CRISPR-cas9 Screening Identified Lethal Genes Enriched in Hippo Kinase Pathway and of Prognosis Significance in primary Low-grade Glioma

doi:10.21203/rs.3.rs-2116553/v1

Download PDF

Research article

CRISPR-cas9 Screening Identified Lethal Genes Enriched in Hippo Kinase Pathway and of Prognosis Significance in primary Low-grade Glioma

https://doi.org/10.21203/rs.3.rs-2116553/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 14 May, 2023

Read the published version in Molecular Medicine →

You are reading this latest preprint version

Background

Lethal genes in low-grade gliomas (LGGs) may have an essential prognostic significance and therefore need to be systematically analyzed. The purpose of this study is to analyze essential genes correlated with cell survival by evaluating CRISPR-cas9 screening data, leading to the identification of novel treatment targets for patients.

Methods

In this study, genes related to cell viability from the Depmap CRISPR-cas9 screen were intersected to differentially expressed genes (DEGs) between tumor and GTEx normal tissues from TCGA. The LASSO regression method was used to construct a signature that used to anticipate overall survival in patients with LGG. An evaluation of the signature was carried out using both multivariate and univariate Cox regression. Then, we determined which key pathways were modulated by this signature by comparing DEGs between low- and high-risk patients. WGCNA was conducted to identify modules associated with high-risk. In this study, we also performed enrichment analysis to identify pathways mediated by DEGs, overlapping genes, and genes shared in the WGCNA. Finally, we used to western blot, qRT-PCR and IHC to detect the expression of hub genes.

Results

Using the TCGA database, 145 oncogenes were identified as overexpressed. These genes were intersected with lethal genes identified in the Depmap database, which are enriched in Hippo pathways. A total of 19 genes were used to construct the gene signature by means of LASSO regression. The median risk score (0.752324) was employed to distinguish between low- and high-risk patients. The patients with high-risk characteristics showed a shorter OS duration in the internal training, the internal validation, and the external validation datasets. Ultimately, the Hippo signaling pathway was the predominantly enriched pathway in 145 genes, DEGs, and 3 modular genes in WGCNA. Finally, we found differences of hub genes expression in different clinical samples. Importantly, protein and mRNA expression of REP65 was significantly up-regulated in tumor cells both in the public cohort and our cohort.

Conclusion

The hippo signaling pathway detected based on CRISPR-cas9 screening is a critical regulator of viability and tumor proliferation and therefore is an innovative new target for treating cancerous brain tumors, including LGG.

CRISPR-cas9 screening

low-grade glioma

Hippo signaling pathway

prognosis

signature

WGCNA

Low-grade gliomas (LGG) have been demonstrated as being among the most prevalent primary tumors affecting the central nervous system and consist of WHO grade II and III gliomas(Louis et al., 2007; Li et al., 2021). Although molecular features including isotope dehydrogenase 1 and 2 genes (IDH1/IDH2), PTEN, EGFR, ATRX, TPP53, coding status of chromosome arms 19q and 1p, Chr 7 gain/Chr 10 loss, Chr19/20 co-gain, have significantly distinguished different classes of LGG(Chiang et al., 2020).LGG has significant heterogeneity that hinders improved patient outcomes(Wang et al., 2021). Until now, several approaches have been used to treat gliomas, such as lytic virus therapy, targeted therapy, immunotherapy, chemotherapy, radiation therapy, and surgery, but clinical outcomes for LGG patients have not significantly improved (Cai et al., 2020). Therefore, it is essential to improve the efficacy of treatment for patients with LGG. In addition to conventional therapies, there is growing interest in the emerging CRISPR/Cas9 gene-editing system.

CRISPR/Cas9 system is widely found in prokaryotic genomes and is an acquired immune defense mechanism that has evolved in bacteria and archaea in response to viral and plasmid invasion(Louradour et al., 2019; Usman et al., 2020). The CRISPR/Cas9 system mainly consists of the Cas9 protein and single-stranded guide RNA (sgRNA)(Peng et al., 2018). The Cas9 protein acts as a guide to cut the DNA double-strand, and the sgRNA acts as a guide, under the guidance of sgRNA. The Cas9 protein can cut the different target sites through the principle of base complementary pairing to achieve the double-strand break of DNA(Moses et al., 2019; Yang et al., 2020).In addition CRISPR/Cas9 can also be used for gene expression regulation (transcriptional activation/repression), epigenetic modifications, and genomic imaging(Amjad et al., 2020; Khanzadi and Khan, 2020). This application relies on the resolution of the Cas9 protein structure(Gangopadhyay et al., 2019). Cas9, a multifunctional protein, possesses two nuclease structural domains, HNH and RuvC. The HNH domain cuts the DNA strand complementarily paired with crRNA, and the RuvC domain cuts the other strand of double-stranded DNA(Stovicek et al., 2017; Wu et al., 2020). Cas9 becomes a single-stranded cleaved protein if one of the two structural domains is mutated, and if both are mutated, Cas9 becomes a protein with only NDA-binding activity(Young et al., 2019). When dCas9 binds to the coding region or promoter region of a gene, it affects the activity of RNA polymerase and thus transcription, a method also known as CRISPRi(Mahas et al., 2018).In humans and yeast, if the dCas9 protein is expressed in fusion with VP64 or KRAB, it brings about transcriptional activation and transcriptional repression, respectively(Wen et al., 2016; Kleinjan et al., 2017).

Genome-wide knockdown technologies developed based on the CRISPR/Cas9 system are making their mark in numerous areas of oncology research. These technologies contribute to the comprehension of the impact of knocking out established genes on biological phenotypes on a genome-wide scale. Recently, it was shown that CRISPR-Cas9-mediated knockdown of the TIM3 gene in human natural killer cells enhanced the growth inhibition of human glioma cells(Morimoto et al., 2021). In addition, it was discovered that the CRISPR/Cas9 system specifically targets EGFR exon 17, resulting in the inhibition of the activation of NF-kB by epigenetically modulating UBXN1 in EGFRwt/vIII glioma cells. Thus, this mechanism indicates that CRISPR/Cas9 is a viable treatment modality for GBM patients with EGFR mutations and EGFR amplification (Huang et al., 2017). Moreover, it was also found in mice with glioblastoma and injected with CRISPR-LNP targeting PLK1, an enzyme essential for cell division, which successfully caused apoptosis in tumor cells after editing the gene encoding this enzyme within the tumor cells. Therefore, the results showed that mice with a single intracerebral injection of CRISPR-LNP compared to the control group had a gene-editing efficiency of 70%. In addition, the median survival of the mice increased from 32.5 days to more than 48 days, and 30% of the mice survived for at least 60 days, while the control mice at 40 days were all dead (Rosenblum et al., 2020). Compared with pooled guide RNA libraries, cas9 may be used in a high-throughput method to filter for genes associated with specific biochemical phenotypes or illnesses(Esvelt et al., 2014). This "phenotype to genotype" strategy plays a role in modifying gene expression by choosing cells that exhibit the phenotype of interest(Pereira and Weinshilboum, 2009), followed by sequence analysis of the desired perturbations, which may identify genes that are involved in cell viability. In parallel, to determine the influence of single-gene knockouts on cell survival, a large-scale cancer-dependent loss-of-function screen was carried out in several well-characterized cancer cell lines. The Cancer Dependency Map (DepMap) website provides these data sets.

