
==== Front
Turk J Biol
Turk J Biol
Turkish Journal of Biology
1300-0152
1303-6092
Scientific and Technological Research Council of Turkey (TUBITAK)

10.55730/1300-0152.2700
tjb-48-04-242
Research Article
Expressions of the satellite repeat HSAT5 and transposable elements are implicated in disease progression and survival in glioma
https://orcid.org/0000-0003-2358-1605
KÖSE Sıla Naz 1α
https://orcid.org/0000-0003-3019-7900
YARAŞ Tutku 23α
https://orcid.org/0000-0001-7050-769X
BURSALI Ahmet 2
https://orcid.org/0000-0002-0158-2693
OKTAY Yavuz 23
https://orcid.org/0000-0002-2050-6186
YANDIM Cihangir 12β
https://orcid.org/0000-0001-6706-1375
KARAKÜLAH Gökhan 23β*
1 Department of Genetics and Bioengineering, Faculty of Engineering, İzmir University of Economics, İzmir, Turkiye
2 İzmir Biomedicine and Genome Center (IBG), İzmir, Turkiye
3 İzmir International Biomedicine and Genome Institute (IBG-İzmir), Dokuz Eylül University, İzmir, Turkiye
* Correspondence: gokhan.karakulah@ibg.edu.tr
α These authors contributed equally.

β These authors contributed equally.

2024
01 7 2024
48 4 242256
05 1 2024
23 8 2024
01 7 2024
© TÜBİTAK
2024
https://creativecommons.org/licenses/by/4.0/ This work is licensed under a Creative Commons Attribution 4.0 International License.
The glioma genome encompasses a complex array of dysregulatory events, presenting a formidable challenge in managing this devastating disease. Despite the widespread distribution of repeat and transposable elements across the human genome, their involvement in glioma’s molecular pathology and patient survival remains largely unexplored. In this study, we aimed to characterize the links between the expressions of repeat/transposable elements with disease progression and survival in glioma patients. Hence, we analyzed the expression levels of satellite repeats and transposons along with genes in low-grade glioma (LGG) and high-grade glioma (HGG). Endogenous transposable elements LTR5 and HERV_a-int exhibited higher expression in HGG patients, along with immune response-related genes. Altogether, 16 transposable elements were associated with slower progression of disease in LGG patients. Conversely, 22 transposons and the HSAT5 satellite repeat were linked to a shorter event-free survival in HGG patients. Intriguingly, our weighted gene coexpression network analysis (WGCNA) disclosed that the HSAT5 satellite repeat resided in the same module network with genes implicated in chromosome segregation and nuclear division; potentially hinting at its contribution to disease pathogenesis. Collectively, we report for the first time that repeat and/or transposon expression could be related to disease progression and survival in glioma. The expressions of these elements seem to exert a protective effect during LGG-to-HGG progression, whereas they could have a detrimental impact once HGG is established. The results presented herein could serve as a foundation for further experimental work aimed at elucidating the molecular regulation of glioma genome.

Glioma
glioblastoma
transposon
satellite repeat
HSAT5
survival
==== Body
pmc1. Introduction

Gliomas are the most common type of central nervous system tumors representing approximately 80% of all malignant brain tumors (Ostrom et al., 2017). They can vary at genetic and epigenetic levels causing differences in tumor aggressiveness and hence a heterogeneous response to therapy (Noushmehr et al., 2010; Sturm et al., 2014; Sakthikumar et al., 2020; Garcia-Fabiani et al., 2021). Gliomas are histologically classified as astrocytic, oligodendroglial, or mixed (astrocytic and oligodendroglial) according to the cell type from which they originate (Louis et al., 2016). Importantly, they can be low-grade (LGG) or high-grade (HGG) according to their malignancy levels. LGGs typically arise de novo without any prior precursor or lower-grade lesions, and they have a better prognosis with approximately 75% five-year survival rates, while HGGs are grade III/IV gliomas generally differentiated from those of lower grades with a much lower five-year survival rate of only 5% (Stupp et al., 2009; Crespo et al., 2015; Teng et al., 2022). Even though gliomas could arise from different cell types as mentioned above, they exhibit similar molecular characteristics based on their grade (i.e. LGG or HGG) and mutational statuses (Arcella et al., 2020).

Numerous studies have attempted to unravel histological and molecular differences between LGG and HGG to facilitate the identification of potential biomarkers and therapeutic targets. LGGs are usually characterized by IDH, ATRX, and TP53 alterations, as well as CDKN2A hypomethylation and 1p/19q codeletion (Parsons et al., 2008; Brat et al., 2015; Louis et al., 2016), whilst progression to HGG can be characterized by PTEN mutations, EGFR amplifications and MGMT promoter hypermethylation (Wang et al., 2016; Brito et al., 2019). It is noteworthy that some of these alterations may have distinct and sometimes opposite prognostic implications in LGG and HGG. Several mutations were found to be associated with better outcomes in LGG whilst presenting worse outcomes in HGG (Eckel-Passow et al., 2015; Juratli et al., 2018). Moreover, dysregulations in signaling pathways, mutations in oncogenes and tumor suppressors, and impaired epigenetic mechanisms may lead to genomic instability and prompt the chromatin of a glioma cell for further malignant progression (Louis et al., 2016; Mackay et al., 2017). Pivotal changes in chromatin architecture were observed for various types of gliomas and they were related to tumor aggressiveness, as well as patient survival (Mack et al., 2019; Wang et al., 2021; Garrett et al., 2022; Yang et al., 2022). Further insights into the dynamics of these chromatin aberrations would therefore pave the way for designing more effective treatment strategies.

There are a multitude of layers that influence chromatin architecture, one of which certainly is the content of DNA. At least 50% of the human genome is made up of repeat elements (International Human Genome Sequencing Consortium, 2001) and their transcription levels reflect the status of chromatin (Probst et al., 2010; Jagannathan and Yamashita, 2017). Even though their exact contribution to subnuclear regulation is yet to be characterized, expression of repeats was generally linked to vital biological processes such as heat stress response (Porokhovnik et al., 2021) and cellular senescence (Karakülah and Yandım, 2020). More importantly, tightly regulated expressions of these elements were particularly implicated in embryonic development (Ulitsky et al., 2011; Yandım and Karakülah, 2019a). Whereas tandemly repeated arrays such as the satellites could affect kinetochore formation and hence chromosomal stability (Vos et al., 2006; Zhu et al., 2011), interspersed elements such as DNA transposons or retrotransposons (i.e. long interspersed nuclear elements [LINEs], short interspersed nuclear elements [SINEs], long terminal repeats [LTRs], and human endogenous retroviruses [HERVs]) may act as a mutagenic force (Lee et al., 2012; Helman et al., 2014). Therefore, not surprisingly, dysregulated transcription levels of both satellites and transposons were reported to be potentially disease-causing in various cancers including but not limited to breast cancer (Zhu et al., 2011; Yandım and Karakülah, 2019b), hepatocellular carcinoma (Karakülah and Yandım, 2021), and pancreatic cancer (Ting et al., 2011; Yandım and Karakülah, 2022).

