Combined treatment of Nimotuzumab and rapamycin is effective against temozolomide-resistant human gliomas regardless of the EGFR mutation status

The use of human glioma cell lines and patient-derived GBMs were approved by the Singhealth Centralized Institutional Review Board, Singapore. GBM tumor specimens from patients, who had undergone surgery, were obtained after written informed consent.

Temozolomide (TMZ; Temodal) was obtained from Schering Plough and dissolved in DMSO to a final concentration of 100 mM. Rapamycin was obtained from Selleckchem.com (Houston, TX) and dissolved in DMSO to a final concentration of 1 M. Nimotuzumab was provided by Innogene Kalbiotech Pte Ltd (Singapore) at a concentration of 5 mg/ml (equivalent to 0.03 mM). For all treatments, except Nimotuzumab, DMSO was added to a final concentration of 0.1% and used as vehicle control.

Human glioma Gli36 cells was kindly provided by A.T. Campagnoni (UCLA School of Medicine, Los Angeles, CA) and ΔGli36 cells was provided by M. Sena-Esteves (University of Massachusetts, Boston, MA). Human glioma U87MG was purchased from American Type Culture Collection (Rockville, MD, USA). Human glioma U87MG.EGFRvIII and U87MG.wtEGFR were engineered to express EGFRvIII and wild-type EGFR proteins respectively and were kindly provided by W. Cavenee, Ludwig Institute of Cancer Research, UCSD, CA). Immortalized normal human astrocytes (iNHA) that overexpress E6, E7, and human telomerase reverse transcriptase (hTERT) were kindly provided by R.O. Pieper (University of California, San Francisco, CA) and was cultured in DMEM supplemented with 10% FBS, 0.5 μg/ml puromycin (Invivogen, San Diego, CA), 25 μg/ml blasticidin and 1.25 μg/ml fungizone (Life Technologies, Grand Island, NY). Primary GBM xenograft cell lines, GBM6 and GBM10, were purchased from Mayo Clinic (Rochester, MN) and maintained as subcutaneous xenografts as previously described [23]. Primary Asian glioma cell lines cultivated from Chinese glioma patients, G5T/VGH, GBM8401 and GBM8901 were purchased from Food Industry Research and Development Institute, Bioresource Collection and Research Center (Hsinchu, Taiwan). GBM8401/TSGH, NDMC (abbreviated as GBM8401) was derived from a patient with a right parietal GBM brain tumor [24]. These cells expressed astrocyte-specific intracytoplasmic marker, glial-acidic fibrillary proteins (GFAP), with a doubling time of 38 h and were capable of forming tumors subcutaneously in athymic nude mice. Astrocytic differentiation and growth inhibition could be induced after dibutyryl (db)-cAMP treatment. GBM8901 was also cultivated from a GBM patient. Similar to GBM8401, these cells were GFAP-positive, primarily bipolar and tripolar cells at initial culture but adopted epitheloid-like cell morphology at confluency. Notably, following subcutaneous transplantation, these cells could metastasize to the lung [25]. Lastly, G5T/VGH cells lacked GFAP and only formed tumors initially which regressed gradually.

All cells were maintained at 37°C in a 5% CO2 - 95% air atmosphere and cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% Fetal Bovine Serum (FBS; Hyclone Laboratories, Logan, UT), penicillin (100 U/ml; Life Technologies, Grand Island, NY), streptomycin (100 μg/ml; Life Technologies) and 2 mM L-glutamine (Life Technologies). ΔGli36 and the U87MG.EGFRvIII cells were further supplemented with 1 μg/ml puromycin (Sigma-Aldrich Corp., St. Louis, MO ) and 500 μg/ml G418 (Life Technologies, Grand Island, NY ), respectively. Primary Asian glioma cell lines GBM8401 and GBM8901 were cultured in RPMI 1640 supplemented with 10% FBS, penicillin/streptomycin and L-glutamine.

High-grade anaplastic astrocytomas, NNI37 and NNI41, were obtained from local patients undergoing tumor resection surgery, following approval of patient informed consent by SingHealth Centralized Institutional Review Board, Singapore. Isolation of cells from patient-derived tumor tissue was performed as follows. In brief, tumor specimens were cut into smaller pieces and washed thoroughly with phosphate-buffered saline (PBS) prior to digestion with 0.25% Trypsin at 37°C for 30 min with constant stirring. Equal volumes of Astrocyte Growth Medium (AGM; Lonza, Basel, Switzerland) were then added to the suspension. Tumor pieces were allowed to settle prior to collecting the supernatant and filtering through a 70-μm membrane filter (BD Biosciences, Franklin Lakes, NJ). Filtered supernatant was centrifuged at 1000 rpm for 5 min at room temperature (r. t). The cell pellet was then resuspended in fresh AGM media and cultured on short-term basis.

