ITGB4 upregulation is associated with progression of lower grade glioma

Gliomas originating in the neuroepithelium account for about 80% of brain malignancies and are the most common cancer of the central nervous system. Clinical management of gliomas remains challenging despite significant advances in comprehensive therapies, including radiotherapy, chemotherapy, and surgery. The ITGB4 (Integrin subunit beta 4) gene encodes a receptor for laminins and its upregulation in tumor tissues is associated with poor prognosis. However, its role in glioma is not well understood. First, we performed a pan cancer analysis of ITGB4 expression in The Cancer Genome Atlas (TCGA) dataset. Survival analysis was done on Chinese Glioma Genome Atlas (CGGA) and TCGA. Immunohistochemistry was then used to validate the expression and role of ITGB4 in glioma. We finally analyzed the possible mechanism by immune infiltration and single-cell sequencing analysis. Here, we found that ITGB4 is upregulated in glioma and accurately predicts the prognosis of lower grade glioma (LGG). Univariate and multivariate Cox regression analyses showed that ITGB4 is a risk factor for LGG. Immunohistochemical analysis confirmed that ITGB4 accurately predicts LGG prognosis. Non-negative matrix factorization (NMF) cluster analysis showed that ITGB4 was closely related to immune related genes. Immune cell infiltration and single cell sequencing analyses indicated that ITGB4 may be closely related to the microenvironment of gliomas, especially tumor-associated fibroblasts. ITGB4 is a promising diagnostic and therapeutic factor in LGG patients.


Prognostic analysis
The TCGA data on the 33 tumors were subjected to univariate cox regression analysis to determine the effects of ITGB4 on OS, progression free interval (PFI) and disease specific survival (DSS).Next, patients with glioma (LGG and GBM) were grouped into the 'high' or 'low' groups based on median value of ITGB4 expression.Patients with LGG were stratified based on clinicopathological features, including age (≤ 40, > 40), gender (female/ male), grade (Grade 2, and Grade 3), IDH status (mutation, wildtype), 1p19q status (codel, non-codel), and MGMT (methylated, unmethylated).Kaplan-Meier analysis was used to determine survival in the low and high ITGB4-expression groups.Univariate and multivariate Cox regression analysis was used to assess if ITGB4 is an independent prognostic factor for patients with LGG.

NMF cluster analysis
We performed NMF cluster analysis on CGGA693 database.First, we obtained the immune related genes from the ImmPort database, then extracted the expression of immune related genes from CGGA693 database, and combined with the survival time and survival status of patients to perform NMF clustering.

Evaluation of the effect of ITGB4 on glioma microenvironment
The proportion of immunocytes in gliomas was calculated using single-sample gene set enrichment analysis (ssGSEA).Infiltration of immune cells was quantified using enrichment scores calculated with R's GSVA package (gene set variation analysis).In order to analyze the relationship between ITGB4 expression and TumorPurity, StromalScore, ESTIMATEScore, and ImmuneScore, Spearman correlation analysis was used.

Database for annotation, visualization and integrated discovery (DAVID)
Our analysis identified differentially expressed genes (DEGs) between glioma patients with high or low expression of ITGB4 using the limma package on R utilizing |logFC|> 1 and FDR < = 0.05.DAVID 6.8 (https:// david.abcc.ncifc rf.gov/) was used to further explore its possible mechanism.

Single cell sequencing analysis
Seurat R package was used to reprocess the count matrix in which the dimensional reduction plot and cell type annotation.ITGB4 expression pattern was visualized using the Feature Plot function.

Statistical analyses
ITGB4 levels in tumor vs normal tissues were compared using a Wilcoxon signed-rank test.Spearman's rank test was used to assess correlation between ITGB4 expression and TumorPurity, StromalScore, ESTIMATEScore, and ImmuneScore.Kaplan-Meier analysis was used assess the effect of ITGB4 on prognosis.Statistical analyses were done on SPSS 24.0 (IBM), GraphPad Prism 6, and R 4.0.1.P = < 0.05 was considered statistically significant.

Ethics approval and consent to participate
The article data is obtained from the network public database in accordance with the database policy.And the informed consent was not required.

Results
ITGB4 expression was assessed in publicly available datasets from TCGA and CGGA.There were 587 and 686 cases in TCGA and CGGA, respectively.A tissue microarray dataset from 169 cases was also included in the analyses.Clinical and molecular features are shown on Table 1.

