Next Article in Journal
Natural Polymorphic Variants in the CYP450 Superfamily: A Review of Potential Structural Mechanisms and Functional Consequences
Previous Article in Journal
When Two-Fold Is Not Enough: Quantifying Uncertainty in Low-Copy qPCR
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJMS
Volume 26
Issue 16
10.3390/ijms26167798
ijms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Multi-Layered Analysis of TGF-β Signaling and Regulation via DNA Methylation and microRNAs in Astrocytic Tumors

by
Klaudia Skóra
1,2,*,
Damian Strojny
2,3,
Dawid Sobański
2,4,
Rafał Staszkiewicz
2,5,6,
Paweł Gogol
2,7,8,9,
Mateusz Miller
10,
Przemysław Rogoziński
11,
Nikola Zmarzły
2 and
Beniamin Oskar Grabarek
2,12
1
Department of Neurological Rehabilitation, District Hospital of St. Padre Pio in Sędziszów Małopolski, 39-120 Sędziszów Małopolski, Poland
2
Collegium Medicum, WSB University, 41-300 Dąbrowa Górnicza, Poland
3
Department of Neurology, New Medical Techniques Specialist Hospital of St. Family in Rudna Mała, 36-060 Rzeszow, Poland
4
Department of Neurosurgery, St. Raphael Hospital, 30-693 Krakow, Poland
5
Department of Neurosurgery, 5th Military Clinical Hospital with the SP ZOZ Polyclinic in Krakow, 30-901 Krakow, Poland
6
Department of Neurosurgery, Faculty of Medicine in Zabrze, Academy of Silesia in Katowice, 40-555 Katowice, Poland
7
Department of Anesthesiology and Intensive Care, Our Lady of Perpetual Help Hospital in Wołomin, 05-200 Wołomin, Poland
8
Department of Trauma and Orthopedic Surgery, Our Lady of Perpetual Help Hospital in Wołomin, 05-200 Wołomin, Poland
9
Pain Treatment Clinic, Our Lady of Perpetual Help Hospital in Wołomin, 05-200 Wołomin, Poland
10
Department of Neurology, Independent Public Healthcare Institution of the Ministry of Internal Affairs and Administration in Rzeszow, 35-111 Rzeszow, Poland
11
Dental Office, Resdent Stomatology Rogozinscy, 35-326 Rzeszow, Poland
12
Faculty of Medicine and Health Sciences, Andrzej Frycz Modrzewski University in Cracow, 30-705 Krakow, Poland
 
add Show full affiliation list
 
remove Hide full affiliation list
*
Author to whom correspondence should be addressed.
Int. J. Mol. Sci. 2025, 26(16), 7798; https://doi.org/10.3390/ijms26167798
Submission received: 15 July 2025 / Revised: 5 August 2025 / Accepted: 11 August 2025 / Published: 12 August 2025
(This article belongs to the Special Issue Cancer Biology: From Genetic Aspects to Treatment, 2nd Edition)
Browse Figures
Versions Notes

Abstract

Astrocytic tumors are a heterogeneous group of glial neoplasms characterized by marked differences in biological behavior and patient prognosis. Transforming growth factor-beta (TGF-β) signaling plays a pivotal role in astrocytoma pathogenesis; however, the extent and mechanisms of its epigenetic regulation remain poorly understood. This study aimed to investigate how promoter methylation and microRNA-mediated mechanisms regulate key genes within the TGF-β signaling pathway across various astrocytoma grades. Tumor tissue samples from 65 patients with WHO grade II–IV astrocytomas were analyzed using Affymetrix gene expression and microRNA microarrays. Promoter methylation of TGF-β signaling genes was assessed using methylation-specific polymerase chain reaction (MSP). Gene expression was validated by reverse transcription quantitative polymerase chain reaction (RT-qPCR), and protein levels were quantified using enzyme-linked immunosorbent assay (ELISA). MicroRNA targets were predicted using bioinformatic tools, and survival analyses were conducted using Kaplan–Meier and Cox regression models. Six genes—SMAD1, SMAD3, SKIL, BMP2, SMAD4, and MAPK1—showed significant upregulation in high-grade tumors (fold change > 5.0, p < 0.05), supported by RT-qPCR and protein-level data. Promoter hypomethylation and reduced expression of regulatory microRNAs (e.g., hsa-miR-145-5p targeting SMAD3) were more common in higher-grade tumors. Protein–protein interaction analysis indicated strong functional interconnectivity among the overexpressed genes. High protein levels of SMAD1, SMAD3, and SKIL were significantly associated with shorter overall survival (p < 0.001). This multi-level analysis reveals that astrocytic tumor progression involves epigenetic derepression and microRNA-mediated dysregulation of TGF-β signaling. Elevated expression of SMAD1, SMAD3, and SKIL emerged as strong prognostic indicators, underscoring their potential as biomarkers and therapeutic targets in astrocytic tumors.