Despite reported changes in global chromatin and genome regulation, there have only been limited investigations into the contribution of repeat elements to the molecular pathogenesis and prognosis of gliomas. Ohka et al. reported that hypomethylation events in LINE elements were associated with shorter overall survival in glioma patients (Ohka et al., 2011). In addition, it was indicated that transcriptions of Alu elements were posttranscriptionally edited in oligodendroglioma (Hwang et al., 2022). Another recent study focusing on endogenous retroviruses showed that 43 HERV elements were upregulated in HGG compared to healthy brain tissue with implicated effects on proximal gene regulation (Koso et al., 2012; Yuan et al., 2021). On the other hand, the specific types of repeats/transposable elements that could be associated with disease progression and survival in glioma have not yet been characterized. This could be partly because most of the published RNA sequencing (RNA-seq) data is prepared with a poly(A) bias in library preparation, which renders the analysis of repeat-arisen transcripts inconclusive (Solovyov et al., 2018).

To elucidate the particular links between repeat/transposable element expression and disease progression and survival in glioma patients, we employed a publicly available dataset (GSE184941) for which total RNA-seq had been performed without an mRNA bias on the tumor resection samples of 32 LGG and 34 HGG patients with IDH mutations (Supplementary file 1: Table S1) (Xiao et al., 2022; Khan et al., 2023). This was the only available relevant dataset that allowed the quantification of total transcripts arisen from repeat elements and transposons whilst also presenting patient survival attributes. We identified differentially expressed satellite repeats and transposons between LGG and HGG samples. We also performed survival analyses considering the expressions of all human RepBase repeats (Bao et al., 2015) with Cox-regression and log-rank tests (Bewick et al., 2004), revealing a subset of repeats that are potentially important for LGG to HGG progression and survival.

2. Results

2.1. Identification of differentially expressed genes and repeat elements in HGG in comparison to LGG

To uncover the differences in gene and repeat/transposon expression, we quantified total RNA transcripts and performed a principal component analysis (PCA) (Figure 1 and Supplementary file 2: Figure S1). Even though a general distinction was observed in PCA analysis both for protein-coding gene expression and repeat arisen transcription levels between LGG and HGG samples (Figures 1A and 1B), the distribution of read percentages across patients were slightly different (Supplementary file 2: Figure S1A) with small changes in the abundance of repeat types giving rise to transcripts (Supplementary file 2: Figure S1B and S1C). When we applied a threshold of | log2(FC)| >0.6 | and false discovery rate (FDR) < 0.05, we identified 605 differentially expressed genes (Figure 1C and Supplementary file 3: Table S2) and 7 repeats (Figure 1D and Supplementary file 3: Table S3) in HGG samples when compared to LGG samples. A gene ontology analysis revealed that differentially expressed genes were mostly associated with biological processes implicating immunological response and cell-cell adhesion (Figure 1E, Supplementary File 4: Table S4), aligning well with previous literature (Xu et al., 2019). Retrotransposons LTR5 and HERVE_a-int were upregulated in HGG samples suggesting a possible role for them in the progression from LGG to HGG.

2.2. Identification of gene and repeat expressions implicated in disease progression and survival

Gene expression signatures that influence survival times of patients provide invaluable information on the carcinogenesis and metastasis process in cancer (Sahu et al., 2021). Therefore, they have the potential to open new treatment avenues. To provide a detailed picture of the survival-related expression events where repeat-arisen transcripts were also considered in glioma, we performed a Cox regression analysis followed by a log-rank test (with a median progression/survival time cut-off) not only for the survival of HGG patients but also for the progression of LGG patients (to HGG). A PCA analysis showed a more pronounced separation between shorter and longer survival in the event-free survival analysis of HGG samples as opposed to progress-free survival analysis of LGG samples (Figure 2).

Based on progress-free survival analysis in LGG samples, we identified 71 genes and 16 repeats (Cox Hazard Ratio [HR] < 0.5, Cox p-value < 0.05 and log-rank p-value < 0.05) that were associated with slower LGG to HGG progression (Table 1, Supplementary file 5: Tables S5 and S7). On the other hand, 111 genes, but none of the repeats, (Cox HR > 1.5, Cox p-value < 0.05 and log-rank p-value < 0.05) were found to be related to faster LGG to HGG progression (Table 1, Supplementary file 5: Tables S6 and S8). In the samples where progression to HGG took place, event-free (event: death) survival analysis revealed 48 genes, but no repeats (Cox HR <0.5, Cox p-value<0.05 and Log-rank p-value<0.05) were associated with longer survival (Table 1, Supplementary file 6: Tables S9 and S11). Interestingly, 1907 genes and 23 repeats (Cox HR >1.5, Cox p-value<0.05 and Log-rank p-value<0.05) were linked to shorter event-free survival in HGG (Table 1, Supplementary file 6: Tables S10 and S12). Only one gene (CASS4) was associated with longer progress-free survival in LGG, as well as longer event-free survival in HGG (Figure 3A). On the other hand, two genes (GPR156 and AGAP1), but no repeats, were associated with shorter survival both in progress-free analysis of LGG samples and event-free analysis of HGG samples at the same time. Intriguingly, 14 genes (IPO9, PLK1, DCAF4L1, ZNF345, LRRC75A, RAD18, BMP8A, LRR1, SETD6, AK2, MAP3K12, LMBR1L, ERCC6L, ZNF776) and one transposon (LTR18C) were related with longer time for LGG-to-HGG progression, yet in opposition, they were linked to shorter event-free survival in HGG samples. Gene ontology analyses revealed that LGG-to-HGG progression could be attributed to neurotransmitter release (Figure 3B, Supplementary file 7: Table S13), whereas survival in HGG could be related to various cell cycle events including spindle organization, chromosome segregation, and nuclear division (Figure 3C, Supplementary file 8: Table S14).

2.3. Weighted gene co-expression network analysis (WGCNA) reveals the expression of HSAT5 satellite as a potential key player in HGG

In order to further elucidate specific gene expression correlation patterns that could be distinctly related to LGG or HGG and to provide a better insight into the cooperation of repeat expression in these patterns, we performed weighted gene co-expression network analysis (WGCNA) (Zhang and Horvath, 2005; Cantone and Fisher, 2013). Our pipeline contained a pool approach where not only the transcript levels of genes but also those of repeats were analyzed altogether. Overall, 13 modules that contained genes and repeats were identified and these were represented with different colors (Figure 4A, Supplementary file 9: Table S15). Tan module was specifically emphasized in LGG samples, and the green module was particularly associated with HGG. Whereas the tan module contained no repeats but included 185 genes, the green module comprised only the satellite repeat HSAT5, no transposons, and 981 genes. The majority (987) of repeat elements/transposons fell into turquoise module, which was emphasized in LGG samples and where all 16 repeats linked to longer LGG progress-free survival (Table 1) were also present. The blue module, on the other hand, was pronounced in HGG and contained 5 repeats, among which only L1M2a was linked to event-free survival in HGG. Greenyellow module was slightly pointing to LGG, and it contained only one repeat.

When a gene ontology analysis was run for all of the WGCNA modules, genes related to ribosome biogenesis and protein translation seemed to be enriched in the tan module (Figure 4B, Supplementary file 10). Interestingly, the green module displayed an enrichment in genes that were linked to nuclear division, chromosome segregation, and double-strand break repair. Turquoise module, where the majority of repeats reside, was enriched with genes linked to cilium organization, methylation and ncRNA metabolic process, splicing and tRNA metabolic process. The blue module was as well in an enrichment with ncRNA processing and ribosome biogenesis genes. On the other hand, the greenyellow module was related to myelination and neuron ensheathment.

