Tissue Proteome Analysis of Different Grades of Human Gliomas Provides Major Cues for Glioma Pathogenesis

Abstract

Gliomas are heterogeneous and most commonly occurring brain tumors. Blood–brain barrier restricts the entry of brain tumor proteins into blood stream thus limiting the usage of serum or plasma for proteomic analysis. Our study aimed at understanding the molecular basis of aggressiveness of various grades of brain tumors using isobaric tagging for relative and absolute quantification (iTRAQ) based mass spectrometry. Tissue proteomic analysis of various grades of gliomas was performed using four-plex iTRAQ. We labeled five sets (each set consists of control, grade-II, III, and IV tumor samples) of individual glioma patients using iTRAQ reagents. Significantly altered proteins were subjected to bioinformatics analysis using Database for Annotation, Visualization and Integrated Discovery (DAVID). Various metabolic pathways like glycolysis, TCA-cycle, electron transport chain, lactate metabolism, and blood coagulation pathways were majorly observed to be perturbed in gliomas. Most of the identified proteins involved in redox reactions, protein folding, pre-messenger RNA (mRNA) processing, antiapoptosis, and blood coagulation were found to be upregulated in gliomas. Transcriptomics data of glioblastoma multiforme (GBM), low-grade gliomas (LGGs), and controls were downloaded from The Cancer Genome Atlas (TCGA) data portal and further analyzed using BRB-Array tools. Expression levels of a few significantly altered proteins like lactate dehydrogenase, alpha-1 antitrypsin, fibrinogen alpha chain, nucleophosmin, annexin A5, thioredoxin, ferritin light chain, thymosin beta-4-like protein 3, superoxide dismutase-2, and peroxiredoxin-1 and 6 showed a positive correlation with increasing grade of gliomas thereby offering an insight into molecular basis behind their aggressive nature. Several proteins identified in different grades of gliomas are potential grade-specific markers, and perturbed pathways provide comprehensive overview of molecular cues involved in glioma pathogenesis.

Introduction

Gliomas are the most common type of central nervous system (CNS) cancers (Shankarkumar and Sridharan, 2011), which are classified into four different grades by WHO, based on their histological and microscopic characteristics. Low-grade gliomas (LGGs; Grade-I and II) exhibit low proliferation rate, while high-grade gliomas (Grade-III and IV) exhibit high proliferation rate. Grade-II gliomas are infiltrative and often recur, while grade-III tumors show nuclear atypia and grade-IV tumors are most aggressive with nuclear atypia, necrosis, and high angiogenesis rate (Louis et al., 2007). The average survival period decreases with increase in grade of the tumors (Cooper et al., 2010).

Cancers are known to carry significant alterations at the molecular level (Calistri et al., 2004), and a number of studies done at the genomic level in the field of glioma (Dunn et al., 2012) indicate that the same applies even in case of glial tumors. Mutations like amplification of EGFR and loss of CDKN2A and PTEN genes are commonly observed in the high-grade gliomas, while mutations in the IDH-1 and IDH-2 genes are common features of LGGs and secondary glioblastomas (Borodovsky et al., 2012; Ohka et al., 2012). These mutations lead to alterations in various signaling pathways like PI3K/Akt/mTOR, PTEN, sonic hedgehog, and notch pathways. However, alterations in posttranslational modifications, which play major role in regulating various metabolic pathways, cannot be detected/identified at the genomic level, making the use of proteomics for better understanding of cancers inevitable (Hanash and Beretta, 2002).

Biological fluids like serum/plasma, cerebrospinal fluid, and urine are good sources for the identification of biomarkers for various diseases (Magni et al., 2010). However, in case of brain tumors, presence of the blood–brain barrier limits the leakage of tumor related proteins into the blood stream, except in high-grade brain tumors where the blood–brain barrier breaches (Kalinina et al., 2011).

Our study aimed at identifying proteomic alterations across different grades of gliomas to understand the pathobiology of the glial tumors and molecular signatures responsible for aggressive nature of high-grade gliomas.

Materials and Methods

Clinical samples

All the proteomic experiments were performed on human biopsy samples obtained after approval from Institutional Review Board of Advanced Centre for Treatment Research and Education in Cancer (ACTREC)—Tata Memorial Hospital (TMH). After obtaining consent from preinformed patients, five tissues each from Grade II, Grade III, and Grade IV were surgically resected. Peritumoral tissues (n = 5), confirmed to be near normal using Immunohistochemistry, were used as control samples. The tissues were snap frozen in liquid nitrogen until further use for experiments. For clinical details refer Supplementary Table S1.

Protein extraction from different grades of glioma tissue

Glioma tissue protein extraction was carried out using TRIzol extraction method (Reddy et al., 2013); the tissue proteins were quantified using 2D-Quant Kit (GE Healthcare) and stored at −20°C until buffer exchange was performed. Detailed methodology is provided in Supplementary Materials and Methods.