The Hippo pathway is a significantly conserved cascade signaling pathway in drosophila and mammals that plays a role in diverse biological activities, such as cell viability, proliferation, apoptosis, differentiation, cell fate determination, and tissue and organ size and homeostasis through the regulation of key target genes(Clattenburg et al., 2015; Seo et al., 2020). The abnormal signaling of the Hippo pathway is also implicated in multiple pathologies, including cancer and immunity-related diseases(Merrick et al., 2019). YAP/TAZ is the predominant transcription co-activator downstream of the Hippo pathway, shuttling between the cytoplasm and the nucleus (Shuttle)(Kim et al., 2020). The YAP/TAZ protein is the hub of the Hippo pathway, where a variety of upstream signaling molecules act directly or indirectly on YAP/TAZ, mainly to regulate the localization of YAP/TAZ, i.e., to regulate YAP/TAZ retention in the cytoplasm or nuclear localization (localization in the nucleus and interaction with corresponding transcriptional factors to modulate the target genes expression)(Le Corre et al., 2019). According to a previous study, NF2-deficient PRCC cancers that have lost the capacity of regulating the Hippo signaling pathway could be treated with dasatinib, which targets Yes in YAP-activated tumors and inhibits its expression(Sourbier et al., 2018). Another study showed that TFAP2C enhances CSCs’ properties and chemoresistance through transcriptional activation of ROCK1 and ROCK2, which are negative regulators of Hippo signaling, leading to deactivation of Hippo signaling in colorectal cancer (CRC) cells(Wang et al., 2018). Moreover, the YAP/TAZ-TEAD transcriptional factor complex is an application target for oncogenic transformation. The YAP locus is shown to be upregulated at different frequencies in human and mouse tumors, such as medulloblastoma, lung, pancreatic, esophageal, hepatocellular, and breast cancers(Pan, 2010). In LGG, LATS2 suppresses the proliferative and metastatic ability of cells via the Hippo signaling pathway(Guo et al., 2019). These studies suggest that the Hippo signaling pathway exerts a critical role in treating cancer.

The biological mechanisms involved in cell viability are complicated. Nonetheless, the cellular vulnerability of LGG has not been investigated in-depth in a systematic manner. Furthermore, the pathways via which these genes function and their prognostic value, have never been fully investigated. Thus, this study aims to identify genes that are differently expressed in tumor tissues and make remarkable contributions to cell viability. Based on the discovered genes, a predictive model with prognostic significance was constructed and validated. Additionally, the pathways, as well as biological activities that were modulated by these genes, were investigated in depth.

2.1 Determination of Vulnerability and Viability

Gene expression, mutation, and copy number data were downloaded from the Depmap database for all 16 primary non-metastatic glioma cell lines, namely CAS1, DKMG, GB1, HS683, KALS1, KNS60, KNS81, LN18, M059K, SF295, SNU201, T98G, U118MG, U251MG U87MG, and YKG1 RNAseq data, which were the result of multiple loss-of-function genomic screens in various cell lines. CERES program package was used for the purpose of calculating the dependency scores, (Meyers et al., 2017) aiming to detect genes that are critical for the ability of the cells lines to proliferate and survive. The relevant CERES input files were downloaded from the Depmap database. Notably, a negative value for a gene score implies that silencing it suppresses the survival condition of a cell line, whereas a positive value demonstrates that silencing it enhances proliferation and survival. The growth inhibitor and growth promotor genes were defined using thresholds of 0.5 and − 0.5, respectively. RNA-Seq data of 491 LGG samples from the TCGA database (https://portal.gdc.cancer.gov) with WHO classification II -III ((Level-3 HTseq-FPKM) and 103 normal cortical samples from the GTEx project were included in this study. These were used as normal sample controls, and after excluding non-coding RNA, differential analysis was performed. In addition, 481 samples with duplicate sequencing, no survival status, overall survival (OS) time < 1 day, no clear WHO classification, and non-primary LGG were excluded for subsequent modeling. DEGs between tumor and normal patients in the TCGA-LGG cohort were calculated using the limma package. Adjusted p-value < 0.05 and absolute fold changes > 1 were used as thresholds for selecting DEGs. Growth inhibitor genes from Depmap were intersected with down-modulated genes obtained from the TCGA-LGG dataset, whereas growth promoter genes were intersected with up-modulated genes.

2.2 Processing of Data in Low-grade glioma

The 481 samples from the TCGA-LGG database were randomly split 7:3 using the caret package, with TCGA-LGG-1 as the modeling cohort. TCGA-LGG-2 was used as the internal validation cohort. Meanwhile, the GSE16011 dataset based on the GPL570 platform was extracted from the GEO database (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc= GSE16011), and 80 of the LGG patients with WHO classification II -III were retained as the external validation group (GEO-LGG). In addition, the CGGA database (http://www.cgga.org.cn/download.jsp) was used to download the CGGA-693 project (CGGA-LGG-1, 332 patients), and the CGGA-325 project (CGGA-LGG-2, 162 patients) containing complete survival information of LGG patients as the external validation group. In summary, the TCGA-LGG-1 cohort was used for modeling, the TCGA-LGG-2 cohort for internal validation, and the GEO-LGG, CGGA-LGG-1, and the CGGA-LGG-2 cohorts for external validation. In addition, the sva package was used to perform background correction, normalization, and expression calculation for the genes involved in the modeling, thus ensuring comparability of validation.

2.3 LASSO Regression Analysis

In the TCGA-LGG cohort, the patients were classified into two datasets at random with a 7:3 ratio, namely the internal training and internal validation datasets. Risk models were constructed with the aid of the LASSO model, aiming to eliminate highly correlated genes. By incorporating gene expression values that their LASSO -Cox coefficients have weighted, risk score formulas were developed. The prognostic significance of the risk scores was evaluated by performing univariate and multivariate Cox regression analyses using the entire dataset as well as the external validation dataset. Time-dependent subject operating characteristic (tROC) curves were used to evaluate the prediction performance of risk scores with conventional clinical and pathological characteristics. For the purpose of plotting ROC curves and calculating the area under the curve (AUC), the survROC computer package was utilized.