Our interesting findings on HSAT5 satellite repeat which was linked to shorter event-free survival in HGG (Figure 5A) and the fact that it appeared in the same WGCNA module (green) with genes that were linked to similar biological phenomena (i.e. nuclear division and chromosome segregation) presented in Figure 3C for HGG event-free survival, prompted us to run a Pearson correlation analysis for the expression levels of HSAT5 and all protein-coding genes in the genome (Supplementary file 11). We found that 310 genes exhibited moderate-to-high level of correlation (r > 0.5) with HSAT5 expression. Strikingly, 195 (63%) of these 310 genes were among those significantly linked to shorter HGG event-free survival as presented in Table S10 (Supplementary file 6). A thorough literature search also showed that most of these genes were reported to be influential in glioma (Table 2). Some of these HSAT5-correlated genes including but not limited to ANKLE2 (Figure 5B) and NDE1 (Figure 5C) could be involved in the regulation of HSAT5 expression due to their crucial functions on chromatin (Chomiak et al., 2022; Sebastian et al., 2022).

3. Discussion

The complexity of molecular pathogenesis underlying gliomas makes them refractory to current treatment strategies. Notably, most of the patients that die due to gliomas suffer from HGG, which in most cases stem from a preceding LGG (Bogdańska et al., 2017). Therefore, understanding the particular events that cause progression to HGG from LGG is important in terms of projecting more insightful approaches toward gliomas. In this study, we uncovered not only differentially expressed protein-coding genes but also transcripts arisen from satellite repeats and transposons in HGG as opposed to LGG, shedding further light into disease progress. We found that genes related to immune response were differentially expressed in HGG as opposed to LGG, agreeing well with previous reports (Gabrusiewicz et al., 2016; Lu et al., 2019; Xiao et al., 2022; Khan et al., 2023). This could be a projected outcome due to immune cell infiltration as reported before (Domingues et al., 2016; Lin et al., 2021). Another interesting finding was the upregulated expression (in HGG) of HERV_a-int and LTR5 transposon, the latter of which was previously linked to cancer (Lee et al., 2012) and human preimplantation development (Yandım and Karakülah, 2019a).

Our progress-free survival analysis of LGG samples on gene expressions revealed that genes related to the regulation of neurotransmitter release, dopamine secretion and exocytosis were linked to faster progression to HGG. Even though genes within this context were not reported in some of the studies that analyzed clinical samples (Wang et al., 2020; Zhang et al., 2020), in vitro and animal studies previously suggested dopamine release as a potential factor in glioma (Li et al., 2014); however, its effect was not entirely understood (Lan et al., 2017). Another interesting finding in our study was that all repeat elements associated with LGG-to-HGG progression were those that had a hazard ratio of less than 0.5. In other words, transposon expression could potentially slow down the progression to HGG. This could be due to increased antigenicity and better immune response caused by transposon expression as suggested earlier (Kong et al., 2019; Yuan et al., 2021). In opposition, event-free survival analysis in HGG revealed that none of the repeat elements were associated with longer survival. However, it identified one satellite (i.e. HSAT5) and 23 transposons linked to shorter survival. Though this result suggests that repeat element expression may lead to opposite consequences in LGG and HGG, only one repeat (i.e. LTR18C) appeared in the survival analyses of both LGG (HR<0.5) and HGG (HR>1.5) groups. In line with this, the genes linked to shorter survival in HGG were also pointing out a different context as opposed to those in LGG. Genes with HR>1.5 were mostly linked to chromosome segregation and nuclear division, generally agreeing well with previous literature (Bredel et al., 2005; Phillips et al., 2006; Murat et al., 2008; Q. Wang et al., 2017).

In an effort to dissect the impact of repeat element expression on the molecular pathogenesis and survival in glioma, we not only presented LGG progress-free survival related repeats and HGG event-free survival linked repeats but also differentially expressed repeats in HGG samples in comparison to LGG. Yuan et al. (2021) reported differentially expressed repeats in tissue samples of HGG (GBM) patients as opposed to normal brain samples; however, the data they employed was prepared with an mRNA bias only detecting poly(A)+ transcripts. Our study utilizes sequencing data obtained from total RNA samples without the poly (A) bias (Khan et al., 2023) as previously recommended (Solovyov et al., 2018). Notably, LTR5 was reported to be upregulated both in a previous study (Yuan et al., 2021), which compares the HGG tissue with normal tissue, and in our study that compares with the HGG tissue with LGG, even though no survival related effect was detected. Another element we detected in HGG event-free survival analysis was HERVK3-int, which was reported to be upregulated in HGG samples with a potential effect on disease progression (Shah et al., 2022). Our holistic approach identified many other repeats/transposons previously unmentioned in the glioma literature. Perhaps, the most interesting one among these is the satellite repeat HSAT5, which predominantly resides in pericentromeric regions of chromosomes (Hillier et al., 2005; Hubley et al., 2016) and was also reported to be upregulated in cellular senescence (Karakülah and Yandım, 2020). Though no studies so far reported HSAT5 within the context of cancer, other satellites such as HSATI (Zhu et al., 2011), HSATII (Ting et al., 2011), and GSATII (Karakülah and Yandım, 2021) were linked to the molecular pathologies of solid cancers. Satellite repeats are known to play pivotal roles in maintaining heterochromatin architecture and kinetochore formation (Rošić et al., 2014; Nishibuchi and Déjardin, 2017) and their abnormal expressions were linked to genomic catastrophes in cell division (Zhu et al., 2011; Bersani et al., 2015; Kishikawa et al., 2016). Their usage as biomarkers is also being valuably considered (Özgür et al., 2021). Our results point out HSAT5 satellite repeat as a potential prognostic biomarker in HGG survival. More importantly, its cooperative expression levels with HGG-specific genes that are known to involve in chromosome segregation and nuclear division could implicate its potential in the molecular pathology of this devastating disease. The relationship of HSAT5 DNA or RNA with chromatin-related factors such as NDE1 and ANKLE2 could illuminate the mechanism of HSAT5 expression further.

4. Conclusion

In this study, we performed a comprehensive analysis on the expressions of repeat elements covering both satellite repeats and transposons to figure out those implicated in glioma progression and survival. Our results interestingly revealed that the expressions of transposable elements could be linked to slower LGG to HGG progression, yet this beneficial effect is overturned once the progress to HGG is manifested. One explanation for the beneficial effect in LGG could be the improved antigenicity caused by repeat elements, which provides better immune response. On the other hand, once the progression to HGG takes place, repeats could be contributing to dysregulated events in chromosome segregation and cell cycle as mostly the results on HSAT5 satellite repeat here suggested. Results presented here could serve as a guide for further experimental work in understanding the molecular basis of gliomas further.

