EGFR wild-type amplification and activation promote invasion and development of glioblastoma independent of angiogenesis
- First Online:
- Cite this article as:
- Talasila, K.M., Soentgerath, A., Euskirchen, P. et al. Acta Neuropathol (2013) 125: 683. doi:10.1007/s00401-013-1101-1
Angiogenesis is regarded as a hallmark of cancer progression and it has been postulated that solid tumor growth depends on angiogenesis. At present, however, it is clear that tumor cell invasion can occur without angiogenesis, a phenomenon that is particularly evident by the infiltrative growth of malignant brain tumors, such as glioblastomas (GBMs). In these tumors, amplification or overexpression of wild-type (wt) or truncated and constitutively activated epidermal growth factor receptor (EGFR) are regarded as important events in GBM development, where the complex downstream signaling events have been implicated in tumor cell invasion, angiogenesis and proliferation. Here, we show that amplification and in particular activation of wild-type EGFR represents an underlying mechanism for non-angiogenic, invasive tumor growth. Using a clinically relevant human GBM xenograft model, we show that tumor cells with EGFR gene amplification and activation diffusely infiltrate normal brain tissue independent of angiogenesis and that transient inhibition of EGFR activity by cetuximab inhibits the invasive tumor growth. Moreover, stable, long-term expression of a dominant-negative EGFR leads to a mesenchymal to epithelial-like transition and induction of angiogenic tumor growth. Analysis of human GBM biopsies confirmed that EGFR activation correlated with invasive/non-angiogenic tumor growth. In conclusion, our results indicate that activation of wild-type EGFR promotes invasion and glioblastoma development independent of angiogenesis, whereas loss of its activity results in angiogenic tumor growth.
Human glioblastoma (GBM) is the most frequent and most malignant primary brain tumor. The majority of GBMs arise de novo and are defined as primary GBMs, while the progression from lower grade astrocytomas results in secondary GBMs . Primary GBMs most frequently harbor the common mutations 9p and 10q loss as well as amplification of epidermal growth factor receptor (EGFR), a tyrosine kinase receptor . Wild-type (wt) EGFR is amplified in 40–50 % of primary GBMs and a fraction of EGFR-amplified tumors in addition express the mutant variant EGFRvIII, a constitutively active receptor . Signaling through the EGFR pathway is a complex process that involves tight regulation of several intracellular cell signaling networks . When these regulatory networks are altered, as in cancer, they have been shown to contribute to malignant transformation and tumor progression through increased cell proliferation, angiogenesis, invasion, and metastasis [23, 37, 39, 43].
The diffuse infiltrative growth of tumor cells within the central nervous system (CNS) still represents a major problem for effective therapeutic intervention as the delivery of active therapeutic agents to the invasive tumor cells is limited by the blood–brain barrier (BBB). While factors that mediate tumor angiogenesis have been well defined [7, 25, 51, 56], the major mechanisms causing non-angiogenic, invasive tumor growth in vivo still remain elusive. This can partly be explained by the lack of representative animal models that reflect the invasive tumor growth seen in patients. To this end, we and others have shown that human GBMs, short-term cultured as multicellular biopsy spheroids, maintain the same DNA copy number as the parental tumors  and can, when xenotransplanted into the CNS of immunodeficient rats, grow invasively for extensive periods without switching to angiogenic tumor growth [53, 59]. Thus, in our model system there appears to be a selection toward a subpopulation of glioma cells, which is capable of initiating and sustaining tumor growth independent of angiogenesis. In the present study, we show that wtEGFR activation is associated with non-angiogenic, infiltrative tumor development both in our animal model as well as human GBMs and that inactivation of the receptor can lead to angiogenic tumor growth.
Materials and methods
Biopsy spheroids were prepared as described previously . After 1–2 weeks in culture, spheroids with diameters between 200 and 300 μm were selected for intracerebral implantation. For functional experiments with cetuximab and EGFR-CD533, spheroids with a standardized cell number were generated as described under “Lentiviral EGFR-CD533 production and infection of glioblastoma cells”.
