Research ArticleExperimental Studies

Endothelial Cells Derived from Non-malignant Tissues Are of Limited Value as Models for Brain Tumor Vasculature

JENNIFER LOHR, ANDREAS MOCK, PHILIPP BECKHOVE and CHRISTEL HEROLD-MENDE
Anticancer Research May 2015, 35 (5) 2681-2690;

Abstract

Background: Human umbilical cord vein endothelial cells (HUVECs) are commonly chosen over freshly isolated endothelial cells from glioblastomas (GECs) due to accessibility and costs. Materials and Methods: To test their suitability for in vitro studies, we comprehensively compared the transcriptomes and responses to major angiogenic cytokines of HUVECs (n=2) and GECs (n=5). Purity of GEC cultures was confirmed by uptake of acetylated low-density protein and immunostaining. Results: Unsupervised analysis revealed a distinct grouping. We identified 854 differentially expressed genes. Pathway and gene ontology enrichment analyses pointed to clear differences in angiogenesis and leukocyte transmigration. Comparing the expression of cell adhesion molecules in five major angiogenic cytokines revealed that HUVECs in contrast to GECs did not exhibit a previously described down-regulation of cell adhesion molecules upon incubation with transforming growth factor betas, but rather with basic fibroblast growth factor. Conclusion: Given our findings, we strongly recommend the use of GECs as model cells for brain tumor endothelium for experiments investigating angiogenesis and immunobiology.

Malignancies of the brain are fatal because of the limited space within the intracranial cavity, their infiltrative growth and therapy resistance (1, 2). The progression of these tumors largely depends on acquisition of new blood vessels (3). As a consequence, these constitute a promising target for cancer therapies. However, tumor vessels are not only essential for nourishing growing tumor masses but also for facilitating access to invading antitumor immune cells. Numerous studies have documented the positive association of immune infiltrates and clinical outcome in several human carcinoma types (4) and in brain tumors (5, 6).

Glioblastomas (GBMs) are the most common and most aggressive type of glial brain tumors, characterized by extensive angiogenesis, and despite multimodal treatment regimens, lead to a median survival of only 14 months (7, 8). This dismal prognosis has been fuelling the development of novel therapeutic approaches, including antiangiogenic and immunotherapeutic strategies. To properly study such treatments in a pre-clinical setting, in vitro experiments on endothelial cells are essential for drug development. However, the isolation and cultivation of tumor endothelial cells are limited by the availability of freshly-operated tumor material and laborious isolation procedures. Moreover, during culturing, tumor endothelial cells change their phenotype very fast and therefore their usage is recommended at low passages, further restricting the obtainable cell numbers (9). Thus model endothelial cells, e.g. human umbilical cord vein endothelial cells (HUVECs), are commonly used to imitate tumor vasculature in vitro (10). These surrogate cells might be problematic because it is well appreciated that endothelial cells are extensively shaped by their surrounding tissue (11). For instance, in the healthy human brain, cerebral endothelial cells comprising the blood–brain barrier (BBB) differ from extracranial endothelial cells in having lower pinocytotic activity, increased expression of tight junction proteins, and the presence of unique transport proteins (12). Tumor blood vessels are believed to differ from their normal counterparts even more substantially: they are more dilated and tortuous, have excessive branching, uneven diameters, chaotic flow patterns, and increased permeability to macromolecules (13).

View this table:
Table I.

Primer sequences used for quantitative PCR. HPRT1, ACTB and GAPDH were used as house-keeping genes.

View this table:
Table II.

Top 10 enrichments of terms by means of p-value depicted for gene ontology and KEGG enrichment analyses.

To determine the suitability of HUVECs as model cells for brain tumor endothelium, we comprehensively compared HUVECs to freshly-isolated endothelial cells from GBM-derived endothelial cells (GECs). We focused on the transcriptional differences between HUVECs and GECs and their response to external stimuli, with emphasis on anchorage molecules necessary for intra-tumoral immune cell infiltration.

Materials and Methods

Patient material. GBM tissues were gathered from newly-diagnosed patients at the Department of Neurosurgery at Heidelberg University, Germany. Informed consent was obtained from each patient according to the research proposals approved by the Institutional Review Board at the Heidelberg Medical Faculty (approval number 005/2003).