Cell viability was determined by cell counting kit-8 (CCK-8) assay (Dojindo, Japan), which measures the number of viable cells based on bioreduction of a water soluble formazan. Human glioma cells (5000 cells/well) were seeded in 96-well dish and 24 h later, Nimotuzumab, rapamycin and respective controls were added to the cells. After an additional 24 h of respective treatments, 10% (v/v) of CCK-8 dye was added into the wells and cells were incubated for 1-2 h. The percentage of viable cells was then determined by measuring absorbance at an optical density (OD) 450 nm with a reference at 650 nm using Victor spectrophotometer (PerkinElmer Life Sciences, Waltham, MA). The percentage cell viability for each cell line and treatment group was normalized to respective vehicle controls.

Equal amounts of proteins were resolved in either 8 or 10% SDS-PAGE and electroblotted onto polyvinylidine difluoride membrane (PVDF) using semi-dry blotting system (Trans-Blot Transfer medium; Bio-Rad Laboratories). Membranes were blocked in 5% BSA in PBS containing 0.1% Tween-20 and incubated overnight at 4°C with antibodies against rabbit anti-phospho-ERK1/2 (Thr202/Tyr204) (1:1000), rabbit anti-ERK1/2 (1:1000), rabbit anti-phospho-AKT (Ser473) (1:500), rabbit anti-AKT (1:1000) from Cell Signaling Technology (Danvers, MA); mouse anti-EGFR (1:200) and anti-pan actin (1:50 000) antibodies from Neomarker (Fremont, CA). After several washes, membrane was incubated with either goat anti-rabbit or goat anti-mouse horseradish peroxidase conjugated secondary antibodies (DakoCytomation, Denmark) (1:10 000). Specific protein bands were visualized with an enhanced chemiluminescence using Western Lightning chemiluminescent kit (Perkin-Elmer, MA).

The band density of specific proteins from each western blot was quantified with MetaVue software (Ver. 6.1) (Molecular Devices Corp.) Briefly, specific protein bands from scanned western films were boxed and integrated intensity values were derived from region measurements. Background was corrected and the intensity of each protein was normalized to the respective loading controls. The activated and total levels of the proteins are expressed as the ratio of intensity of each protein to the respective loading controls.

Statistical differences between values were determined by either one way ANOVA or Student’s t-test. A value of p < 0.05 was considered as statistically significant. To determine the differences between single and combined treatment’s killing efficacy, one-way ANOVA followed by Tukey’s multiple comparison test (p < 0.05) was done using software package Prism 3.0 (Graphpad Software Inc., San Diego, CA).

As TMZ is routinely used as a first-line therapy for GBMs, we sought to first determine the concentration required to achieve 50% inhibition of cell proliferation, i.e., IC50 of TMZ, Nimotuzumab and rapamycin in immortalized human astrocytes (iNHA) which are the most common cell types that give rise to glioma. The results showed that IC50 was achieved at 500 μM of TMZ concentration (Figure 1A). The IC50 for Nimotuzumab was determined to be 2 μg/μl (0.013 mM) (Figure 1B) and these cells exhibited resistance to rapamycin even up to a concentration of 0.5 mM (Figure 1C). However, at this concentration, rapamycin was far too toxic for most of the glioma cells we have tested (data not shown). At a concentration of 0.1 mM, rapamycin exerted differential cytotoxicity levels in iNHA versus human glioma, and thus was chosen for subsequent experiments. Next, we determined the cell kill efficiency of rapamycin and Nimotuzumab in comparison to TMZ in the EGFR-null Gli36 cells, as confirmed by western blot analysis (Figure 1D). TMZ dose response curve indicated that Gli36 cells were sensitive to TMZ treatment with an IC50 of 250 μM (Figure 1E). As shown in Figure 1F, the cell viability in rapamycin and Nimotuzumab treatment groups was approximately 12% and 42% respectively. In combination treatment group, only 7% of the cells were viable.

To gain insight on possible molecular mechanisms involved after respective treatments, western blot analysis was performed to examine the expression levels of activated AKT and ERK1/2, which are downstream targets of EGFR. In these non-EGFR expressing glioma cells, rapamycin treatment reduces activation of AKT at Serine 473 residue when compared to the untreated or DMSO control group (Figure 1G). In contrast, Nimotuzumab treatment did not inhibit pAKT when compared with the control group. Instead, the results showed an increase in activated AKT and ERK1/2 pathways. Despite the detection of similar amount of actin proteins, we are unable to detect the presence of AKT/ERK in combination treatment. In EGF-treated A-431 cells, an epidermoid carcinoma cell line overexpressing wtEGFR, Nimotuzumab did not induce activation of AKT and rapamycin treatment marginally reduces pAKT levels [21]. These data suggested human GBM lacking EGF and its corresponding receptors trigger a non-AKT-dependent pathway in respond to the mono-and combination treatment.