5. Methods

5.1. Data acquisition and preprocessing

We obtained raw sequencing data from the Sequence Read Archive (SRA) database (Leinonen et al., 2011) for 66 glioblastoma patients (SRA accession: SRP339173; GEO accession: GSE184941) using the “fastq-dump -gzip -skip-technical -readids -dumpbase -clip -split-3” command of the SRA Tool Kit v.2.9.0. We collected the GRCh38 human genome assembly and comprehensive gene annotation (release 34) in gene transfer format from the GENCODE1 website (Frankish et al., 2021). We obtained genomic coordinates of repeat instances across the genome from the RepeatMasker2 website. We aligned paired-end reads from each tumor sample to the human reference genome using the Rsubread package v1.34.7 (Liao et al., 2014) in the R statistical computation environment with default parameters. We used SAMtools suite v1.3.1 (Li et al., 2009) to sort and index the binary alignment map files. We counted reads using the featureCounts function of the Rsubread package (Liao et al., 2014), which allowed us to differentiate between exonic and repetitive regions. We merged repeat and gene counts into a single expression matrix, and calculated counts per million (CPM) values for each gene and repeat element across all tumor samples. We excluded any gene or repeat element that was not expressed above 1 CPM in at least 25% of the tumor samples from further analysis. Besides, when filtering for HGG and LGG samples, IDH mutants and percentage of total mapped reads greater than 70 were included in both samples. To improve the accuracy of estimated repeat element expressions, we only considered sequencing reads that overlapped nonexonic regions and uniquely mapped to DNA, LINE, SINE, LTR, and satellite repeat regions.

5.2. Coexpression network analysis of protein-coding genes and repeat elements

To perform coexpression network analysis, we used the R package WGCNA v1.47 (Langfelder and Horvath, 2008). Adjacencies between all protein-coding genes and repeats across samples were calculated. In this step, CPM values of genomic features were used as input. We designated the soft threshold value as 12 and the average linkage hierarchical clustering method was selected for grouping the genomic features with similar expression patterns. Additionally, the dynamic tree cut algorithm (Langfelder et al., 2008) was utilized to determine network modules and minimum network module size was set to 30. Gene ontology (GO) enrichment analysis of each network module was performed with the R package clusterProfiler v3.18.0 (Yu et al., 2012).

5.3. Statistical analysis and graphical representation

Statistical analysis and graphical representation were performed using the R environment3 for the LGG and HGG samples. Differential expression (DE) analysis of genomic features was performed using the edgeR package (version 3.28.2) (Robinson et al., 2010). We obtained a PCA plot of samples to visualize clusters of LGG and HGG samples based on their similarities (quantile median = 0.50). We used the log rank test (logrank_test function of the coin package, version 1.4.1) (Royston et al., 2019) to test the statistical significance of survival time differences between groups and the Cox regression analysis (coxph function of the survival package, version 3.2.11) to estimate hazard ratios (HR). The Kaplan-Meier curves were drawn using the survfit function to show the probabilities of survival for a certain time interval. We used the ggplot2 package (version 3.3.6) to create other graphical representations.

Supplementary Data

Availability of data and material

The dataset analyzed during the current study is available in the GEO repository, with accession number GSE184941.

Figure 1 Comparison of gene and repeat expressions between low-grade (LGG) and high-grade (HGG) gliomas. A Principal component analysis (PCA) of protein-coding genes. B PCA of repeat elements C MA plot of protein coding gene expressions D MA plot of repeat element expressions. E Gene ontology analysis results on biological processes for genes differentially expressed in HGG relative to LGG.

Figure 2 Principal component analyses (PCA) of the expressions of genes implicated in glioma progress and survival with a median cut-off. A PCA of samples associated with short and long progress-free survival in LGG considering the protein-coding gene expressions. B PCA of samples associated with short and long progress-free survival in LGG considering repeat element expressions. C PCA of samples associated with short and long event-free survival in HGG considering the protein-coding gene expressions. D PCA of samples associated with short and long progress-free survival in LGG considering repeat element expressions.

Figure 3 Progress-free survival genes in LGG and event-free survival genes in HGG. A Hazardous (HR>1.5, Log-rank p<0.05) and protective (HR<0.5, Log-rank p<0.05) genes based on Cox regression analyses and Log-rank test in LGG and HGG. B Gene ontology analysis of hazardous genes in LGG-to-HGG progression. C Gene ontology analysis of hazardous genes in HGG event-free survival.

Figure 4 Weighted Gene Co-expression Network Analysis (WGCNA) of gene and repeat expressions. A WGCNA modules B Gene ontology analysis of the genes in WGCNA modules.

Figure 5 Survival analysis of HSAT5 expression and its correlation with key chromatin factors. A Kaplan-Meier event-free survival graph of HGG patients based on HSAT5 satellite repeat expression with a median cut-off. B,C Scatter plot representations showing the correlation of HSAT expression with the expressions of ANKLE2 and NDE1 genes in HGG patients.

Table 1 Significant repeat elements linked to patient survival in LGG and HGG.

Low-grade glioma (LGG)	High-grade glioma (HGG)	
Longer progress-free survival (Log-rank p<0.05)	Repeat	Cox HR (<0.5)	Shorter progress-free survival (Log-rank p < 0.05)	Repeat	Cox HR (>1.5)	Longer event-free survival (Log-rank p < 0.05)	Repeat	Cox HR (<0.5)	Shorter event-free survival (Log-rank p < 0.05)	Repeat	Cox HR (>1.5)	
LTR1F2	0.226566207	None	None	hAT-16_Crp	3.841230728	
MER107	0.246705973	HERV1_LTRb	3.772926264	
MER1B	0.253616041	MLT1E3-int	3.693884586	
LTR57	0.342074886	L1M2a	3.661849157	
LTR1C	0.362824242	L1M3f	3.640586269	
MER130	0.379093156	CR1-11_Crp	3.092362862	
LTR40c	0.388071793	MER95	2.958200736	
LTR1	0.394057275	UCON97	2.91316852	
LTR1A2	0.394722746	MSTC-int	2.886032148	
LTR2752	0.398195734	MER70-int	2.688166503	
LTR18C	0.421946618	HERVI-int	2.664554289	
LTR27C	0.449130349	L1MDa	2.613437037	
LTR35A	0.458301493	HSAT5	2.598739703	
MER9B	0.465923328	LTR75	2.456202539	
LTR27	0.476049565	HERVK3-int	2.414286886	
LTR7C	0.491857327	L1MB2	2.396560074	
		L1M8	2.340619317	
LTR70	2.3387068	
L1M4a1	2.31165162	
LTR18C	2.18835578	
MER51-int	2.165050709	
HERVKC4-int	1.908960599	
Kanga1c	1.852908386	

Table 2 Genes, whose expressions significantly correlated with that of HSAT5 satellite repeat, and their known associations with gliomas. Gene names written in bold characters indicate a direct link with survival in HGG as represented in Figures 1A and 1C.