Cell isolation and cell culture. The method for isolation of endothelial cells from GBM tissues has been described in detail elsewhere (6). In brief, fresh GBM specimens were dissociated manually with scissors followed by a 1-h digestion with 0.45 mg/ml Liberase Blendzyme I (Roche Diagnostics, Mannheim, Germany) and 0.04% DNase I (Sigma-Aldrich, Saint Louis, Missouri, USA). Myelin was removed by centrifugation for 10 min at 4,600×g in 15% Dextran (Sigma) diluted in Phosphate-Buffered Saline (PBS). Endothelial cells were positively selected with Cluster of Differentiation (CD) 31 Dynabeads (Dynal, Invitrogen, Paisly, UK) and maintained in MV2 medium (Promocell, Heidelberg, Germany) on gelatin-coated flasks. HUVECs were purchased from ProVitro (Berlin, Germany) and maintained in Endothelial Growth Medium (ProVitro). All cells were cultured under standard culture conditions (at 37°C, with 95% humidified air and 5% CO2). Media changes were performed twice a week.

Characterization of cell cultures. Endothelial cell cultures were characterized by uptake of acetylated low-density protein (AcLDL). Sub-confluent cultures were incubated with fluorochrome-labeled AcLDL (Life Technologies, Carlsbad, CA, USA) at a ratio of 1:500 for 4 h, washed once with PBS and photographed. To further test the purity of freshly isolated GEC cultures, cells were immunostained with mouse antibodies to human CD31 (1:100; BD Pharmingen, San Jose, CA, USA), mouse antibodies to human glial fibrillary protein (GFAP,1:2, Progen, Heidelberg, Germany) and human mouse anti-CD68 antibodies (1:150; Dako, Hamburg, Germany). Secondary labeling was performed with an anti-mouse Alexa-Flour488 antibody (Molecular Probes, Life Technologies, Carlsbad, CA, USA; used at 1:400). Nuclei were counterstained with 4’,6-diamidino-2-phenylindol (DAPI) (1:1000; Life Technologies).

RNA extraction. HUVECs and GECs were harvested in passage 1 or 2 by trypsinization. Total RNA was isolated using the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to manufacturer's protocol. Total RNA concentration and nucleic acid purity was determined by Nanodrop (Thermo Scientific, Waltham, MA, USA).

Figure 1.
Figure 1.

Characterization of glioblastoma endothelial cells and human umbilical cord vein endothelial cells. Photographs of HUVEC (A, C) and GEC (B, D) cultures showing typical endothelial-like morphology (A, B;) in bright-field microscopy (BF) and uptake of acetylated low-density lipoprotein (AcLDL) as overlay with corresponding BF (C, D). Immunofluorescence stainings of a representative GEC culture showing negativity for glial marker glial fibrillary acidic protein (GFAP) (E-G) and leukocyte marker CD45 (H-J). Scale bars: 100 μm.

Figure 2.
Figure 2.

Unsupervised and supervised comparative transcriptomics. A: Consensus clustering. Eucledian distances between the microarrays were clustered hierarchically. B: 3-Dimensional principal component analysis. C: Volcano plot. Visualization of the result of the Student's t-test. Data points based on the actual grouping (glioblastoma endothelial cells vs. human umbilical cord vein endothelial cells; green) were plotted together with those basing on a random grouping (blue). Estimated false discovery rate (FDR)=2.7%. D: Hierarchical clustering of the top differentially expressed genes at a p-value below 0.01 (854 genes).

Figure 3.
Figure 3.

Hierarchical clustering of genes of candidate Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. The top 854 differentially expressed genes belonging to the GO terms cell migration (A) and cell adhesion (C), as well as the KEGG pathways leukocyte migration (B) and cell adhesion molecules (CAMs) (D) are depicted.

Microarray analysis and data normalization. Total RNA (1 μg) of HUVEC (n=2) and GEC (n=5) cultures were submitted to the Genomics Core Facilities of the German Cancer Research Center (DKFZ, Heidelberg, Germany) for microarray analyses. After purification and labeling according to the Illumina protocol, samples were hybridized to Human Sentrix-6 BeadChip arrays (Illumina, San Diego, CA, USA).

Raw intensity data were obtained after image analysis of the fluorescent spot intensity reads. Next, intra-array normalization was conducted using quantile normalization and intra-array normalization using median-centering. Lastly, data were log2-transformed.

cDNA synthesis and quantitative polymerase chain reaction (qPCR). One microgram of total RNA was reverse transcribed with a Transcriptor cDNA First Strand Synthesis Kit (Roche, Mannheim, Germany) and random hexamer primers. mRNA expression analysis was performed in triplicates on a LightCycler 480 (Roche) using the LightCycler 480 Probes Master and probes from the Universal Probe Library (Roche) as described [www.roche-applied-science.com]. Normalized expression ratios were determined for each sample using the housekeeping genes glycerinaldehyd-3-phosphat-dehydrogenase (GAPDH), beta-acin (ACTB) and hypoxanthine phosphoribosyltransferase 1 (HPRT1). For primer sequences, seeTable I.

