The glioblastoma multiforme (GBM) is the most malignant form of brain tumors and one of the most aggressive types of cancer [1
]. As the median survival time for patients with glioblastoma ranges from 12 to 15 months, and the 2-year overall survival rate averages only 26%, early diagnosis, and treatment is important to improve outcome [2
]. Despite the efforts of the medical and scientific community, a reliable non-invasive method for early diagnosis of glioblastoma has not been developed up to date. Computed tomography and brain biopsy are still the main methods for the diagnosis of this disease. Especially, the need for biomarkers is becoming more apparent as the clinical management of GBM moves towards personalized medicine. There is a hope that EV secreted by glioblastoma cells could be a source of biomarkers detectable in the peripheral circulation [3
]. EV are the particles released either from the cell surface–microvesicles having a size of 100–1000 nm or from an endosomal route–exosomes having a size of 40–150 nm [4
]. EV contain an array of different types of macromolecules–proteins, nucleic acids, lipids, and metabolites–which could be considered as the candidates for biomarkers [5
]. Available information indicates that more invasive GBM cells secrete higher amounts of exosomes. This strategy allows tumors to hijack their microenvironment and modulate anti-tumor immunity [3
]. As EV can cross the hematoencephalic barrier and enter the peripheral circulation, macromolecular profiling of GBM-derived EV could be a noninvasive way of assessing tumors in situ. Extensive studies of RNA content in glioma-derived EV were performed [7
]. In several studies, an array of protein cargo of GBM-derived EV has been identified, however, a slender picture of the EV protein profiles is not yet available [10
]. Previously, we have performed a comparative analysis of proteins from glioblastoma and normal cells trying to select the potential cancer biomarkers. Based on this analysis, a list of proteins overexpressed in glioblastoma cells as “potential glioblastoma biomarkers” was generated [13
]. We have constructed a database of proteoforms for glioblastoma as well as for normal fibroblasts [14
]. Here, we provide a graphical protein pattern (virtual 2DE) of exosomes secreted from five glioblastoma cell lines using high-resolution mass spectrometry (MS) and discuss the presence and the functional roles of the proteins from glioblastoma cells in exosomes. Additionally, we show that many proteins with enhanced expression levels in glioblastoma cells are presented in the exosome pattern. The gene ontology (GO) analysis of the exosome protein pattern was also performed.
Exosomes promote tumor formation and development by mediating the intercellular transport of miRNAs, mRNA, and proteins, and there is growing interest in their use as biomarkers for disease diagnosis and the monitoring of disease recurrence [22
]. Exosomes secreted by cancer cells stimulate the malignancy of gliomas, including suppressing the immune response or affecting the tumor microenvironment [23
]. Exosomes released by glioblastoma cells may cross the hematoencephalic barrier and carry molecular cargo into the peripheral circulation. Therefore, the determination of the molecular composition of exosomes secreted by glioblastoma cells seems to be an extremely promising direction for the development of non-invasive methods for diagnosing this disease [26
]. In this study, we performed a proteomic analysis of vesicles isolated from the conditioned medium of five glioma cell lines. EV from glioma cells have been isolated by ultracentrifugation and characterized in size, quantity, and morphology by commonly used in nanoparticle size and shape interpretation methods, including NTA, DLS, atomic force, and cryo-EM microscopy. According to the International Society for Extracellular Vesicles (ISEV) guidelines, isolated EV are considered to be exosomes if they are in the size range of 30–200 nm, have a typical spherical form, contain a bilayer membrane, and are enriched with exosomal markers [27
]. The results obtained from DLS or NTA demonstrate the presence of particles with a characteristic exosome size in the final samples [4
]. Cytometric and MS detection of the CD9 or CD81 markers on the surface of the particles confirms the exosomal nature of the vesicles. Moreover, most particles detected by both cryo-EM and AFM have a vesicular structure and match the dimensional characteristics of exosomes. Summarizing the data, it can be stated that most vesicles isolated from CCM of glioma cells can be defined as exosomes.
A set of 133 proteins that are common in all the five samples of exosomes secreted by GBM cell lines was detected. Correlating each protein with its main function, whole groups of proteins were identified that were associated both with processes occurring in tumor cells and with possible functions of exosomes secreted by them. First, we drew attention to proteins that may be related to the formation or traffic of exosomes. These included structural vesicular proteins, for example, clathrin (CLTC), chaperones involved in the assembly of vesicles, cytoskeleton components, small G-proteins and other proteins involved in the movement of vesicles and their fusion with the membrane. Some of them are of particular interest—for example, vimentin (VIME) involved in the positioning of organelles is a marker of the epithelial-mesenchymal transition of tumor cells [28
]. Programmed cell death 6-interacting protein, PDCD6IP, interacts with endophilins that regulate the shape of the membrane during endocytosis and protects cells from Ca2 + -dependent apoptosis [29
]. It was shown that increased expression of this gene and endophilins leads to vacuolization of cells, protecting them from programmed death [30
It was surprising that the whole set of the nucleic acid interacting proteins, including several chromatin and ribosomal proteins, fell into the set of the proteins common to exosomes from glioma cell lines. As part of exosomes, these proteins can be absorbed by macrophages. Presented to lymphocytes, fragments of these proteins can participate in the formation of immunological tolerance to tumor cells. Among the proteins that regulate the behavior of vesicles, we identified a vasolin-containing protein, TERA(VCP), ATPase, which is needed to remove improperly assembled or spent peptides from ribosomes, the endoplasmic reticulum, mitochondria, and chromatin. TERA(VCP)is involved in membrane fusion and the regulation of vesicle traffic from the plasma membrane to the lysosome. We assume that a sorting mechanism is associated with the TERA(VCP) function, which explains the presence of the above proteins in exosomes.