Gene name	p-values	Pearson’s corr. coef. (r)	Direct association with glioma	
ANKRD20A1	6.48E+02	0.723955	Membrane shaping (Wolf et al., 2019)	
QTRT2	2.01E+02	0.712178	Ferroptosis (Sallam et al., 2022)	
ANKRD20A3P	1.72E+04	0.688139	-	
IGF2BP2	2.47E+04	0.683868	Chemoresistance (Han et al., 2022)	
RALY	2.09E+04	0.656957	-	
CCDC150	3.09E+05	0.651718	-	
TRIO	3.79E+05	0.648938	Cell migration, invasion and disease outcome (Salhia et al., 2008)	
ODF2	4.04E+04	0.648070	Hypoxia (Baghbani et al., 2013)	
ANKLE2	6.30E+05	0.641924	Apoptosis and survival (Wang et al., 2018)	
SHC4	7.97E+05	0.638604	EMT, disease progression and outcome (Tilak et al., 2021)	
TTF2	1.27E+06	0.631903	Cell cycle and repair mechanisms, extracellular matrix organization (Chai et al., 2019)	
ANKRD11	1.53E+06	0.629146	-	
GSDMB	1.88E+06	0.626186	Immune cell infiltration (Lin J et al, 2022)	
MCM8	1.91E+06	0.625905	Stemness (Wang et al., 2021)	
SGCA	2.22E+06	0.623704	ECM signaling (Balvers et al., 2013)	
NDE1	2.28E+06	0.623247	Cell proliferation, migration, and disease prognosis (Chen et al., 2017; Song et al., 2019)	
DIAPH3	2.94E+06	0.619450	Chemoresistance (Chehade et al. 2022)	
PIAS2	3.46E+06	0.616934	Posttranslational modification (Li and Meng, 2021)	
PLCG1	3.74E+06	0.615770	Disease prognosis (Li et al., 2022)	
FUS	3.76E+06	0.615690	Angiogenesis (He et al., 2019)	
QRICH2	4.18E+06	0.614060	-	
CAPRIN2	4.62E+06	0.612492	-	
WDR62	4.98E+05	0.611313	-	
SFI1	5.52E+06	0.609695	-	
GNL3L	6.79E+06	0.606424	Survival (Sun et al., 2021)	
POLE	7.04E+06	0.605863	Genomic stability and disease progression (Erson-Omay et al., 2015)	
ZBTB49	7.97E+06	0.603872	-	
ANAPC7	8.56E+06	0.602737	Disease progression (de Groot et al., 2011)	
PSMC3IP	9.16E+06	0.601632	Angiogenesis (Zhang et al., 2017)	

1 https://www.gencodegenes.org

2 http://www.repeatmasker.org/

3 v3.6.0; https://www.r-project.org/

Ethical approval and consent to participate: Not applicable.

Consent for publication: Not applicable.

Competing interests: The authors declare that there are no competing interests.

Supplementary files: Supplementary files can be downloaded from the following link:

https://aperta.ulakbim.gov.tr/record/273725

Supplementary File 1:

Table S1. SRR numbers of data used for the study.

Supplementary File 2:

Figure S1. Distribution of read percentages of LGG and HGG samples. A. Violin plot representing the distribution of read percentages of protein-coding genes and repeat elements in LGG and HGG samples. B. Pie chart representing the percentages of repeat classes in LGG samples. C. Pie chart representing the percentages of repeat classes in HGG samples.

Supplementary File 3:

Table S2. Differentially expressed genes in HGG samples in comparison to LGG samples.

Table S3. Differentially expressed repeats in HGG samples in comparison to LGG samples.

Supplementary File 4:

Table S4. Gene ontology analysis result list on biological processes of differentially expressed genes in HGG relative to LGG.

Supplementary File 5:

Table S5. Protective genes (HR<0.5) obtained in LGG progress-free survival analysis.

Table S6. Hazardous genes (HR>1.5) obtained in LGG progress-free survival analysis.

Table S7. Protective repeats (HR<0.5) obtained in LGG progress-free survival analysis.

Table S8. Hazardous repeats (HR>1.5) obtained in LGG progress-free survival analysis.

Supplementary File 6:

Table S9. Protective genes (HR<0.5) obtained in HGG event-free survival analysis.

Table S10. Hazardous genes (HR>1.5) obtained in HGG event-free survival analysis.

Table S11. Protective repeats (HR<0.5) obtained in HGG event-free survival analysis.

Table S12. Hazardous repeats (HR>1.5) obtained in HGG event-free survival analysis.

Supplementary File 7:

Table S13. Gene ontology analysis result list on biological processes of hazardous genes identified in LGG progress-free analysis.

Supplementary File 8:

Table S14. Gene ontology analysis result list on biological processes of hazardous genes identified in HGG event-free analysis.

Supplementary File 9:

Table S15. Genes listed in WGCNA modules.

Supplementary 10: Gene ontology analysis of genes identified in WGCNA modules pertaining to biological processes.

Supplementary File 11:

Table S16. Correlations of the expressions of protein-coding genes with HSAT5 satellite repeat expression.