1. Introduction

Astrocytic tumors, which arise from glial astrocytes in the central nervous system, represent a heterogeneous group of neoplasms with diverse degrees of malignancy, histological complexity, and clinical behavior. According to the World Health Organization (WHO), these tumors are graded from I to IV, with grades II–IV classified as infiltrating astrocytomas. Grade IV, known as glioblastoma (GBM), represents the most aggressive form, with a median survival of only 12–18 months despite the use of multimodal therapy [1,2,3]. Lower-grade astrocytomas (grades II and III) typically exhibit a more indolent clinical course but possess the potential for malignant transformation, thereby worsening patient prognosis over time [4].
The molecular characterization of astrocytic tumors has revolutionized their classification and clinical management. The 2016 WHO classification was the first to integrate key genetic alterations—such as IDH mutations and 1p/19q codeletion—into tumor nomenclature [5]. The updated 2021 WHO classification further prioritizes molecular stratification, refining diagnostic categories based primarily on genomic and epigenetic features rather than histology alone [6]. This paradigm shift has led to the redefinition of specific astrocytic tumor subtypes and emphasized the essential role of molecular diagnostics in guiding treatment decisions and predicting outcomes [7,8,9].
Among the most clinically relevant molecular alterations are IDH1/2 mutations, MGMT promoter methylation, 1p/19q codeletion, TP53 mutations, and EGFR amplification. Importantly, IDH-mutant astrocytomas are associated with significantly longer survival compared to their IDH-wildtype counterparts, even at higher histological grades [8,9,10]. These findings support the emerging view that tumor behavior is more accurately predicted by molecular profiles than by histological grade alone.
Transforming growth factor-beta (TGF-β) signaling plays a critical and paradoxical role in both normal central nervous system physiology and oncogenesis. The TGF-β family consists of three isoforms in humans—TGF-β1, TGF-β2, and TGF-β3—each encoded by a distinct gene but sharing high structural homology and overlapping biological functions [11]. These cytokines regulate numerous cellular processes, including proliferation, differentiation, immune regulation, and extracellular matrix remodeling [12]. In the healthy brain, TGF-β supports immune homeostasis and neuroprotection; however, in malignant contexts, it exhibits a dual role [13]. During early tumorigenesis, TGF-β acts as a tumor suppressor, but in advanced disease stages, it promotes tumor progression through immunosuppression, epithelial–mesenchymal transition (EMT), and neovascularization—collectively referred to as the “TGF-β paradox”.
In gliomas, overexpression and dysregulation of TGF-β isoforms have been linked to increased tumor invasiveness, therapeutic resistance, and poor prognosis. Furthermore, recent studies highlight that TGF-β activity within the glioma microenvironment influences not only tumor cells but also immune and stromal components, positioning TGF-β as a central modulator of tumor plasticity [14].
In recent years, epigenetic modifications have emerged as pivotal regulators of glioma development and progression [15,16]. These heritable but reversible changes in gene expression—independent of alterations in the DNA sequence—include DNA methylation, histone modifications, chromatin remodeling, and the activity of non-coding RNAs such as microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) [17,18]. Epigenetic dysregulation disrupts numerous oncogenic and tumor-suppressive pathways, including TGF-β signaling, thereby contributing to gliomagenesis, immune escape, and therapeutic resistance [19].
In astrocytic tumors, aberrant DNA methylation patterns are known to influence tumor classification and clinical outcomes. For example, MGMT promoter methylation status is routinely assessed to predict responsiveness to alkylating chemotherapy in glioblastoma [20]. However, the broader implications of epigenetic regulation on key signaling pathways such as TGF-β remain insufficiently explored [21]. Enzymes such as histone deacetylases (HDACs), methyltransferases, and chromatin remodelers can repress or activate TGF-β target genes, thereby modulating proliferation, stemness, and invasive capacity [22]. Additionally, regulatory RNAs—particularly miRNAs—are capable of targeting multiple nodes within the TGF-β signaling, including ligands, receptors (e.g., TGFBR1/2), intracellular SMADs, and downstream effectors, adding further complexity to its role in glioma biology [23].
A growing body of research, including work by Naik et al. [19] and Ding et al. [23], underscores the significance of epigenetic control over TGF-β signaling as a major determinant of tumor progression and therapeutic resistance. These investigations demonstrate how non-coding RNAs and histone modifications modulate TGF-β transcriptional activity, thereby influencing key processes such as EMT, angiogenesis, and immune evasion [19,23]. Nevertheless, much of this evidence is derived from non-CNS tumor models, highlighting a critical knowledge gap regarding the specific epigenetic regulation of TGF-β signaling in astrocytic tumors.
To date, relatively few studies have systematically investigated the epigenetic regulation of TGF-β isoforms across astrocytoma malignancy grades. A notable exception is the study by Kurowska et al. [24], which offered valuable insights into the differential expression of TGF-β1, TGF-β2, and TGF-β3 isoforms in astrocytic tumors of varying malignancy. Based on microarray and RT-qPCR analyses of 43 tumor samples, the authors reported significant upregulation of TGF-β2 in high-grade tumors (G3/G4) compared to low-grade gliomas (G2), along with increased expression of TGF-β1 in the microarray dataset [24]. However, while this study confirmed transcriptional alterations of TGF-β isoforms and associated genes, it did not investigate the epigenetic mechanisms responsible for these changes. Specifically, methylation profiles, histone modifications, and non-coding RNA expression were not assessed [24]. Accordingly, the present study aimed to evaluate differences in the expression patterns of mRNAs, promoter methylation status, and microRNAs related to TGF-β signaling pathways in astrocytic tumors, with particular attention to variation across malignancy grades.

2. Results

2.1. mRNA Microarray and RT-qPCR Analysis

Out of the 110 genes associated with the TGF-β signaling pathway (KEGG hsa04350), one-way ANOVA identified 27 genes as significantly differentially expressed between grade II (G2) and higher-grade (G3/G4) astrocytomas (fold change [FC] > 1.50 or <–1.50; p < 0.05; Table A1). To further refine the analysis and emphasize the most biologically relevant alterations, a more stringent threshold was applied (FC > 5.0 or <–5.0; p < 0.05), yielding a subset of six genes with markedly altered expression: SKIL, BMP2, SMAD1, SMAD3, SMAD4, and MAPK1. Notably, SMAD1, SMAD3, and SMAD4 exhibited consistent expression changes in both G3 and G4 tumors compared to G2. Microarray profiling revealed a clear upregulation of all six genes in G3/G4 astrocytomas relative to G2, indicating a potential enhancement of TGF-β-mediated transcriptional activity in higher-grade tumors (Table 1). In the following sections, we focus specifically on these six prominently dysregulated genes, as they serve as key effectors or modulators within the TGF-β signaling cascade and are likely to contribute critically to the molecular shifts associated with malignant progression.
As shown in Figure 1, RT-qPCR analysis confirms the observed expression patterns, highlighting the consistent overexpression of selected transcripts.

