We described a promising scheme for discovery of potential protein biomarkers using various in silico tools along with evidence-based literature search. This strategy mined nine proteins as potential circulatory biomarkers for GBM. To evaluate their biological role and impact as credible potential biomarkers, these proteins were assessed through reported experimental confirmations and literature study.
The study mined contactin-2 (CNTN2) as a secretory protein which is a neural-specific glycoprotein found in exosomes and involved in neurodevelopment (Marell et al. 2019). Interestingly, CNTN2 RNAi inhibits the proliferation of U87-Glioma stem cells and substantially reduces the expression of epidermal growth factor receptor and HES1, a Notch-targeting gene (Guo et al. 2017). CNTN2 is also associated with maximum level in the oligodendrogliomas that is also considerably increased in high-grade glioma tissues in contrast to low grade glioma tissues (Yu and Fu 2015). Similarly, human glioma cell lines (LN-18, U-118 MG, SW1783 and SW1088) showed upregulation of CNTN2 protein as compared to normal human astrocytes (NHA) (Yan and Jiang 2016). Being a crucial activator of RTK/ Ras/MAPK signaling pathway in glioma cells, CNTN2 increase the activity of RTKs that give rise to signals through Ras/MAPK cassette, resulting in GBM and other cancers (Wan et al. 2006, Carracedo et al. 2008). Inhibition of the CNTN2 and subsequently of RTK/ Ras/MAPK signaling pathway may repress the progression of GBM. Since overexpression of CNTN2 protein has been confirmed in GBM cells and tissues but no evidence in secretome of GBM, therefore the current in silico analysis, suggests CNTN2 as a potential circulating protein indicative of GBM.
Another physiologically important protein, LY6H is also predicted as a secretory protein. It belongs to LY6 gene family of proteins located on human chromosomes 6, 8, 11, and 19. The LY6 family members such as LY6D, LY6E, LY6K and LY6H are positively correlated with poor outcome on patient survival in multiple cancer types (Luo et al. 2016). An over expression of LY6H mRNA has also been detected in GBM tissues compared to normal brain (Lee et al. 2006). A direct interaction of LY6H with prostate stem cell antigen (PSCA), a unique biomarker for grade III-IV gliomas (Geiger et al. 2011), is predicted in our study (Fig. 3A), which indicates the involvement of LY6H in mediating the cellular mechanisms leading to progression of gliomas. In conclusion, the deregulated gene expression of LY6H at mRNA level, its oncogenic potential, secretory nature and interaction with GBM biomarker i.e. PSCA, encourages to identify the protein expression of LY6H in the GBM cells secretome and body fluids of GBM patients, for its confirmation as a viable circulating diagnostic protein.
Opioid binding protein/cell adhesion molecule-like (OPCML) is a tumor suppressor gene, reported in multiple cancers (Lian et al. 2021). CSF proteome profiling of glioma patients observed down-regulation of OPCML (Saratsis et al. 2012), while its mRNA expression was also considerably low in GBM cell lines (U87MG) and tissues (Reed et al. 2007). PPI interaction analysis depicts a strong interaction of OPCML with an oncogene, Ras-related protein R-Ras2 (R-RAS2) (Figure. 3B). RRAS2 mRNA and protein is overexpressed in a multiple human CNS tumors, including GBM, suggesting it an initial step in neural cell transformation (Gutierrez-Erlandsson et al. 2013). Since a meager data is available on the aberrant protein expression of OPCML in CSF of glioma patients, therefore we suggest further investigations to be carried out on the body fluids of GBM patients and cells secretome to confirm OPCML as a circulating candidate protein signature.
Serpin peptidase inhibitor-clade I (neuroserpin)-member 1 (SERPINI1) is involved in the development and normal function of the nervous system (Han et al. 2021). SERPINI1 promote the mechanism for the initiation of brain metastasis in lung and breast cancers, by guarding tumor cells from death signals and promoting vascular cooption (Valiente et al. 2014). Other members of this family are also involved in GBM pathogenesis; where SERPINE1 showed decreased expression in GBM cells (Tarassishin et al. 2014), whereas SERPINA4, SERPINA10, SERPINC1 and SERPINF1 proteins exhibit significant downregulation in plasma of GBM patients (Gautam et al. 2012; Gupta et al. 2013). A strong interaction is evident for SERPINI1 with myc proto-oncogene protein (MYC) and protein max (Max), two prominent oncoproteins (Fig. 3C) indicating a suggestive role of SERPINI1 in tumor progression through a strong interaction mechanism with other oncoproteins.
A neuronal secreted protein, Leucine-rich glioma inactivated-1 (LGI1) is also identified as a plausible secretory protein for GBM. It was initially isolated through its link with a chromosome translocation breakpoint in GBM cell line, since then it has proved to have a controlling function in the suppression of cell migration and invasion of the cells by down regulating the MEK/ERK pathway (Chernova et al. 1998). LGI1 gene is significantly downregulated in GBM tissue and cells (Chernova et al. 1998; Gu et al. 2002; Besleaga et al. 2003) while reduced level of LGI1 transcript is also evident in both primary GBM tissues and GBM cells. LGI1 utilizes the phosphatidylinositol 3-kinase/ERK pathway and suppresses the production of MMP1/3; therefore loss of LGI1 expression results in the invasive phenotype in GBM (Besleaga et al. 2003). The predicted protein interacting partners of LGI1 i.e. disintegrin and metalloproteinase domain-containing protein 11 (ADAM11) and disintegrin and metalloproteinase domain-containing protein 22 (ADAM22) (Fig. 3D), also showed differential expression in gliomas. The ADAM22 that regulates the growth inhibition through integrin-dependent pathway, is significantly up-regulated in the high-grade gliomas, whereas ADAM11 is identified in both low and high-grade gliomas (D'Abaco et al. 2006). Till date, LGI1 has been studied only at transcript level in GBM, no proteomic studies has been carried out regarding its status in GBM. Therefore, a further analysis of GBM secretome proteome and other biofluids will be helpful for its confirmation as a circulating protein biomarker.