Funding: Not applicable.
==== Refs
References

Apridita Sebastian W Shiraishi H Shimizu N Umeda R Lai S 2022 Ankle2 deficiency-associated microcephaly and spermatogenesis defects in zebrafish are alleviated by heterozygous deletion of vrk1 Biochemical and Biophysical Research Communications 624 95 101 10.1016/j.bbrc.2022.07.070 35940133
Arcella A Limanaqi F Ferese R Biagioni F Oliva MA 2020 Dissecting molecular features of gliomas: genetic loci and validated biomarkers International Journal of Molecular Sciences 21 2 10.3390/ijms21020685
Baghbani F Raoofian R Hasanzadeh NM Hamzehloei T Soukhtanloo M 2013 Identification of novel hypoxia response genes in human glioma cell line a172 Iranian Journal of Basic Medical Sciences, 16 5 675 82 23826488
Bao W Kojima KK Kohany O 2015 Repbase Update, a database of repetitive elements in eukaryotic genomes Mobile DNA 6 11 10.1186/s13100-015-0041-9 26045719
Balvers RK Kleijn A Kloezeman JJ French PJ Kremer A 2013 Serum-free culture success of glial tumors is related to specific molecular profiles and expression of extracellular matrix-associated gene modules Neuro-Oncology 15 12 1684 95 10.1093/neuonc/not116 24046260
Bersani F Lee E Kharchenko PV Xu AW Liu M 2015 Pericentromeric satellite repeat expansions through RNA-derived DNA intermediates in cancer Proceedings of the National Academy of Sciences of the United States of America 112 49 15148 15153 10.1073/pnas.1518008112
Bewick V Cheek L Ball J 2004 Statistics review 12: survival analysis Critical Care (London, England) 8 5 389 394 10.1186/cc2955 15469602
Bogdańska MU Bodnar M Piotrowska MJ Murek M Schucht P 2017 A mathematical model describes the malignant transformation of low grade gliomas: prognostic implications PloS One 12 8 e0179999 10.1371/journal.pone.0179999 28763450
Brat DJ Verhaak RGW Aldape KD Yung WKA Salama SR 2015 Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas The New England Journal of Medicine 372 26 2481 2498 10.1056/NEJMoa1402121 26061751
Bredel M Bredel C Juric D Harsh GR Vogel H 2005 Functional network analysis reveals extended gliomagenesis pathway maps and three novel MYC-interacting genes in human gliomas Cancer Research 65 19 8679 8689 10.1158/0008-5472.CAN-05-1204 16204036
Brito C Azevedo A Esteves S Marques AR Martins C 2019 Clinical insights gained by refining the 2016 WHO classification of diffuse gliomas with: EGFR amplification, TERT mutations, PTEN deletion and MGMT methylation BMC Cancer 19 1 968 10.1186/s12885-019-6177-0 31623593
Cantone I Fisher AG 2013 Epigenetic programming and reprogramming during development Nature Structural & Molecular Biology, 20 3 282 289 10.1038/nsmb.2489
Chai RC Li YM Zhang KN Chang YZ Liu YQ 2019 RNA processing genes characterize RNA splicing and further stratify lower-grade glioma JCI Insight 5 17 e130591 10.1172/jci.insight.130591 31408440
Chehade G Ruiz-Reig N Raftopoulos C Tajeddine N Tissir F 2022 P10.02.B The fragile balance between proliferation and neoplasia in neural stem cells: role of DIAPH3 in glioblastoma Neuro-Oncology 24 Issue Supplement_2 ii48 10.1093/neuonc/noac174.167
Chen HY Lin LT Wang ML Laurent B Hsu CH 2017 Musashi-1 enhances glioblastoma cell migration and cytoskeletal dynamics through translational inhibition of Tensin3 Scientific Reports 7 1 8710 10.1038/s41598-017-09504-7 28821879
Chomiak AA Guo Y Kopsidas CA McDaniel DP Lowe CC 2022 Nde1 is required for heterochromatin compaction and stability in neocortical neurons IScience 25 6 104354 10.1016/j.isci.2022.104354 35601919
Crespo I Vital AL Gonzalez-Tablas M Patino MDC Otero A 2015 Molecular and Genomic Alterations in Glioblastoma Multiforme The American Journal of Pathology 185 7 1820 1833 10.1016/j.ajpath.2015.02.023 25976245
Domingues P González-Tablas M Otero ÁPascual D Miranda D 2016 Tumor infiltrating immune cells in gliomas and meningiomas Brain, Behavior, and Immunity 53 1 15 10.1016/j.bbi.2015.07.019 26216710
Eckel-Passow JE Lachance DH Molinaro AM Walsh KM Decker PA 2015 Glioma Groups Based on 1p/19q, IDH, and TERT Promoter Mutations in Tumors The New England Journal of Medicine 372 26 2499 2508 10.1056/NEJMoa1407279 26061753
Erson-Omay EZ Çağlayan AO Schultz N Weinhold N Omay SB 2015 Somatic POLE mutations cause an ultramutated giant cell high-grade glioma subtype with better prognosis Neuro-Oncology 17 10 1356 64 10.1093/neuonc/nov027 25740784
Frankish A Diekhans M Jungreis I Lagarde J Loveland JE 2021 GENCODE 2021 Nucleic Acids Research 49 D1 D916 D923 10.1093/nar/gkaa1087 33270111
Gabrusiewicz K Rodriguez B Wei J Hashimoto Y Healy LM 2016 Glioblastoma-infiltrated innate immune cells resemble M0 macrophage phenotype JCI Insight 1 2 10.1172/jci.insight.85841
Garcia-Fabiani MB Haase S Comba A Carney S McClellan B 2021 Genetic Alterations in Gliomas Remodel the Tumor Immune Microenvironment and Impact Immune-Mediated Therapies Frontiers in Oncology 11 631037 10.3389/fonc.2021.631037 34168976
Garrett MC Albano R Carnwath T Shah S Woo D 2022 Chromatin structure predicts survival in glioma patients Scientific Reports 12 1 8221 10.1038/s41598-022-11019-9 35581287
de Groot JF Lamborn KR Chang SM Gilbert MR Cloughesy TF 2011 Phase II study of aflibercept in recurrent malignant glioma: a North American Brain Tumor Consortium study Journal of Clinical Oncology 29 19 2689 95 10.1200/JCO.2010.34.1636 21606416
Han J Yu X Wang S Wang Y Liu Q 2022 IGF2BP2 induces U251 glioblastoma cell chemoresistance by inhibiting FOXO1-mediated PID1 expression through stabilizing lncRNA DANCR Frontiers in Cell and Developmental Biology 9 659228 10.3389/fcell.2021.659228 35141227
He Z Ruan X Liu X Zheng J Liu Y 2019 FUS/circ_002136/miR-138-5p/SOX13 feedback loop regulates angiogenesis in Glioma Journal of Experimental & Clinical Cancer Research 38 1 65 10.1186/s13046-019-1065-7 30736838
Helman E Lawrence MS Stewart C Sougnez C Getz G 2014 Somatic retrotransposition in human cancer revealed by whole-genome and exome sequencing Genome Research 24 7 1053 1063 10.1101/gr.163659.113 24823667
Hillier LW Graves TA Fulton RS Fulton LA Pepin KH 2005 Generation and annotation of the DNA sequences of human chromosomes 2 and 4 Nature 434 7034 724 731 10.1038/nature03466 15815621
Hubley R Finn RD Clements J Eddy SR Jones TA 2016 The Dfam database of repetitive DNA families Nucleic Acids Research 44 D1 D81 9 10.1093/nar/gkv1272 26612867
Hwang T Kim S Chowdhury T Yu HJ Kim KM 2022 Genome-wide perturbations of Alu expression and Alu-associated post-transcriptional regulations distinguish oligodendroglioma from other gliomas Communications Biology 5 1 62 10.1038/s42003-022-03011-w 35042936
International Human Genome Sequencing Consortium 2001 Initial sequencing and analysis of the human genome Nature 409 6822 860 921 10.1038/35057062 11237011
Jagannathan M Yamashita YM 2017 Function of junk: pericentromeric satellite DNA in chromosome maintenance Cold Spring Harbor Symposia on Quantitative Biology 82 319 327 10.1101/sqb.2017.82.034504 29610245
Juratli TA Stasik S Zolal A Schuster C Richter S 2018 TERT promoter mutation detection in cell-free tumor-derived DNA in patients with IDH wild-type glioblastomas: a pilot prospective study Clinical Cancer Research: An Official Journal of the American Association for Cancer Research 24 21 5282 5291 10.1158/1078-0432.CCR-17-3717 29941484