In vitro stimulation of endothelial cells and flow cytometric analysis. HUVECs were cultivated in cell culture flasks until reaching a confluency of 70%. Cells were incubated with cytokines hepatocyte growth factor (HGF), vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), transforming growth factor beta 1 (TGFβ1) and TGFβ2 (all ReliaTech, Wolfenbüttel Germany) diluted in basal medium (Endothelial Growth Medium basal, 3% fetal calf serum) and basal medium as a control. After 72 h, cells were trypsinized and stained with monoclonal mouse antibodies to intercellular adhesion molecule 1 (ICAM1; 1:1000; Acris, Herford, Germany) and vascular cell adhesion molecule 1 (VCAM1; 1:100; Acris) in PBS containing 0.5% bovine serum albumin and 2 mM EDTA for 1 h at 4°C, then washed and stained with PE-conjugated anti-mouse IgG monoclonal antibody (1:100; Dianova, Hamburg, Germany). Flow cytometric analyses were performed using FACSCalibur (BD Biosciences, San Jose, CA, USA). All samples were evaluated with appropriate isotype controls and analyzed using the FlowJo software (TreeStar Inc. Ashland, OR, USA).

Statistical analysis. Unless otherwise stated, statistical analyses and data normalization were conducted in R (www.r-project.org). Differential gene expression in GECs and HUVECs were assessed using Student's t-test. To assess validity of the grouping (HUVEC vs. GEC), we estimated the false discovery rate (FDR) by randomly permuting the class labels (14).

Results

GEC cultures display typical characteristics and are of high purity. Five different endothelial cell cultures freshly-isolated from GBM tissues (GECs) were used for our studies in an anonymized manner (NCH700, NCH714, NCH716, NCH718 and NCH723). All tumor specimens were primary GBMs and patients received neither radiotherapy nor chemotherapy before tumor resection. Morphologically, GECs demonstrated a typical endothelial-like appearance similar to that of HUVECs (Figure 1A and B). Purity and functional properties were tested by uptake of AcLDL, a very reliable parameter for the identification of endothelial cells (Figure 1C and D) (15). In GECs, cell numbers with AcLDL uptake were comparable to those for HUVECs. Contamination with cells of glial or leukocytic origin were excluded by lack of GFAP and CD45 staining (Figure 1 E-J). Only GEC and HUVEC cultures in passage 1 or 2 were used for further experiments to avoid phenotypic and functional changes upon passaging.

Comparative transcriptome analysis reveals profound differences between GECs and HUVECs. Comparing the transcriptome of HUVECs (n=2; HUVEC1 and HUVEC2) and GECs (n=5; NCH700, NCH714, NCH716, NCH718 and NCH723) in an unsupervised fashion revealed a marked separation. Consensus clustering clearly segregated the samples in the corresponding group (Figure 2A). This separation could also be observed in a PCA plot (Figure 2B). By applying Student's t-test, we identified 854 differentially expressed genes (p<0.01) with an estimated FDR of only 2.7% (Figure 2C and D).

Gene ontology and pathway analysis. Next, we investigated the alteration of biological functions by means of functional genomics. Table II summarizes the enrichment of the top 854 differentially expressed genes for gene ontology (GO) terms and KEGG pathways. Both analyses indicate that GECs and HUVECs markedly differ in their expression of genes involved in cell migration, cell motility, cell adhesion, and vasculature and blood vessel development, as well as leukocyte transmigration (Table II, Figure 3).

To validate our microarray analysis, we determined the mRNA expression level of six genes in the same samples by qPCR. CD248 and TGFB2 were validated as being significantly up-regulated in GECs. Relative mRNA expression of CD248 and platelet-derived growth factor receptor beta (PDGFRB) was more than 1000-fold higher in GECs. In contrast, CDH5 and endothelial cell-selective adhesion molecule (ESAM) were found to be significantly down-regulated in GECs with fold-changes of more than 25 (Figure 4).

Cell adhesion molecules are differentially regulated by cytokines in GECs and HUVECs. Our gene ontology and pathway analyses showed that genes involved in neoangiogenesis and transendothelial migration of T-cells are differentially expressed in HUVECs and GECs. In a next step, we aimed to explore these findings on the post-transcriptional and functional level. We designed an in vitro experiment addressing both aspects of neoangiogenesis and T-cell transmigration at the same time.