2.2. In Silico Identification of miRNA Regulators of TGF-β Signaling-Associated Genes

To investigate the post-transcriptional regulation of TGF-β signaling components, we first performed miRNA profiling using the Affymetrix GeneChip™ (Santa Clara, CA, USA) miRNA 2.0 array on astrocytic tumor tissue samples. Subsequently, in silico prediction tools—TargetScan and miRDB—were employed to identify high-confidence miRNA–mRNA interactions corresponding to the differentially expressed TGF-β-related transcripts (Table 2). Notably, SKIL was predicted as a target of hsa-miR-1277-3p (target score: 100), and this miRNA was significantly downregulated in both G3 and G4 tumors (fold change: –3.18 ± 0.98 and –3.81 ± 0.12, respectively), supporting a potential derepression of SKIL in higher-grade tumors.
SMAD1 and SMAD3 were identified as putative targets of hsa-miR-30a-5p and hsa-miR-145-5p, respectively, both exhibiting strong negative fold changes in G3/G4 samples, consistent with their roles as suppressive regulators of these transcripts. Similarly, SMAD4 was predicted to be regulated by hsa-miR-155-3p, which showed moderate downregulation in both high-grade groups.
For BMP2, two regulatory miRNAs—hsa-miR-587 and hsa-miR-302c-5p—were predicted, both displaying decreased expression, particularly in G4 tumors. Interestingly, MAPK1 exhibited an inverse regulatory pattern, with significant upregulation of hsa-miR-130b-3p observed in G3 (+2.87 ± 0.37) and G4 (+3.16 ± 0.12) tumors, suggesting potential feedback or compensatory regulatory mechanisms.

2.3. Methylation Patterns of Selected Genes Related to TGF-β Signaling Pathway in Astrocytic Tumor Samples

Figure 2 illustrates the promoter methylation status of six genes involved in the TGF-β signaling pathway—SKIL, SMAD1, SMAD3, SMAD4, BMP2, and MAPK1—across astrocytic tumors of increasing malignancy (grades G2, G3, and G4). The data reveal a clear trend toward reduced promoter methylation frequency with increasing tumor grade. Notably, MAPK1 and BMP2 exhibit complete or near-complete loss of promoter methylation in G4 samples, indicative of strong epigenetic derepression. SKIL also demonstrates a pronounced shift, with 32 out of 36 G4 cases showing a non-methylated promoter status, suggesting transcriptional upregulation driven by loss of epigenetic silencing. SMAD3 and SMAD4 similarly display decreased promoter methylation in higher-grade tumors; however, the degree of demethylation is more variable, implying potential heterogeneity in their epigenetic regulation. These findings support the notion that promoter hypomethylation contributes to the overexpression of key TGF-β signaling components during astrocytoma progression and may underlie their functional activation in the high-grade tumor microenvironment.
SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2).

2.4. Analysis of Differences in Concentration of SKIL, BMP2, SMAD1, SMAD3, SMAD4, and MAPK1 in Astrocytic Tumor Obtained via ELISA

ELISA-based quantification of selected proteins involved in the TGF-β signaling pathway revealed a significant increase in expression levels correlating with tumor grade (Table 3). Compared to G2 tumors, high-grade astrocytomas (G3 and G4) exhibited elevated levels of SKIL, SMAD1, SMAD3, BMP2, and MAPK1, with most differences reaching statistical significance (p < 0.05). Notably, SMAD4 showed a marked increase only in G4 tumors (Table 3).

2.5. Functional Enrichment Analysis (PPI) of TGF-β Signaling-Associated Genes

Figure 3 presents a protein–protein interaction (PPI) network comprising six key proteins—SKIL, SMAD1, SMAD3, SMAD4, BMP2, and MAPK1—involved in the TGF-β signaling pathway. The network consists of six nodes connected by 12 edges, resulting in an average node degree of 4, indicating that each protein interacts with approximately four others. The high average local clustering coefficient (0.85) reflects a strong degree of interconnectivity, suggesting that these proteins not only engage in pairwise interactions but also form a tightly integrated functional module. As anticipated, STRING-based analysis revealed a high level of interaction confidence, consistent with well-established connections among SMAD proteins, BMP2, MAPK1, and SKIL within the TGF-β signaling cascade. Gene Ontology (GO) enrichment analysis further emphasized canonical TGF-β-related functions, including SMAD complex assembly and receptor-mediated signaling. While these results validate the biological coherence of the selected gene set, they largely reaffirm known pathway architecture rather than reveal novel interactions, as the initial selection was based on curated KEGG pathway membership.
SKIL, SKI-like proto-oncogene; BMP2, Bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, Mitogen-activated protein kinase 1 (ERK2).
The dot plot visualizes enriched biological processes based on similarity grouping at 0.8 cutoff. Dot size corresponds to the number of genes involved, while the color intensity represents the false-discovery rate (FDR). Key enriched processes include SMAD protein complex assembly, TGF-β receptor signaling, BMP signaling, and positive regulation of transcription and miRNA expression.

2.6. Kaplan–Meier Survival Analysis and Cox Proportional Hazards Model for SKIL, BMP2, SMAD1, SMAD3, SMAD4, and MAPK1 in Astrocytic Tumors

Survival analyses revealed a consistent pattern linking elevated protein expression of TGF-β signaling components with worse prognosis in patients with astrocytic tumors. Among the examined markers, SKIL stood out as a significant negative prognostic factor. Higher SKIL levels were associated with an increased risk of death (HR = 1.006; 95% CI: 1.003–1.009; p < 0.001), where every 100 ng/mL increment in concentration translated to an approximately 82% increase in mortality. The Cox model fit was robust (χ2 = 49.49, p < 0.0001), and the Kaplan–Meier curve demonstrated a clear separation in survival between high and low expression groups (Figure 4A).
A similarly strong prognostic value was observed for SMAD1, with higher expression levels linked to a markedly increased risk of mortality (HR = 3.79; 95% CI: 2.21–6.48; p < 0.000001). The average SMAD1 concentration in the cohort was 6.06 ng/mL. The model’s excellent fit (χ2 = 92.22, p < 0.000001) and the pronounced drop in survival between 20 and 40 months underscore SMAD1’s relevance as a marker of aggressive disease (Figure 4B).
The unfavorable survival pattern extended to SMAD3, which was also significantly associated with poor outcomes. Patients with higher SMAD3 expression faced more than a threefold increase in mortality risk (HR = 3.71; 95% CI: 1.80–7.65; p = 0.00039), with a mean concentration of 10.63 ng/mL. The survival curve for SMAD3 mirrored that of SMAD1, with substantial divergence emerging in the 20–40 month follow-up window (χ2 = 88.25, p < 0.000001; Figure 4C).
Further supporting this trend, elevated SMAD4 levels were linked to increased mortality as well (HR = 1.74; 95% CI: 1.47–2.06; p < 0.000001). The average SMAD4 concentration was 5.41 ng/mL. Survival probability dropped significantly beyond 20 months in the high-expression group, as shown by the Kaplan–Meier analysis (χ2 = 58.50, p < 0.000001; Figure 4D).
This negative prognostic pattern also applied to BMP2, whose overexpression was associated with an elevated risk of death (HR = 1.014; 95% CI: 1.006–1.023; p = 0.0012). With a mean level of 465.54 pg/mL, BMP2’s model performance was strong (χ2 = 63.29, p < 0.000001), and survival curves showed a gradual decline in the high-expression group after the 20-month mark (Figure 4E).
Lastly, MAPK1 overexpression was significantly predictive of poor survival outcomes (HR = 1.30; 95% CI: 1.12–1.52; p = 0.0006). The average concentration of MAPK1 was 4.52 ng/mL, and the model remained statistically valid (χ2 = 13.05, p = 0.0003). Kaplan–Meier curves confirmed a steady decline in survival for patients with high MAPK1 levels (Figure 4F).