Karakülah G Yandım C 2020 Signature changes in the expressions of protein-coding genes, lncRNAs, and repeat elements in early and late cellular senescence Turkish Journal of Biology 44 6 356 370 10.3906/biy-2005-21 33402863
Karakülah G Yandım C 2021 Identification of differentially expressed genomic repeats in primary hepatocellular carcinoma and their potential links to biological processes and survival Turkish Journal of Biology 45 5 599 612 10.3906/biy-2104-13 34803457
Khan AB Lee S Harmanci AS Patel R Latha K 2023 CXCR4 expression is associated with proneural-to-mesenchymal transition in glioblastoma International Journal of Cancer 152 4 713 724 10.1002/ijc.34329 36250346
Kishikawa T Otsuka M Yoshikawa T Ohno M Ijichi H 2016 Satellite RNAs promote pancreatic oncogenic processes via the dysfunction of YBX1 Nature Communications 7 13006 10.1038/ncomms13006
Kong Y Rose CM Cass AA Williams AG Darwish M 2019 Transposable element expression in tumors is associated with immune infiltration and increased antigenicity Nature Communications 10 1 5228 10.1038/s41467-019-13035-2
Koso H Takeda H Yew CCK Ward JM Nariai N 2012 Transposon mutagenesis identifies genes that transform neural stem cells into glioma-initiating cells Proceedings of the National Academy of Sciences 109 44 E2998 E3007 10.1073/pnas.1215899109
Lan YL Wang X Xing JS Lou JC Ma XC 2017 The potential roles of dopamine in malignant glioma Acta Neurologica Belgica 117 3 613 621 10.1007/s13760-016-0730-2 27995487
Langfelder P Horvath S 2008 WGCNA: an R package for weighted correlation network analysis BMC Bioinformatics 9 559 10.1186/1471-2105-9-559 19114008
Langfelder P Zhang B Horvath S 2008 Defining clusters from a hierarchical cluster tree: the Dynamic Tree Cut package for R Bioinformatics 24 5 719 720 10.1093/bioinformatics/btm563 18024473
Lee E Iskow R Yang L Gokcumen O Haseley P 2012 Landscape of somatic retrotransposition in human cancers Science 337 6097 967 971 10.1126/science.1222077 22745252
Leinonen R Sugawara H Shumway M on behalf of the International Nucleotide Sequence Database Collaboration 2011 The sequence read archive Nucleic Acids Research 39 suppl_1 D19 D21 10.1093/nar/gkq1019 21062823
Li H Handsaker B Wysoker A Fennell T Ruan J 2009 The sequence alignment/map format and SAMtools Bioinformatics 25 16 2078 2079 10.1093/bioinformatics/btp352 19505943
Li J Zhu S Kozono D Ng K Futalan D 2014 Genome-wide shRNA screen revealed integrated mitogenic signaling between dopamine receptor D2 (DRD2) and epidermal growth factor receptor (EGFR) in glioblastoma Oncotarget 5 4 882 893 10.18632/oncotarget.1801 24658464
Li T Yang Z Li H Zhu J Wang Y 2022 Phospholipase Cγ1 (PLCG1) overexpression is associated with tumor growth and poor survivalIDH wild-type lower-grade gliomas in adult patients Laboratory Investigation 102 2 143 153 10.1038/s41374-021-00682-7 34697421
Li X Meng Y 2021 SUMOylation regulator-related molecules can be used as prognostic biomarkers for glioblastoma Frontiers in Cell and Developmental Biology 9 658856 10.3389/fcell.2021.658856 33898460
Liao Y Smyth GK Shi W 2014 featureCounts: an efficient general purpose program for assigning sequence reads to genomic features Bioinformatics 30 7 923 930 10.1093/bioinformatics/btt656 24227677
Lin J Lai X Liu X Yan H Wu C 2022 Pyroptosis in glioblastoma: a crucial regulator of the tumour immune microenvironment and a predictor of prognosis Journal of Cellular and Molecular Medicine 26 5 1579 1593 10.1111/jcmm.17200 35083859
Lin WW Ou GY Zhao WJ 2021 Mutational profiling of low-grade gliomas identifies prognosis and immunotherapy-related biomarkers and tumour immune microenvironment characteristics Journal of Cellular and Molecular Medicine 25 21 10111 10125 10.1111/jcmm.16947 34597473
Louis DN Perry A Reifenberger G von Deimling A Figurearella-Branger D 2016 The 2016 World Health Organization classification of tumors of the central nervous system: a summary Acta Neuropathologica 131 6 803 820 10.1007/s00401-016-1545-1 27157931
Lu J Li H Chen Z Fan L Feng S 2019 Identification of 3 subpopulations of tumor-infiltrating immune cells for malignant transformation of low-grade glioma Cancer Cell International 19 265 10.1186/s12935-019-0972-1 31632199
Mack SC Singh I Wang X Hirsch R Wu Q 2019 Chromatin landscapes reveal developmentally encoded transcriptional states that define human glioblastoma The Journal of Experimental Medicine 216 5 1071 1090 10.1084/jem.20190196 30948495
Mackay A Burford A Carvalho D Izquierdo E Fazal-Salom J 2017 Integrated molecular meta-analysis of 1,000 pediatric high-grade and diffuse intrinsic pontine glioma Cancer Cell 32 4 520 537 e5 10.1016/j.ccell.2017.08.017 28966033
Murat A Migliavacca E Gorlia T Lambiv WL Shay T 2008 Stem cell-related ‘self-renewal’ signature and high epidermal growth factor receptor expression associated with resistance to concomitant chemoradiotherapy in glioblastoma Journal of Clinical Oncology 26 18 3015 3024 10.1200/JCO.2007.15.7164 18565887
Nishibuchi G Déjardin J 2017 The molecular basis of the organization of repetitive DNA-containing constitutive heterochromatin in mammals Chromosome Research 25 1 77 87 10.1007/s10577-016-9547-3 28078514
Noushmehr H Weisenberger DJ Diefes K Phillips HS Pujara K 2010 Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma Cancer Cell 17 5 510 522 10.1016/j.ccr.2010.03.017 20399149
Ohka F Natsume A Motomura K Kishida Y Kondo Y 2011 The global DNA methylation surrogate LINE-1 methylation is correlated with MGMT promoter methylation and is a better prognostic factor for glioma PloS One 6 8 e23332 10.1371/journal.pone.0023332 21829728
Ostrom QT Gittleman H Liao P Vecchione-Koval T Wolinsky Y 2017 CBTRUS Statistical Report: Primary brain and other central nervous system tumors diagnosed in the United States in 2010–2014 Neuro-Oncology 19 suppl_5 v1 v88 10.1093/neuonc/nox158 29117289
Özgür E Mayer Z Keskin M Yörüker EE Holdenrieder S 2021 Satellite 2 repeat DNA in blood plasma as a candidate biomarker for the detection of cancer Clinica Chimica Acta 514 74 79 10.1016/j.cca.2020.12.008
Parsons DW Jones S Zhang X Lin JCH Leary RJ 2008 An integrated genomic analysis of human glioblastoma multiforme Science 321 5897 1807 1812 10.1126/science.1164382 18772396
Phillips HS Kharbanda S Chen R Forrest WF Soriano RH 2006 Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis Cancer Cell 9 3 157 173 10.1016/j.ccr.2006.02.019 16530701
Porokhovnik LN Veiko NN Ershova ES Kostyuk SV 2021 The role of human satellite III (1q12) copy number variation in the adaptive response during aging, stress, and pathology: a pendulum model Genes 12 10 10.3390/genes12101524
Probst AV Okamoto I Casanova M El Marjou F Le Baccon P 2010 A strand-specific burst in transcription of pericentric satellites is required for chromocenter formation and early mouse development Developmental Cell 19 4 625 638 10.1016/j.devcel.2010.09.002 20951352
Robinson MD McCarthy DJ Smyth GK 2010 edgeR: a bioconductor package for differential expression analysis of digital gene expression data Bioinformatics 26 1 139 140 10.1093/bioinformatics/btp616 19910308
Rošić S Köhler F Erhardt S 2014 Repetitive centromeric satellite RNA is essential for kinetochore formation and cell division The Journal of Cell Biology 207 3 335 349 10.1083/jcb.201404097 25365994
Royston P Choodari-Oskooei B Parmar MKB Rogers JK 2019 Combined test versus logrank/Cox test in 50 randomised trials Trials 20 1 172 10.1186/s13063-019-3251-5 30885277
Sahu D Chang YL Lin YC Lin CC 2021 Characterization of the survival influential genes in carcinogenesis International Journal of Molecular Sciences 22 9 10.3390/ijms22094384
Sakthikumar S Sakthikumar S Roy A Haseeb L Pettersson ME 2020 Whole-genome sequencing of glioblastoma reveals enrichment of non-coding constraint mutations in known and novel genes Genome Biology 21 1 127 10.1186/s13059-020-02035-x 32513296
Salhia B Tran NL Chan A Wolf A Nakada M 2008 The guanine nucleotide exchange factors trio, Ect2, and Vav3 mediate the invasive behavior of glioblastoma The American Journal of Pathology 173 6 1828 38 10.2353/ajpath.2008.080043 19008376
Sallam M Mysara M Baatout S Guns P-J Ramadan R 2022 Meta-analysis of RNA-seq datasets identifies novel players in glioblastoma Cancers 14 23 5788 10.3390/cancers14235788 36497269
Shah AH Govindarajan V Doucet-O’Hare TT Rivas S Ampie L 2022 Differential expression of an endogenous retroviral element [HERV-K(HML-6)] is associated with reduced survival in glioblastoma patients Scientific Reports 12 1 6902 10.1038/s41598-022-10914-5 35477752
Solovyov A Vabret N Arora KS Snyder A Funt SA 2018 Global Cancer Transcriptome Quantifies Repeat Element Polarization between Immunotherapy Responsive and T Cell Suppressive Classes Cell Reports 23 2 512 521 10.1016/j.celrep.2018.03.042 29642008
Song X Wan X Huang T Zeng C Sastry N 2019 SRSF3-regulated RNA alternative splicing promotes glioblastoma tumorigenicity by affecting multiple cellular processes Cancer Research 79 20 5288 5301 10.1158/0008-5472.CAN-19-1504 31462429
Stupp R Hegi ME Mason WP van den Bent MJ Taphoorn MJB 2009 Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial The Lancet Oncology 10 5 459 466 10.1016/S1470-2045(09)70025-7 19269895
Sturm D Bender S Jones DTW Lichter P Grill J 2014 Paediatric and adult glioblastoma: multiform (epi)genomic culprits emerge Nature Reviews Cancer 14 2 92 107 10.1038/nrc3655 24457416
Sun R Pan Y Mu L Ma Y Shen H 2021 Development of a 3 RNA binding protein signature for predicting prognosis and treatment response for glioblastoma multiforme Frontiers in Genetics 12 768930 10.3389/fgene.2021.768930 34733320
Teng C Zhu Y Li Y Dai L Pan Z 2022 Recurrence- and malignant progression-associated biomarkers in low-grade gliomas and their roles in immunotherapy Frontiers in Immunology 13 10.3389/fimmu.2022.899710
Tilak M Alural B Wismer SE Brasher MI New LA 2021 Adaptor protein ShcD/SHC4 interacts with Tie2 receptor to synergistically promote glioma cell invasion Molecular Cancer Research 19 5 757 770 10.1158/1541-7786.MCR-20-0188 33495401
Ting DT Lipson D Paul S Brannigan BW Akhavanfard S 2011 Aberrant overexpression of satellite repeats in pancreatic and other epithelial cancers Science 331 6017 593 596 10.1126/science.1200801 21233348
Ulitsky I Shkumatava A Jan CH Sive H Bartel DP 2011 Conserved function of lincRNAs in vertebrate embryonic development despite rapid sequence evolution Cell 147 7 1537 1550 10.1016/j.cell.2011.11.055 22196729
Vos LJ Famulski JK Chan GKT 2006 How to build a centromere: from centromeric and pericentromeric chromatin t kinetochore assembly Biochemistry and Cell Biology 84 4 619 639 10.1139/o06-078 16936833
Wang H Wang X Xu L Zhang J Cao H 2020 Prognostic significance of age related genes in patients with lower grade glioma Journal of Cancer 11 13 3986 3999 10.7150/jca.41123 32328202
Wang L Shang Z Zhou Y Hu X Chen Y 2018 Autophagy mediates glucose starvation-induced glioblastoma cell quiescence and chemoresistance through coordinating cell metabolism, cell cycle, and survival Cell Death & Disease 9 2 213 10.1038/s41419-017-0242-x 29434213
Wang Q Hu B Hu X Kim H Squatrito M 2017 Tumor evolution of glioma-intrinsic gene expression subtypes associates with immunological changes in the microenvironment Cancer Cell 32 1 42 56 e6 10.1016/j.ccell.2017.06.003 28697342
Wang X Zhang L Song Y Jiang Y Zhang D 2021 MCM8 is regulated by EGFR signaling and promotes the growth of glioma stem cells through its interaction with DNA-replication-initiating factors Oncogene 40 27 4615 4624 10.1038/s41388-021-01888-1 34131285
Wang J Cazzato E Ladewig E Frattini V Rosenbloom DIS 2016 Clonal evolution of glioblastoma under therapy Nature Genetics 48 7 768 776 10.1038/ng.3590 27270107
Wang J Huang TYT Hou Y Bartom E Lu X 2021 Epigenomic landscape and 3D genome structure in pediatric high-grade glioma Science Advances 7 23 10.1126/sciadv.abg4126
Wolf D Hofbrucker-MacKenzie SA Izadi M Seemann E Steiniger F 2019 Ankyrin repeat-containing N-Ank proteins shape cellular membranes Nature Cell Biology 21 10 1191 1205 10.1038/s41556-019-0381-7 31548610
Xiao F Shen J Zhou L Fang Z Weng Y 2022 ZNF395 facilitates macrophage polarization and impacts the prognosis of glioma American Journal of Cancer Research 12 9 4312 4325 https://pubmed.ncbi.nlm.nih.gov/36225626/ 36225626
Xu Y Geng R Yuan F Sun Q Liu B 2019 Identification of differentially expressed key genes between glioblastoma and low-grade glioma by bioinformatics analysis PeerJ 7 e6560 30867991
Yandım C Karakülah G 2019a Expression dynamics of repetitive DNA in early human embryonic development BMC Genomics 20 1 439 10.1186/s12864-019-5803-1 31151386
Yandım C Karakülah G 2019b Dysregulated expression of repetitive DNA in ER+/HER2− breast cancer Cancer Genetics 239 36 45 10.1016/j.cancergen.2019.09.002 31536958
Yandım C Karakülah G 2022 Repeat expression is linked to patient survival and exhibits single nucleotide variation in pancreatic cancer revealing LTR70:r.879A>G Gene 822 146344 10.1016/j.gene.2022.146344 35183687
Yang Q Jiang N Zou H Fan X Liu T 2022 Alterations in 3D chromatin organization contribute to tumorigenesis of EGFR-amplified glioblastoma Computational and Structural Biotechnology Journal 20 1967 1978 10.1016/j.csbj.2022.04.007 35521558
Yu G Wang LG Han Y He QY 2012 clusterProfiler: an R package for comparing biological themes among gene clusters Omics: A Journal of Integrative Biology 16 5 284 287 10.1089/omi.2011.0118 22455463
Yuan Z Yang Y Zhang N Soto C Jiang X 2021 Human endogenous retroviruses in glioblastoma multiforme Microorganisms 9 4 10.3390/microorganisms9040764
Zhang B Horvath S 2005 A general framework for weighted gene co-expression network analysis Statistical Applications in Genetics and Molecular Biology 4 Article17 10.2202/1544-6115.1128 16646834
Zhang D Zhao J Han C Liu X Liu J 2020 Identification of hub genes related to prognosis in glioma Bioscience Reports 40 5 10.1042/BSR20193377
Zhang L Wang Y Rashid MH Liu M Angara K 2017 Malignant pericytes expressing GT198 give rise to tumor cells through angiogenesis Oncotarget 8 31 51591 51607 10.18632/oncotarget.18196 28881671
Zhu Q Pao GM Huynh AM Suh H Tonnu N 2011 BRCA1 tumour suppression occurs via heterochromatin-mediated silencing Nature 477 7363 179 184 10.1038/nature10371 21901007
