Identification of COL6A1 as the Key Gene Associated with Antivascular Endothelial Growth Factor Therapy in Glioblastoma Multiforme

Abstract

Background:

Vascular endothelial growth factors (VEGFs) are important for glioblastoma multiforme (GBM) growth and development. However, the effects of VEGF-targeting drugs in primary GBM remain poorly understood.

Aim:

We aimed to explore the key genes correlated with VEGF expression and prognosis and elucidate their potential implications in GBM anti-VEGF therapy.

Materials and Methods:

RNA-seq data with the corresponding clinicopathological information was retrieved from The Cancer Genome Atlas and the Chinese Glioma Genome Atlas. Weighted gene coexpression network analyses was performed on differentially expressed genes to construct coexpression modules and investigate their correlation with VEGFs. Functional enrichment analyses were performed based on the coexpressed genes from the most promising modules. CytoHubba and Kaplan-Meier analyses were implemented to identify the key genes in the modules of interest. The oncomine database, quantitative reverse transcription PCR, and the Human Protein Atlas were used to investigate the expression characteristics of the identified key genes.

Results:

Four modules (cyan, green, purple, and tan) correlated significantly with VEGF expression. Enrichment analyses suggested that extracellular matrix-receptor interaction, growth factor binding, and the PI3K-Akt pathways were involved in VEGF expression. Four hub genes (COL6A1, SNRPG, COL3A1, and AHI1) associated with VEGF were identified. Among them, COL6A1 was regarded as the key gene associated with anti-VEGF therapy. Further, COL6A1 was upregulated in GBM compared to that in normal brain tissues. COL6A1 overexpression was associated with a poor prognosis.

Conclusion:

COL6A1 was identified as the key gene associated with anti-VEGF therapy and may provide novel insight into GBM targeted therapy.

Introduction

Glioma is the most common neuroepithelial malignant tumor characterized by high disability and recurrence rates and poor prognosis. The WHO classification of CNS tumors categorizes gliomas into four grades (I-IV) based on the degree of malignancy, with glioblastoma multiformes (GBMs) belonging to grade IV (Louis et al., 2016).The Central Brain Tumor Registry of the United States Statistical Report (2015) predicted a yearly incidence of 5.12 glioma cases per 100,000 individuals (Ostrom et al., 2018). The overall survival (OS) of patients with GBM is only 14 months, despite the standard therapeutic approach (maximal surgical resection in combination with adjuvant chemotherapy and radiotherapy) (Johnson and O'Neill, 2012).

Vascular endothelial growth factors (VEGFs), first described as vascular permeability factors in 1983, are secreted by tumor cells and modify the tumor extracellular matrix (ECM) to promote the ingrowth of new blood vessels and peritumoral edema (Senger et al., 1983, 1994-1995).

The VEGF family includes several subtypes, namely, VEGF-A, VEGF-B, VEGF-C, VEGF-D, VEGF-E, VEGF-F, and placental growth factor. Of them, VEGF-A and VEGF-B regulate angiogenesis mainly by binding to VEGF receptor-1 (VEGFR1) and VEGFR2, whereas VEGF-C and VEGF-D promote lymphangiogenesis by binding to VEGFR3 (Connolly et al., 1989; Shibuya, 1995; Joukov et al., 1996; Olofsson et al., 1996; Achen et al., 1998; Shibuya and Claesson-Welsh, 2006). Further, PIGF can stimulate endothelial cell (EC) proliferation although it is mainly expressed in the placenta (Maglione et al., 1991). The VEGF-like proteins, VEGF-E and VEGF-F, are expressed in nonmammals only.

These chemoattractant are secreted by glioma cells via autocrine and paracrine mechanisms. They induce the migration of fibroblasts and brain capillary ECs by interacting with their specific receptors and promote the formation of capillary-like structures (Abramovitch et al., 1995). Therefore, they promote the growth of both ECs and glioma cells themselves (Plate et al., 1992; Samoto et al., 1995).