3. Discussion

The malignant progression of astrocytic tumors is driven by the cumulative dysregulation of transcriptional, epigenetic, and post-transcriptional mechanisms. In this study, we present multi-level evidence demonstrating that key mediators of the TGF-β signaling pathway—SKIL, BMP2, SMAD1, SMAD3, SMAD4, and MAPK1—are consistently upregulated in high-grade astrocytomas. This coordinated overexpression occurs at both the transcript and protein levels and is accompanied by promoter hypomethylation and suppression of regulatory miRNAs [25,26,27]. Together, these converging alterations amplify TGF-β signaling activity and are strongly associated with poor clinical outcomes, reflecting a functionally significant reprogramming of the tumor signaling architecture [28,29,30,31].
These findings expand upon our previous research, which identified epigenetic deregulation of SMAD isoforms in relation to tumor grade and prognosis. The current study broadens this framework by integrating transcriptomic, epigenetic (promoter methylation), post-transcriptional (miRNA), and protein-level data. This multi-omics approach reveals a network-wide dysregulation involving both canonical (SMAD-dependent) and non-canonical (MAPK-mediated) branches of TGF-β signaling [32].
In non-neoplastic glial cells, canonical TGF-β signaling through SMAD2/3 and SMAD4 is predominantly cytostatic and homeostatic [33,34]. However, in gliomas, this pathway is hijacked to promote tumorigenic processes such as epithelial–mesenchymal transition (EMT), immune evasion, and extracellular matrix remodeling. In our analysis, the upregulation of SMAD3 and SMAD4 was associated with increased transcriptional activation of oncogenic targets [35,36].
In addition to canonical signaling, the BMP branch—particularly BMP2, SMAD1, and SMAD4—was significantly upregulated in high-grade tumors. This finding contrasts with earlier studies that linked BMP2 signaling to cellular differentiation and favorable prognosis in glioblastoma. For example, Zhou et al. identified BMP2 mRNA as a component of a favorable glioma grading model, with higher expression linked to lower-grade tumors and improved survival [37]. Similarly, Li et al. showed that a BMP2-mimicking peptide promoted differentiation and reduced self-renewal in glioblastoma stem-like cells [38].
This apparent contradiction may reflect the context-dependent roles of BMP signaling. While BMP2 promotes differentiation in glioma stem-like cells under specific conditions, its upregulation in bulk tumor tissue—as observed here—may instead reflect a non-functional or dysregulated compensatory response, possibly driven by an immunosuppressive or mesenchymal microenvironment. Alternatively, BMP2 expression may be a failed compensatory attempt at differentiation that is overridden by tumor-intrinsic resistance mechanisms or epigenetic silencing of downstream effectors.
These complexities highlight the need for mechanistic studies at single-cell or spatial resolution to delineate whether BMP2 exerts tumor-promoting or tumor-suppressive effects in different tumor compartments or stages. Our findings suggest that BMP2 signaling in glioblastoma does not follow a unidirectional paradigm but may exhibit dual or paradoxical roles depending on molecular subtype, tumor microenvironment, and differentiation status.
Consistent with this, SMAD1, a principal downstream effector of BMP2, was also significantly overexpressed. Although BMP signaling can inhibit growth in normal neuroepithelial cells, SMAD1 activation in gliomas has been linked to mesenchymal transformation and enhanced plasticity [39,40]. The simultaneous upregulation of SMAD1, SMAD3, and SMAD4 suggests a functional convergence between canonical TGF-β and BMP signaling arms, which may cooperatively reinforce transcriptional programs that drive stemness, invasion, and therapeutic resistance [41,42,43].
The role of SKIL in this context warrants particular attention. Traditionally considered a repressor of SMAD-mediated transcription [44,45], SKIL was consistently overexpressed in high-grade tumors and associated with poor prognosis [46]. This suggests that SKIL may act as a context-dependent oncogene, potentially modulating co-activators or chromatin accessibility in response to elevated TGF-β signaling [47]. Indeed, prior studies have linked SKIL overexpression to apoptosis resistance in other tumor types [45,48], underscoring its potential as a therapeutic target.
MAPK1, a key effector of the non-canonical TGF-β signaling axis, was also markedly upregulated in higher-grade tumors [49]. Given its role in ERK signaling and its promoter hypomethylation status, MAPK1 likely supports mitogenic and pro-survival signaling that bypasses canonical SMAD-mediated controls [50,51]. This ERK pathway activation has been previously implicated in glioma cell proliferation, radioresistance, and metabolic reprogramming [52,53,54]. Our data further support the hypothesis that the interplay between canonical and non-canonical TGF-β branches establishes a robust signaling redundancy [55,56] that sustains malignant phenotypes despite potential therapeutic blockade of individual nodes [57,58,59].
On the epigenetic level, promoter hypomethylation of BMP2, SMAD1, and MAPK1 was especially pronounced in grade IV tumors, where promoter regions were almost entirely unmethylated. This pattern reflects transcriptional derepression of genes that are typically silenced in differentiated neural tissue [60,61,62]. Such epigenetic deregulation aligns with the glioma-CpG island methylator phenotype (G-CIMP), previously associated with high-grade, biologically aggressive gliomas [63], and further supports the potential utility of DNA methylation profiling in astrocytoma stratification [64,65,66,67,68].
Post-transcriptional modulation via miRNAs further shaped the expression landscape of the TGF-β signaling [69,70]. The marked downregulation of hsa-miR-145-5p, a validated repressor of SMAD3, permits the accumulation of SMAD3 protein and may amplify oncogenic signaling. Similarly, hsa-miR-199a-5p and hsa-miR-125b-5p, targeting SMAD1 and MAPK1, respectively, were suppressed in higher-grade tumors, facilitating overexpression of their targets. These miRNAs are known to function in neural development and glial differentiation, and their silencing may contribute to dedifferentiation and treatment resistance [71]. Additional repression of hsa-miR-29c-3p and hsa-miR-145-3p, which regulate SKIL and SMAD4, respectively, suggests a broader miRNA-mediated disruption of TGF-β signaling regulation [72,73,74]. Together, these changes represent a network-level release of inhibitory control, promoting sustained signal transduction and transcriptional output.
PPI revealed an enriched and interconnected network linking the overexpressed gene products. SMAD4 emerged as a central scaffold protein that integrates SMAD1 and SMAD3 signaling complexes [75,76]. The dual engagement of SMAD4 by both canonical TGF-β and BMP pathways highlights its functional centrality in coordinating diverse transcriptional outputs relevant to invasion, immune suppression, and therapy resistance [39]. MAPK1, positioned outside the SMAD axis, forms functional interactions with regulators of cell cycle progression and apoptosis, further reinforcing the signaling redundancy characteristic of high-grade tumors [77,78]. SKIL, while canonically a repressor, may act in gliomas through alternative protein–protein interactions not yet fully elucidated but potentially relevant to chromatin accessibility or transcriptional pausing [79]. The overall structure of the PPI network underscores the cohesive functional integration of canonical and non-canonical pathways that drive glioma pathogenesis [80].
Clinically, the prognostic impact of these molecular alterations was striking. Elevated protein levels of SMAD1 and SMAD3 were among the strongest independent predictors of poor overall survival, with hazard ratios exceeding 3.7. High SKIL expression also conferred significant prognostic disadvantage, despite its traditional classification as a transcriptional repressor [81]. These findings suggest that in the glioma context, even canonical repressors may adopt tumor-supportive roles, depending on cellular context and cofactor availability.
The clinical relevance of these proteins extends beyond prognostication. Their expression profiles could guide biomarker-driven therapeutic strategies, including combination treatments that simultaneously inhibit SMAD-mediated transcription and MAPK-driven escape pathways [82,83,84,85,86]. Although Kaplan–Meier analyses demonstrated strong associations between high expression of SKIL, SMAD1, SMAD3, SMAD4, BMP2, and MAPK1 and poor overall survival, it is important to acknowledge that these genes were also significantly upregulated in higher-grade tumors. Thus, the observed survival disadvantage may partly reflect underlying differences in tumor grade rather than purely independent prognostic effects. Future studies incorporating multivariate survival analyses, adjusted for histological grade, are warranted to dissect the relative contributions of gene expression and tumor grade to patient outcomes.
While this study provides a comprehensive, multi-level characterization of TGF-β signaling dysregulation in astrocytic tumors, several limitations must be acknowledged. First, the analysis was conducted on bulk tumor homogenates, limiting cellular resolution and precluding the discrimination of molecular alterations between tumor cells, reactive astrocytes, and non-neoplastic components such as endothelial or immune cells. Single-cell RNA sequencing or spatial transcriptomics would be required to resolve cell-type–specific expression patterns. Second, although the cohort size was sufficient for discovery-phase analyses and statistical modeling, the sample size may limit the generalizability of the findings. Therefore, validation in larger, independent, or multicenter cohorts is necessary to confirm the prognostic significance and biological robustness of the identified alterations. Third, functional validation of predicted microRNA–mRNA interactions was not performed; specifically, luciferase reporter assays and gain/loss-of-function experiments are needed to verify direct regulatory relationships, particularly for key axes such as miR-145-5p–SMAD3. Lastly, assays of downstream pathway activity—such as SMAD nuclear translocation or transcriptional reporter assays—would provide deeper mechanistic insights into TGF-β signaling activation beyond expression-level analyses. While these limitations highlight avenues for future investigation, they do not diminish the mechanistic or clinical relevance of our findings.