High ITGB4 expression correlates with poor prognosis
The results above showed that ITGB4 is highly expressed in tumors and is an independent prognostic factor in gliomas, especially LGG.Next, we analyzed TCGA and CGGA datasets to determine the impact of ITGB4 in the survival of LGG and GBM.First, we divided all glioma patients on TCGA into the high and the low expression groups based on median ITGB4 expression level.This analysis revealed that patients with high ITGB4 expression had significantly worse OS than those with low expression (HR = 2.04, 95% CI 1.49-2.79,p = < 0.001, Fig. 2A).
We then performed a subgroup analysis by dividing LGG patients into the high and low expression groups based on ITGB4 expression.Survival analysis of this group showed that high ITGB4 levels correlated with significantly lower OS relative to low ITGB4 expression (HR = 2.21, 95% CI 1.39-3.53,p = 0.001, Fig. 2B).However, this was not the case in GBM in which ITGB4 was not a prognostic factor (HR = 1.18, 95% CI 0.78-1.80,p = 0.435, Fig. 2C).

High ITGB4 expression correlated with high immune infiltration
We obtained 2,483 immune related genes from the ImmPort database, and then performed NMF clustering on glioma patients according to gene expression, survival time and survival status of patients.Through NMF cluster analysis, we divided glioma patients into two categories, C1 and C2 (Fig. 5A).Survival analysis confirmed that there was a significant difference in survival between the two groups (HR = 3.37, 95% CI 2.74-4.14, p = < 0.001, Fig. 5B), C1 had worse survival time.Interestingly, ITGB4 expression in C1 was significantly higher than that in C2 (Fig. 5C).Next, we used single simple gene set enrichment analysis (ssGSEA) to assess immune cell infiltration and observed that ITGB4 expression level negatively correlated with tumor purity (r = -0.600,p = < 0.001, Fig. 6A), and positively correlated with stromal score (r = 0.570, p = < 0.001, Fig. 6B), ESTIMATE score (r = 0.600, P < 0.001, Fig. 6C) and immune score (r = 0.600, P < 0.001, Fig. 6D), indicating that ITGB4 expression is closely relevant to the tumor microenvironment (TME).

ITGB4 functions in immune related pathways in glioma
Analysis of immune cell infiltration revealed that ITGB4 is closely related to the TME in LGG.Next, investigation of the potential underlying mechanism on DAVID indicated that the DEGs were enriched in signal pathways associated with immune responses, such as neutrophil activation, collagen-containing extracellular matrix, complement and coagulation cascades, and neutrophil activation (Fig. 7).This is consistent with findings from numerous studies showing that the TME is closely associated with tumor related immunity and indicates that ITGB4 may contribute to poor LGG prognosis via its effects on tumor immune responses.

Single cell sequencing showed that ITGB4 was closely related with tumor related fibroblasts
Next, we examined ITGB4 expression in single cell sequencing data (dataset GSE117891) obtained from GEO.
Filtering and standardizing the single-cell sequencing data revealed that ITGB4 is highly expressed in tumorassociated fibroblasts, indicating that ITGB4 may regulate the function of fibroblasts in the TME (Fig. 8A,B).