Since neovascularization is vital for glioma growth, targeting VEGF and inhibiting neovascularization is a major therapeutic approach for tumors (Kim et al., 1993). The anti-VEGF monoclonal antibody (bevacizumab; Avastin)-neutralized VEGF inhibited the growth of ECs and angiogenesis activity induced by VEGFs (Kim et al., 1992; Presta et al., 1997; Ferrara et al., 2005). Bevacizumab used alone or in combination with cytotoxic chemotherapy in patients had a good performance status at the time of GBM recurrence. However, a few studies do not support the use of bevacizumab to prolong progression-free survival and OS in the primary site of GBMs (Diaz et al., 2017; Moriya et al., 2018). Therefore, to elucidate the key molecules and signaling pathways involved with the expression and function of VEGFs in GBM is of utmost importance.

Weighted gene coexpression network analysis (WGCNA) is a commonly used computational method that explores the complicated relationships between genes and phenotypes (such as brain imaging, genetics, and cancer data analysis) and helps in identifying therapeutic targets or candidate biomarker. In brief, gene expression data are transformed into a coexpression module by WGCNA, thereby providing insights into gene networks that may be responsible for phenotypic traits of interest (Langfelder and Horvath, 2008).

In the present study, we aimed to explore the gene cluster(s) associated with the expression of VEGF in GBM using WGCNA. Further, we analyzed the biological functions and pathways associated with the interested gene clusters to understand the mechanisms by which VEGFs modulate GBM. Consequently, we identified one key gene correlated with the expression of VEGFs and GBM patient prognosis, which provides novel strategies for the anti-VEGF treatment of GBM.

Materials and Methods

Data mining

Figure 1 illustrates the study design. RNA-seq data of 5 normal brain tissues and 167 GBM cases as well as the corresponding clinicopathological information were obtained from The Cancer Genome Atlas (TCGA) database. From the Chinese Glioma Genome Atlas (CGGA), RNA-seq data and their corresponding clinicopathological information of 343 GBMs cases were downloaded for validation.

FIG. 1.

Overview of the procedures for analyzing VEGF and VEGF-associated key gene in GBM. GBM, glioblastoma multiforme; VEGF, vascular endothelial growth factors.

WGCNA for identifying clustered gene modules

Normalization was performed using the Limma R package. The edge R package was utilized to calculate the significantly differentially expressed genes (DEGs) in GBMs, and log-fold change| >1 and p-value <0.05 were set as the thresholds for differential gene screening based on GBM in TCGA dataset. A scale-free network was constructed using the “blockwiseModules” function of WGCNA R package (version 3.6.1). The soft threshold for network construction was chosen for the adjacency matrix to be the continuous value between 0 and 1 so that the constructed network can be closer to the real biological network state. Following the module partition analysis, the gene coexpression modules were identified. Then, the module eigengene, which represents the expression level for each module, was calculated. These modules were analyzed by their eigengene correlation to VEGF expression.

Functional enrichment analyses

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed on the interested modules to investigate their functional role in VEGF expression.

Identification of key genes correlated with VEGF expression

Using STRING database protein-protein interactions (PPIs) between the interested module genes were assessed. In addition, CytoHubba plugin of the Cytoscape software was used to rank nodes by their network features, selecting the top genes as hub genes in the “Maximum Correlation Criteria.” Furthermore, we evaluated the impact of expression level of hub genes and their corresponding VEGF members on GBM prognosis. Together with the expression levels of VEGFs, the hub genes that correlated with GBM prognosis were regarded as key genes.

Verification of expression and prognostic value of the key gene

We assessed the expression of key gene in GBM and the corresponding normal brain tissues using Oncomine™ database. Utilizing the Human Protein Atlas (HPA) online database, we detect the expression of key gene at a translational level. Further, we performed Kaplan-Meier survival analysis to detect the prognostic value of the key gene followed by result validation using ROC curve. To investigate the biofunction of the key gene, gene Set Enrichment Analysis (GSEA) was performed in the enrichment of the MSigDB collection (c2.cp.kegg and h.all. v6.2. symbols) based on expression levels of the key gene.

