and the TV was calculated using the formula: TV=length × width²/2. TV was expressed on the indicated days as the median TV ± standard deviation for the indicated groups of mice.

The protein expression levels of p-MET, p-AKT, p-ERK, MET, AKT and ERK were analyzed by western blot analysis. On day 21 of the efficacy study, at 2 h following the final injection of GA, the mice were humanely sacrificed. The tumors were harvested in lysis buffer (Cell Signaling Technology, Inc.) and homogenized using Misonix Sonicator 4000 (Misonix Inc., Farmingdale, NY, USA); protein lysates were generated and protein concentrations were determined using a bicinchoninic acid assay (Pierce Biotechnology, Inc., Rockford, IL, USA). Equal amounts of protein (50 µg) were then separated by SDS-PAGE on 10% gels, blotted on polyvinylidene difluoride membranes (Sigma-Aldrich) and probed with p-MET, p-AKT, p-ERK, MET, AKT and ERK rabbit polyclonal primary antibodies (dilution, 1:1,000; incubation, overnight at 4°C) and subsequently with goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (#sc-2040; dilution, 1:1,000; incubation, 1 h at room temperature), and detected with an enhanced chemiluminescence kit (Sigma-Aldrich).

The tumor specimens were fixed in 10% buffered formalin for 24 h prior to being transferred to 70% ethanol. The tumor samples were subsequently paraffin-embedded, and 4-mm sections were cut and baked onto microscope slides (formalin, paraffin and slides from Sigma-Aldrich). The slides were incubated with the primary Ki-67 antibody (dilution, 1:1,000; incubation, overnight at 4°C), then secondary antibodies (dilution, 1:1,000; incubation, 1 h at room temperature), and visualized using a colorimetric method (EnVision+ System-HRP DAB kit; Dako North America, Inc., Carpinteria, CA, USA). All of the immunostained sections were counterstained using hematoxylin. An automated Ventana Discovery XT Staining Module (Ventana Medical Systems, Inc., Tucson, AZ, USA) was used to conduct histological staining. The stained sections were analyzed using an Olympus BX46 microscope (Olympus Corporation, Tokyo, Japan), and quantitative analysis of section staining was performed using the Automated Cellular Imaging system (GE Healthcare Life Sciences, Chalfont, UK). The number of Ki-67-positive nuclei was counted regardless of the immunointensity in 4 random fields at ×100 magnification (60% center field).

As a measure of the level of apoptosis, caspase-3, −8 and −9 activity were measured using caspase colorimetric protease kits (Abnova Corporation, Walnut, CA, USA). Fresh tumors in each group were resected following the final treatment with 10, 20 or 30 mg/kg GA for 2 h on day 21 of the efficacy study, and the tumor lysis containing 200 µg protein was incubated with 5 µl 4 mM pNA-conjugated caspase substrates (4-amino acid sequences; DEVD-pNA, IETD-pNA and LEHD-pNA) at 37°C for 2 h. The quantity of pNA released was measured at 405 nm using a FLx800™ Multi-Detection microplate reader (BioTek Instruments, Inc., Winooski, VT, USA).

All the results and data were confirmed in at least three separate experiments. The data are expressed as the mean ± standard deviation, and were analyzed by Student's t-test using SPSS software, version 13.0 (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

In order to investigate the tumor growth inhibition effect of GA on NCI-H1993 xenografts in vivo, NCI-H1993 tumor-bearing mice received i.p. injection with 10, 20 or 30 mg/kg GA once a day for 21 days. As demonstrated in Fig. 1A and B, treatment with 10 mg/kg GA only slightly inhibited tumor growth, however, 20 mg/kg GA markedly inhibited tumor growth (P=0.021) and 30 mg/kg GA treatment almost completely inhibited tumor growth (P=0.008) compared with the vehicle control group. Throughout the duration of the efficacy study, no body weight loss was observed in any of the groups (Fig. 1C). The MET selective inhibitor PHA665752 was used as a positive control.

To ascertain whether the expression of p-MET in NCI-H1993 tumors was affected by GA treatment, the tumor tissues were analyzed by western blotting. The results demonstrated that the protein expression levels of p-MET in the NCI-H1993 tumor tissues were downregulated by GA in a concentration-dependent manner compared with those of the vehicle group. In addition, the protein expression levels of the downstream signaling molecules p-AKT and p-ERK were also markedly downregulated in a concentration-dependent manner, compared with those of the vehicle group (Fig. 2). The MET selective inhibitor PHA665752 was used as a positive control.

GA was also assessed for its effect on the tumor mitotic index (Ki-67) using IHC. A significant 2–3-fold reduction in Ki-67 levels was observed 2 h following the administration of 20 or 30 mg/kg GA in the NCI-H1993 tumor tissues (P=0.046 and 0.009, respectively; Fig. 3A and B), which was consistent with the results of the efficacy study. The MET selective inhibitor PHA665752 also significantly reduced the expression of Ki-67 in tumor tissues (P=0.007).

In order to investigate whether GA induces apoptosis in the NCI-H1993 xenograft model, the activities of caspase-3, −8 and −9 were measured using a colorimetric assay. The results demonstrated that GA had no effect on caspase-3, −8 and −9 activities in the NCI-H1993 xenograft model (Fig. 4).