The Immune Tumor Microenvironment in Gliomas: May CITED2 Play a Role? ^†

Abstract

Gliomas are the most common brain cancers, resulting from transformed glial cells. CITED2 is a co-transcriptional regulator previously implicated in several types of cancer, affecting both cell-intrinsic processes and the microenvironment. Because in breast cancer it was shown to contribute to the recruitment of macrophages and their polarization to an immunosuppressive phenotype, a potential similar role was explored in gliomas. By analyzing publicly available databases using a set of bioinformatics tools, it was found that CITED2 is overexpressed in higher-grade gliomas and contributes to an adverse prognosis. In addition, CITED2 expression correlates with macrophage infiltration and a M2 phenotype.

1. Introduction

Gliomas are the most common malignant tumors of the central nervous system [1,2]. Of these, 50% are classified as glioblastoma (GBM), a highly aggressive glioma (grade 4) with a poor prognosis [3].

The current standard of care for GBM is resection surgery, followed by radiotherapy and chemotherapy with temozolomide. Lomustine and bevacizumab are generally used when the tumor progresses, but these therapeutic approaches are marked by tumor resistance and do not lead to an increase in overall survival [2]. More recently, immunotherapies have revolutionized the treatment of cancer, but they have been shown to be ineffective in GBM, at least as monotherapy [4]. This appears to be mainly due to the existence of the blood–brain barrier and the characteristic immunosuppressive tumor microenvironment (TME), which is still poorly characterized [5].

Among the cells that make up the TME, GBM-associated macrophages/microglia (GAM) constitute the most abundant cell population [6,7]. In this context, GAMs present a pro-tumor/immunosuppressive M2 phenotype, to the detriment of M1 with pro-inflammatory and anti-tumor activity [8], which contributes to the immunosuppressive microenvironment.

The protein CITED2 (CBP/p300-interacting transactivator with ED-rich tail 2) is a transcriptional modulator with key roles in processes such as self-renewal and cell differentiation [9]. It has also been implicated in the development of diverse types of cancer, with either pro-tumorigenic and anti-tumorigenic roles [9,10]. Nevertheless, a direct role for CITED2 in gliomas, whether cell intrinsic or microenvironmental, has not previously been reported.

Within the scope of TME composition, CITED2 has been shown to induce CCL20 expression and macrophage recruitment to breast cancer tumors, [11] and, in macrophages, it restricts NF-κB activation by negatively interfering with the expression of pro-inflammatory mediators [12,13]. Therefore, CITED2 may be involved in the recruitment and polarization of macrophages, favoring their immunosuppressive effects. Modeling its expression could be a potential strategy to reprogramming GAMs to a pro-inflammatory phenotype, which has been indicated as a promising antitumor strategy for glioblastoma [14].

2. Materials and Methods

Data from the publicly available databases Genotype–Tissue Expression (GTEx) Data Set and The Cancer Genome Atlas (TCGA), including the lower-grade glioma (LGG) and glioblastoma (GBM) cohorts, were analyzed [15]. RNA sequencing (Illumina HiSeq) and survival data were accessed through the UCSC Xena platform [16] (https://xenabrowser.net/heatmap/, accessed on 5 January 2024). Differential expression analyses between LGG vs. normal tissue and GBM vs. normal tissue were performed using the GEPIA2 Gene Expression Profiling Interactive Analysis platform [17] (http://gepia2.cancer-pku.cn/#degenes, accessed on 5 January 2024). Gene set enrichment (GSEA) analyses were performed using Enrichr [18,19,20] (https://maayanlab.cloud/Enrichr/, accessed on 5 January 2024). TIMER2.0 web server [21,22,23] (http://timer.cistrome.org/, accessed on 5 January 2024) was used to analyze tumor-infiltrating immune cells in LGG and GBM.

Statistical analyses were performed using Prism 5 (GraphPad Software 8.4.0, La Jolla, CA, USA). Statistical tests were used as indicated in figure legends, and a p < 0.05 was considered statistically significant.

3. Results

3.1. CITED2 Is Significantly More Expressed in GBM Than in LGG

To determine whether CITED2 expression differs between different glioma stages, data from TCGA’s lower-grade glioma (grades 2 and 3) and glioblastoma (grade 4) were compared. CITED2 levels were found to be significantly increased in GBM (Figure 1a). Higher CITED2 expression was also associated with an adverse prognosis in glioma (Figure 1b).

CITED2 was previously shown to be expressed in cancer cells but also in tumor-infiltrating cells, like immune cells or endothelial cells [24]. Therefore, to explore the biological processes enriched in LGG and GBM vs. normal tissues, the differential expression was assessed and the top 50 genes with the highest fold-change were analyzed. Interestingly, genes involved in immunological processes were significantly more expressed in GBM than in normal tissues, differing from LGG. In accordance with this result, GBM was previously shown to present higher immune infiltration than LGG [25].

3.2. CITED2 Expression Correlates with Macrophage Infiltration

To evaluate whether CITED2 expression can influence macrophage infiltration, the TIMER2.0 tool was used. As expected, CITED2 expression was correlated with macrophage infiltration, both M1 and M2 (

Table 1. Correlation between CITED2 expression and macrophage infiltration in GBM (n = 153).

Macrophage Subtype	Rho	p Value
All	0.181	p < 0.05 1
M1	0.253	p < 0.05 1
M2	0.266	p < 0.01 2

¹ TIMER algorithm; ² CIBERSORT-ABS algorithm.

). In LGG, only M2 macrophage infiltration was correlated with CITED2 expression (

Table 2. Correlation between CITED2 expression and macrophage infiltration in LGG (n = 516).

Macrophage Subtype	Rho	p Value
All	0.209	p < 0.05 1
M1	−0.011	ns 2
M2	0.143	p < 0.05 3

¹ TIMER algorithm; ² QUANTISEC algorithm; ³ CIBERSORT-ABS algorithm; ns, non-significant.

).

Finally, to explore the clinical relevance of macrophage infiltration, a multivariable Cox proportional hazard model was used. Although in GBM, only age and macrophage infiltration were shown to be relevant for survival (age, HR 1.031, p < 0.001; macrophage infiltration XCELL, HR 290.680, p = 0.047; CITED2 expression, HR 1.082, p = 0.579), regarding LGG, the same two variables and CITED2 expression were shown to significantly impact survival (age, HR 1.057, p < 0.001; macrophage infiltration TIMER, HR 9.675, p < 0.001; CITED2 expression, HR 1.538, p < 0.001), leading to an adverse prognosis.

Altogether, these results suggest that CITED2 expression may be relevant in attracting macrophages and polarizing them to an M2 phenotype in an early phase of glioma development.