4. Materials and Methods

This study methodology was built upon the work conducted in our previous papers [32,87].

4.1. Patient Enrollment and Tissue Collection

Tumor tissue samples were obtained from 65 patients (35 women, 30 men) diagnosed with astrocytic brain tumors classified as grade II (G2; low-grade), grade III (G3; anaplastic), or grade IV (G4; glioblastoma), according to the WHO criteria [4]. All patients underwent elective surgical resection at one of two neurosurgical centers in Krakow, Poland: the Department of Neurosurgery at the 5th Military Clinical Hospital with SP ZOZ Polyclinic or the Neurosurgery Department of St. Raphael’s Hospital. The mean age of patients ranged from 56 to 59 years, depending on tumor grade. Female patients included 10 with grade 2, 7 with grade 3, and 18 with grade 4 tumors; male patients included 7 (G2), 5 (G3), and 18 (G4).
All surgeries were elective and performed between 8:00 and 11:00 AM. A standardized fasting protocol was implemented, with the final preoperative meal scheduled at 6:00 PM the day before surgery. Eligibility criteria included histologically confirmed astrocytoma, age ≥ 18 years, and provision of informed consent. Patients with additional primary or metastatic malignancies, urgent surgical indications, or inadequate imaging or tissue quality were excluded.
Initial tumor suspicion was based on contrast-enhanced CT and confirmed by MRI, which included T1- and T2-weighted, FLAIR, and when indicated, diffusion tensor imaging. Functional MRI and tractography supported preoperative planning for tumors near eloquent brain areas. Intraoperative tools included neuronavigation, 5-ALA-guided fluorescence, and cortical stimulation when appropriate. Final tumor grading was determined via histopathological evaluation based on WHO guidelines [7,8,9].