The human embryonic kidney cell line 293T (ATCC number CRL-11268) and the U87 cell line were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and maintained in Dulbecco’s modified eagle medium (DMEM) supplemented with 10 % fetal calf serum (FCS) and 1 % glutamine. All cell lines were grown at 37 °C in a humidified atmosphere of 5 % CO2.
In vivo experiments
Nude immunodeficient rats (rnu/rnu Rowett) were fed a standard pellet diet and were provided with water ad libitum. All procedures were approved by the Norwegian National Animal Research Authority. Biopsy spheroids were stereotactically implanted into the right brain hemisphere as described previously . Rats were euthanized with CO2, perfused intracardially with 0.9 % NaCl and killed when symptoms developed.
Intracranial convection-enhanced delivery (CED) of cetuximab was started 6 weeks after tumor implantation and was given for 4 weeks. CED was performed using osmotic minipumps (Alzet mini-osmotic pump, model 2ML4, Durect Corp., Cupertino, CA), which maintain a constant flow of 2.5 μl/h over 28 days. Pumps were filled with the antibody at a concentration of 5 mg/ml, consequently the rats received 300 μg of the antibody per day. The pumps were connected to an intracranial catheter (Alzet Brain Infusion Kit 2). Pumps were placed subcutaneously at the back of the rats. The catheter tip was inserted through the same burr hole that had been created to inject the tumor cells and was placed approximately at the injection site of tumor cells. Control groups for cetuximab received pumps loaded with PBS.
For pimonidazole analysis, animals were injected with hypoxyprobe-1 (HPI, Burlington, MA) 30 min prior to euthanasia. Brains were removed and fixed in 4 % formalin for 1–7 days, or tumors were excised and snap frozen in liquid nitrogen for protein isolation.
Immunohistochemistry of paraffin sections was performed as described previously . The following primary antibodies were used: anti-human nestin diluted 1:200 (Millipore, Billerica, MA), anti-human sox2 diluted 1:200 (R&D Systems, Minneapolis, MN), anti-CD31 diluted (Santa Cruz, Santa Cruz, CA), anti-wtEGFR diluted 1:500 (Santa Cruz), anti-pEGFR (Tyr1173) diluted 1:250 (Cell Signaling, Danvers, MA), anti-pimonidazole diluted 1:200 (HPI), anti-vWF, diluted 1:500 (DAKO), anti-angiopoietin2 diluted 1:200 (Santa Cruz), anti-EGFRvIII diluted 1:200 (clone L8A, a gift kindly provided by S. Clayton, Duke University, Durham, NC) and anti-GFP diluted 1:200 (Millipore). The H&E and immunohistochemical stainings were analyzed on a Nikon light microscope (Nikon, Tokyo, Japan) using Nikon imaging software. The quantification of vessel area fractions was performed using the Nikon imaging software. Overview pictures of histological slides were taken using a digital slide scanner and Imagescope software (Aperio, Vista, CA).
Protein extraction and western blotting were performed as described previously . Primary antibodies used were anti-pAkt (Ser-473) diluted 1:500 (Cell Signaling), anti-pStat3 (Tyr-705) diluted 1:2,000 (Cell Signaling), anti-pMAPK (Thr-202/Tyr-204) diluted 1:2,000 (Cell Signaling), anti-EGFR diluted 1:500 [Life Technologies (Biosource)], anti-EGFRvIII diluted 1:1,000 (clone L8A, a gift kindly provided by S. Clayton, Duke University, Durham, NC), anti-VEGF diluted 1:200 (Santa Cruz), anti-HIF-1α diluted 1:500 (Becton–Dickinson, San Jose, CA), anti-angiopoietin1 diluted 1:300 (Santa Cruz), anti-angiopoietin2 diluted 1:500 (Santa Cruz), anti-FGF2 diluted 1:500 (Santa Cruz), anti-CD133/1 clone AC133 diluted 1:100 (Miltenyi, Bergisch-Gladbach, Germany), anti-vimentin diluted 1:500 (DAKO), anti-snail diluted 1:100 (Abgent, San Diego, CA), anti-Twist diluted 1:100 (Santa Cruz), anti-beta-Actin diluted 1:1,000 (Abcam, Cambridge, UK) and anti-GAPDH diluted 1:2,500 (Abcam).