Quantitative reverse transcription PCR

A total of 55 samples, including 40 tumor samples from patients with GBM and 15 peritumoral brain tissues, were subjected to quantitative reverse transcription PCR (qRT-PCR) to verify the expression of key gene. Total RNAs was extracted using TRIzol reagent (AG21102; Accurate Biotechnology, Hunan, China). cDNA was synthesized from the extracted RNA (0.5 μg) using the RT Premix for qPCR (AG11706; Accurate Biotechnology) in Reverse Transcription System (Biometra Advanced). The RT-PCR was performed at 37°C for 15 min, 85°C for 5 s, and 4°C hold. qRT-PCR was performed on cDNA, using the SYBR Green Kit (AG11701; Accurate Biotechnology) on the Bio-Rad CFX96 connect Real-Time PCR System. The mixes were predenatured at 95°C for 30 min, followed by 40 cycles of denaturation at 95°C for 15 s, and 60°C for 30 s. The 2^−ΔCt method was utilized to calculate the relative expression level of COL6A1. The primers used are as follows: COL6A1 forward 5′-GGCCTGGACTGGACATGA GA-3′, reverse 5′-AAAGTCAGCAACATGGATATGGTT C-3′; GAPDH, forward 5′-GCCATCACAGCAACACAGAA-3′, reverse 5′-GCCATACCAGTAAGCTTGCC-3′.

Statistical analysis

Wilcoxon test was used to compare the difference of VEGFs and expressions of key genes between the groups. OS is presented as the Kaplan-Meier curve. Kaplan-Meier survival curve with the log-rank test was calculated and plotted using survminer R-package. p < 0.05 was regarded as statistically significant.

Ethics Approval and Informed Consent

The study was approved by the Research Ethics Committee of Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Science [No. GDREC20190145H(R2)]. All subjects provided written informed consent.

Results

VEGF expression in GBMs

Expression of VEGF-A and VEGF-B was upregulated, whereas that of VEGF-D was reduced in GBM tissue compared to normal brain tissue (Supplementary Fig. S1A). Although VEGFs influenced the angiogenesis in glioma, no strong correlation between VEGF expression and OS of patients with GBM was observed in TCGA cohort (Supplementary Fig. S1B-F).

WGCNA of DEGs in GBM

Totally, 6676 DEGs (3227 downregulated and 3450 upregulated genes) were identified, screened out, and analyzed (Fig. 2). After eliminating any substandard samples, soft power was set as 4, and DEGs were clustered into 18 modules (Fig. 3 and Supplementary Fig. S2). The cyan module showed significant association with VEGF-A expression (Cor = 0.43, p = 4e-08), whereas the purple module was associated with VEGF-C expression (Cor = 0.46, p = 2e-09). The green module was significantly associated with the expression of VEGF-B (Cor = 0.32 p = 4e-05) and PIGF (Cor = 0.7, p = 2e-24), while the tan module had an intensive association with VEGF-D expression (Cor = 0.29, p = 3e-04) (Fig. 4).

FIG. 2.

DEGs between GBM and normal brain tissues. (A) Volcano plot and (B) heatmap visualizations of the DEGs between GBM cases and normal brain tissues. Red represents upregulation; green, downregulation. NORMAL represents normal brain tissue; TUMOR represents GBM sample. Log₂-fold change (FC) |>1|, p < 0.01. DEGs, differentially expressed genes. Color images are available online.

FIG. 3.

WGCNA analysis. Application of (A) free fit index and (B) mean connectivity determines the adjacency and topological overlap matrixes. (C) Cluster dendrogram of genes based on a dissimilarity measure with assigned module colors. The colored rows show the module membership obtained by the dynamic tree cut method and after merging modules. (D) Eigengene dendrogram and heatmap plot of correlated eigengenes term metamodules. Each row and column is in accordance with one module. Red represents positive correlation; blue, negative. WGCNA, weighted gene coexpression network analysis. Color images are available online.

FIG. 4.

Module-trait associations. (A) Each row corresponds to a ME, and each column corresponds to a trait. Each cell contains the corresponding correlation and p-value based on the correlation between MEs and traits. (B-F) Scatterplot depicting correlation between gene significance for corresponding traits (VEGF) and module membership of interested ME. p < 0.05, significant difference. ME, module eigengene. Color images are available online.