Buffer exchange, in-solution digestion, and isobaric tagging for relative and absolute quantification labeling

To label the proteins with isobaric tagging for relative and absolute quantification (iTRAQ) reagents and to perform liquid chromatography–mass spectrometry (LC-MS), the protein samples in rehydration buffer were made LC-MS compatible by exchanging them with 0.5 M TEAB buffer (Triethylammonium bicarbonate buffer) using 3 kDa molecular weight cutoff (MWCO) filters. Fifty micrograms of buffer exchanged protein samples were subjected to denaturation, reduction, and alkylation followed by digestion using trypsin. The trypsin digested control, grade-II, III, and IV tissue protein samples were labeled with iTRAQ reagent 114, 115, 116, and 117, respectively. Labeling reaction was quenched by adding Milli-Q water, and the labeled peptides were pooled.

Offgel fractionation and LC-MS/MS of the labeled peptides

To reduce the complexity of the labeled peptide samples, offgel fractionation was performed using Agilent 3100 offgel fractionator. IPG strips of length 24 cm, pH 3–10 were used, and the peptides were separated based on their pI. Offgel fractionated peptides were subjected to LC-MS/MS in an Agilent 6550 iFunnel Q-TOF equipped with Chip cube and interfaced with nanoflow LC system (Agilent Technologies), and data were analyzed using Spectrum Mill software (Agilent Technologies). For more details refer Supplementary Materials and Methods.

Bioinformatics analysis to identify pathways perturbed in gliomas

Significantly altered proteins were subjected to bioinformatics analysis using Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.7 online tool, which revealed the pathways and biological processes in which the significantly altered proteins are involved (Huang da et al., 2009). Statistical analysis was performed using MetaboAnalyst 3.0 online tool for discriminating different grades of gliomas and control samples (Xia et al., 2015).

Messenger RNA (mRNA) expression data for glioblastoma multiforme (GBM), LGGs, and control samples were downloaded from The Cancer Genome Atlas (TCGA) data portal (https://gdc-portal.nci.nih.gov), and the mRNA expression data were analyzed using BRB-Array tools (http://brb.nci.nih.gov/BRB-ArrayTools) (Simon et al., 2007). Using class comparison module of BRB-Array tools mRNA expression of LGGs and GBMs was compared with control samples. Furthermore, the mRNA expression data were correlated with protein expression data obtained from our quantitative mass spectrometry studies.

Cell line details

Human glioblastoma cell line LN18 was purchased from cell repository, National Centre for Cell Science (NCCS, Pune, India). The cells were cultured in sterile Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% of fetal bovine serum (FBS), 100 U of Penicillin, and 100 μg/mL of Streptomycin. Cell lines were allowed to grow at 37°C and 5% CO₂ in an incubator.

Transfection

The human genes CRYAB and SNCA were subcloned into mammalian expression vector pcDNA3-EGFP, and the gene constructs were transfected into LN18 cells. Transfection was performed using BBS-CaCl2 method (Chen and Okayama, 1987). After 48 h posttransfection, cells were observed under fluorescence microscope for the positively transfected cells. For more details refer Supplementary Materials and Methods.

Proliferation assay

Using MTT assay proliferation rates of transfected and control cell lines were measured (Xu et al., 2017). In a 96-well culture plate 1 × 10⁴ cells (per 100 μL of complete media), which were transiently transfected with CRYAB_pcDNA3-EGFP or SNCA_ pcDNA3-EGFP or pcDNA3-EGFP (control) constructs, were seeded into each well in triplicates and cultured for 5 days. Twenty microliters of MTT (stock concentration = 5 mg/mL) was added to each well and incubated for 4 h followed by addition of 150 μL of DMSO (Dimethyl sulfoxide) to dissolve the formazan crystals. Absorbance was recorded at 490 and 570 nm using enzyme-linked immunosorbent assay (ELISA) plate reader (Thermo Fisher). Net optical density (OD) at 490 nm was plotted against time (number of days). All the values were background corrected using the absorbance values obtained for blank. For more details refer Supplementary Materials and Methods.

Results

Comparison of different grades of glioma tumor tissue proteome using iTRAQ method

Tissue proteomic analysis of grade II, III, and IV gliomas with respect to peritumoral tissues (control) was performed using iTRAQ based mass spectrometry approach. Total 171 proteins were identified with at least two unique peptide signatures (false discovery rate [FDR] ≤1%) using LC-MS/MS. Proteins common in a minimum of three out of five sets were subjected to further bioinformatics analysis. Extensive literature survey for proteins identified in our study revealed the expression profiles to be in accordance with those reported by other groups (Supplementary Fig. S1). Partial list of a few differentially expressed proteins is provided in Table 1 and complete list in Supplementary Table S2. Peptide sequences and length of the peptides for the identified proteins are represented in Supplementary Table S3. Box plots of few differentially expressed proteins and representative mass spectra in various grades of gliomas are shown in Figure 1.

FIG. 1.

Tissue proteomic analysis of different grades of gliomas using iTRAQ method. (A) Box plots representing significantly altered proteins in different grades of gliomas and control tissues. (B) Mass spectra for two significantly altered proteins obtained using iTRAQ method in the individual patients of gliomas. iTRAQ, isobaric tagging for relative and absolute quantification.

Table 1.