2.4 Enrichment Analysis

Gene Ontology (GO) is a classification system that is used for annotating cell components, molecular functions, and biological processes. Gene pathways were annotated using the Kyoto Encyclopedia of Genes and Genomes (KEGG). The clusterProfiler software was employed to conduct the GO and KEGG analyses. The q-value < 0.05 and p-value < 0.05 were established as the criteria for identifying significantly enriched pathways.

2.5 Creation of the PPI Network and Detection of Hub

The STRING website provides information about available and anticipated protein-protein interactions, aiding the creation of PPI networks. Subsequently, the overlapping gene interactions were visualized with the aid of the Cytoscape software.

2.6 Weighted Correlation Network Analysis

In the WGCNA analysis, all DEGs satisfying the p-value < 0.05 in normal and tumor samples in the CGGA-LGG-1 cohort were used as input. Each sample was clustered well, with only a single outlier sample excluded using a shear line of 120 as the threshold. Subsequently, a soft threshold from 1 to 20 was used for topology calculation, and the optimal soft threshold was determined to be 10. After converting the correlation matrix into an adjacency matrix according to the soft threshold, it was then converted into a topological overlap matrix (TOM) for mean linkage hierarchical clustering. Furthermore, the associated modules were subjected to classification based on the TOM, with the number of genes in each module not below 50. In the present research, the shear height of the gene modules was 0.2, and similar modules were integrated. In addition, GS and MM within each module were calculated for scatter plots. Finally, the correlation between the merged modules and the high-risk population was obtained with the help of the Pearson method.

2.7 Vitro assays in clinical samples

In our study, we collected the 10 LGG tissues and 10 normal brain tissues were prepared for extracting RNA and preotein in qRT-PCR and western blot assays. Moreover, we enrolled another 10 LGG tissues and 10 intractable epilepsy tissues for IHC section. The study of human samples was approved by the Medical Ethical Committee of the first affiliated hospital of xinjiang medical university. The detailed protocol about western blot, qRT-PCR and IHC were displayed in previous references. In brief, paraffin-embedded tissue sections were baked, dehydrated, and subjected to antigen repair on tissue slides. Then, endogenous peroxidase was inactivated with 3% hydrogen peroxide for 15 minutes. The non-specific antigen was blocked using 5% BSA for 30 minutes at room temperature and incubated with primary antibody overnight at 4°C. The next day, slides are incubated with the cognate secondary antibody for 1 hour at room temperature. Then, DAB staining, hematoxylin staining, followed by dehydration in gradient ethanol and clear treatment of slides in xylene. For western blot, proteins from different tissues were extracted with RIPA buffer, then separated on SDS/PAGE and transferred to PVDF membranes, incubated overnight with the corresponding primary antibodies and incubated with HRP-conjugated secondary antibodies, and finally the luminescence signal was detected by ECL. For qRT-PCR, total RNA was extracted from the tissues using TRIzol and then converted to cDNA using Servicebio®RT First Strand cDNA Synthesis Kit. Real-time PCR was performed using SYBR Green qPCR Master Mix (None ROX). The following antibodies were used: GAPDH (1:2000, GB15002, Servicebio), SOX9 (1:1000, 67439-1-IG, proteintech), RPE65 (1:1000, 17939-1-AP, proteintech), LSM2 (1:1000, 46289, SAB). GAPDH was used for normalization. ImageJ software was used to evaluate and quantify the gray value for 20 samples. Relative mRNA expression was determined based on CT values and normalized by the GAPDH expression levels. Sequences of primers were as follows:

GAPDH forward: 5'-GGAAGCTTGTCATCAATGGAAATC-3', GAPDH reverse: 5'-TGATGACCCTTTTGGCTCCC-3';

SOX9 forward: 5'-GTCAACGGCTCCAGCAAGAA-3', SOX9 reverse: 5'-CGTTCTTCACCGACTTCCTCC-3';

RPE65 forward: 5'-TGGGCCAGGACTCTTTGAAG-3', RPE65 reverse: 5'-TGCGGATGAACCTTCTGTGG-3';

LSM2 forward: 5'-TCGTGGAACTAAAGAATGACCTGA-3', LSM2 reverse: 5'-CATCCTGTAGCAACTGTGTGTCG-3'.

2.8 Data Analysis

The Kaplan-Meier analysis, as well as the Log-rank test, were conducted to perform the survival analysis. The p-value < 0.05 was established as the criterion for determining statistical significance. The survROC software was employed for the purpose of calculating the time-dependent AUC values.

3.1 Determination of Functional Genomic Genes in low-grade glioma

Cell lines of 16 primary gliomas were scored for dependency. The genes with dependency scores below − 0.5 in all glioma cell lines intersected with genes that were up-modulated in TCGA, and 145 genes were found. Moreover, genes whose dependency scores were above 0.5 in all glioma cell lines were intersected with genes downmodulated in TCGA, yielding only six genes (Fig. 1A). The 145 genes identified were designated as the oncogenes. Figure 1B depicts the scores assigned to oncogenes according to their dependence on other genes. Heat map only shows the Top 10 differential genes expressed in TCGA (Fig. 1C). The three most enriched pathways in GO analysis were spindle organization, spliceosomal complex, and ribonucleoprotein complex binding (Fig. 1D). The PPI network shows that these proteins contain 150 nodes with 794 edges between them (Fig. 1E). Overall, the mutation rates of all these genes were low, with only PDGFRA, SRBD1, CCDN1, CDC73, and ACLY having > 1% mutation frequency among the 145 oncogenes (Figure S1A), and no significant mutations in the other six intersecting genes (Figure S1B). Interestingly, the findings obtained from KEGG analysis demonstrated enrichment of these oncogenes in the cell cycle, Hippo signaling pathway, and other pathways (Fig. 1F).