4.2. Extraction of Total Ribonucleic Acid (RNA) from Tissues

Tumor tissues were homogenized using a T18 Digital Ultra-Turrax system (IKA Poland Ltd., Warsaw, Poland). Total RNA was isolated with TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA), followed by purification using the RNeasy Mini Kit (QIAGEN, Hilden, Germany). DNase I treatment (Fermentas International Inc., Burlington, ON, Canada) was used to eliminate residual genomic DNA.
RNA integrity was assessed by electrophoresis on 1% agarose gels stained with ethidium bromide. RNA concentrations were determined by measuring absorbance at 260 n.

4.3. Gene Expression Microarrays

A curated set of genes involved in the TGF-β signaling cascade was compiled using pathway annotations obtained from the Kyoto Encyclopedia of Genes and Genomes (KEGG), specifically through the PathCards integrative platform (hsa04350; accessed 19 January 2025), corresponding to the “TGF-beta signaling pathway.” [88]. Transcriptomic profiling was conducted using Affymetrix HG-U133A_2.0 arrays (Affymetrix, Santa Clara, CA, USA) and the GeneChip™ 3′ IVT PLUS Reagent Kit (Catalog No. 902416). Protocols were conducted according to the manufacturer’s instructions and previously described methodologies [32,87,89].

4.4. miRNA Profiling and Target Prediction

MicroRNA expression profiling was performed using the GeneChip™ miRNA 2.0 Array (Affymetrix, Santa Clara, CA, USA). Target prediction was performed with TargetScan (http://www.targetscan.org/) [90] and miRDB (http://mirdb.org) [91]. Predicted interactions with confidence scores >80 were considered high-confidence, whereas those with lower scores were marked for future validation [91,92].

4.5. qRT-PCR Validation

Gene expression changes identified by microarrays were validated using qRT-PCR with the SensiFast SYBR No-ROX One-Step Kit (Bioline, London, UK). The thermal cycling protocol consisted of an initial reverse transcription step at 45 °C, followed by polymerase activation at 95 °C for 2 min. This was succeeded by 40 amplification cycles comprising denaturation at 95 °C for 5 s, annealing at 60 °C for 10 s, and extension at 72 °C for 5 s. Gene expression levels were normalized to endogenous controls using the 2−ΔΔCt method. Primer sequences are provided in Table 4.

4.6. DNA Methylation Assessment

Promoter CpG islands were identified using the MethPrimer tool (http://www.urogene.org/cgi-bin/methprimer/methprimer.cgi; accessed on 19 January 2025) [93]. Bisulfite conversion preceded methylation-specific PCR (MSP), performed using the QuantiTect SYBR Green PCR Kit (Qiagen GmbH, Hilden, Germany). Cycling parameters included denaturation (94 °C), annealing (65 °C), and extension (72 °C).
PCR products were resolved on 1% agarose gels stained with ethidium bromide. Fragment identification used pBR322/HaeIII markers. Methylated and unmethylated DNA controls were included (EpiTect Control DNA, Qiagen GmbH, Hilden, Germany). The characteristics of the primers designed for MSP are presented in Abbreviation (Table 5).

4.7. Enzyme-Linked Immunosorbent Assay (ELISA) Quantification

Quantitative assessment of selected signaling and inflammatory proteins in astrocytic tumor samples (grades G2, G3, and G4) was conducted using ELISA, strictly adhering to the manufacturers’ protocols. Each assay was performed in triplicate when sample volume allowed. Absorbance was measured at the appropriate wavelength using a calibrated microplate reader, and protein concentrations were calculated based on standard curves generated with known concentrations of the respective analytes. Protein levels were determined using commercial ELISA kits from MyBioSource Inc. (San Diego, CA, USA). The following targets were analyzed: SMAD1 (Cat. No. MBS2025476), SMAD3 (MBS161553), SMAD4 (MBS450115), SKIL (SKI-like oncogene, MBS8806352), BMP2 (bone morphogenetic protein 2, MBS137998), and MAPK1/ERK2 (mitogen-activated protein kinase 1, MBS2127197). To ensure inter-sample comparability, all protein concentrations were normalized to the total protein content of each lysate, expressed as ng/mL or pg/mL of total protein extract.

4.8. Statistical Methods

Data were processed using StatPlus (v. 1.1) and the Affymetrix Transcriptome Analysis Console. Normality of distributions was assessed using the Shapiro–Wilk test. Differences between groups were assessed using one-way ANOVA followed by Benjamini–Hochberg correction and Scheffé’s post hoc test. The assumption of homogeneity of variance was evaluated using Levene’s test prior to performing ANOVA. Although Scheffé’s test is relatively robust to violations of this assumption, appropriate diagnostics were conducted to confirm the validity of the statistical approach. Student’s t-test was used for selected pairwise comparisons. Survival analyses were performed using Kaplan–Meier curves with the log-rank test, and multivariate associations were assessed using Cox proportional hazards regression. For Kaplan–Meier survival analysis, patients were stratified into “high” and “low” expression groups using the median protein concentration as the cutoff of each biomarker. This median-based approach is widely used for exploratory biomarker analysis in oncology when no established clinical threshold exists. Survival plots were generated using R (V4.4.0) [94] and the survival (V3.8.3) [95,96] and survminer (V0.5.0) [97] packages. Based on national statistics indicating approximately 3270 new brain tumor cases in Poland during 2020–2021 [98], the cohort of 65 participants was considered acceptable. Assuming maximal variance and a 95% confidence level, the estimated margin of error was approximately 12%.
Data were processed using StatPlus (v. 1.1) and the Affymetrix Transcriptome Analysis Console. Normality of distributions was assessed using the Shapiro–Wilk test. Differences between groups were assessed using one-way ANOVA followed by Benjamini–Hochberg correction and Scheffé’s post hoc test. The assumption of homogeneity of variance was evaluated using Levene’s test prior to performing ANOVA. Although Scheffé’s test is relatively robust to violations of this assumption, appropriate diagnostics were conducted to confirm the validity of the statistical approach. Student’s t-test was used for selected pairwise comparisons. Survival analyses were performed using Kaplan–Meier curves with the log-rank test, and multivariate associations were assessed using Cox proportional hazards regression. For Kaplan–Meier survival analysis, patients were stratified into “high” and “low” expression groups using the median protein concentration as the cutoff of each biomarker. This median-based approach is widely used for exploratory biomarker analysis in oncology when no established clinical threshold exists. Survival plots were generated using R (V4.4.0) [94] and the survival (V3.8.3) [95,96] and survminer (V0.5.0) [97] packages. Based on national statistics indicating approximately 3270 new brain tumor cases in Poland during 2020–2021 [98], the cohort of 65 participants was considered acceptable. Assuming maximal variance and a 95% confidence level, the estimated margin of error was approximately 12%.