The primary antibody was detected using a goat F(ab′)2 fragment anti-rabbit IgG (H + L)-peroxidase diluted 1:100,000 (Beckman Coulter, Brea, CA), or goat anti mouse IgG-HRP diluted 1:25,000 (Santa Cruz) or HRP-conjugated goat anti-rabbit/mouse secondary antibody (Immunotech, Fullerton, CA) diluted 1:2,500.
Cloning of EGFR-CD533
The EGFR-CD533 construct was a gift from Joseph Contessa, Yale University School of Medicine, New Haven, CT. From this plasmid, EGFR-CD533 was amplified by PCR using 5′-GCATCATCTAGAGCCACCATGCGACCCTCCGGG-3′ as forward and 5′-GCATCACTCGAGTCAGCGCTTCCGAACGATG-3 as reverse primer. The primers were designed to insert XbaI and XhoI restriction sites flanking the EGFR-CD533 gene. The lentiviral vector pRRL.sinCMVeGFPpre  was cut with XbaI and SalI to remove the eGFP gene. The PCR product was cut with XbaI and XhoI and ligated into the lentiviral vector.
Lentiviral EGFR-CD533 production and infection of glioblastoma cells
Lentiviral vectors carrying EGFR-CD533 or GFP were produced in 293T cells using FuGene HD transfection reagent (Life technologies, Paisley, UK) according to the manufacturer’s instructions. The production and titration of lentiviral vectors were performed as described previously . For infection, spheroids were dissociated using the Neuronal dissociation kit (Miltenyi, Bergisch-Gladbach, Germany), plated in round-bottomed 96 wells with 3,000 cells/well in culture medium with 4 % methylcellulose, and infected with viral supernatants at an MOI of 5–30. 96-well plates were centrifuged for 1.5 h at 31 °C. Medium was changed 2 days after infection. 6 days after infection, reformed spheroids were stereotactically implanted using 10 spheroids/rat.
Array CGH was performed as previously described .
Gene expression analysis
RNA was purified from tissue samples using Ambion Tri-reagent (life technologies) following the manufacturer’s instructions. RNA samples were then DNAse treated using Ambions turbo DNA Free kit to remove any contaminating genomic DNA. Microarray analysis of EGFR-CD533 and control animals were carried out as specified in .
Functional analysis of gene expression data
Data were analyzed using IPA (Ingenuity Systems, http://www.ingenuity.com). Right-tailed Fisher’s exact test was used to calculate a p value determining the probability that each biological function assigned to that data set is due to chance alone. Upstream regulator analysis was based on prior knowledge of expected effects between transcriptional regulators and their target genes stored in the Ingenuity® Knowledge Base. Two statistical measures, standard in IPA, were used to detect potential transcriptional regulators: an overlap P value and an activation z score. First, the analysis examined how many known targets of each transcriptional regulator were present in our data set, resulting in an estimation of an overlap P value. We set a threshold of an overlap P value <0.05 to identify significant upstream regulators. Second, the known effect (activation or suppression) of a transcriptional regulator on each target gene was compared with observed changes in gene expression. Based on concordance between them, an activation z score was assigned, showing whether a potential transcriptional regulator was in “activated” (z score > 2), “inhibited” (z score < −2) or uncertain state.
Fluorescence in situ hybridization (FISH)
FISH analyses of paraffin sections were performed with the Vysis LSI EGFR SpectrumOrange/CEP 7 SpectrumGreen probe (Abbott Molecular, Des Plaines, IL) using the DAKO Histology FISH Accessory Kit (DAKO, Glostrup, Denmark).
Magnetic resonance imaging (MRI)
Axial T1-weighted (T1w) RARE sequences and (T2w) RARE sequences were acquired as described previously . Tumor volumes were calculated using a volumetric approach, where masks were created in Bruker’s Paravision 5.0 software, by delineating tumor in consecutive sections of the T2-weighted images. A region growing algorithm was used to assist in finding the contours of the tumor, where the seed point was placed centrally in the tumor, and the parameters of the algorithm were optimized to include all hyperintense pixels from the tumor area.