3.2 Creation and Verification of the Gene Signature

The sva program package was used for background correction, normalization, and expression calculation for the genes involved in their modeling (Figure S2). With the aid of the Lasso regression model, a sum of 19 genes was filtered and used to construct risk equations. Supplementary Table 1 contains the coefficients derived. In the internal training dataset, a median risk score of 0.752324 was used as the threshold to classify patients into low- and high-risk groups, and the same threshold was used in the other validation datasets. In the modeling set, the patients’ OS was found to be remarkably improved in the low-risk group as opposed to that of the high-risk patients. In Fig. 2, these result indicates that the prognostic marker of risk score was valid (p-value < 0.001). The analytic results of the survival curves demonstrated that the LGG patients with a high-risk profile had a 3-year survival rate of about 56.34% with a 95% CI confidence interval of [47.18%-67.3%]. LGG patients with a low-risk profile had a 3-year survival rate of about 97.2% with a 95% CI of [93.9%-100%]; In the high-risk LGG patients, the 5-year survival rate was found to be about 37.58% with 95% CI [27.45%-51.4%], much lower than that of the low-risk LGG patients who had a rate of about 89.3% with 95% CI confidence interval [78.86%-100%]; For LGG patients with a high-risk profile, the 10-year survival rate was found to be about 12.54% with 95% CI [5%-31.4%], whereas that of the low-risk LGG patients was about 65.1% with a 95% CI of [43.50%-97.3%]. The AUC for 1, 3, and 5 years was 0.865, 0.910, and 0.874, respectively, according to the analysis results of the ROC curves for the different periods. Validation of the above results was performed in both the internal and external validation datasets. As anticipated, in the internal validation dataset, the findings recorded from the Kaplan-Meier curve demonstrated a short OS duration in patients belonging to the high-risk group as opposed to patients belonging to the low-risk group, with a statistically significant difference (p-value < 0.05). The analytical findings from the survival curves demonstrated that the high-risk LGG patients exhibited a 3 and 5-year survival of about 60.3% and 36.3%, respectively. In addition, 95% CI was [47.2%-76.9%] and [20.4%-64.5%], respectively. In low-risk LGG patients, the survival rates were about 89.1% and 62.6% at 3 and 5 years, respectively. 95% CI were [79.2%-100%] and [44.0%-89.0%], respectively. The analysis of ROC curves over 1, 3, and 5 years generated AUC values of 0.861, 0.878, and 0.690, respectively.

In the CGGA-LGG-1 cohort, the results obtained by the survival curves demonstrated a remarkably shortened OS duration in the high-risk patients as opposed to that of patients belonging to the low-risk group (p-value < 0.001). For the high-risk LGG patients, the survival rates at 3 and 5 years were about 59.4% and 46.4%, respectively. Moreover, 95% CI was [52.6%-67.0%] and [39.2%-54.9%], respectively. For the low-risk patients, the survival rates at 3 and 5 years were about 74.9% and 66.9%, respectively. Moreover, 95% CI were [68.1%- 82.4%] and [59.3%-75.5%]. Overall, the AUC values recorded from the ROC curve analysis of the CGGA-LGG-1 cohort for 1, 3, and 5 years were 0.726, 0.702, and 0.703, respectively. Similarly, in the CGGA-LGG-2 cohort, high-risk patients exhibited an unfavorable OS in contrast with that of the low-risk patients (p-value < 0.001), with AUC values obtained from ROC curves for 1, 3, and 5 years being of 0.736, 0.767, and 0.743, respectively. Interestingly, no significant differences were observed in the GEO-LGG cohort between patients in the high- and low-risk groups in terms of OS (p-value = 0.051). Following the adjustment for clinical and pathological factors, Cox regression demonstrated that risk score independently served as a prognostic feature across the training set. In the entire training TCGA-LGG set (Fig. 3A, B, p-value < 0.001, HR: 1.060, 95% CI: 1.036–1.085) and the external validation CGGA-LGG set (Fig. 3C,D, p < 0.05, HR: 1.016, 95% CI: 1.003–1.029), the GEO-LGG set (Fig. 3E,F, p < 0.001, HR: 1.171, 95% CI: 1.074–1.276), the risk score was shown to independently function as prognostic feature. In the TCGA-LGG full set, differences in OS stratified by clinicopathological characteristics of the universals, i.e., high- and low-risk groups, were analyzed by stratification. For the low-risk cohort, the OS remained superior in the high-risk group (Fig. 4A-F) according to subgroups classified by sex, age, tumor grade, and mutational status of ATRX, EGFR, PTEN, TP53, and IDH (Figure S3).

In addition, the predictive efficacy of independent prognostic factors was predicted in multifactorial Cox regression in the CGGA cohort and TCGA cohort. In Fig. 5A, for the 3-year tROC curve, the AUC areas for age, risk score, and tumor grade were 0.487, 0.858, and 0.527respectively for the entire TCGA-LGG dataset. For the 5-year tROC curve, the AUC areas for risk score, age, and tumor grade in the whole TCGA-LGG dataset were 0.882, 0.485, and 0.435, respectively. For the 10-year tROC curve, the AUC areas for risk score, age, and tumor grade in the whole TCGA-LGG dataset were 0.797, 0.486, and 0.465, respectively. In Fig. 5B, for the 3-year tROC curve, the AUC areas for risk score, IDH mutation status, and chromosome 1p19q joint deletion in the externally validated CGGA-LGG dataset were 0.723, 0.354, and 0.371, respectively. For the 5-year tROC curve, the AUC areas for risk score, IDH mutation status, and 1p19q joint chromosome deletion in the externally validated CGGA-LGG dataset were 0.718, 0.366, and 0.340, respectively. For the 10-year tROC curve, the AUC areas for risk score, IDH mutation status, and 1p19q joint chromosome deletion in the externally validated CGGA-LGG dataset were 0.708, 0.394, and 0.318, respectively.

3.3 DEGs Between Low- and High-Risk Subgroups

328 DEGs were identified in the CGGA-LGG-1 dataset by calculating DEGs between the high- and low-risk groups (Fig. 6A-B). These findings demonstrated an enrichment of the above DEGs in GO analysis for extracellular matrix structural constituent, collagen-containing extracellular matrix, and extracellular matrix organization (Fig. 6C). Moreover, they were also enriched in KEGG analysis in the p53 signaling pathway, Hippo pathway, and P13K-Akt signaling pathway (Fig. 6D). Interestingly, the Hippo pathway was shown to be a crucial mechanism in regulating cell viability both in our signature and in the DEG enrichment analysis.

3.4 WGCNA

The WGCNA method was adopted for the purpose of identifying hub genes in the CGGA-LGG-1 dataset. The soft threshold power was set as 10 to guarantee that the network was scale-free (Fig. 6E). With a cutoff height of 0.2, 16 clusters were clustered (Fig. 6F). It was discovered that different modules were correlated with the high-risk group (Fig. 6G), which included the green-yellow module (correlation coefficient = 0.29, p-value < 0.001), pink module (correlation parameter = 0.28, p-value < 0.001), and grey module (correlation coefficient = 0.34, p-value < 0.001). Genes from the three CGGA-LGG-1 datasets overlapped with 145 oncogenes, yielding a total of 13 genes (Fig. 6H). The topmost 3 enriched pathways identified by the KEGG enrichment analysis comprised the Hippo pathway, spliceosome, and Coronavirus Disease 2019 (COVID-19) (Fig. 7A). In GO analysis, the topmost 3 genes with high enrichment levels among these 13 genes were ribosome, ribosomal subunit, and structural constituent of ribosome (Fig. 7B). Among the 13 overlapping genes, three genes (SOX9, RPE65, LSM2) were enriched in the Hippo pathway.