5. Conclusions

This integrative study demonstrates that astrocytic tumor progression is marked by coordinated dysregulation at the transcriptional, epigenetic, and post-transcriptional levels within the TGF-β signaling pathway. Specifically, the upregulation of SMAD1, SMAD3, SMAD4, BMP2, MAPK1, and SKIL—driven by promoter hypomethylation and reduced expression of tumor-suppressive microRNAs—leads to enhanced activation of both canonical and non-canonical branches of TGF-β signaling. These molecular alterations not only reflect the aggressive biological behavior of high-grade astrocytomas but are also significantly associated with unfavorable patient survival, highlighting their relevance as prognostic biomarkers. The findings offer a compelling rationale for the development of targeted therapeutic approaches aimed at TGF-β pathway components, potentially in combination with epigenetic or microRNA-based interventions. Future research should include functional validation and prospective, multicenter studies to translate these molecular insights into clinically applicable strategies in neuro-oncology.

Author Contributions

Conceptualization, K.S., B.O.G. and D.S. (Dawid Sobański); methodology, D.S. (Damian Strojny) and D.S. (Dawid Sobański); software, B.O.G.; validation, K.S., D.S. (Dawid Sobański) and B.O.G.; visualization, N.Z.; formal analysis, D.S. (Damian Strojny); investigation, K.S. and R.S.; resources, M.M. and P.R.; data curation, K.S. and B.O.G.; writing—original draft preparation, K.S. and P.G.; writing—review and editing, K.S., B.O.G. and D.S. (Dawid Sobański); supervision, B.O.G.; project administration, B.O.G. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

This study was conducted in accordance with the Declaration of Helsinki and received ethical approval from two institutions: the Bioethical Committee of the Regional Medical Chamber in Krakow (reference number: 247/KBL/OIL/2021; dated 11 June 2021) and the Bioethics Committee of the State Academy of Applied Sciences in Przemyśl (approval number: 7/2024; dated 1 August 2024).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patient(s) to publish this paper.

Data Availability Statement

The data presented in this study are available in this article.

Acknowledgments

We thank Oskar Ogloszka for language editing support.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

BMPBone Morphogenetic Protein
BMP2Bone Morphogenetic Protein 2
CNSCentral Nervous System
CNS WHOCNS WHO Classification
Co-SMADCommon Mediator SMAD
CpGCytosine-phosphate-Guanine
DOAJDirectory of Open Access Journals
ELISAEnzyme-Linked Immunosorbent Assay
EMTEpithelial-to-Mesenchymal Transition
ERKExtracellular Signal-Regulated Kinase
FCFold Change
G-CIMPGlioma CpG Island Methylator Phenotype
GOGene Ontology
HRHazard Ratio
IDHIsocitrate Dehydrogenase
LDLinear Dichroism
MGMTO6-Methylguanine-DNA Methyltransferase
miRNAMicroRNA
MSPMethylation-Specific Polymerase Chain Reaction
mRNAMessenger RNA
PCRPolymerase Chain Reaction
PPIProtein–Protein Interaction
qPCRQuantitative Polymerase Chain Reaction
RefSeqNCBI Reference Sequence Database
R-SMADReceptor-Regulated SMAD
RT-qPCRReverse Transcription Quantitative PCR
SKILSKI-like Proto-Oncogene
SMADMothers Against Decapentaplegic Homolog
STRINGSearch Tool for the Retrieval of Interacting Genes
TGF-βTransforming Growth Factor-Beta
TLAThree-Letter Acronym
WHOWorld Health Organization
MAPKMitogen-Activated Protein Kinase
MAPK1Mitogen-Activated Protein Kinase 1