Proportion of positive tumor cells (P): 0 % (0); 1–10 % (1); 11–50 % (2); >50 % (3)
Intensity of staining (I): negative (0); weak (1); moderate (2); strong (3)
Staining index (SI): Proportion (P) × intensity (I)
The mean SI was assessed from both scorings and served as the final score displayed in the results section.
The Norwegian Data Inspectorate and the Regional Committee for Ethics in Research have approved this project. The study was performed in accordance with the Helsinki Declaration.
Survival was analyzed by a log-rank test based on the Kaplan–Meier test using SPSS software. Differences between pairs of groups were determined by the Student’s t test. P values <0.05 were considered significant.
Two human GBM xenograft phenotypes can be defined by different angiogenic and invasive growth properties
When comparing the invasive capabilities of both tumor subtypes, we found that the non-angiogenic phenotype showed extensive single cell infiltration into cortical brain areas in addition to migration into the contralateral hemisphere while tumor growth in the angiogenic phenotype was largely confined to the white matter with a more defined border and significantly fewer cells invading into the cortex (Fig. 1e, f).
Since various stem cell markers have been associated with a tumorigenic phenotype in human gliomas , we analyzed invasive and angiogenic xenografts for the expression of the stem cell markers nestin, sox2 and CD133. All tumors of either phenotype showed expression of both nestin and sox2. However, CD133, a controversial marker for stem-like cancer cells, was expressed at low to intermediate levels in two out of three xenografts of either phenotype (Fig. S2). Thus, the expression of stem cell markers was not linked to a specific phenotype.
Amplification and activation of wild-type EGFR is associated with non-angiogenic, invasive tumor growth
Genomic and histological profile of xenografts derived from patient biopsies
Highly invasive, non-angiogenic
6q, 9p, 10, 13q, 14p, 18q
9p, 10, 13
Angiogenic, less invasive
9p, 10, 14, 6q
9p, 10, 11q
We have previously shown that non-angiogenic tumors can spontaneously switch to an angiogenic phenotype upon in vivo passaging . To determine whether wtEGFR amplification, expression, and activation were changed during the angiogenic switch, we analyzed four EGFR-amplified, non-angiogenic tumors by serial in vivo passaging. We observed that two tumors switched to an angiogenic phenotype, while the other two maintained a stable invasive phenotype (Fig. 2c). Both phenotypes harbored EGFR amplification; however, while the stably invasive tumors showed high levels of wtEGFR expression and phosphorylation (pEGFR), both proteins were downregulated in the core of tumors that switched to an angiogenic phenotype (Fig. 2c). In addition to the wtEGFR protein, various mutants exist, such as EGFRvIII . However, EGFRvIII is expressed in fewer tumor cells than wtEGFR in most patient samples and amplification of EGFRvIII occurs to a much lesser extent compared to wtEGFR . By evaluating EGFRvIII in our xenografts, we found its expression only in tumors with wtEGFR amplification (Fig. S4; Table 1), suggesting a strong association between wtEGFR amplification and EGFRvIII expression which has also been described in patient biopsies . Interestingly, EGFRvIII was lost in stably invasive tumors after serial in vivo passages while it was expressed in xenografts that switched to an angiogenic phenotype (Fig. 2d, e). This shows opposite regulations of wtEGFR and EGFRvIII in our xenograft system and indicates that wtEGFR expression is necessary to drive the invasive phenotype, while EGFRvIII is dispensable and might be involved in stimulating angiogenic tumor growth when wtEGFR expression is lost. The relevance of EGFRvIII for tumor angiogenesis has been previously described [6, 11, 36].
Activation of EGFR in patient samples correlates with invasive/non-angiogenic tumor growth
Cetuximab inhibits tumor growth and invasion of EGFR-amplified xenografts
Stable inactivation of EGFR by a dominant-negative receptor induces an angiogenic switch in EGFR-amplified xenografts
Kaplan–Meier analysis showed a significantly prolonged survival of animals in the EGFR-CD533 group compared to the control group (Fig. 5b; log-rank: P < 0.05), indicating a slower development of the modified tumors. To assess the tumor growth characteristics, we performed MRI. The images revealed circumscribed, contrast-enhancing tumors in the EGFR-CD533 group, while controls showed highly invasive, non-enhancing lesions (Fig. 5c, Fig. S6). The MRI findings were confirmed both at the macroscopic and microscopic level. Angiogenic growth was clearly evident in the EGFR-CD533 tumors, which were circumscribed and harbored microvascular proliferations and necroses. Angiogenesis was absent in the diffusely invasive growing tumors of the control group (Fig. 5d, Fig. S6). Importantly, EGFR activation (pEGFR) was significantly inhibited in the EGFR-CD533 tumors compared to the control tumors (Fig. 5d, e, Fig. S6).