3.5 Validation of SOX9, RPE65, LSM2 in clinical samples

To verify the reliability of the hub genes (SOX9, RPE65, LSM2) obtained from the bioinformatics analysis, we explored the abnormal expression of hub genes in LGG samples combined with normal cortical samples from the GTEx database. The results showed three hub genes were highly expressed in tumor samples (Fig. 7C). Moreover, we collected 10 LGG samples and 10 normal brain tissues to detect the expression level of hub genes by qRT-PCR and western blot. In the results of qRT-PCR, the mRNA level of RPE65 were highly expressed in LGG samples, however, there was no difference in the expression of other genes in different samples (Fig. 7D). In the results of western blot, the protein level of RPE65, LSM2 were highly expressed in LGG samples (Fig. 7E). Finally, we used IHC section to further detect the expression of SOX9, RPE65, LSM2 (Fig. 8). There is no doubt that the expression of the three genes was significantly different for both IHC in the public database and in our cohort. Importantly, protein and mRNA expression of RPE65 was significantly up-regulated in tumor cells in different cohorts.

The incidence of LGG is increasing worldwide (Zhang et al., 2020). LGG patients have a relatively favorable prognosis. However, these gliomas frequently advance to high-grade gliomas, which may cause reoccurrence or death due to malignant biological characteristics, including aggressive proliferation and radiotherapy resistance. Recent studies reported that LGG tumor markers are highly predictive of prognosis and successful treatment (Smith and Schwartz, 1984). These molecular markers are EGFR mutation status, PTEN mutation status, MGMT promoter methylation, TP53 mutation status, 1p/19q chromosome deletion, ATRX mutation status, and IDH1 mutation status (Roszkowski et al., 2016). The cause of LGG is characterized by various alterations in oncogenes. Crispr-cas9 screening, as a foundational technique, is an excellent approach to identify synthetic lethal genes in a systematic manner(Hewitt et al., 2021). The present research combined the findings of DepMap's low-grade glioma CRISPR-cas9 screen with the TCGA dataset, yielding a total of 145 oncogenes that are abundantly expressed and lethal to LGG. According to the results of the enrichment analysis, the Hippo signaling pathway was shown to be significantly enriched in these genes. In addition, a gene signature was constructed from the 145 genes, which might be used to classify patients into 2 groups, namely the high- and low-risk groups. Moreover, the Hippo signaling pathway was also significant when analyzing DEGs between high- and low-risk groups and genes. The intersection between the 145 oncogenes, WGCNA analysis, and DEGs also highlighted the Hippo pathway, indicating its importance for the viability of LGG.

In this study, three Hippo signaling pathway-related genes are mainly involved in the functional regulation of signaling pathways that inhibit cell growth, cancer genesis and metastasis, tissue regeneration, and variable shear. Furthermore, the mechanisms of these genes were investigated. SOX9 (SRY-Box Transcriptional Factor 9) is a Protein Coding gene that performs an integral role in encoding a protein that distinguishes the CCTTGAG sequence in addition to other members of the HMG box class of DNA binding proteins (Wang et al., 2007). It also plays a role in chondrocyte differentiation and is involved in the protein kinase activity and DNA-binding transcriptional factor (Ye et al., 2020). It has been demonstrated that SOX9 enhances the epithelial-mesenchymal transitions in gastric cancer cells via the mechanism of activating the Hippo-YAP signaling pathways(Zhou et al., 2019). In addition, the deletion of the YAP1 or SOX9 genes through CRISPR/Cas9 inhibited PPARδ-induced oncogenic activity. PPARδ collaborates with the hippo coactivator YAP1 to elevate the expression level of SOX9 and the progression of gastric cancer (Song et al., 2020). SOX9 expression is correlated with an unfavorable prognosis and TMZ resistance in GBM patients(Xu et al., 2018). Despite the fact that SOX2-SOX9 serves as an oncogenic axis that modulates the characteristics of stem cells and the resistance to chemotherapy. On the other hand, rapamycin attenuated the production of SOX protein, and it was found that combining rapamycin and temozolomide suppressed the progression of glioma in cells that expressed high levels of SOX2/SOX9(Garros-Regulez et al., 2016). In patients with glioma, a higher level of SOX9 expression is correlated with an unfavorable prognosis(Wang et al., 2012). The results show that SOX9 is essential to the survival of LGG. Further research discovered that Hippo coactivator YAP1 stimulates SOX9 expression in esophageal cancer cells, resulting in these cancer cells exhibiting stem cell-like characteristics(Song et al., 2014). SOX9 trans-activated long non-coding RNA NEAT1, which stimulates self-regeneration of hepatocellular carcinoma stem cells via the PKA/Hippo pathway(Cheng et al., 2021). Similarly, RPE65 (retinoic acid isomerase RPE65) is a protein-coding gene. The activity of RPE65 could be essential to the elimination of all-trans retinal, which are substrates for the synthesis of retinoic acid from skin cells. Therefore, a decrease in the cellular population of retinoic acid and its antecedent molecules may lead to altered gene regulation(Amann et al., 2012).In addition, deletion of retinal pigment epithelial extracellular signal-regulated kinase 1/2 leads to reduced RPE65 and retinal degeneration(Pyakurel et al., 2017). LSM2 (LSM2 homolog, U6 small nuclear RNA, and mRNA degradation-related) belongs to the LSM family of encoded RNA-binding proteins. It has been shown to perform a function in pre-mRNA splicing as a constituent of the U4/U6-U5 tri-snRNP complex involved in the assembly of the spliceosome and as a constituent of the catalytic pre-snRNP complex (spliceosome B complex)(Bertram et al., 2017). A particular binding site for the heptameric LSM2-8 complex is the 3'-terminal U-bundle of snRNA U6(Achsel et al., 1999). A previous study shows that LSM2-8 and XRN-2 contribute to the silencing of H3K27me3 marker genes by targeting RNA decay(Mattout et al., 2020). The findings of the CRISPR-cas9 screen indicate that the suppression of these genes resulted in cell death in all kinds of LGG. The antagonism of these genes could be a feasible method for the treatment of LGG. The overexpression of these genes in LGG may be a promising target for drug development in the future.

Despite previous research creating predictive signatures based primarily on gene expression, gene expression was integrated with functional genomic screening in the present research. Irrespective of their tumor grade, the genetic profile can distinguish between high and low-risk individuals. Therefore, suitable targeted medications may be used for the purpose of enhancing the prognosis of patients with a high risk. The prediction performance of the proposed signature is superior to classical clinicopathological parameters, such as IDH1 mutation status, tumor grade, or 1p19q staining combined with ablation. However, molecular subtyping and tumor grading are widely used in clinical practice. Apart from age and tumor grade, this signature is the only independent prognostic variable in multivariable logistic regression.