In Singapore and Asian countries, the molecular signature of gliomas is not well- documented. The latest information available in PubMed is an article dated more than a decade ago by a local neurosurgeon in the “Journal of Neurooncology” where he reported that the genetic profiles of Asian glioma patients do not appear to follow the conventional molecular pathways [26]. Herein, we first determined whether the combination treatment is equally effective in inhibiting tumor cell proliferation as compared to single drug treatment alone in Asian gliomas. All of the Asian patient-derived glioma cells, i.e. GBM8401, G5T/VGH and GBM8901 expressed similar levels of wtEGFR (Figure 2A), and were TMZ-resistant. These cells were only marginally sensitive to Nimotuzumab and rapamycin as a single agent (Figure 2B-D). However, cell viability was reduced 3-fold in GBM8401 and 1.5-fold in GBM8901 with combination treatment in comparison to TMZ (Figure 2C and D). In fact, rapamycin and Nimotuzumab combination treatment significantly reduced the percentage of viable cells by at least 12-50% with respect to the monotherapies (Figure 2; p < 0.01, one-way ANOVA). Taken together, the results showed that combination treatment of Nimotuzumab and rapamycin was more efficacious than TMZ treatment as a single agent.

EGFRvIII is the most common deletion mutant found in human gliomas [27,28]. However, overexpression of EGFRvIII is usually not observed in isolation, but in combination with amplification of the wild-type receptor, suggesting selective pressure for both species in gliomagenesis [29]. Most recently, wild-type EGFR has been shown to phosphorylate EGFRvIII, leading to phosphorylation of STAT proteins and progression in tumorigenesis [22]. In the earlier part of the study, we showed that combined treatment is effective in Asian glioma cells expressing wild-type EGFR; herein, we sought to determine whether the combination of Nimotuzumab and rapamycin may also be effective in mutant EGFRvIII-expressing human glioma cells. For this, human U87MG glioma cells overexpressing wild-type EGFR (U87MG.wtEGFR; 170 kDa) and mutant EGFRvIII (U87MG.EGFRvIII, 145 kDa), as validated by western blot analysis (Figure 3A), were used. U87MG.wtEGFR and U87MG.EGFRvIII were treated with 500 μM of TMZ, rapamycin, Nimotuzumab and the combination of rapamycin and Nimotuzumab. The enhanced cytotoxic effect in combination treatment was independently confirmed in U87MG.wtEGFR (Figure 3B) and U87MG.EGFRvIII (Figure 3C), respectively. In these glioma cells expressing either wt or EGFRvIII receptor, elevated cell death observed in the Nimotuzumab group correlated with reduced levels of activated AKT in both cell types (Figure 3D), providing further support that Nimotuzumab acts regardless of EGFR status. U87MG.EGFRvIII has lower percentage of viable cells in comparison to U87MG.wtEGFR, suggesting that EGFRvIII-positive human glioma cells were less resistant to TMZ compared to their wtEGFR counterpart. Enhanced cell death in combination treatment versus mono-treatment was also observed in half the normal Nimotuzumab concentration was used in U87MG.wtEGFR (Figure 3E). Similar findings were observed in U87MG.EGFRvIII. Taken together, the data showed that combined treatment is effective regardless of the EGFR status in human glioma.

With the aim of translating the combination treatment regimen for clinical use, we studied the effect of combined treatment in patient-derived glioma cells. These tumors were histologically high grade glioma derived from Singaporean patients of Chinese descent and expressed only wtEGFR by western blot analysis (Figure 4A). GBM6 and 10 are primary human GBM xenografts that were originally derived from consented Mayo clinic patients’ tumor specimens. In line with previous studies, our western blot results showed that GBM6 and 10 were EGFRvIII and wild-type EGFR expressing cell lines respectively [30,31]. EGFRvIII-expressing GBM6 was more resistant to TMZ treatment (Figure 4B) when compared to the wtEGFR-expressing GBM10 (Figure 4C). This was not surprising because GBM6 contains unmethylated MGMT promoter [23]. The results showed that although both cell lines were resistant to TMZ treatment, co-treatment of Nimotuzumab and rapamycin was more effective in cell kill when compared to single treatment regardless of the EGFR status of the tumor (Figure 4B and C). These findings were reproducible in wtEGFR expressing high-grade glioma derived from Singaporean patients (Figure 4D and E).