This was based on the finding that the cell adhesion molecules (CAMs) ICAM1 and VCAM1, which are important for transmigration of T-cells into tumor, are regulated by angiogenic growth factors, such as bFGF, VEGF, HGF and TGFβs (16-20). Therefore, we compared the influence of these cytokines on protein expression of ICAM1 and VCAM1 on the surface of HUVECs and GECs.

As shown in our previous study on GECs, TGFβs reduced ICAM1 and VCAM1 expression down to less than 50% and 10% respectively, whereas VEGF, HGF and bFGF had no significant effect [Figure 5B and D; (6)]. In contrast, in HUVECs, expression of ICAM1 and VCAM1 was severely down-regulated by bFGF (ICAM1 <7%; VCAM1 <51%) but not by TGFβs (Figure 5A and C). Additionally, ICAM1 was down-regulated by incubation with VEGF, HGF and TGFß2 in HUVECs. TGFβ1 did not significantly regulate ICAM1 nor VCAM1 expression. Together, these data clearly show that CAMs essential for T-cell infiltration are differently regulated in HUVECs as compared to GECs.

Figure 4.
Figure 4.

Validation of candidate genes from comparative microarray analysis. Three candidate genes up-regulated (A) and three down-regulated (B) in glioblastoma endothelial cells (GECs) were illustrated. Fold differences were calculated using the median expression of GECs and human umbilical cord vein endothelial cells (HUVECs). Heteroscedastic Student's t-test was used for group comparison. *p<0.05 **p<0.01, and ***p<0.001.

Discussion

There is a huge need for model cells for in vitro experiments that properly recapitulate the brain tumor endothelium. HUVECs have been the gold-standard cell lines for these purposes (10). In contrast to GECs, HUVECs are easily accessible. However, this study strongly indicates that HUVECs are not adequate substitutes for GECs for addressing angiogenesis and studying immune-relevant processes in GBM.

As a major concern, we found that HUVECs and GECs differ markedly in their transcriptome. Previous studies investigating the GEC transcriptome aimed to identify tumor vessel markers, which can be used for diagnosis or therapeutic targeting (21-23). Cell isolates were obtained via laser microdissection of vessels from tissue sections and analyzed by DNA microarrays (22, 23), or purified from fresh tissue dissociations and analyzed by serial analyses of gene expression (21). Gene expression of GECs was compared to that of vessels derived from non-malignant brain tissue (21-23), to endothelial cells from WHO grade II gliomas (23), or to endothelial cells from colonic cancer (21). They all came to the same conclusion that tumor and normal (brain) endothelium are distinct at the molecular level. Remarkably, Dieterich and colleagues showed high similarities between GECs analyzed, while non-malignant and grade II glioma samples were similar to each other as well, and clearly differed from GBM samples (23). Comparing GECs to colonic cancer endothelial cells Madden and colleagues reported limited conservation in respect to both tumor-induced and tumor-repressed vessel markers (21). Taken together, these findings indicate that GBM vasculature displays a unique gene expression clearly distinct from that of HUVECs as shown in our study but also from that of endothelial cells of lower grade gliomas, normal brain and other peripheral types of cancer.

Moreover, we performed gene ontology analyses of our microarray data and particularly discovered gene groups involved in angiogenesis (e.g. cell migration, tube morphogenesis, tube development, vascular/blood vessel development) to be differentially expressed in GECs compared to HUVECs. In line with our findings, the above-mentioned studies showed similar alterations with respect to angiogenesis-related terms (21-23). In addition, KEGG pathway analysis comparing GECs and HUVECs showed alterations in genes involved in leukocyte transendothelial migration. Of note, to our knowledge, this is the first study to describe alterations in this critical step for immune cell migration into the tumor stroma. Considering these results, in vitro experiments using HUVECs as surrogates for GECs addressing angiogenic and immunological questions might lead to misleading results.