To determine whether the tumors were hypoxic, pimonidazole staining was performed. While hypoxic areas were identified around necroses in the EGFR-CD533 group, no hypoxic regions were observed in control tumors (Fig. 5d). We further determined the expression of HIF1A and VEGFA to verify induction of an angiogenic switch at the molecular level. Upregulation of HIF1A and VEGFA occurred in the EGFR-CD533 transduced, angiogenic tumors whereas both proteins were absent/less expressed in the control tumors (Fig. 5f, Fig. S6). Functional analysis of gene expression data using Ingenuity Pathway Analysis (IPA) showed that the biological function “angiogenesis” was highly enriched (p 2.22 × 108) in the dataset. Z score analysis revealed that angiogenesis was significantly increased (z score 3,305) in EGFR-CD533 tumors (Table S3). Subsequently, we used IPA to query for significantly altered upstream regulators in the dataset. HIF1A was the top enriched upstream regulator (P 3.45 × 1011; Fig. S7).
To analyze whether inhibition of EGFR activation affected major downstream signaling pathways, we performed western blots for phosphorylated Akt, MAPK and Stat3. While Akt and MAPK activation were not changed, phosphorylation of Stat3 was enhanced in the EGFR-CD533 group (Fig. 5g). Thus, major downstream signaling was not blocked by inhibiting EGFR activation which is in line with previous clinical studies using EGFR inhibitors [24, 33, 41] and might in part explain the resistance to anti-EGFR therapy.
Stable inactivation of EGFR induces a mesenchymal to epithelial-like transition in EGFR-amplified xenografts
In summary, our data show that tumor cell invasion is strongly associated with wtEGFR amplification and activation and that this process is independent of angiogenesis in human GBMs. The selection for EGFR amplification in our animal model, the coincidence of wtEGFR expression and invasion as an early process in tumorigenesis, and the stem-like properties of these cells described previously  suggest a functional role for wtEGFR in cancer development. This is supported by a recent study, demonstrating the importance of wtEGFR for tumor development and invasive growth within a stem-like human GBM xenograft model . Taken together, this study and our results show that the non-angiogenic wtEGFR-amplified population clearly represents a different subset of cancer cells as compared to the highly angiogenic cancer stem-like cells, which have been proposed to be the only tumorigenic cells within GBMs [2, 20, 32, 55].
Although the inhibition of EGFR activation significantly reduced tumor growth and invasion in our model, tumor cells had the capacity to induce an angiogenic program and thereby escape the invasion block as shown by stable expression of dominant-negative EGFR. In this experimental set-up, we observed a mesenchymal to epithelial-like transition, which might explain the inability of EGFR-CD533 expressing tumor cells to escape from hypoxic areas through invasion and instead induce an angiogenic program by upregulating HIF1A. This was verified by a microarray analysis showing that genes which are transcriptionally activated by HIF1A were upregulated in the EGFR-CD533 expressing tumors. Recently, Lu et al.  observed that c-Met induced an epithelial to mesenchymal transition and invasive phenotype after VEGF inhibition in glioblastoma. This indicates that in addition to EGFR other tyrosine kinase receptors might be important for invasion and a mesenchymal phenotype in high-grade glioma. In contrast, low grade gliomas often do not show amplification of tyrosine kinase receptors , but are also invasive. Thus, in these tumor types other mechanisms driving tumor invasion might be responsible.