A Few Significantly Altered Proteins in Various Grades of Gliomas as Obtained by Isobaric Tagging for Relative and Absolute Quantification Based Mass Spectrometry

S. No.	Entry name	Uniprot accession	Protein mol. wt. (in Da)	Avg. fold change (Gr2/Nor.)	Avg. fold change (Gr3/Nor.)	Avg. fold change (Gr4/Nor.)
1	Brain acid soluble protein 1	P80723	27305.9	0.67 ± 0.14	0.90 ± 0.52	0.30 ± 0.08
2	Endophilin-A1	Q99962	44169.1	0.64 ± 0.06	0.82 ± 0.25	0.36 ± 0.06
3	Creatine kinase U-type, mitochondrial	P12532	49885.9	0.54 ± 0.16	0.54 ± 0.14	0.37 ± 0.05
4	Tubulin polymerization-promoting protein	O94811	27612.5	0.64 ± 0.10	1.06 ± 0.44	0.40 ± 0.06
5	Synapsin-1	P17600	78750	0.56 ± 0.17	0.46 ± 0.09	0.41 ± 0.14
6	Tyrosine-protein phosphatase nonreceptor type substrate 1	P78324	59083.2	0.59 ± 0.13	0.60 ± 0.11	0.45 ± 0.08
7	Alpha-synuclein	P37840	16622.2	0.55 ± 0.08	0.68 ± 0.21	0.45 ± 0.11
8	High-mobility group protein B1	P09429	31263.1	1.60 ± 0.36	1.63 ± 0.29	2.03 ± 0.68
9	Ras-related protein Rab-7a	P51149	26225.2	1.99 ± 0.07	2.62 ± 0.57	2.07 ± 0.51
10	Heat shock protein HSP 90-alpha	P07900	96589.9	1.67 ± 0.30	1.90 ± 0.37	2.07 ± 0.67
11	Nucleophosmin	P06748	37502.7	1.57 ± 0.16	1.54 ± 0.33	2.09 ± 0.60
12	Peptidyl-prolyl cis-trans isomerase B	P23284	27547.2	1.35 ± 0.24	1.86 ± 0.39	2.13 ± 0.56
13	Vimentin	P08670	56879.2	2.15 ± 0.47	2.78 ± 0.68	2.14 ± 0.42
14	Acyl-CoA-binding protein	P07108	11918.2	3.54 ± 0.89	4.09 ± 1.25	2.15 ± 0.78
15	Histone H2B type 1-M	Q99879	16872.1	2.02 ± 0.55	2.08 ± 0.55	2.22 ± 0.88
16	Nestin	P48681	186253.1	1.98 ± 0.21	2.28 ± 0.55	2.24 ± 0.40
17	Stress-70 protein, mitochondrial	P38646	81460.5	2.32 ± 0.64	2.12 ± 0.77	2.25 ± 0.85
18	Superoxide dismutase [Mn], mitochondrial	P04179	27055.1	1.20 ± 0.17	1.29 ± 0.22	2.29 ± 0.33
19	Annexin A1	P04083	43554.7	2.37 ± 0.54	1.89 ± 0.29	2.35 ± 0.42
20	Thioredoxin	P10599	13752.4	1.33 ± 0.38	1.81 ± 0.71	2.37 ± 0.53
21	Tenascin	P24821	258318.1	2.27 ± 0.27	2.59 ± 0.80	2.51 ± 0.43
22	Elongation factor 1-alpha 1	Q5VTE0	57445.8	1.78 ± 0.50	1.81 ± 0.34	2.56 ± 0.91
23	Annexin A5	P08758	39164.6	2.47 ± 0.42	2.54 ± 0.41	2.59 ± 0.50
24	Profilin-1	P07737	16666.7	1.40 ± 0.15	1.88 ± 0.32	2.62 ± 0.71
25	Ig lambda-3 chain C regions	P0CG06	12705.8	1.23 ± 0.17	1.54 ± 0.33	2.64 ± 0.53
26	Histone H3.1t	Q16695	17496.1	2.25 ± 0.70	2.13 ± 0.52	2.69 ± 1.02
27	Keratin, type I cytoskeletal 10	P13645	62369.8	2.49 ± 0.90	1.54 ± 0.33	2.74 ± 0.74
28	Alpha-1 antitrypsin	P01009	51808.1	1.40 ± 0.22	1.49 ± 0.27	2.80 ± 0.61
29	Calnexin	P27824	75952	2.01 ± 0.37	1.46 ± 0.29	3.00 ± 0.72
30	Ig kappa chain C region	P01834	12933	1.39 ± 0.28	1.53 ± 0.33	3.05 ± 0.62
31	Thymosin beta-4-like protein 3	Q08EQ4	6368	1.30 ± 0.22	1.97 ± 0.28	3.30 ± 0.95
32	Nucleolin	P19338	89643.9	2.17 ± 0.32	2.37 ± 0.70	3.31 ± 1.05
33	l-lactate dehydrogenase A chain	P00338	41009.6	0.85 ± 0.06	1.25 ± 0.21	3.32 ± 1.03
34	Ig alpha-1 chain C region	P01876	40383.8	1.13 ± 0.14	1.41 ± 0.46	3.35 ± 1.43
35	Ig gamma-1 chain C region	P01857	40655	1.50 ± 0.28	1.85 ± 0.33	3.44 ± 0.81
36	Serotransferrin	P02787	87705.6	1.72 ± 0.62	1.32 ± 0.17	3.44 ± 1.37
37	Serum albumin	P02768	80011.6	1.98 ± 0.58	1.89 ± 0.73	3.65 ± 1.31
38	Protein S100-A6	P06703	11534	1.43 ± 0.30	2.25 ± 0.76	3.89 ± 1.96
39	Ig gamma-2 chain C region	P01859	39987.3	2.28 ± 0.64	2.14 ± 0.70	4.37 ± 1.07
40	Ferritin light chain	P02792	21806.2	1.83 ± 0.40	2.11 ± 0.12	4.85 ± 2.29
41	Annexin A2	P07355	43444.5	3.29 ± 1.14	1.91 ± 0.48	4.96 ± 1.41
42	Alpha-1-antichymotrypsin	P01011	51569.2	3.39 ± 2.12	1.85 ± 0.36	5.05 ± 2.70

