Surgical specimens from 79 consecutive GMB patients who underwent total tumor removal surgery from 2007 to 2015 at the Department of Neurosurgery, Kagoshima University Hospital, were available for immunohistochemical study. All were treated with TMZ (75 mg/m²/day) concurrently with conventional radiation therapy conventional radiation 40 Gy plus CyberKnife 35 Gy/5 fractions. This was followed by maintenance TMZ treatment (150 mg/m² for 5 consecutive days of every 28 days). We retrospectively reviewed the clinical records of all patients. Of these, 12 underwent the second operation followed by same adjuvant therapy due to the recurrence; tissue samples from the 1st and 2nd surgical procedure were studied.

Our retrospective analysis was approved by the ethics committee of Kagoshima University Hospital (reference 27–05). The authors certify that the study involving human subjects was conducted in accordance with the 2013 Declaration of Helsinki. The data we collected are routinely obtained and are essential for the adequate management of glioblastoma. They were analyzed anonymously in a linkable fashion to protect patient privacy. Informed patient consent was waived due to the retrospective nature of the analysis; the information was obtained from anonymized medical charts and records.

All surgical specimens were fixed in 10% formaldehyde, embedded in paraffin, and cut into 3-µm slices. After microwave antigen retrieval in citrate buffer (pH 6.0) the samples were incubated with anti-ATP7B rabbit antibody (Novus Biologicals, Littleton, CO. USA) (1:100 dilution) and mouse anti-O6-methylguanine-DNA methyltransferase monoclonal antibody (MGMT, Chemicon international, USA) (1:200 dilution) as the primary antibody. The cells in three microscopic fields (magnification ×400) were counted independently by two researchers (M. F and S.Y). The criterion for a positive reaction was staining of a vesicle or vesicles in cytoplasm for ATP7B and clear nuclear staining for MGMT. The ratio of positive cells (%) was obtained by dividing the number of immunopositive cells by the total number of cells per field. The mean of six ratios of positive cells (2 readers, 3 fields each) was used for analysis. When more than 10% of the tumor cells were positive, the tumor was recorded as high-ATP7B GBM; tumors with less than 10% ATP7B-positive cells were categorized as low-ATP7B GBM. Anti-isocitrate dehydrogenase 1 (IDH1) R132H (Dianova GmbH, Hamburg, Germany) (1:50 dilution) was used as primary antibody following same immunohistochemical staining protocol mentioned above.

All cells were cultured in DMEM (Thermo Fisher Scientific, Grand Island, NE, USA) containing 10% fetal calf serum (Thermo Scientific, Logan, UT, USA) and 100 units/ml antibiotic-antimycotic solution (Thermo Fisher Scientific) in 5% CO₂ at 37°C. Wild type (WT) ATP7B-expressing cell lines (U251/WD-1 and U251/WD-2) were established by WT ATP7B transfection of the human glioblastoma cell line U251 using lipofectamin-2000 (ThermoFisher Scientific) as described previously (5). The control, U251/EV, was established from U251 cells transfected with vector (pRc/CMV) only.

Protein from cell lysates was prepared as described (9). The protein concentration was determined with the Bio-Rad protein assay kit according to the manufacturer's protocol (Bio-Rad Laboratories, Hercules, CA, USA). Total protein (100 µg) was subjected to 7.5% SDS-PAGE under reducing conditions. Immunoblotting was as described elsewhere (10). The polyclonal anti-ATP7B antibody (Novus Biologicals, Littleton, CO, USA) or anti-GAPDH monoclonal antibody (Cell Signaling Technology, Danvers, MA, USA) was used as the primary antibody. Horseradish peroxidase-conjugated anti-rabbit IgG and anti-mouse IgG (GE Healthcare, Buckinghamshire, UK) were the secondary antibody respectively. Immunoreactive bands were visualized with enhanced chemiluminescence using the ECL prime Western blotting system (GE Healthcare).

Equal numbers of cells (2×10³) were inoculated into each well of multi-well plates and treated for 7 days with TMZ (Sigma-Aldrich, St., Louis, MO, USA). Then their sensitivity to each of the administered TMZ dose was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) (Sigma-Aldrich) colorimetric assay (9). The drug concentration that reduced the number of cells to 50% of cells grown in control medium was recorded as the IC₅₀ value.

Data were analyzed using the Statistical Package for the Social Sciences (SPSS, v22; IBM, Armonk, NY, USA). For intergroup comparisons we used the Student t-test for continuous- and the Chi square test for categorical variables. Overall survival was plotted by applying Kaplan-Meier methods. Multivariable Cox proportional hazards regression analysis was used to obtain hazard ratios and 95% confidence intervals (CIs). Quantitative data were expressed as the mean ± SD. P<0.05 was considered to indicate a statistically significant difference.

In all 79 GBM tissue samples obtained at the initial surgery, ATP7B expression were immunohistochemically assessed. The degree of cytoplasmic staining in tumor cells varied (Fig. 1A); the average rate of ATP7B protein expression was 4.4±0.98%. Two representative cases of GBM tumors exhibiting high (>10%) and low ATP7B expression (<10%) are shown in Fig. 1B and C, respectively.

Of our 79 patients, 12 (15.2%) harbored high-ATP7B GBM; 14.4±3.0% of the tumor cells were ATP7B positive. The other 67 patients had low-ATP7B GBM; 2.6±0.64% of the cells were ATP7B positive (Fig. 2A). Investigation of the relationship between patient characteristics and ATP7B expression revealed no significant association with the patient gender and age, and the GBM tumor location. None of our patients younger than 50 years bore high-ATP7B GBM (Table I). The Kaplan-Meier estimate of overall survival in Fig. 2B was significantly shorter in patients with high-ATP7B GBM (hazard ratio 0.452, 95% CI 0.206–0.994, P=0.048). The median overall survival of patients with high- and low ATP7B GBM was 14.6 and 24.7 months, respectively. In addition, the MGMT expression in low-ATP7B GBM (5.7±1.2%) and high-ATP7B GBM (5.3±1.3%) has no significant difference. IDH1 mutant is negative in all 79 GBM cases.

Of our 79 patients, 12 underwent the second operation due to the recurrence. Fig. 3A and B show the immunohistochemical staining for ATP7B of tissues removed at the 1st and 2nd surgical procedure from the identical patient. In this representative case, the mean ATP7B positivity ratio was 0.93±0.15% and 20.93±7.79% in samples from the 1st and 2nd operation, respectively. In all 12 recurrent cases the expression of ATP7B in the 2nd surgical samples was significantly higher than the 1st ones (9.17±2.56% vs. 2.75±0.55%, P=0.008 by the paired t-test) (Fig. 3C).

We detected ATP7B 165 kDa bands in the both transfected cell lines but not in U251 and U251/EV cells by immunoblotting (Fig. 4). Lower bands (105 kDa) are likely to be specifically digested ATP7B protein.

The sensitivity of the transfected cells to TMZ was evaluated by MTT assay. Interestingly, U251/WD-1 cells and U251/WD-2 cells were 3.8 and 1.7 times more resistant to TMZ than U251/EV cells (Table II), indicating that the resistance to TMZ was correlated with ATP7B expression in these cells.