Slit homolog 1 protein (SLIT1) belongs to SLIT family of proteins. These proteins are extremely conserved glycoproteins that are secreted and the central ligands for roundabout receptors (ROBOs) (Kidd et al. 1999). The SLIT/ROBO pathway has a substantial role in cell-signaling, cell motility, angiogenesis and cell migration. Moreover, SLIT have significant role in oncogenesis, cancer development and metastasis (Mehlen et al. 2011; Gara et al. 2015). It is also linked with development and advancement of numerous cancers including lung, brain, colon and breast (Dallol et al. 2002). Rossi et al. observed significant down-regulation of SLIT1 and LGI1 gene expression in GBM cell lines (U87) (Rossi et al. 2005). The unusual gene expression of SLIT1 in GBM cell lines and its secretory nature provides ample evidence to further validate and identify this protein in the secretome of GBM cell lines and in the biofluids of GBM patients.
NPTX1 is a potential plasma biomarker of synaptic dysfunction, and regulates mitochondrial dynamics and trafficking during apoptotic neurodegeneration (Clayton et al. 2012; Ma et al. 2018) It is associated with various types of cancers including hepatocellular carcinoma where it inhibits the proliferation and promotes apoptosis through AKT-mediated signaling mechanism (Zhao et al. 2019). It is also suggested that the specific small interfering RNA like miR-128-3p act via NPTX1/IRS-1/PI3K/AKT signaling pathway to suppress glioma (Huo et al. 2019). Moreover, up regulation of NPTX1 gene in invasive GBM cells while downregulation of neuronal pentraxin receptor (NPTXR) in CSF of glioma provides interesting regulatory mechanisms of NPTX1 regulation in gkiomas (Hoelzinger et al. 2005). Likewise, an insilio identification of 240 upregulated genes in GBM brain tissues including NPTX1, further support our findings for NPTX1 as possible circulating brain specific biomarker for GBM (Li et al. 2019). The PPI network analysis also shows a strong interaction of NPTX1 with an oncogene, reticulocalbin-2 (RCN2) (Fig. 3E), which has been previously identified in global proteome analysis of GBM stem cells (Kozuka-Hata et al. 2012).
GDF1 is a secreted glycoprotein that belongs to the transforming growth factor-β (TGF-β) superfamily (Bao et al. 2015). In brain, TGF-βs are significantly associated with the development of cancer (Luwor et al. 2008). Yang et al reported that epigenetic silencing of GDF1 nullify the growth-inhibitory SMAD signaling and renders proliferation advantage to gastric epithelial cells during carcinogenesis (Yang et al. 2016). GDF15, another member of the GDFs subfamily is highly enriched in GBM tissues and upregulated in CSF of GBM patients (Shnaper et al. 2009). It substantially regulates the expression of programmed death-ligand 1 (PD-L1), a co inhibitory ligand predominantly expressed by tumor cells, in GBMs (Peng et al. 2019). Functional association predicted through STRING indicated plausible mechanistic regulation of GDF1, SMAD2 and SMAD4 (Fig. 3F). These two interacting partners of GDF1 are also related to gliomas where the protein expression level of SMAD2 was downregulated in glioma cell lines compared to normal astrocytes (Zhang et al. 2006). Similarly, reduced gene and protein expression of SMAD4 is also associated with adverse outcome in GBM patients (He et al. 2011). Moreover, strong upregulation of GDF1 was observed in GBM and astrocytoma via network-based analysis (Feng et al. 2012). Collectively the presence of various other members of GDFs in the body fluids of GBM patients, indicates the role of GDF1 in cancer cell survival and provides the basis to investigate GDF1 protein in body fluids of GBM patients that will further confirm it as a circulating protein biomarker, helpful for the diagnosis and prognosis of GBM.
The cellular repressor of E1A-stimulated genes 2 (CREG2) is a member of CREG family of proteins and a secreted glycoprotein expressed specifically in the brain exhibiting varied function (Kunita et al. 2002). Although the role of CREG in vascular diseases is highly established (Li et al. 2017), it is also involved in the differentiation of neurons in human teratocarcinoma NTERA-2 cells (Veal et al. 2000; Han et al. 2010) and showed an increased expression in malignant gastric cancer indicating its significance in gastric cell development (Xu et al. 2011). CREG2 gene is also differentially expressed in breast and kidney cancer–derived circulating endothelial colony forming cells (ECFCs) (Moccia et al. 2017), demonstrating its oncogenic potential. The present in silico analysis suggested a central role of CREG in oncogenesis however further investigations are warranted.