Quantitative proteomic data obtained from the tissue proteomic analysis of different grades of gliomas were subjected to statistical analysis using MetaboAnalyst 3.0. Partial least square discriminant analysis yielded a panel of 15 proteins with variable importance of projection (VIP) >1.5 that could differentiate between control, grade-II, III, and IV gliomas on three-dimensional (3D) score plots (Fig. 2).

FIG. 2.

Statistical analysis of brain tissue proteins identified in the current study using MetaboAnalyst 3.0. (A) PLS-DA 3D-score plots showing separate clusters for control, grade-II, III, and IV gliomas. (B) Significantly altered proteins, which are important in discriminating the four different groups of samples identified by PLS-DA. 3D, three-dimensional.

Protein expression data from the current study were compared with TCGA transcriptomics data of LGGs and GBM. Total 44 common genes/proteins were identified in the comparison of TCGA LGG transcriptomics data with our iTRAQ data, of which 32 genes/proteins exhibited similar trends and the remaining 12 exhibited opposite trends (Supplementary Table S4). Comparison of TCGA GBM mRNA expression data with our iTRAQ data revealed 70 common genes/proteins of which 51 genes/proteins showed similar trends and the remaining 19 showed opposite trends (Supplementary Table S5).

Proliferation assay was performed using MTT reagent to check the effect of overexpression of CRYAB and SNCA genes on LN18 cells, where empty pcDNA3-EGFP transfected LN18 cells were used as control. Cells overexpressing SNCA exhibited lower proliferation rates than those cells which were overexpressing CRYAB or EGFP (control) (Fig. 3).

FIG. 3.

Proliferation assay in LN18 cells transfected with control and CRYAB and SNCA gene constructs. (A, B) Graphs representing proliferation rates of metabolically active transfected LN18 cells in two different passages.

Discussion

In this study we performed a comprehensive proteomic analysis of different grades of glioma patients using iTRAQ-based mass spectrometry approach. Significantly altered proteins were subjected to gene functional annotation analysis using DAVID. Comparison of tissue proteome of various grades of gliomas indicated a significant alteration of proteins involved in several metabolic pathways associated with neurodegenerative disorders, integration of energy metabolism, carbohydrate, amino acid, pyruvate metabolism, antigen processing, and presenting pathways (Fig. 4 and Supplementary Table S6). A few proteins associated with the cell cycle regulation and proliferation were found to be altered in gliomas. Pathway mapping of proteins identified from the iTRAQ based mass spectrometry experiments was performed using KEGG Mapper (Supplementary Fig. S2) (Kanehisa et al., 2017).

FIG. 4.

Pathway analysis of significantly altered proteins using DAVID, an online tool. Comprehensive pathway analysis using differentially expressed proteins in various grades of gliomas suggests major perturbation in metabolic pathways. The proteins were mapped in these pathways to identify their role in cell metabolism, particularly energy metabolism. Boxes on top of the proteins (left to right) indicate their expression in glioma grade-2 (left hand side box), grade-3 (middle box), and grade-4 (right hand side box) with respect to control brain tissue. DAVID, Database for Annotation, Visualization and Integrated Discovery.

Most cancer cells depend on glycolysis to meet their energy demands (Pelicano et al., 2006). In our study, we found differential expression of various proteins associated with energy metabolism such as triose phosphate isomerase, glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate mutase 1, phosphoglycerate kinase 1, and l-lactate dehydrogenase A and B chains. Our study revealed upregulation of Lactate dehydrogenase (LDH) in glial tumors. LDH converts pyruvate to lactate in anaerobic conditions. However, tumor tissues convert pyruvate to lactate despite the presence of sufficient amount of oxygen. This results in a decrease of pH in the tumor microenvironment, thereby leading to the death of normal neuronal cells. Gliomas exhibit increased expression of LDH, ultimately helping in tumor cell invasion by killing the nearby normal cells (Baumann et al., 2009), thereby suggesting its role in glioma pathogenesis.