Discussion
Brain malignancies are estimated to account for 1.6% of new malignancies and 2.5% of annual deaths worldwide 12 .
Gliomas account for 80% of brain malignancies and are mainly treated using surgery, radiation, and chemotherapy.However, glioma prognosis remains poor, especially glioblastoma which has a 5-year survival rate of 3% 13 .The poor prognosis is because glioma growth is infiltrative, difficulty in making clean glioma resections, and glioma resistance to chemoradiotherapy.Lower grade glioma (LGG) is characterized by a propensity for transformation into high-grade glioma.Thus, better understanding of glioma pathogenesis and novel diagnostic and therapeutic targets for glioma are urgently needed.
Integrins are a family of cell adhesion receptors that bind to extracellular matrix ligands, cell surface ligands, and soluble ligands.These abundant proteins participate in processes involved in the tumor biological behavior 14 .Integrins possess a large extracellular domain, a transmembrane area, and a cytoplasmic area.The extracellular domain binds to extracellular matrix (ECM) proteins like laminin while the cytoplasmic area interacts with kinases, like cytoplasmic tyrosine kinase.Its special structure and the diversity of the internal and external structures of the connected cells, confers integrin α6β4 with unique cytoskeleton and signal transduction functions.ITGB4, a member of the integrin family, has been implicated in tumor progression 15 .Mounting evidence indicates that ITGB4 promotes immune escape by tumors.However, the role of ITGB4 in glioma is poorly understood.
Here, we first performed a pan cancer analysis of ITGB4 expression on TCGA and found that it is highly expressed in various tumors, including colorectal cancer 16 , gastric cancer 17 and non-small lung cancer 18 .Univariate Cox regression analysis showed that ITGB4 is strongly associated with LGG prognosis and that LGG cases with high ITGB4 had a worse prognosis than those with lower levels.Immunohistochemical validation of the bioinformatics analysis results indicated that LGG cases with high ITGB4 levels had worse prognosis.Similar observations have been made in some epithelial cancers.TP53 mutations and ITGB4 upregulation co-occur in many aggressive malignancies.Squamous cell carcinomas of the lung have a higher frequency of TP53 mutations.ITGB4 overexpression can also lead to venous invasion and reduced overall survival in non-small cell lung cancer patients 18 .ITGB4 has also been shown to play a crucial role in the development of prostate cancer found that ITGB4 promotes prostate tumorigenesis by co-expression with ErbB2 and c-Met in tumor progenitor cells 19 .These findings are consistent with our analysis and indicate ITGB4 promotes tumor initiation and progression.www.nature.com/scientificreports/Some studies have found that ITGB4 expression was increased in glioma stem cells and human glioma tissues.Upregulation of ITGB4 was correlated with glioma grades.Inhibition of ITGB4 in glioma cells decreased the self-renewal abilities of glioma stem cells and suppressed the malignant behaviours of glioma cells in vitro and in vivo 20 .Among the ten genes selected, HLA-E, MSN, GNG-5, MYL12A, ITGB4, PDPN, AGTRAP, S100A4, PLSCR1, VAMP5 were selected as the most significant genes related to purity and prognosis in glioma.The risk score model based on the 10 genes could moderately predict gliomas' overall survival 21 .
Numerous studies have investigated the mechanisms by which ITGB4 influences carcinogenesis.Recently, the tumor microenvironment (TME) has emerged as an anticancer therapeutic target.Our immune infiltration data show that ITGB4 expression negatively correlates with tumor purity and positively correlates with immune and stromal scores, indicating that ITGB4 may be a potential therapeutic target in the TME.Functional enrichment analysis on DAVID also indicated that ITGB4 is closely associated with immune response.Tumor angiogenesis is an important component of TME.ITGB4 may induce angiogenesis via the NF-κB and phospho-ERK nuclear translocation pathway, or the SRC, PI3K/Akt, and iNOS pathways 22 .Our analysis of single-cell sequencing data showed that ITGB4 is highly expressed on tumor associated fibroblasts.Interestingly, a recent study showed that ITGB4 mediates metabolic reprogramming of tumor associated fibroblasts 23 .These results indicate that ITGB4 closely interacts with the TME, particularly with tumor associated fibroblasts that promote tumor development.

Figure 2 .
Figure 2. High ITGB4 levels correlate with poor prognosis in lower grade glioma.(A-C) Survival analysis revealed that the survival time of patients with high ITGB4 expression was significantly lower relative to those in the low expression group in (A) (TCGA), (B) (TCGA-LGG), (D) (CGGA), and (E) (CGGA-LGG), but there was no significant difference in (C) (TCGA-GBM) and (F) (CGGA-GBM).

Figure 3 .
Figure 3. Prediction of outcome of the ITGB4 in stratified patients in the CGGA dataset.Survival analysis of the signature in patients stratified by gender (A,B), age (C,D), grade (E,F), IDH (G,H), 1p19q status (I,J), and MGMT promoter (K,L).

Figure 4 .Figure 5 .
Figure 4. ITGB4 expression in gliomas and its prognostic significance were analyzed by immunohistochemistry. (A) Shows that ITGB4 is weakly, moderately, and strongly positive in gliomas respectively.High ITGB4 expression was related to poor OS in glioma and lower grade glioma (B,C).

Figure 6 .
Figure 6.Immune infiltration patterns in CGGA lower grade glioma datasets with low and high ITGB4 levels was analyzed using ssGSEA.(A-D) Scatter plot showing correlation between ITGB4 and tumor purity, stromal, ESTIMATE, and immune scores.

Figure 7 .
Figure 7. KEGG and GO analysis of the mechanism by which ITGB4 promotes tumor progression.

Table 1 .
Clinicopathological characteristics of glioma patients from the TCGA, CGGA database and tissue microarray.

Table 2 .
Univariate and multivariate analysis for overall survival of lower grade glioma.