This prompted us to comprehensively perform a functional experiment with GECs and HUVECs combining aspects of angiogenesis and immunobiology. We investigated the influence of five of the most frequent angiogenic cytokines in GBM [VEGF, bFGF, HGF, TGFβ1 and TGFβ2; (6)] on major CAMs for T-cell transmigration, which are ICAM1 and VCAM1 (6, 24, 25). Interestingly, HUVECs down-regulated their CAM protein expression in the presence of VEGF, bFGF and HGF, whereas TGFβs had nearly no effect. In contrast, GECs responded exclusively to TGFβ1 and TGFβ2 but not to VEGF, bFGF and HGF, with a dramatic down-regulation of ICAM1 and VCAM1. Regulation of CAMs and subsequently lymphocyte adhesion by angiogenic cytokines has been under investigation for more than two decades using HUVECs as models for tumor vasculature (16-20). There exist hints that TGFβs might repress T-cell adherence to HUVECs pre-stimulated with TNF-α by a mechanism independent from ICAM and VCAM regulation (26). TGFβs play an important role in various key tumorigenic processes and are addressed in several phase II trials targeting TGFβs or TGFβ signaling for glioma and other types of cancer (27).

Figure 5.
Figure 5.

Regulation of cell adhesion molecules on human umbilical cord vein endothelial surface in response to angiogenic cytokines compared to glioblastoma endothelial cells. Flow cytometric quantification of intercellular adhesion molecule 1 (ICAM1) and vascular cell adhesion molecule 1 (VCAM1) expression in HUVECs (A, C) and GECs (B, D) after 72 h incubation with 100 ng/ml hepatocyte growth factor (HGF), vascular endothelial growth factor (VEGF), and basic fibroblast growth factor (bFGF), respectively, or 10 ng/ml transforming growth factor beta 1 (TGFβ1) and TGFβ2, respectively. Data are presented as the mean and SEM of independent experiments on three different cultures each. Asterisks indicate significant inhibition of ICAM1 and VCAM1 protein expression. (Graph adapted by permission from the American Association for Cancer Research: Lohr et al., Clin. Cancer Res. 2011; 17(13), 4296-4308. DOI: 10.1158/1078-0432.CCR-10-2557).

Of note, CAM repression by bFGF and VEGF has been most extensively studied (16, 18-20). Likewise, bFGF was the strongest regulator of ICAM1 and VCAM1 in our HUVEC experiments. VEGF down-regulating effects were reported for ICAM1, congruent with our results (16, 18-20). Remarkably, an in vivo study implanting a colonic carcinoma cell line in mice verified that tumor-induced CAM down-regulation significantly diminished leukocyte adhesion. This observation was further investigated on endothelial cells isolated from these tumors, identifying bFGF and VEGF as possible causes (17). In a subsequent study, using the same model, they showed that anti-angiogenic compounds can restore CAM expression and lymphocyte infiltration (28). Because of these and other encouraging preclinical data, various angiogenesis inhibitors are being tested or are already in use for the treatment of different types of cancer, including colonic cancer and GBM (29).

The VEGF antibody to bevacizumab, as the most prominent example, was the first anti-angiogenic agent achieving approval in 2004 for combination treatment of metastatic colonic cancer (30). In 2009, it was approved as a monotherapy for recurrent GBM, although disease in most patients became resistant and developed more aggressive relapses (31, 32). Hence, the understanding over resistance mechanisms was acknowledged to be crucial to the improvement of treatment outcomes. Research has not only been focusing on the effect of anti-angiogenic treatment on GBM cell lines or mouse models (29), but also on the endothelial compartment, for which HUVECs were mostly used. Experiments were conducted on viability, proliferation, apoptosis, tube formation and interaction with GBM cells (33-38). However, only one of these studies verified their results on a single GEC culture (36), which might explain the overlooked treatment resistance.

In conclusion, if model endothelial cells are selected for experiments in cancer, especially for brain tumors, the suitability of these cells needs to be unequivocally proven. For future preclinical in vitro testing, our findings strongly indicate the use of HUVECs as model cells for tumor endothelium and GBMs should be avoided, in particular for experiments investigating the efficacy of anti-angiogenic compounds on endothelial cells, interaction with tumor cells, or leukocyte adhesion to tumor endothelium and leukocyte transmigration. Finally, the influence of new compounds used for GBM therapy should be confirmed on GECs.

Acknowledgements

The Authors would like to thank Farzaneh Kashfi, Hildegard Göltzer, Ilka Hearn und Melanie Greibich for excellent technical support.

Footnotes

  • ↵* These Authors contributed equally to this study.

  • Conflicts of Interest

    The Authors confirm that they have no conflicts of interest in regard to this article.

  • Received February 13, 2015.
  • Revision received February 26, 2015.
  • Accepted March 2, 2015.