Proteins associated with cell cycle regulation and cell proliferation like ubiquitin carboxyl-terminal hydrolase isozyme L1 (UCHL1), Peroxiredoxin-1, 60 kDa heat shock protein, mitochondrial (HSPD1), and 14-3-3 family of proteins were found to be differentially expressed in gliomas. UCHL1 is a deubiquitinating enzyme found to be differentially expressed in various cancers (Hibi et al., 1999; Takase et al., 20l3; Tezel et al., 2000), although its exact role in cancers remains controversial. UCHL1 gene knockdown studies in ovarian cancer cell lines showed increased cell proliferation and decreased apoptosis (Jin et al., 2013), whereas overexpression of UCHL1 in gastric cancer cells increases cell migration and invasion through the activation of Akt and Erk1/2 pathways (Gu et al., 2015). Increased expression of UCHL1 protein in breast cancer cell lines reduced the rate of cell proliferation by arresting the cells in G0/G1 phase of cell cycle (Xiang et al., 2012). Our results provide a plausible explanation for the high proliferative potential of gliomas arising from perturbation of key cell cycle regulatory proteins.

Proteins associated with coagulation pathway like alpha-1 antitrypsin (SERPINA 1) and fibrinogen alpha and gamma chains were found to be upregulated in gliomas. SERPINA 1 is a glycoprotein, which irreversibly inhibits the activity of serine proteases and is found to be expressed in various human tissues like liver, brain, kidney, testes, pancreas, and gastrointestinal tract (Carlson et al., 1988). Increased levels of SERPINA 1 reported earlier in case of brain tumors were found to correlate with greater invasiveness of the tumor cells. The increase in SERPINA 1 levels is probably in response to proteolysis of extracellular matrix, which further reduces the inflammatory response generated (Sawaya et al., 1987).

AAT showed positive correlation with the grade of the tumor in our study where expression was observed to increase with progression from low- to high-grade tumors. Fibrinogen forms fibrin clots at sites of injury and is an important component of blood coagulation pathway. Soluble fibrinogen interacts with integrins, nonintegrins, and other extracellular components and has been known to enhance the metastatic potential of tumor cells. Degradation products of fibrin increase angiogenesis, suppress immunological reactions, and increase cell proliferation (Hatzfeld et al., 1982; Plow and Edgington, 1986; Thompson et al., 1992). In our study, we observed a positive correlation between the expression of fibrinogen alpha and gamma chains during the progression from low- to high-grade tumors. Thus, perturbation of proteins involved in blood coagulation seems to be a key event in gliomagenesis, and further validation of these proteins could provide a better understanding of disease pathobiology.

Protein disulfide isomerase (PDI), a chaperone protein, plays a role in formation of functional protein by catalyzing disulfide bond breakage and reformation in the endoplasmic reticulum. Increased levels of PDI have been reported in various metastatic and invasive cancers like breast, prostate, hepatocellular carcinoma, and glioblastoma (Goplen et al., 2006; Maresh et al., 2010; Salmans et al., 2013; Shuda et al., 2003). The significance of PDI in glioma cell migration and invasion was first shown by Goplen et al. (2006) by carrying out PDI inhibitory studies in human glioma models.

Our study revealed an increase in PDI levels in gliomas. Nucleophosmin (NPM) is a nucleolar chaperone protein, which prevents the aggregation of denatured proteins (Szebeni and Olson, 1999). NPM regulates cell proliferation by binding to unduplicated centrosomes and prevents the cell's entry into S-phase of cell cycle. Once the centrosomes are duplicated, NPM is removed from the centrosomes by the action of CDK2/Cyclin E mediated phosphorylation (Okuda et al., 2000). NPM binds to a tumor suppressor protein Arf at the N-terminus, preventing the binding of Mdm2 to Arf protein resulting in the free Mdm2 protein. Mdm2, a negative regulator of p53, suppresses the expression of p53 gene thus resulting in the uncontrolled cell proliferation and negative regulation of apoptosis (Bertwistle et al., 2004; Sherr, 2001). NPM has been reported to be upregulated in the early stages of thyroid cancers (Pianta et al., 2010), and it's overexpression in HT29 cell lines was observed to have resulted in increased cell migration and invasiveness (Liu et al., 2012).

In our study, we identified increased expression of chaperone proteins like NPM, protein disulfide-isomerase A3, HSP-90, and calnexin in various grades of tumors. The roles of chaperones in regulating processes like apoptosis, cell migration, and invasion suggest their involvement in glioma pathogenesis.

Accumulation of free radicals in the mitochondria results in the oxidative damage of the mitochondrial membranes and releases cytochrome c into the cytosol, which leads to cell death by apoptosis. Actively proliferating cancer cells circumvent this problem by producing large amounts of superoxide dismutase (SOD), which scavenges oxygen radicals and protects the cancer cells from oxidative damage (Huang et al., 2000). In our study we observed a positive correlation between SOD2 and different grades of glioma. Overexpression of alpha-synuclein in U373 glioma cell lines showed increased apoptosis due to the activation of caspases (Stefanova et al., 2003). Annexin A1, reported to be upregulated in grade-IV astrocytomas (Mulla et al., 2004), is a calcium and phospholipid binding protein and plays a major role in immune suppression by inhibiting the synthesis of eicosanoids and phospholipase A2 (Weyd et al., 2013).