Our results highlight the dynamic nature of highly malignant tumor cells that have a number of genetic changes in common such as 9p and 10q deletions that disrupt PTEN, p53 and RB tumor suppressor pathways [9, 49]. The inactivation of these pathways is probably sufficient to drive tumor growth as also verified in genetic mouse models of GBMs [1, 62]. Inactivation of these tumor suppressor genes may also activate major downstream signaling events such as AKT, MAPK and Stat3 which, as shown in the present study, is not dependent on wtEGFR activation; however, the EGFR status as demonstrated here has an important impact on the balance between invasive and angiogenic tumor growth. In particular, patient tumors are highly heterogenous and contain both, cells with and without EGFR amplification. Cells with high EGFR amplification are more frequent in invasive areas as compared to the main angiogenic tumor mass [48, 54]. Additional support has been provided in a recent study showing that the invasive areas of GBMs with co-amplification of EGFR and PDGFR exclusively contain EGFR-amplified cells, while PDGFR amplified cells are only found in the main tumor mass . Accordingly, the angiogenic switch in human tumors might be induced by less migratory cells in which EGFR signaling is absent/low. In this context, we showed that only a subset of EGFR-amplified tumor cells within GBM biopsies had a highly activated EGFR and importantly, these cells were found in non-angiogenic/invasive areas.
In our xenograft model system, wtEGFR was the main driver of invasion, whereas mutated EGFRvIII was lost after serial passaging. In contrast, wtEGFR was downregulated in tumors that switched to an angiogenic phenotype, while EGFRvIII was stably expressed. This shows opposite regulations of wtEGFR versus EGFRvIII in our xenograft system and strong associations to either invasion or angiogenesis, respectively. Several studies have shown that EGFRvIII is responsible for angiogenic growth within GBM animal models and cell lines using overexpression approaches [6, 11, 28, 36] and that EGFRvIII differs from wtEGFR signaling [29, 44–46]. Although it has been demonstrated that wtEGFR also can induce upregulation of angiogenic factors in glioma cell lines in vitro [21, 28, 37, 52], there is lack of evidence that this can be a mechanism in vivo. In contrast, by preserving naturally occuring EGFR-amplified cells in vivo, we have clearly shown that wtEGFR is a driver of invasion in human GBM in vivo. The mechanism of how the gene expression of EGFRvIII and wtEGFR are regulated in our animal model and also in human GBM still needs to be identified. However, a recent study suggests that at least EGFRvIII is epigenetically regulated .
The EGFR-activated tumor subpopulation is an important target for therapy as these cells are highly invasive and, accordingly, have the capacity to escape current therapies. In addition, an intact BBB inherently impedes drug delivery to invasive and non-angiogenic tumor regions. In our xenograft model, we clearly demonstrate that local delivery of an anti-EGFR antibody significantly inhibits tumor growth and invasion. These results highlight the importance of anti-EGFR therapy as an anti-invasive treatment strategy for GBM and most likely also explain why systemic administration of otherwise effective anti-EGFR drugs fails to show substantial effects in the clinic . Although it has been postulated that small molecule inhibitors may successfully circumvent the drug penetration problem often associated with antibody therapies involving the CNS, recent observations show that drug transporters in endothelial cells of intact vessels prevent effective penetration of these molecules [17, 31, 38]. Thus, a major challenge in targeting these invasive, EGFR-activated tumor subpopulations will be to effectively deliver bioactive molecules across the BBB and at the same time inhibit potential angiogenic escape mechanisms.
We thank I. Gavlen, B. Hansen, E. Fick, L. Vårdal, E. Jensen, A. S. Herdlevær, R. Owen, S. Leh, K. Mangseth and E. Mutlu for expert technical assistance and the Molecular imaging center (MIC) in Bergen, Norway for technical support. We thank A. Muller from the Genomics Research Unit of CRP-Santé in Luxembourg for expert assistance with the use of Ingenuity Pathway Analysis. K. Talasila was supported by a PhD fellowship from the University of Bergen. This work was supported by the Research Council of Norway (grant 213630 to HM), The Norwegian Cancer Society, Helse Vest, Haukeland University Hospital, the Bergen Medical Research Foundation, the European Commission 6th Framework Programme (Contract 504743) and by the CORE program of the Fonds National de la Recherche (FNR) in Luxembourg (ESCAPE C10/BM/784322).
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.