Functional and pathway enrichment analyses

Genes in the cyan module were mainly enriched in extracellular structure organization, regulation of receptor binding, extracellular space, and ECM structural constituent (Fig. 5A), while those in the purple module were mainly enriched in the collagen metabolic process, ECM organization, and growth factor binding (Fig. 5B). Further, genes in the green module were mainly enriched in translational elongation, mitochondrial intermembrane space, glucosyltransferase activity, and protein phosphatase inhibitor activity (Fig. 5C), while those of tan module were mainly enriched in positive regulation of fatty acid beta-oxidation, glycolipid metabolic process, microtubule organizing center part, and monocarboxylic acid transmembrane transporter activity (Fig. 5D).

FIG. 5.

Function enrichment analyses of interested modules. GO analysis of the (A) cyan, (B) green, (C) purple, and (D) tan module. KEGG analysis of the (E) cyan module, (F) green module, (G) purple module, and (H) tan modules. The brighter the color is, the greater the correlation is. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes. Color images are available online.

KEGG analysis showed that genes in cyan module were enriched in the proteoglycans in cancer, ECM-receptor interaction, and Wnt signaling pathway (Fig. 5E). Genes in green module were enriched in ribosome, spliceosome, and carbon metabolism (Fig. 5F). Genes in purple module showed were mainly enriched in the PI3K-Akt pathway, ECM-receptor interaction, small cell lung cancer, and proteoglycan pathways in cancer (Fig. 5G). Further, genes in tan module were enriched in glycerophospholipid metabolism and PPAR signaling pathway (Fig. 5H).

COL6A1 can be regarded as key gene associated with VEGF expression in GBMs

Protein-protein interaction networks of the genes in the interested modules are shown in Figure 6. CytoHubba plugin screened COL6A1, SNRPG, COL3A1, and AHI1 as the hub genes from four modules, respectively. Of them, alterations in the expression level of COL6A1 and VEGF-A closely correlated with GBM progress (Fig. 7A, B). Therefore, COL6A1 were regarded as the key gene associated with VEGFs in GBMs. Oncomine analysis and qRT-PCR showed that the mRNA expression of COL6A1 in GBM is higher than that in the corresponding normal brain tissues (Fig. 8A, B). HPA immunohistochemistry data showed that COL6A1 was moderately upregulated in high-grade gliomas compared to normal brain tissues (Fig. 8C). Furthermore, survival analysis showed that COL6A1 was associated with poor prognosis of GBM in TCGA cohort; this finding was validated in CGGA cohort (Fig. 8D, F). The 5-year survival ROC curve indicated the expression level of COL6A1 has good performance in predicting OS (Fig. 8E, G). Further, GSEA reveal that the most significantly enriched signaling pathways were based on their NES. Genes in high-expression phenotype of the key gene were significantly enriched with ECM-receptor interaction and focal adhesion in GBM (Fig. 8H).

FIG. 6.

Visualization of modules. PPI network of the (A) cyan, (B) green, (C) purple, and (D) tan modules. Color images are available online.

FIG. 7.

Effect of expression level of hub genes and VEGFs on the prognosis of GBM prognosis. Effect of COL6A1 and VEGF-A on GBM patients in (A) TCGA cohort and (B) CGGA cohort; Effect of SNRPG and VEGF-B on GBM patients in (C) TCGA cohort and (D) CGGA cohort; Effect of COL3A1 and VEGF-C on GBM patients in (E) TCGA cohort and (F) CGGA cohort; Effect of AHI1 and VEGF-D on GBM patients in (G) TCGA cohort and (H) CGGA cohort; Effect of SNRPG and PIGF on GBM patients in (I) TCGA cohort and (J) CGGA cohort; p < 0.05, significant difference. *p < 0.05. AHI1, Cadherin-23; COL6A1, collagen IV α1; COL3A1, collagen III α1; CGGA, Chinese Glioma Genome Atlas; VEGF-A, vascular endothelial growth factor A; VEGF-B, vascular endothelial growth factor B; VEGF-C, vascular endothelial growth factor C; VEGF-D, vascular endothelial growth factor D; PlGF, placenta growth factor; SNRPG, small nuclear ribonucleoprotein G; TCGA, The Cancer Genome Atlas database. Color images are available online.

FIG. 8.