In addition to its role in apoptosis, annexin A5 promotes glioma cell invasion, chemoresistance, and increases the activity/expression of matrix metalloproteinase-2 (Wu et al., 2014). A study by Qing et al. (2008) using NPM gene knockdown showed that reduced NPM protein levels induced apoptosis and decreased proliferation rates in neural stem cells. In our study SOD2, annexin A1 and A5, and NPM levels were observed to be upregulated in gliomas, while alpha synuclein was found to be downregulated suggesting an important role of oxygen radical scavenging and oxidative damage in glioma progression.

We observed an increase in the expression of proteins involved in redox reactions such as thioredoxin, peroxiredoxin-1, and ferritin light chain with increase in the grade of the tumor. Thioredoxin is a disulfide reducing protein, and in addition to its role as a growth promoting factor in leukemic cells (Rosen et al., 1995), it also helps cancer cells escape from apoptosis (Nakamura et al., 1997) and increases angiogenesis by inducing expression of HIF-1 and VEGF (Welsh et al., 2002).

Peroxiredoxin is involved in various biological processes such as scavenging of hydrogen peroxide radicals (O'Leary et al., 2014), cell motility, and cell invasion (Taniuchi et al., 2015). Ferritin is an iron binding protein, which sequesters the free iron ions thus protecting the cells from oxidative damage (Cohen et al., 2005). The expression of ferritin has been reported to be directly proportional to cellular iron levels (Cairo et al., 1985). Tumor cells require large amounts of iron, as cofactor for proteins involved in energy production, DNA synthesis, and cell cycle regulation to meet with their metabolic processes. Accumulation of large amounts of iron in tumor cells (Freitas et al., 2007) has been established to play a role in tumor immunosuppression thus helping them escape from the host immune system (Elliott and Head, 2012). Involvement of the above redox proteins in cell cycle and angiogenesis provides an explanation to the increase in the levels of these proteins with increase in grade of the tumor.

Proliferation rate of LN18 cells transfected with SNCA gene construct was observed to be decreased compared to that of LN18 cells transfected with either CRYAB gene construct or control vector. This reduction in rate of proliferation can be attributed to the interference of the SNCA gene product with that of the cell cycle progression or its role in induction of apoptosis. However, LN18 cells transfected with CRYAB gene construct showed no alteration in the rate of proliferation compared with that of the cells transfected with control vector. These results are in sync with the iTRAQ based mass spectrometry experiments performed in the current study which revealed that SNCA was downregulated in all the grades of glioma compared to its expression in the control tissues. This downregulation might be responsible for increasing the tumor growth either by increasing the rate of proliferation or by inhibiting cell death.

Concluding Remarks

In summary, this comprehensive proteomics analysis of different grades of gliomas revealed proteins associated with major pathways like glycolysis/gluconeogenesis, lactate metabolism, oxidative phosphorylation, coagulation cascades, apoptosis, and cell cycle. The increased expression of proteins like ferritin light chain, alpha-1 antitrypsin, thymosin beta-4-like protein 3, and LDH-A chain with increase in the grades of the tumors provides a plausible explanation toward the aggressive nature of high-grade tumors over low-grade tumors and could be potential markers of disease progression. Ectopic expression of SNCA gene construct in LN18 cell line revealed role of SNCA in suppression of tumor growth. Although the current study provides mechanistic cues of increased aggressive nature of high-grade brain tumors, further validation of several candidate proteins identified from this study may offer identification of grade-specific protein markers and better insight into the molecular basis of glioma pathogenesis.

Footnotes

Acknowledgments

This research was supported by Department of Biotechnology (DBT) (No. BT/PR14359/MED/30/916/2010, BT/PR4599/BRB/10/1042/2012) Government of India. K.G. was supported by Council of Scientific and Industrial Research (CSIR) Senior Research Fellowship.

Author Disclosure Statement

The authors declare that no conflicting financial interests exist.

Abbreviations Used

References

Baumann

, Leukel

, Doerfelt

, et al. (2009). Lactate promotes glioma migration by TGF-beta2-dependent regulation of matrix metalloproteinase-2. Neuro Oncol, 11, 368–380.

Bertwistle

, Sugimoto

, and Sherr

. (2004). Physical and functional interactions of the Arf tumor suppressor protein with nucleophosmin/B23. Mol Cell Biol, 24, 985–996.

Borodovsky

, Seltzer

, and Riggins

. (2012). Altered cancer cell metabolism in gliomas with mutant IDH1 or IDH2. Curr Opin Oncol, 24, 83–89.

Cairo

, Bardella

, Schiaffonati

, Arosio

, Levi

, and Bernelli-Zazzera

. (1985). Multiple mechanisms of iron-induced ferritin synthesis in HeLa cells. Biochem Biophys Res Commun, 133, 314–321.

Calistri

, Rengucci

, Lattuneddu

, et al. (2004). Detection of colorectal cancer by a quantitative fluorescence determination of DNA amplification in stool. Neoplasia, 6, 536–540.