References

  1. ↵
    1. Bao S,
    2. Wu Q,
    3. McLendon RE,
    4. Hao Y,
    5. Shi Q,
    6. Hjelmeland AB,
    7. Dewhirst MW,
    8. Bigner DD,
    9. Rich JN
    : Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature 444(1720): 756-760, 2006.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Wen PY,
    2. Kesari S
    : Malignant gliomas in adults. N Engl J Med 359: 492-507, 2008.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Folkman J
    : Tumor angiogenesis: therapeutic implications. N Engl J Med 18: 1182-1186, 1971.
    OpenUrl
  4. ↵
    1. Fridman WH,
    2. Galon J,
    3. Pages F,
    4. Tartour E,
    5. Sautes-Fridman C,
    6. Kroemer G
    : Prognostic and predictive impact of intra- and peritumoral immune infiltrates. Cancer Res 71: 5601-5610, 2011.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Yang I,
    2. Tihan T,
    3. Han SJ,
    4. Wrensch MR,
    5. Wiencke J,
    6. Sughrue ME,
    7. Parsa AT
    : CD8+ T-cell infiltrate in newly diagnosed glioblastoma is associated with long-term survival. J Clin Neurosci 17: 1381-1385, 2010.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Lohr J,
    2. Ratliff T,
    3. Huppertz A,
    4. Ge Y,
    5. Dictus C,
    6. Ahmadi R,
    7. Grau S,
    8. Hiraoka N,
    9. Eckstein V,
    10. Ecker RC,
    11. Korff T,
    12. von Deimling A,
    13. Unterberg A,
    14. Beckhove P,
    15. Herold-Mende C
    : Effector T-cell infiltration positively impacts survival of glioblastoma patients and is impaired by tumor-derived TGF-β. Clin Cancer Res 17(13): 4296-4308, 2011.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Norden AD,
    2. Drappatz J,
    3. Wen PY
    : Antiangiogenic therapies for high-grade glioma. Nat Rev Neurol 5: 610-620, 2009.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Stupp R,
    2. Mason WP,
    3. van den Bent MJ,
    4. Weller M,
    5. Fisher B,
    6. Taphoorn MJ,
    7. Belanger K,
    8. Brandes AA,
    9. Marosi C,
    10. Bogdahn U,
    11. Curschmann J,
    12. Janzer RC,
    13. Ludwin SK,
    14. Gorlia T,
    15. Allgeier A,
    16. Lacombe D,
    17. Cairncross JG,
    18. Eisenhauer E,
    19. Mirimanoff RO
    : Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352(10): 987-996, 2005.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Miebach S,
    2. Grau S,
    3. Hummel V,
    4. Rieckmann P,
    5. Tonn JC,
    6. Goldbrunner RH
    : Isolation and culture of microvascular endothelial cells from gliomas of different WHO grades. J Neurooncol 76: 39-48, 2006.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Richardson MR,
    2. Lai X,
    3. Witzmann FA,
    4. Yoder MC
    : Venous and arterial endothelial proteomics: mining for markers and mechanisms of endothelial diversity. Expert Rev Proteomics 7(6): 823-831, 2010.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Aird WC
    : Molecular heterogeneity of tumor endothelium. Cell Tissue Res 335(1): 271-281, 2009.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Liebner S,
    2. Czupalla CJ,
    3. Wolburg H
    : Current concepts of blood-brain barrier development. Int J Dev Biol 55: 467-476, 2011.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Craig LE,
    2. Spelman JP,
    3. Strandberg JD,
    4. Zink MC
    : Endothelial cells from diverse tissues exhibit differences in growth and morphology. Microvasc Res 55: 65-76, 1998.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Carmeliet P,
    2. Jain RK
    : Angiogenesis in cancer and other diseases. Nature 407: 249-257, 2000.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Storey JD,
    2. Tibshirani R
    : Statistical significance for genomewide studies. PNAS 100: 9440-9445, 2003.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Bouma-ter Steege JC,
    2. Baeten CI,
    3. Thijssen VL,
    4. Satijn SA,
    5. Verhoeven IC,
    6. Hillen HF,
    7. Wagstaff J,
    8. Griffioen AW