Key gene in module. (A) mRNA expression of COL6A1 in GBM and normal brain tissue. (B) The results of qRT-PCR of GBM samples and peritumoral brain tissue samples. (C) Immunohistochemistry data from HPA showed that COL6A1 is upregulated in high-grade glioma at translational level. (D, F) Kaplan-Meier survival curve depicting the effect of COL6A1 expression on overall survival of GBM patients in TCGA cohort and CGGA cohort. (E, G) ROC curve evaluates the performance of the expression level of COL6A1 in predicting overall survival. (H) Enrichment plots of COL6A1 of GBM from GSEA. p < 0.05, significant difference. *p < 0.05; **p < 0.01. GSEA, gene Set Enrichment Analysis; HPA, Human Protein Atlas; qRT-PCR, quantitative reverse transcription PCR. Color images are available online.

Discussion

Michaelsen et al. (2018) reported that VEGF-C sustains VEGFR-2 activation, thereby promoting tumor growth and that VEGF-C is involved in counteracting bevacizumab therapy in GBM. Although anti-VEGF therapy is promising in GBM management, its application has been limited. Further, escape mechanisms exhibited by GBM cells requires further exploration (Kim et al., 2018).

In the present study, we applied WGCNA to construct coexpression modules associated with VEGF expression and functional enrichment analyses to explore the biofunction of the four interested modules, namely cyan, green, purple, and tan. The biological processes or pathways that the interested module genes mainly enriched in were ECM-receptor interaction, growth factor binding, and PI3k-Akt pathway. ECM stiffness increases angiogenesis by stimulating VEGF signaling in ECs, thereby supporting growth and survival of the ECs. Besides, ECM hampers drug penetration to this niche and promotes the cancer cells proliferation (Najafi et al., 2019).

It is not surprising that the growth factor-binding pathway was an enriched process in the interested modules. Interestingly, VEGF-stimulated irregular tumor flow induces further tumor hypoxia and subsequent increases in VEGF production (Carmeliet, 2005). In addition, the PI3K-Akt pathway is pivotal for tumor progression, invasion, metastasis and angiogenesis (Martini et al., 2014). Ping et al. (2011) demonstrated that the interaction of CXCL12 with its receptor CXCR4 promotes GCS (glioma stem cells) -initiated glioma growth and angiogenesis by stimulating VEGF production via the PI3K-Akt pathway. Targeting the PI3K-Akt pathway results in the downregulation of VEGF expression and the inhibition of glioma proliferation (Zhang et al., 2012; Paul-Samojedny et al., 2015; Ramezani et al., 2017).

In our study, COL6A1 (collagen type VI α1 chain) was identified as a candidate biomarker related to VEGF expression and GBM prognosis via integrated bioinformatics analysis. COL6A1 assists in the synthesis of COL6 (collagen VI), a component of the ECM and constructs microfibrillar networks in the connective tissues of blood vessels and muscles. Previous studies have proven that COL6A1 is associated with myopathy and cardiovascular diseases (Bushby et al., 2014; Chen et al., 2019). Further, studies have also indicated that COL6A1 is significantly correlated with multifarious malignancies. Reportedly, COL6A1 expression is an independent prognostic marker of patient survival in cervical cancer (Hou et al., 2016). Owusu-Ansah et al. (2019) revealed that knockdown of COL6A1 suppresses metastasis and invasion of pancreatic cancer. These studies suggest that COL6A1 plays significant role in tumor progression.

Our studies have showed that high expression of COL6A1 is associated with worse prognosis of GBM. A previous study has revealed that infiltrative or proliferative growth of GBM results in cell compaction, thereby changing collagen and VEGF expression. Disruption of the structure and expression of collagen inhibit tumor angiogenesis, but does not promote glioma invasion and necrosis (Mammoto et al., 2013). However, fewer studies illustrated the relationship of these two genes in GBM and angiogenesis.

In the present study, we found that the high expression of COL6A1 group was significantly associated with ECM-receptor interaction as demonstrated by GSEA. An anchoring meshwork constructed by COL6 bridges cells to the surrounding environment and organizes the tissue architecture of the vasculature via directly interacting with COL6 (Marchand et al., 2019). In summary, our finding indicates that COL6A1 may regulate VEGF-associated angiogenesis and promote GBM tumorigenesis via mediating ECM-tumor interaction. More experimental support is needed to validate our result in the future.

Conclusion

In the present study, we showed that the expression of VEGFs was strongly correlated with ECM-receptor interaction, growth factor binding, and PI3K-Akt pathway. Overexpression of COL6A1 was associated with VEGF expression and poor prognosis in GBM. Therefore, we consider COL6A1 as holding significant potential to be considered as a novel strategy of anti-VEGF treatment in GBM management.

Footnotes

Authors' Contributions

H.L., C.X.H., and Y.Y. designed the study, checked the data, and prepared the article. J.T.Z. and G.Z.L. performed data collection and statistical analysis. R.M. and S.W.C. searched the literature. P.H.X. and Y.J.Z. performed the experiments. W.P. and D.Z. supervised this project.

Acknowledgments

We thank Mr. Zongtai Zheng, Mr. Zesen Chen, and Mr. Edison Zhang for the data processing.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This program was financially supported by Natural Science Foundation of China (No. 81901250), High-level Hospital Construction Project of Guangdong Province of China (No. DFJH201924), and Natural Science Foundation of Guangdong Province of China (No.2018A0303130236).

Supplementary Material

References

Abramovitch

, Meir

, Neeman

(1995) Neovascularization induced growth of implanted C6 glioma multicellular spheroids: magnetic resonance microimaging. Cancer Res, 55:1956-1962.

Achen

, Jeltsch

, Kukk

, et al. (1998) Vascular endothelial growth factor D (VEGF-D) is a ligand for the tyrosine kinases VEGF receptor 2 (Flk1) and VEGF receptor 3 (Flt4). Proc Natl Acad Sci U S A, 95:548-553.

Bushby

, Collins

, Hicks

(2014) Collagen type VI myopathies. Adv Exp Med Biol, 802:185-199.

Carmeliet P (2005) VEGF as a key mediator of angiogenesis in cancer. Oncology 69:Suppl 3, 4-10.

Chen

, Wu

, Yan

, Du

(2019) COL6A1 knockdown suppresses cell proliferation and migration in human aortic vascular smooth muscle cells. Exp Ther Med, 18:1977-1984.

Connolly

, Heuvelman

, Nelson

, et al. (1989) Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis. J Clin Invest, 84:1470-1478.

Diaz

, Ali

, Qadir

, et al. (2017) The role of bevacizumab in the treatment of glioblastoma. J Neurooncol, 133:455-467.

Ferrara

, Hillan

, Novotny

(2005) Bevacizumab (Avastin), a humanized anti-VEGF monoclonal antibody for cancer therapy. Biochem Biophys Res Commun, 333:328-335.

Hou

, Tong

, Kazobinka

, et al. (2016) Expression of COL6A1 predicts prognosis in cervical cancer patients. Am J Transl Res, 8:2838-2844.

10.

Johnson

, O'Neill

(2012) Glioblastoma survival in the United States before and during the temozolomide era. J Neurooncol, 107:359-364.

11.

Joukov

, Pajusola

, Kaipainen

, et al. (1996) A novel vascular endothelial growth factor, VEGF-C, is a ligand for the Flt4 (VEGFR-3) and KDR (VEGFR-2) receptor tyrosine kinases. EMBO J, 15:1751.

12.

Kim

, Li

, Houck

, et al. (1992) The vascular endothelial growth factor proteins: identification of biologically relevant regions by neutralizing monoclonal antibodies. Growth Factors, 7:53-64.

13.

Kim

, Li

, Winer

, et al. (1993) Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumour growth in vivo. Nature, 362:841-844.

14.

Kim

, Umemura

, Leung

(2018) Bevacizumab and glioblastoma: past, present, and future directions. Cancer J, 24:180-186.

15.

Langfelder

, Horvath

(2008) WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics, 9:559.

16.

Louis

, Perry

, Reifenberger

, et al. (2016) The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary. Acta Neuropathol, 131:803-820.

17.

Maglione

, Guerriero

, Viglietto

, et al. (1991) Isolation of a human placenta cDNA coding for a protein related to the vascular permeability factor. Proc Natl Acad Sci U S A, 88:9267-9271.

18.

Mammoto

, Jiang

, et al. (2013) Role of collagen matrix in tumor angiogenesis and glioblastoma multiforme progression. Am J Pathol, 183:1293-1305.

19.

Marchand

, Monnot

, Muller

, Germain

(2019) Extracellular matrix scaffolding in angiogenesis and capillary homeostasis. Semin Cell Dev Biol, 89:147-156.

20.

Martini

, De Santis

, Braccini

, et al. (2014) PI3K/AKT signaling pathway and cancer: an updated review. Ann Med, 46:372-383.

21.

Michaelsen

, Staberg

, Pedersen

, et al. (2018) VEGF-C sustains VEGFR2 activation under bevacizumab therapy and promotes glioblastoma maintenance. Neurooncol, 20:1462-1474.

22.

Moriya

, Ohba

, Adachi

, et al. (2018) A retrospective study of bevacizumab for treatment of brainstem glioma with malignant features. J Clin Neurosci, 47:228-233.

23.

Najafi

, Farhood

, Mortezaee

(2019) Extracellular matrix (ECM) stiffness and degradation as cancer drivers. J Cell Biochem, 120:2782-2790.

24.

Olofsson

, Pajusola

, Kaipainen

, et al. (1996) Vascular endothelial growth factor B, a novel growth factor for endothelial cells. Proc Natl Acad Sci U S A, 93:2576-2581.

25.

Ostrom

, Gittleman

, Truitt

, et al. (2018) CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2011-2015. Neurooncol, 20,(suppl_4)::iv1-iv86.

26.

Owusu-Ansah

, Song

, Chen

, et al. (2019) COL6A1 promotes metastasis and predicts poor prognosis in patients with pancreatic cancer. Int J Oncol, 55:391-404.

27.

Paul-Samojedny

, Pudełko

, Suchanek-Raif

, et al. (2015) Knockdown of the AKT3 (PKBγ), PI3KCA, and VEGFR2 genes by RNA interference suppresses glioblastoma multiforme T98G cells invasiveness in vitro. Tumour Biol, 36:3263-3277.

28.

Ping

, Yao

, Jiang

, et al. (2011) The chemokine CXCL12 and its receptor CXCR4 promote glioma stem cell-mediated VEGF production and tumour angiogenesis via PI3K/AKT signalling. J Pathol, 224:344-354.

29.

Plate

, Breier

, Weich

, Risau

(1992) Vascular endothelial growth factor is a potential tumour angiogenesis factor in human gliomas in vivo. Nature, 359:845-848.

30.

Presta

, Chen

, O'Connor

, et al. (1997) Humanization of an anti-vascular endothelial growth factor monoclonal antibody for the therapy of solid tumors and other disorders. Cancer Res, 57:4593-4599.

31.

Ramezani

, Vousooghi

, Kapourchali

, et al. (2017) Rolipram potentiates bevacizumab-induced cell death in human glioblastoma stem-like cells. Life Sci, 173:11-19.

32.

Samoto

, Ikezaki

, Ono

, et al. (1995) Expression of vascular endothelial growth factor and its possible relation with neovascularization in human brain tumors. Cancer Res, 55:1189-1193.

33.

Senger

, Brown

, Claffey

, Dvorak

(1994-1995) Vascular permeability factor, tumor angiogenesis and stroma generation. Invas Metastasis,, 14:385-394.

34.

Senger

, Galli

, Dvorak

, et al. (1983) Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science, 219:983-985.

35.

Shibuya

(1995) Role of VEGF-flt receptor system in normal and tumor angiogenesis. Adv Cancer Res, 67:281-316.

36.

Shibuya

, Claesson-Welsh

(2006) Signal transduction by VEGF receptors in regulation of angiogenesis and lymphangiogenesis. Exp Cell Res, 312:549-560.

37.

Zhang

, Ding

, Yu

, et al. (2012) Dexmedetomidine protects against oxygen-glucose deprivation-induced injury through the I2 imidazoline receptor-PI3K/AKT pathway in rat C6 glioma cells. J Pharm Pharmacol, 64:120-127.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.26 MB

1.05 MB