Carlson

, Rogers

, Sifers

, Hawkins

, Finegold

, and Woo

. (1988). Multiple tissues express alpha 1-antitrypsin in transgenic mice and man. J Clin Invest, 82, 26–36.

Chen

, and Okayama

. (1987). High-efficiency transformation of mammalian cells by plasmid DNA. Mol Cell Biol, 7, 2745–2752.

Cohen

, Dafni

, Meir

, Harmelin

, and Neeman

. (2005). Ferritin as an endogenous MRI reporter for noninvasive imaging of gene expression in C6 glioma tumors. Neoplasia, 7, 109–117.

Cooper

, Gutman

, Long

, et al. (2010). The proneural molecular signature is enriched in oligodendrogliomas and predicts improved survival among diffuse gliomas. PloS One, 5, e12548.

10.

Dunn

, Rinne

, Wykosky

, et al. (2012). Emerging insights into the molecular and cellular basis of glioblastoma. Genes Dev, 26, 756–784.

11.

Elliott

, and Head

. (2012). Cancer: Tumor iron metabolism, mitochondrial dysfunction and tumor immunosuppression; “A tight partnership—was warburg correct?.”. J Cancer Ther, 3, 278–311.

12.

Freitas

, Boncompagni

, Vaccarone

, Fenoglio

, Barni

, and Baronzio

. (2007). Iron accumulation in mammary tumor suggests a tug of war between tumor and host for the microelement. Anticancer Res, 27, 3059–3065.

13.

Goplen

, Wang

, Enger

, et al. (2006). Protein disulfide isomerase expression is related to the invasive properties of malignant glioma. Cancer Res, 66, 9895–9902.

14.

, Yang

, Zhao

, et al. (2015). The de-ubiquitinase UCHL1 promotes gastric cancer metastasis via the Akt and Erk1/2 pathways. Tumour Biol, 36, 8379–8387.

15.

Hanash

, and Beretta

. (2002). Operomics: Integrated genomic and proteomic profiling of cells and tissues. Brief Funct Genomic Proteomic, 1, 10–22.

16.

Hatzfeld

, Hatzfeld

, and Maigne

. (1982). Fibrinogen and its fragment D stimulate proliferation of human hemopoietic cells in vitro. Proc Natl Acad Sci U S A, 79, 6280–6284.

17.

Hibi

, Westra

, Borges

, Goodman

, Sidransky

, and Jen

. (1999). PGP9.5 as a candidate tumor marker for non-small-cell lung cancer. Am J Pathol, 155, 711–715.

18.

Huang da

, Sherman

, and Lempicki

. (2009). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc, 4, 44–57.

19.

Huang

, Feng

, Oldham

, Keating

, and Plunkett

. (2000). Superoxide dismutase as a target for the selective killing of cancer cells. Nature, 407, 390–395.

20.

Jin

, Yu

, Lou

, et al. (2013). UCHL1 is a putative tumor suppressor in ovarian cancer cells and contributes to cisplatin resistance. J Cancer, 4, 662–670.

21.

Kalinina

, Peng

, Ritchie

, and Van Meir

. (2011). Proteomics of gliomas: Initial biomarker discovery and evolution of technology. Neuro Oncol, 13, 926–942.

22.

Kanehisa

, Furumichi

, Tanabe

, Sato

, and Morishima

. (2017). KEGG: New perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res, 45, D353–D361.

23.

Liu

, Zhang

, et al. (2012). Expression of nucleophosmin/NPM1 correlates with migration and invasiveness of colon cancer cells. J Biomed Sci, 19, 53.

24.

Louis

, Ohgaki

, Wiestler

, et al. (2007). The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol, 114, 97–109.

25.

Magni

, Van Der Burgt

, Chinello

, et al. (2010). Biomarkers discovery by peptide and protein profiling in biological fluids based on functionalized magnetic beads purification and mass spectrometry. Blood Transfus, 8 (Suppl. 3), s92–s97.

26.

Maresh

, Mah

, Alavi

, et al. (2010). Differential expression of anterior gradient gene AGR2 in prostate cancer. BMC Cancer, 10, 680.

27.

Mulla

, Christian

, Solito

, Mendoza

, Morris

, and Buckingham

. (2004). Expression, subcellular localization and phosphorylation status of annexins 1 and 5 in human pituitary adenomas and a growth hormone-secreting carcinoma. Clin Endocrinol, 60, 107–119.

28.

Nakamura

, Nakamura

, and Yodoi

. (1997). Redox regulation of cellular activation. Annu Rev Immunol, 15, 351–369.

29.

Ohka

, Natsume

, and Wakabayashi

. (2012). Current trends in targeted therapies for glioblastoma multiforme. Neurol Res Int, 2012, 878425.

30.

Okuda

, Horn

, Tarapore

, et al. (2000). Nucleophosmin/B23 is a target of CDK2/cyclin E in centrosome duplication. Cell, 103, 127–140.

31.

O'Leary

, Terrile

, Bajor

, et al. (2014). Peroxiredoxin-1 protects estrogen receptor alpha from oxidative stress-induced suppression and is a protein biomarker of favorable prognosis in breast cancer. Breast Cancer Res, 16, R79.