In summary, this study systematically investigates genes susceptible to cell viability, and the Hippo signaling pathway is crucial to this process. The gene signature proposed in this study was combined with a functional genomic screening that predicted the prognosis of patients with LGG more accurately in contrast with conventional clinicopathological markers. These genes associated with the Hippo signaling pathway can be targeted for the treatment of LGG.

LGG: Low-grade glioma

DEGs: Differentially expressed genes

IDH1/IDH2: Isotope dehydrogenase 1 and 2 genes

sgRNA: Single-stranded guide RNA

DepMap: The Cancer Dependency Map

SOX9: SRY-Box Transcriptional Factor 9

Funding

This work was supported by the Xinjiang Key Laboratory of Neurological Disorder Research(grant number: XJDX1711-2206), Xinjiang Uygur Autonomous Region Regional Collaborative Innovation Project (Shanghai Cooperation Organization Science and Technology Partnership Program and International Science and Technology Cooperation Program) (grant number: 2021E00131), Xinjiang Natural Science Foundation youth science fund Project (grant number: 2021D01C341).

Author Contributions

MM& AM: Conceptualization, Methodology, Validation, Investigation, Supervision, Software, Visualization, Writing - original draft, Writing- Reviewing and Editing. YW: Supervision, Visualization, Investigation, Validation. XC, MT, GZ, HW, YL, MIT, AA, NM, NY and MK participated in the coordination of data acquisition and data analysis and reviewed the manuscript.

Ethics Statement

The studies involving human participants were reviewed and approved by The Ethics Committee of First Affiliated Hospital of Xinjiang Medical University. The patients/participants provided their written informed consent to participate in this study.

Consent for publication

Not applicable.

Data Availability Statement

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Conflict of Interest

The authors declare no conflicts of interest.

Acknowledgments

We are grateful to the contributors to the public databases used in this study.

Achsel T, Brahms H, Kastner B, Bachi A, Wilm M, Lührmann R. A doughnut-shaped heteromer of human Sm-like proteins binds to the 3'-end of U6 snRNA, thereby facilitating U4/U6 duplex formation in vitro. EMBO J. 1999;18(20):5789–802. doi:10.1093/emboj/18.20.5789.
Amann P, Luo C, Owen R, Hofmann C, Freudenberger M, Schadendorf D, et al. Vitamin A metabolism in benign and malignant melanocytic skin cells: importance of lecithin/retinol acyltransferase and RPE65. J Cell Physiol. 2012;227(2):718–28. doi:10.1002/jcp.22779.
Amjad F, Fatima T, Fayyaz T, Khan M, Qadeer M. Novel genetic therapeutic approaches for modulating the severity of β-thalassemia (Review). Biomedical Rep. 2020;13(5):48. doi:10.3892/br.2020.1355.
Bertram K, Agafonov D, Dybkov O, Haselbach D, Leelaram M, Will C, et al. Cryo-EM Structure of a Pre-catalytic Human Spliceosome Primed for Activation. Cell. 2017;170(4):701–13.e711. doi:10.1016/j.cell.2017.07.011.
Cai H, Yu Y, Ni X, Li C, Hu Y, Wang J, et al. LncRNA LINC00998 inhibits the malignant glioma phenotype via the CBX3-mediated c-Met/Akt/mTOR axis. Cell Death Dis. 2020;11(12):1032. doi:10.1038/s41419-020-03247-6.
Cheng Z, Liang X, Zhang C, Wang R, Wei T, Ning B, et al. SOX9-transactived long non-coding RNA NEAT1 promotes the self-renewal of liver cancer stem cells through PKA/Hippo signaling. Signal Transduct Target therapy. 2021;6(1):87. doi:10.1038/s41392-021-00466-x.
Chiang J, Diaz A, Makepeace L, Li X, Han Y, Li Y, et al. Clinical, imaging, and molecular analysis of pediatric pontine tumors lacking characteristic imaging features of DIPG. Acta Neuropathol Commun. 2020;8(1):57. doi:10.1186/s40478-020-00930-9.
Clattenburg L, Wigerius M, Qi J, Rainey J, Rourke J, Muruganandan S, et al. NOS1AP Functionally Associates with YAP To Regulate Hippo Signaling. Mol Cell Biol. 2015;35(13):2265–77. doi:10.1128/mcb.00062-15.
Esvelt K, Smidler A, Catteruccia F, Church G. (2014). Concerning RNA-guided gene drives for the alteration of wild populations. eLife 3. doi: 10.7554/eLife.03401.
Gangopadhyay S, Cox K, Manna D, Lim D, Maji B, Zhou Q, et al. Precision Control of CRISPR-Cas9 Using Small Molecules and Light. Biochemistry. 2019;58(4):234–44. doi:10.1021/acs.biochem.8b01202.
Garros-Regulez L, Aldaz P, Arrizabalaga O, Moncho-Amor V, Carrasco-Garcia E, Manterola L, et al. mTOR inhibition decreases SOX2-SOX9 mediated glioma stem cell activity and temozolomide resistance. Expert Opin Ther Targets. 2016;20(4):393–405. doi:10.1517/14728222.2016.1151002.
Guo C, Liang C, Yang J, Hu H, Fan B, Liu X. LATS2 inhibits cell proliferation and metastasis through the Hippo signaling pathway in glioma. Oncol Rep. 2019;41(5):2753–61. doi:10.3892/or.2019.7065.
Hewitt G, Borel V, Segura-Bayona S, Takaki T, Ruis P, Bellelli R, et al. Defective ALC1 nucleosome remodeling confers PARPi sensitization and synthetic lethality with HRD. Mol Cell. 2021;81(4):767–83.e711. doi:10.1016/j.molcel.2020.12.006.
Huang K, Yang C, Wang Q, Li Y, Fang C, Tan Y, et al. The CRISPR/Cas9 system targeting EGFR exon 17 abrogates NF-κB activation via epigenetic modulation of UBXN1 in EGFRwt/vIII glioma cells. Cancer Lett. 2017;388:269–80. doi:10.1016/j.canlet.2016.12.011.
Khanzadi M, Khan A. CRISPR/Cas9: Nature's gift to prokaryotes and an auspicious tool in genome editing. J Basic Microbiol. 2020;60(2):91–102. doi:10.1002/jobm.201900420.
Kim Y, Lee S, Kim S, Kim S, Kim S, Lee C. YAP Activity is Not Associated with Survival of Uveal Melanoma Patients and Cell Lines. Sci Rep. 2020;10(1):6209. doi:10.1038/s41598-020-63391-z.
Kleinjan D, Wardrope C, Sou N, S., and Rosser S. Drug-tunable multidimensional synthetic gene control using inducible degron-tagged dCas9 effectors. Nat Commun. 2017;8(1):1191. doi:10.1038/s41467-017-01222-y.
Le Corre D, Ghazi A, Balogoun R, Pilati C, Aparicio T, Martin-Lannerée S, et al. The cellular prion protein controls the mesenchymal-like molecular subtype and predicts disease outcome in colorectal cancer. EBioMedicine. 2019;46:94–104. doi:10.1016/j.ebiom.2019.07.036.
Li Z, Guan Y, Zhang G. Genomic Analysis of Glioblastoma Multiforme Reveals a Key Transcription Factor Signature Relevant to Prognosis and the Immune Processes. Front Oncol. 2021;11:657531. doi:10.3389/fonc.2021.657531.
Louis D, Ohgaki H, Wiestler O, Cavenee W, Burger P, Jouvet A, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol. 2007;114(2):97–109. doi:10.1007/s00401-007-0243-4.
Louradour I, Ghosh K, Inbar E, Sacks D. (2019). CRISPR/Cas9 Mutagenesis in Phlebotomus papatasi: the Immune Deficiency Pathway Impacts Vector Competence for Leishmania major. mBio 10(4). doi:10.1128/mBio.01941-19.
Mahas A, Neal Stewart C, Mahfouz M. Harnessing CRISPR/Cas systems for programmable transcriptional and post-transcriptional regulation. Biotechnol Adv. 2018;36(1):295–310. doi:10.1016/j.biotechadv.2017.11.008.
Mattout A, Gaidatzis D, Padeken J, Schmid C, Aeschimann F, Kalck V, et al. LSM2-8 and XRN-2 contribute to the silencing of H3K27me3-marked genes through targeted RNA decay. Nat Cell Biol. 2020;22(5):579–90. doi:10.1038/s41556-020-0504-1.
Merrick D, Mistry K, Wu J, Gresko N, Baggs J, Hogenesch J, et al. Polycystin-1 regulates bone development through an interaction with the transcriptional coactivator TAZ. Hum Mol Genet. 2019;28(1):16–30. doi:10.1093/hmg/ddy322.
Meyers R, Bryan J, McFarland J, Weir B, Sizemore A, Xu H, et al. Computational correction of copy number effect improves specificity of CRISPR-Cas9 essentiality screens in cancer cells. Nat Genet. 2017;49(12):1779–84. doi:10.1038/ng.3984.
Morimoto T, Nakazawa T, Matsuda R, Nishimura F, Nakamura M, Yamada S, et al. (2021). CRISPR-Cas9-Mediated TIM3 Knockout in Human Natural Killer Cells Enhances Growth Inhibitory Effects on Human Glioma Cells. Int J Mol Sci 22(7). doi:10.3390/ijms22073489.
Moses C, Nugent F, Waryah C, Garcia-Bloj B, Harvey A, Blancafort P. Activating PTEN Tumor Suppressor Expression with the CRISPR/dCas9 System. Mol therapy Nucleic acids. 2019;14:287–300. doi:10.1016/j.omtn.2018.12.003.
Pan D. The hippo signaling pathway in development and cancer. Dev Cell. 2010;19(4):491–505. doi:10.1016/j.devcel.2010.09.011.
Peng H, Zheng Y, Blumenstein M, Tao D, Li J. CRISPR/Cas9 cleavage efficiency regression through boosting algorithms and Markov sequence profiling. Bioinf (Oxford England). 2018;34(18):3069–77. doi:10.1093/bioinformatics/bty298.
Pereira N, Weinshilboum R. Cardiovascular pharmacogenomics and individualized drug therapy. Nat Rev Cardiol. 2009;6(10):632–8. doi:10.1038/nrcardio.2009.154.
Pyakurel A, Balmer D, Saba-El-Leil M, Kizilyaprak C, Daraspe J, Humbel B, et al. (2017). Loss of Extracellular Signal-Regulated Kinase 1/2 in the Retinal Pigment Epithelium Leads to RPE65 Decrease and Retinal Degeneration. Mol Cell Biol 37(24). doi:10.1128/mcb.00295-17.
Rosenblum D, Gutkin A, Kedmi R, Ramishetti S, Veiga N, Jacobi A, et al. (2020). CRISPR-Cas9 genome editing using targeted lipid nanoparticles for cancer therapy. Sci Adv 6(47). doi:10.1126/sciadv.abc9450.
Roszkowski K, Furtak J, Zurawski B, Szylberg T, Lewandowska M. (2016). Potential Role of Methylation Marker in Glioma Supporting Clinical Decisions. Int J Mol Sci 17(11). doi:10.3390/ijms17111876.
Seo Y, Park S, Kim H, Nam J. (2020). The Hippo-YAP Signaling as Guardian in the Pool of Intestinal Stem Cells. Biomedicines 8(12). doi:10.3390/biomedicines8120560.
Smith R, Schwartz A. Diltiazem prophylaxis in refractory migraine. N Engl J Med. 1984;310(20):1327–8. doi:10.1056/nejm198405173102015.
Song S, Ajani J, Honjo S, Maru D, Chen Q, Scott A, et al. Hippo coactivator YAP1 upregulates SOX9 and endows esophageal cancer cells with stem-like properties. Cancer Res. 2014;74(15):4170–82. doi:10.1158/0008-5472.Can-13-3569.
Song S, Wang Z, Li Y, Ma L, Jin J, Scott A, et al. PPARδ Interacts with the Hippo Coactivator YAP1 to Promote SOX9 Expression and Gastric Cancer Progression. Mol cancer research: MCR. 2020;18(3):390–402. doi:10.1158/1541-7786.Mcr-19-0895.
Sourbier C, Liao P, Ricketts C, Wei D, Yang Y, Baranes S, et al. NF2Targeting loss of the Hippo signaling pathway in -deficient papillary kidney cancers. Oncotarget. 2018;9(12):10723–33. doi:10.18632/oncotarget.24112.
Stovicek V, Holkenbrink C, Borodina I. (2017). CRISPR/Cas system for yeast genome engineering: advances and applications. FEMS Yeast Res 17(5). doi:10.1093/femsyr/fox030.
Usman B, Nawaz G, Zhao N, Liao S, Qin B, Liu F, et al. (2020). Oryza sativaProgrammed Editing of Rice (L.) Gene Using CRISPR/Cas9 Improves Grain Yield by Modulating the Expression of Pyruvate Enzymes and Cell Cycle Proteins. Int J Mol Sci 22(1). doi:10.3390/ijms22010249.
Wang H, McKnight N, Zhang T, Lu M, Balk S, Yuan X. SOX9 is expressed in normal prostate basal cells and regulates androgen receptor expression in prostate cancer cells. Cancer Res. 2007;67(2):528–36. doi:10.1158/0008-5472.Can-06-1672.
Wang H, Wang X, Xu L, Zhang J, Cao H. RUNX1 and REXO2 are associated with the heterogeneity and prognosis of IDH wild type lower grade glioma. Sci Rep. 2021;11(1):11836. doi:10.1038/s41598-021-91382-1.
Wang L, He S, Yuan J, Mao X, Cao Y, Zong J, et al. Oncogenic role of SOX9 expression in human malignant glioma. Med Oncol (Northwood Lond Engl). 2012;29(5):3484–90. doi:10.1007/s12032-012-0267-z.
Wang X, Sun D, Tai J, Chen S, Yu M, Ren D, et al. TFAP2C promotes stemness and chemotherapeutic resistance in colorectal cancer via inactivating hippo signaling pathway. J experimental Clin cancer research: CR. 2018;37(1):27. doi:10.1186/s13046-018-0683-9.
Wen W, Yuan Z, Ma S, Xu J, Yuan D. CRISPR-Cas9 systems: versatile cancer modelling platforms and promising therapeutic strategies. Int J Cancer. 2016;138(6):1328–36. doi:10.1002/ijc.29626.
Wu X, Ma W, Mei C, Chen X, Yao Y, Liu Y, et al. Description of CRISPR/Cas9 development and its prospect in hepatocellular carcinoma treatment. J experimental Clin cancer research: CR. 2020;39(1):97. doi:10.1186/s13046-020-01603-0.
Xu X, Wang Z, Liu N, Cheng Y, Jin W, Zhang P, et al. Association between SOX9 and CA9 in glioma, and its effects on chemosensitivity to TMZ. Int J Oncol. 2018;53(1):189–202. doi:10.3892/ijo.2018.4382.
Yang B, Kuang J, Wu C, Zhou W, Zhu S, Jiang H, et al. (2020). Screening Genes Promoting Exit from Naive Pluripotency Based on Genome-Scale CRISPR-Cas9 Knockout. Stem cells international 2020, 8483035. doi: 10.1155/2020/8483035.
Ye Z, Zheng M, Zeng Y, Wei S, Wang Y, Lin Z, et al. Bioinformatics Analysis Reveals an Association Between Cancer Cell Stemness, Gene Mutations, and the Immune Microenvironment in Stomach Adenocarcinoma. Front Genet. 2020;11:595477. doi:10.3389/fgene.2020.595477.
Young C, Pyle A, Spencer M. CRISPR for Neuromuscular Disorders: Gene Editing and Beyond. Physiology (Bethesda). 2019;34(5):341–53. doi:10.1152/physiol.00012.2019.
Zhang X, Li C, Xiao L, Gao C, Zhao W, Yang M, et al. Predicting Individual Prognosis and Grade of Patients with Glioma Based on Preoperative Eosinophil and Neutrophil-to-Lymphocyte Ratio. Cancer Manage Res. 2020;12:5793–802. doi:10.2147/cmar.S260695.
Zhou H, Li G, Huang S, Feng Y, Zhou A. SOX9 promotes epithelial-mesenchymal transition via the Hippo-YAP signaling pathway in gastric carcinoma cells. Oncol Lett. 2019;18(1):599–608. doi:10.3892/ol.2019.10387.

1bar.png
2bar.png
3bar.png
Datasheetpcr.xls
FigureS1.tif
Figure S1. A. Rate of gene mutation in the TCGA-LGG dataset regarding 145 genes (Only PDGFRA, SRBD1, CCDN1, CDC73, and ACLY have > 1% mutation frequency). B. No significant mutations in the six overlapped genes.
FigureS2.tif
Figure S2. A-B. Patients with LGG were shown to be distributed in two separate distribution clusters, according to the results of principal component analysis (PCA).
FigureS3.tif
Figure S3.(A–B) Analysis of survival time using the Kaplan-Meier curve for patients with ATRX wild type and ATRX mutant in the high- and low-risk groups. (C–D) Patients with EGFR wild-type and EGFR mutation groups were classified into high- and low-risk groups, and their overall survival (OS) was analyzed using a Kaplan-Meier curve. (E–F) Patients with PTEN wild-type and PTEN mutant groups were analyzed using the Kaplan-Meier method for the purpose of determining their OS.(G-H) Patients with TP53 wild-type and TP53 mutant groups were classified into high- and low-risk groups, and their OS was analyzed using a Kaplan-Meier curve. (I) Kaplan-Meier curve study of OS in patients with IDH2 R132 wild-type in the high- and low-risk groups. (J-K) Patients with IDH2 R172 wild-type and IDH2 R172 mutant subgroups were divided into high- and low-risk groups, and their OS was analyzed using a Kaplan-Meier curve.
GAPDH.png
GAPDH.png
GAPDH.tif
GO.txt
GSMM.xls
KEGG.txt
LSM2.png
LSM2.png
LSM2.tif
RCODE.txt
RPE65.png
RPE65.png
RPE65.tif
SOX9.png
SOX9.png
SOX9.tif
SupplementaryTable1.docx
TCGALGGrisk.txt
TCGAclinical.txt
TMEscores.xlsx
VENN.xlsx
adj.CERES.xlsx
all.xls
celllinesamples.xls
diff.xls
tumor1bar.png
tumor2bar.png
tumor3bar.png
tumorGAPDH.tif
tumorLSM2.tif
tumorRPE65.tif
tumorSOX9.tif

Download PDF

Journal Publication

published 14 May, 2023

Read the published version in Molecular Medicine →

Reviewers invited by journal
18 Oct, 2022
Reviewers agreed at journal
17 Oct, 2022
Editor assigned by journal
06 Oct, 2022
Submission checks completed at journal
05 Oct, 2022
Editor invited by journal
05 Oct, 2022
First submitted to journal
30 Sep, 2022

You are reading this latest preprint version

CRISPR-cas9 Screening Identified Lethal Genes Enriched in Hippo Kinase Pathway and of Prognosis Significance in primary Low-grade Glioma

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1. Introduction

2. Materials And Methods

2.1 Determination of Vulnerability and Viability

2.2 Processing of Data in Low-grade glioma

2.3 LASSO Regression Analysis

2.4 Enrichment Analysis

2.5 Creation of the PPI Network and Detection of Hub

2.6 Weighted Correlation Network Analysis

2.7 Vitro assays in clinical samples

2.8 Data Analysis

3. Result

3.1 Determination of Functional Genomic Genes in low-grade glioma

3.2 Creation and Verification of the Gene Signature

3.3 DEGs Between Low- and High-Risk Subgroups

3.4 WGCNA

3.5 Validation of SOX9, RPE65, LSM2 in clinical samples

4. Discussion

5. Conclusions

Abbreviations

Declarations

References

Supplementary Files

Status:

Journal Publication

Version 1