    : Angiogenic profile of breast carcinoma determines leukocyte infiltration. Clin Cancer Res 10(21): 7171-7178, 2004.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Dirkx AE,
    2. oude Egbrink M,
    3. Kuijpers M,
    4. van der Niet S,
    5. Heijnen V,
    6. Bouma-ter Steege S,
    7. Wagstaff J,
    8. Griffioen AW
    : Tumor angiogenesis modulates leukocyte-vessel wall interactions in vivo by reducing endothelial adhesion molecule expression. Cancer Res 63: 2322-2329, 2003.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Griffioen AW,
    2. Damen CA,
    3. Martinotti S,
    4. Blijham GH,
    5. Groenewegen G
    : Endothelial intercellular adhesion molecule-1 expression is suppressed in human malignancies: the role of angiogenic factors. Cancer Res 56(5): 1111-1117, 1996.
    OpenUrlAbstract/FREE Full Text
    1. Kitayama J,
    2. Nagawa H,
    3. Yasuhara H,
    4. Tsuno N,
    5. Kimura W,
    6. Shibata Y,
    7. Muto T
    : Suppressive effect of basic fibroblast growth factor on transendothelial emigration of CD4(+) T-lymphocyte. Cancer Res 54(17): 4729-4733, 1994.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Tromp SC,
    2. oude Egbrink MG,
    3. Dings RP,
    4. van Velzen S,
    5. Slaaf DW,
    6. Hillen HF,
    7. Tangelder GJ,
    8. Reneman RS,
    9. Griffioen AW
    : Tumor angiogenesis factors reduce leukocyte adhesion in vivo. Int Immunol 12(5): 671-676, 2000.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Madden SL,
    2. Cook BP,
    3. Nacht M,
    4. Weber WD,
    5. Callahan MR,
    6. Jiang Y,
    7. Dufault MR,
    8. Zhang X,
    9. Zhang W,
    10. Walter-Yohrling J,
    11. Rouleau C,
    12. Akmaev VR,
    13. Wang CJ,
    14. Cao X,
    15. St Martin TB,
    16. Roberts BL,
    17. Teicher BA,
    18. Klinger KW,
    19. Stan RV,
    20. Lucey B,
    21. Carson-Walter EB,
    22. Laterra J,
    23. Walter KA
    : Vascular gene expression in nonneoplastic and malignant brain. Am J Pathol 165(2): 601-608, 2004.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Pen A,
    2. Moreno MJ,
    3. Martin J,
    4. Stanimirovic DB
    : Molecular markers of extracellular matrix remodeling in glioblastoma vessels: microarray study of laser-captured glioblastoma vessels. Glia 55(6): 559-572, 2007.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Dieterich LC,
    2. Mellberg S,
    3. Langenkamp E,
    4. Zhang L,
    5. Zieba A,
    6. Salomäki H,
    7. Teichert M,
    8. Huang H,
    9. Edqvist PH,
    10. Kraus T,
    11. Augustin HG,
    12. Olofsson T,
    13. Larsson E,
    14. Söderberg O,
    15. Molema G,
    16. Pontén F,
    17. Georgii-Hemming P,
    18. Alafuzoff I,
    19. Dimberg A
    : Transcriptional profiling of human glioblastoma vessels indicates a key role of VEGFA and TGFβ2 in vascular abnormalization. J Pathol 228(3): 378-390, 2012.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Lyck R,
    2. Reiss Y,
    3. Gerwin N,
    4. Greenwood J,
    5. Adamson P,
    6. Engelhardt B
    : T-cell interaction with ICAM1/ICAM2 double-deficient brain endothelium in vitro: the cytoplasmic tail of endothelial ICAM1 is necessary for transendothelial migration of T-cells. Blood 102: 3675-3683, 2003.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Mazzolini G,
    2. Narvaiza I,
    3. Bustos M,
    4. Duarte M,
    5. Tirapu I,
    6. Bilbao R,
    7. Qian C,
    8. Prieto J,
    9. Melero I
    : ανβ3 Integrin-mediated adenoviral transfer of interleukin-12 at the periphery of hepatic colon cancer metastases induces VCAM1 expression and T-cell recruitment. Mol Ther 3: 665-672, 2001.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Gamble JR,
    2. Vadas MA
    : Endothelial cell adhesiveness for human T-lymphocytes is inhibited by transforming growth factor-beta 1. J Immunol 146(4): 1149-1154, 1991.
    OpenUrlAbstract
  27. ↵
    1. Sheen YY,
    2. Kim MJ,
    3. Park SA,
    4. Park SY,
    5. Nam JS
    : Targeting the transforming growth factor-β signaling in cancer therapy. Biomol Ther 21(5): 323-331, 2013.