Among the proteins identified in the five exosome samples, the cell adhesion proteins include integrins, annexins, tetraspanins (including exosome markers CD9 and CD81), proteases, and protease inhibitors. These proteins can affect the microenvironment of tumor cells, promoting processes such as invasion, angiogenesis, and the modulation of the immune response (e.g., vitronectin, VTN, which inhibits the complement system). Among these proteins, ANXA2, CD44, and tenascin-C (TENA) were identified. Annexin A2 (ANXA2) plays a significant role in the invasion, metastasis, angiogenesis, proliferation, actin polymerization, and endosomal sorting [32
]. CD44 is a receptor for hyaluronic acid and can also interact with other ligands, such as osteopontin, collagens, and matrix metalloproteinases. It is implicated in various malignant processes including cell motility, tumor growth, and angiogenesis, and has been suggested as a cancer stem cell marker [33
]. Tenascin-C (TENA) is a hexameric glycoprotein expressed by the neural stem and progenitor cells and endothelium during embryogenesis. It is located in the extracellular matrix and influences cell migration, inhibits focal contact formation, and promotes angiogenesis. TNC is overexpressed and correlated with tumor malignancy observed in many cancer types and is associated with glioma outcomes [34
]. Previously, we have shown the increased levels of stem cell genes, including SOX2
, and GLI1
in these glioma cell lines [35
] that coincide with the presence of the CD44 and TNC in all the exosome samples from CCM.
Proteins involved in metabolic processes are mainly represented by groups of enzymes, ion channels, amino acid transporters, and subunits of G-proteins that regulate adenylate cyclase and phospholipase C. The set of enzymes contained in exosomes of glioma cells well reflects the characteristics of the metabolism of tumor cells, most of which catalyze various stages of glycolysis and are involved in the Warburg effect. So, among the proteins found in all the exosome samples, we identified lactate dehydrogenase B (LDHB), which converts pyruvate to lactic acid, and ATP, as well as phosphoglycerate kinase 1 (PGK1), which not only is responsible for the synthesis of 3-phosphoglycerate, but also mediates the inhibition of citric acid synthesis [37
]. It was shown that glioma cell-derived exosomes activate glycolysis in human bone marrow mesenchymal stem cells, resulting in their tumor-like phenotype transformation [25
]. Interestingly, the glioma exosomes contained ATP-citrate synthase (ACLY), which converts citric acid into acetyl coenzyme A, and fatty acid synthase (FASN), which synthesizes palmitate from acetyl coenzyme A in the presence of NADPH. This indicates the possibility of a transition between different branches of energy metabolism, which probably creates an additional opportunity for tumor cells to adapt to adverse conditions.
Additionally, we draw attention to two proteins that can participate in the development of glioblastomas and their resistance to treatment. Firstly, it is the DNA-dependent protein kinase catalytic subunit (PRKDC), which is involved in the repair of double-strand breaks with a nonhomologous DNA end joining (NHEJ) contributing to radioresistance [38
]. It has been shown that the degradation of this protein is regulated by the vasolin-containing protein, VCP, another protein found in the exosome samples [39
]. Secondly, it is a major vault protein (MVP), which is associated with nuclear pores, regulates nuclear-cytoplasmic transport and determines resistance to certain drugs [40
]. It should be noted that its gene is a target of the transcription factor GLI1 [42
], which according to our previous data is abnormally active in these glioma lines [36
An important conclusion comes after a comparative analysis of exosomal proteins and “potential glioblastoma biomarkers”. Based on our previous analysis of the proteomes of the glioma and normal cells, a list of proteins overexpressed in glioma cell lines was generated [13
]. It is important that 13 of them (ANXA1, ANXA2, ENOA, G3P, HS90B, KPYM, PRDX1, TPIS, TERA, VIME, 1433E, COF1, NPM) were detected by MS in glioblastoma exosome samples (Figure 5
and Supplementary Table S1
). We can call these proteins “exosomal glioblastoma biomarkers”. The data obtained in the present study allows us to expand the list of exosomal potential biomarkers of gliomagenesis by adding CD44 and Tenascin-C (TNC). Both these proteins are among the 133 common exosomal proteins identified in our study and are associated with malignancy of gliomas [33
]. Surprisingly, such proteins as PCNA and p53, which are at the top of the list of “potential glioblastoma biomarkers” [13
], were not detected by MS in exosome preparations. These proteins had been previously detected in glioblastoma exosomes by Western blotting [19
]. These proteins were detected in exosomes from ovarian cancer cells and are presented in Vesiclepedia http://microvesicles.org/index.html#
. We can conclude that our list of exosomal proteins of glioblastoma detected by MS is not complete, and immunological methods of detection can sometimes be more sensitive than MS.
In summary, we can conclude that the set of 133 proteins common to glioblastoma exosomes detected in our study is only partial but represents the main functional aspects of exosome cargo proteins and could be representative for exosomes functions. Our study suggests the use of specific brain tumor exosomes and several protein markers to help to create non-invasive techniques to diagnose disease. According to our data, exosomes contain reliable protein biomarkers of glioblastoma.