32.

Pelicano

, Martin

, Xu

, and Huang

. (2006). Glycolysis inhibition for anticancer treatment. Oncogene, 25, 4633–4646.

33.

Pianta

, Puppin

, Franzoni

, et al. (2010). Nucleophosmin is overexpressed in thyroid tumors. Biochem Biophys Res Commun, 397, 499–504.

34.

Plow

, and Edgington

. (1986). Lymphocyte suppressive peptides from fibrinogen are derived predominantly from the A alpha chain. J Immunol, 137, 1910–1915.

35.

Qing

, Yingmao

, Lujun

, and Shaoling

. (2008). Role of Npm1 in proliferation, apoptosis and differentiation of neural stem cells. J Neurol Sci, 266, 131.

36.

Reddy

, Rao

, Malhotra

, et al. (2013). A simple protein extraction method for proteomic analysis of diverse biological specimens. Curr Proteomics, 10, 298–311.

37.

Rosen

, Lundman

, Carlsson

, et al. (1995). A CD4⁺ T cell line-secreted factor, growth promoting for normal and leukemic B cells, identified as thioredoxin. Int Immunol, 7, 625–633.

38.

Salmans

, Zhao

, and Andersen

. (2013). The estrogen-regulated anterior gradient 2 (AGR2) protein in breast cancer: A potential drug target and biomarker. Breast Cancer Res, 15, 204.

39.

Sawaya

, Zuccarello

, and Highsmith

. (1987). Alpha-1-antitrypsin in human brain tumors. J Neurosurg, 67, 258–262.

40.

Shankarkumar

, and Sridharan

. (2011). Glioma Indian scenario: Is there a human leucocyte antigen association?. J Nat Sc Biol Med, 2, 205–208.

41.

Sherr

. (2001). The INK4a/ARF network in tumour suppression. Nat Rev Mol Cell Biol, 2, 731–737.

42.

Shuda

, Kondoh

, Imazeki

, et al. (2003). Activation of the ATF6, XBP1 and grp78 genes in human hepatocellular carcinoma: A possible involvement of the ER stress pathway in hepatocarcinogenesis. J Hepatol, 38, 605–614.

43.

Simon

, Lam

, Li

, Ngan

, Menenzes

, and Zhao

. (2007). Analysis of gene expression data using BRB-Array tools. Cancer Inform, 3, 11–17.

44.

Stefanova

, Schanda

, Klimaschewski

, Poewe

, Wenning

, and Reindl

. (2003) Tumor necrosis factor-alpha-induced cell death in U373 cells overexpressing alpha-synuclein. J Neurosci Res, 73, 334–340.

45.

Szebeni

, and Olson

. (1999). Nucleolar protein B23 has molecular chaperone activities. Protein Sci, 8, 905–912.

46.

Takase

, Hibi

, Yamazaki

, et al. (2003). PGP9.5 overexpression in esophageal squamous cell carcinoma. Hepatogastroenterology, 50, 1278–1280.

47.

Taniuchi

, Furihata

, Hanazaki

, et al. (2015). Peroxiredoxin 1 promotes pancreatic cancer cell invasion by modulating p38 MAPK activity. Pancreas, 44, 331–340.

48.

Tezel

, Hibi

, Nagasaka

, and Nakao

. (2000). PGP9.5 as a prognostic factor in pancreatic cancer. Clin Cancer Res, 6, 4764–4767.

49.

Thompson

, Smith

, Stirk

, Marshall

, Stout

, and Kocchar

. (1992). Angiogenic activity of fibrin degradation products is located in fibrin fragment E. J Pathol, 168, 47–53.

50.

Welsh

, Bellamy

, Briehl

, and Powis

. (2002). The redox protein thioredoxin-1 (Trx-1) increases hypoxia-inducible factor 1alpha protein expression: Trx-1 overexpression results in increased vascular endothelial growth factor production and enhanced tumor angiogenesis. Cancer Res, 62, 5089–5095.

51.

Weyd

, Abeler-Dorner

, Linke

, et al. (2013). Annexin A1 on the surface of early apoptotic cells suppresses CD8⁺ T cell immunity. PloS One, 8, e62449.

52.

, Yang

, Xiong

, et al. (2014). Annexin A5 promotes invasion and chemoresistance to temozolomide in glioblastoma multiforme cells. Tumour Biol, 35, 12327.

53.

Xia

, Sinelnikov

, Han

, and Wishart

. (2015). MetaboAnalyst 3.0-making metabolomics more meaningful. Nucleic Acids Res, 43, W251–W257.

54.

Xiang

, Li

, Yin

, et al. (2012). The ubiquitin peptidase UCHL1 induces G0/G1 cell cycle arrest and apoptosis through stabilizing p53 and is frequently silenced in breast cancer. PloS One, 7, e29783.

55.

, Hu

, and Qiu

. (2017). Potential mechanisms of microRNA-129-5p in inhibiting cell processes including viability, proliferation, migration and invasiveness of glioblastoma cells U87 through targeting FNDC3B. Biomed Pharmacother, 87, 405–411.

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.

0.56 MB

0.00 MB