    OpenUrl
  28. ↵
    1. Dirkx AE,
    2. oude Egbrink MG,
    3. Castermans K,
    4. van der Schaft DW,
    5. Thijssen VL,
    6. Dings RP,
    7. Kwee L,
    8. Mayo KH,
    9. Wagstaff J,
    10. Boumater Steege JC,
    11. Griffioen AW
    : Anti-angiogenesis therapy can overcome endothelial cell anergy and promote leukocyte-endothelium interactions and infiltration in tumors. FASEB J 20(6): 621-630, 2006.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Cea V,
    2. Sala C,
    3. Verpelli C
    : Antiangiogenic therapy for glioma. J Signal Transduct 2012: 483040, 2012.
    OpenUrlPubMed
  30. ↵
    1. Hurwitz H,
    2. Fehrenbacher L,
    3. Novotny W,
    4. Cartwright T,
    5. Hainsworth J,
    6. Heim W,
    7. Berlin J,
    8. Baron A,
    9. Griffing S,
    10. Holmgren E,
    11. Ferrara N,
    12. Fyfe G,
    13. Rogers B,
    14. Ross R,
    15. Kabbinavar F
    : Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350(23): 2335-2342, 2004.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Cohen MH,
    2. Shen YL,
    3. Keegan P,
    4. Pazdur R
    : FDA drug approval summary: bevacizumab (Avastin) as treatment of recurrent glioblastoma multiforme. Oncologist 14(11): 1131-1138, 2009.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Norden AD,
    2. Young GS,
    3. Setayesh K,
    4. Muzikansky A,
    5. Klufas R,
    6. Ross GL,
    7. Ciampa AS,
    8. Ebbeling LG,
    9. Levy B,
    10. Drappatz J,
    11. Kesari S,
    12. Wen PY
    : Bevacizumab for recurrent malignant gliomas: efficacy, toxicity, and patterns of recurrence. Neurology 70(10): 779-787, 2008.
    OpenUrlCrossRef
  33. ↵
    1. Chen Z,
    2. Htay A,
    3. Dos Santos W,
    4. Gillies GT,
    5. Fillmore HL,
    6. Sholley MM,
    7. Broaddus WC
    : In vitro angiogenesis by human umbilical vein endothelial cells (HUVEC) induced by three-dimensional co-culture with glioblastoma cells. J Neurooncol 92(2): 121-128, 2009.
    OpenUrlCrossRefPubMed
    1. Gorski DH,
    2. Beckett MA,
    3. Jaskowiak NT,
    4. Calvin DP,
    5. Mauceri HJ,
    6. Salloum RM,
    7. Seetharam S,
    8. Koons A,
    9. Hari DM,
    10. Kufe DW,
    11. Weichselbaum RR
    : Blockage of the vascular endothelial growth factor stress response increases the antitumor effects of ionizing radiation. Cancer Res 59(14): 3374-3378, 1999.
    OpenUrlAbstract/FREE Full Text
    1. Hoang T,
    2. Huang S,
    3. Armstrong E,
    4. Eickhoff JC,
    5. Harari PM
    : Enhancement of radiation response with bevacizumab. J Exp Clin Cancer Res 31: 37, 2012.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Jahangiri A,
    2. De Lay M,
    3. Miller LM,
    4. Carbonell WS,
    5. Hu YL,
    6. Lu K,
    7. Tom MW,
    8. Paquette J,
    9. Tokuyasu TA,
    10. Tsao S,
    11. Marshall R,
    12. Perry A,
    13. Bjorgan KM,
    14. Chaumeil MM,
    15. Ronen SM,
    16. Bergers G,
    17. Aghi MK
    : Gene expression profile identifies tyrosine kinase c-Met as a targetable mediator of antiangiogenic therapy resistance. Clin Cancer Res 19(7): 1773-1783, 2013.
    OpenUrlAbstract/FREE Full Text
    1. Kurzen H,
    2. Schmitt S,
    3. Näher H,
    4. Möhler T
    : Inhibition of angiogenesis by non-toxic doses of temozolomide. Anticancer Drugs 14(7): 515-522, 2003.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Mathieu V,
    2. De Nève N,
    3. Le Mercier M,
    4. Dewelle J,
    5. Gaussin JF,
    6. Dehoux M,
    7. Kiss R,
    8. Lefranc F
    : Combining bevacizumab with temozolomide increases the antitumor efficacy of temozolomide in a human glioblastoma orthotopic xenograft model. Neoplasia 10(12): 1383-1392, 2008.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Endothelial Cells Derived from Non-malignant Tissues Are of Limited Value as Models for Brain Tumor Vasculature
JENNIFER LOHR, ANDREAS MOCK, PHILIPP BECKHOVE, CHRISTEL HEROLD-MENDE
Anticancer Research May 2015, 35 (5) 2681-2690;
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords