Chapter Six - Molecular determinants of the interaction between glioblastoma CD133+ cancer stem cells and the extracellular matrix

https://doi.org/10.1016/bs.irn.2020.03.005Get rights and content

Abstract

Glioblastoma multiforme (GBM) is the most common primary tumor of the human brain. It is characterized by invasive growth and strong resistance to treatment, and the median survival time of patients is 15 months. The invasive growth of this tumor type is associated with tumor cells with an aggressive phenotype, while its treatment resistance is attributed to cancer stem cells (CSCs). It remains unclear if CSCs have a more invasive nature than differentiated glioblastoma cells (DGCs), and what contribution CSCs make to the aggressive phenotype of GBM. Interaction with the extracellular matrix (ECM) is a key factor in the development of invasion. The aim of the present study was to compare the expression levels of signaling pathway proteins involved in interaction of receptors with the ECM in CSCs and DGCs. The U-87MG GBM cell line was used in the present study CSCs were extracted from gliomaspheres through magnetic-activated cell sorting based on the expression of cluster of differentiation 133 (CD133); CD133-negative DCGs were used as a control. HPLC and mass spectrometry were also used, and biological and molecular functions, signaling pathways and protein-protein interactions were analyzed using publicly available databases. Increased expression levels of the following 10 proteins involved in interaction with the ECM were identified in CSCs, compared with expression levels in DGCs: COL6A1, COL6A3, FN1, ITGA2, ITGA5, ITGAV, ITGB1, ITGB3, LAMB1 and LAMC1. The proteome of CSCs was observed to have > 2-fold higher expression of these key proteins, when compared with the DGC proteome. Increased expression levels of four proteins (FERMT2, LOXL2, HDAC2 and FBN1) involved in activating signaling in response to receptor interaction with the ECM was also observed, indicating that CSCs may have highly invasive nature. LOXL2 expression level was > 9-fold higher in CSCs compared to DGCs, suggesting that this protein may have potential as an marker for CSCs and as a target for this cell type in GBM.

Introduction

Glioblastoma multiforme (GBM) is one of the most aggressive types of human brain cancer. It is characterized by rapid invasive growth, significant brain infiltration and strong resistance to treatment. Despite advances in surgery, radical elimination of the tumor without causing severe and irreversible neurological damage to the patient remains a challenge (Costa, Lawson, Lelotte, et al., 2019). Therefore, the primary treatment type is high-dose radiation with multiple rounds of chemotherapy (Lukas, Wainwright, Ladomersky, et al., 2019). Prognosis for patients with this type of tumor is unfavorable, as the survival rate at 2 years following diagnosis is 27–43%, and the median survival time of patients with GBM is 15 months (Stupp & Ram, 2018). Due to their invasive ability, cancer cells that have deeply penetrated the brain tissue from the original lesion inevitably cause relapses, requiring new approaches and treatment protocols for patients with this disease.

The ability to penetrate the brain matter (Perrin, Samuel, Koszyca, et al., 2019) is due to the mesenchymal-like aggressive phenotype of cancer cells, and is influenced by coordinated intercellular interactions with the local microenvironment (Roos, Ding, Loftus, et al., 2017). The key properties of this phenotype include switching from adhesive E-cadherins to migratory N-cadherins, extending the range of integrin receptors on the cellular surface, developing the ability to produce extracellular matrix (ECM) components, synthesizing matrix metalloproteases (Bryukhovetskiy & Shevchenko, 2016).

Focal adhesion proteins are critical to underpinning the invasive potential of cancer cells with an aggressive phenotype (Bolteus, Berens, & Pilkington, 2001). GBM cells penetrate the brain tissue, adhere to ECM proteins and create cellular-matrix connections with structural and signaling functions that interact with the microenvironment (Huttenlocher & Horwitz, 2011). These connections between the ECM and the actin cytoskeleton of the cells also generate tension that is required for migration beyond the primary lesion. Inhibition of the interaction between cancer cells and the ECM is one of the aims of GBM treatment.

GBM relapse and progression is also attributed to cancer stem cells (CSCs) that have unique signaling and morphological properties (Singh, Clarke, Terasaki, et al., 2003), including an ability to initiate and support rapid tumor growth. CSCs occupy the dominant position in the GBM cellular hierarchy, and are detected at the anterior border of the tumor (Gimple, Bhargava, Dixit, et al., 2019), indicating that they are involved in invasive processes. However, it remains unclear if CSCs are more invasive than differentiated tumor cells, and if they contribute to the development of GBM with an aggressive phenotype.

The CD133 antigen is the most reliable marker for CSCs (Singh, Hawkins, Clarke, et al., 2004). This marker is introduced into GBM tissue by the symmetrical division of CSCs and by reprogramming of differentiated CD133-negative (CD133 −) tumor cells resulting in phenotypic plasticity (Li, Zhou, Xu, & Xiao, 2013) in the tumor cell population. The ability of to recruit normal neural CD133-positive (CD133 +) stem cells further complicates our understanding of its heterogeneity (Okawa, Gagrica, & Blin, 2017). The aim of the present study was, therefore, to compare the expression profiles of proteins involved in the interaction with the ECM, and signaling pathways in CD133 + CSCs and differentiated glioblastoma cells (DGCs). The present study may lead to the discovery of new molecular targets for regulating the invasive activity of CSCs.

Section snippets

Human GBM cells

For the present study, the U-87MG GBM cell line was obtained from the American Type Culture Collection (cat no. HTB-14™). This cell line is not the original U-87 line established at the University of Uppsala, but derived from a human glioblastoma of unknown origin (Allen, Bjerke, Edlund, et al., 2016). However, as demonstrated in our previous study, the stimulation of GBM U-87MG cells with transforming growth factor (TGF)-β1 led to a significant increase in the expression levels of proteins

CSCs have altered proteomes compared to DGCs

Proteome analysis identified 1990 unique proteins. A total of 1891 proteins were identified in the CSC sample, and 1748 proteins were found in the sample of differentiated GBM cells (DGCs). Identified proteins showed a high percentage of overlap between the two cell populations: 1649 proteins were present in all cell lysates; 242 proteins were found only in CSCs; and 99 proteins were observed only in DGCs. Among the discovered proteins, 589 had significantly different expression levels in CSCs (

Discussion

The present study investigated proteins that are upregulated in CSCs, and that serve a crucial role in their interaction with the ECM. Activation of hyaluronic acid receptors, CD44 (Table 1) and hyaluronan-mediated motility receptor (HMMR) in CSCs has been found to be a key factor for invasion (Mooney, Choy, Sidhu, et al., 2016). CD44 glycoprotein is described as a CSC marker in various cancer types, including GBM, and it is frequently used as an indicator of tumor cells with aggressive

Funding

This study was funded by the Ministry of Science and Higher Education of Russia (Contract no. 14.584.21.0027 ID:RFMEFI58417X0027).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

Ethics approval and consent to participate.

Informed consent was obtained from all individual participants included in the study and all procedures performed in studies involving human participants were in accordance with the ethical standards of the Far Eastern Federal University and University of Uppsala.

Authors' contributions

V.S. and N.A. prepared and analyzed the samples, as well as performed the cell lysis, chromatography and mass spectrometry, and contributed to the bioinformatics analysis. S.A. provided and performed the statistical analysis and was responsible for the mathematical process of the results. Y.K. and H.S. discussed, analyzed and interpreted the results of the study, and also worked on the manuscript. I.B. wrote the manuscript, proposed the study idea, designed the study, offered support with the

Conflict of interest

The authors declare there are no conflicts of interests.

References (51)

  • R. Bauer et al.

    Inhibition of collagen XVI expression reduces glioma cell invasiveness

    Cellular Physiology and Biochemistry

    (2011)
  • A.J. Bolteus et al.

    Migration and invasion in brain neoplasms

    Current Neurology and Neuroscience Reports

    (2001)
  • D.V. Brown et al.

    Coexpression analysis of CD133 and CD44 identifies proneural and mesenchymal subtypes of glioblastoma multiforme

    Oncotarget

    (2015)
  • D.V. Brown et al.

    Expression of CD133 and CD44 in glioblastoma stem cells correlates with cell proliferation, phenotype stability and intra-tumor heterogeneity

    PLoS One

    (2017)
  • I. Bryukhovetskiy et al.

    Molecular mechanisms of the effect of TGF-β1 on U87 human glioblastoma cells

    Oncology Letters

    (2016)
  • A. Bryukhovetskiy et al.

    To the novel paradigm of proteome-based cell therapy of tumors: Through comparative proteome mapping of tumor stem cells and tissue-specific stem cells of humans

    Cell Transplantation

    (2014)
  • Z. Cierna et al.

    Fibrillin-1 (FBN-1) a new marker of germ cell neoplasia in situ

    BMC Cancer

    (2016)
  • E. Costa et al.

    Long-term survival after glioblastoma resection: Hope despite poor prognosis factors

    Journal of Neurosurgical Sciences

    (2019)
  • E.H. Danen

    Integrins: Regulators of tissue function and cancer progression

    Current Pharmaceutical Design

    (2005)
  • X.G. Du et al.

    Clinical relevance of lysyl oxidase-like 2 and functional mechanisms in glioma

    Oncotargets and Therapy

    (2018)
  • R.C. Gimple et al.

    Glioblastoma stem cells: Lessons from the tumor hierarchy in a lethal cancer

    Genes & Development

    (2019)
  • J.S. Hartheimer et al.

    Targeting hyaluronan interactions for glioblastoma stem cell therapy

    Cancer Microenvironment

    (2019)
  • J.D. Humphries et al.

    Integrin ligands at a glance

    Journal of Cell Science

    (2006)
  • A. Huttenlocher et al.

    Integrins in cell migration

    Cold Spring Harbor Perspectives in Biology

    (2011)
  • E. Kawamura et al.

    Fermitin family homolog-2 (FERMT2) is highly expressed in human placental villi and modulates trophoblast invasion

    BMC Developmental Biology

    (2018)
  • Cited by (11)

    View all citing articles on Scopus
    View full text