Appendix A

Table A1. Differentially expressed genes between high-grade (G3/G4) and low-grade (G2) astrocytomas identified by microarray profiling (FC > 2.0 or <–2.0; p < 0.05).
Table A1. Differentially expressed genes between high-grade (G3/G4) and low-grade (G2) astrocytomas identified by microarray profiling (FC > 2.0 or <–2.0; p < 0.05).
IDmRNAlog2FC G3 Vs. G2log2FC G4 Vs. G2
206675_s_atSKIL7.43 ± 0.549.18 ± 0.43
215889_at7.42 ± 0.439.54 ± 0.76
217591_at7.18 ± 0.879.71 ± 0.43
225227_at7.21 ± 0.329.81 ± 0.51
210993_s_atSMAD15.76 ± 0.878.18 ± 0.98
227798_at5.87 ± 0.717.34 ± 1.02
205396_atSMAD36.13 ± 0.777.81 ± 1.03
205397_x_at6.21 ± 0.548.12 ± 0.76
205398_s_at6.43 ± 0.918.43 ± 0.76
202526_atSMAD48.16 ± 0.789.12 ± 0.65
202527_s_at8.23 ± 0.609.87 ± 0.17
235725_at8.43 ± 0.869.12 ± 1.09
1565702_at8.11 ± 0.819.54 ± 1.02
1565703_at8.71 ± 0.449.88 ± 1.04
205289_atBMP25.11 ± 0.317.65 ± 0.87
205290_s_at6.09 ± 0.617.53 ± 0.56
208351_s_atMAPK110.11 ± 0.9814.13 ± 1.23
212271_at9.98 ± 1.0714.55 ± 1.24
224620_at9.73 ± 0.4316.12±1.32
224621_at10.16 ± 0.7615.65 ± 0.98
229847_at10.54 ± 0.1114.76 ± 1.19
1552263_at10.17 ± 0.7614.61 ± 1.25
1552264_a_at10.78 ± 0.1914.19 ± 1.28
207530_s_atCDKN2B−2.51 ± 0.32−4.67 ± 0.65
236313_at−2.32 ± 0.34−4.87 ± 0.43
202248_atE2F42.01 ± 0.122.98 ± 0.19
38707_r_at2.12 ± 0.193.09 ± 0.23
221586_s_atE2F52.19 ± 0.283.81 ± 0.51
203304_atBAMBI3.17 ± 0.324.87 ± 0.43
218468_s_atGREM12.18 ± 0.434.12 ± 0.32
218469_at2.22 ± 0.654.18 ± 0.91
206516_atAMH2.02 ± 0.192.16 ± 0.13
204926_atINHBA2.87 ± 0.324.32 ± 0.16
210511_s_at2.91 ± 0.314.51 ± 0.19
227140_at2.71 ± 0.544.81 ± 0.32
205258_atINHBB3.18 ± 0.433.91 ± 0.39
207687_atINHBC2.11 ± 0.712.12 ± 0.21
203075_atSMAD23.43 ± 0.194.32 ± 0.32
203076_s_at3.32 ± 0.214.54 ± 0.22
203077_s_at3.51 ± 0.324.32 ± 0.12
226563_at3.76 ± 0.314.38 ± 0.76
235598_at3,81 ± 0.194.44 ± 0.18
239271_at3.55 ± 0.374.79 ± 0.23
205187_atSMAD53.43 ± 0.434.31 ± 0.43
205188_s_at3.43 ± 0.214.51 ± 0.32
225219_at3.15 ± 0.814.48 ± 0.19
225223_at3.81 ± 0.194.65±0.23
235451_at3.45 ± 0.184.71 ± 0.43
207069_s_atSMAD6−4.13 ± 0.17−4.81 ± 0.18
209886_s_at−4.32 ± 0.31−4.65 ± 0.43
209887_at−4.19 ± 0.54−4.71 ± 0.44
213565_s_at−4.32 ± 0.43−4.91 ± 0.23
204790_atSMAD7−3.21 ± 0.51−4.13 ± 0.17
206320_s_atSMAD92.12 ± 0.182.87 ± 0.32
227719_at2.10 ± 0.652.51 ± 0.12
213044_atROCK13.31 ± 0.254.18 ± 0.19
214578_s_at3.32 ± 0.324.56 ± 0.25
230239_at3.71 ± 0.544.32 ± 0.19
235854_x_at3.45 ± 0.514.51 ± 0.23
1561486_at3.81 ± 0.674.44 ± 0.44
205596_s_atSMURF23.12 ± 0.713.98 ± 0.51
227489_at3.18 ± 0.194.05 ± 0.62
230820_at3.65 ± 0.284.09 ± 0.71
232020_at3.23 ± 0.543.98 ± 0.32
211518_s_atBMP4−3.91 ± 0.19−4.54 ± 0.18
205430_atBMP5−3.49 ± 0.31−3.71 ± 0.32
205431_s_at−4.01 ± 0.29−3.87 ± 0.47
206176_atBMP6−2.19 ± 0.71−3.09 ± 0.61
215042_at−2.11 ± 0.18−3.32 ± 0.54
241141_at−2.87 ± 0.15−3.21 ± 0.43
209590_atBMP7−4.54 ± 0.71−4.98 ± 0.17
209591_s_at−4.32 ± 0.29−4.71 ± 0.65
211259_s_at−4.87 ± 0.34−4.76 ± 0.23
211260_at−4.54 ± 0.71−4.87 ± 0.35
209747_atTGFB32.76 ± 0.533.45 ± 0.46
SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2); CDKN2B, cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4); E2F4, E2F transcription factor 4; E2F5, E2F transcription factor 5; BAMBI, BMP and activin membrane-bound inhibitor; GREM1, gremlin 1, DAN family BMP antagonist; AMH, anti-Müllerian hormone; INHBA, inhibin subunit beta A; INHBB, inhibin subunit beta B; INHBC, inhibin subunit beta C; SMAD2, SMAD family member 2; SMAD5, SMAD family member 5; SMAD6, SMAD family member 6; SMAD7, SMAD family member 7; SMAD9, SMAD family member 9 (also known as SMAD8); ROCK1, Rho-associated coiled-coil containing protein kinase 1; SMURF2, SMAD specific E3 ubiquitin protein ligase 2; BMP4, bone morphogenetic protein 4; BMP5, bone morphogenetic protein 5; BMP6, bone morphogenetic protein 6; BMP7, bone morphogenetic protein 7; TGFB3, transforming growth factor beta 3.

References

  1. Cano-Valdez, A.M.; Sevilla-Lizcano, D.B. Pathological Classification of Brain Tumors. In Principles of Neuro-Oncology; Monroy-Sosa, A., Chakravarthi, S.S., De La Garza-Salazar, J.G., Meneses Garcia, A., Kassam, A.B., Eds.; Springer International Publishing: Cham, Switzerland, 2021; pp. 75–105. ISBN 978-3-030-54878-0. [Google Scholar]
  2. Pienkowski, T.; Kowalczyk, T.; Kretowski, A.; Ciborowski, M. A Review of Gliomas-Related Proteins. Characteristics of Potential Biomarkers. Am. J. Cancer Res. 2021, 11, 3425. [Google Scholar]
  3. Machnik, G.; Bułdak, Ł.; Zapletal-Pudełko, K.; Grabarek, B.O.; Staszkiewicz, R.; Sobański, D.; Okopień, B. The Impact of Wound-Healing Assay, Phorbol Myristate Acetate (PMA) Stimulation and siRNA-Mediated FURIN Gene Silencing on Endogenous Retroviral ERVW-1 Expression Level in U87-MG Astrocytoma Cells. Adv. Med. Sci. 2024, 69, 113–124. [Google Scholar] [CrossRef] [PubMed]
  4. Chen, Z.; Hambardzumyan, D. Immune Microenvironment in Glioblastoma Subtypes. Front. Immunol. 2018, 9, 1004. [Google Scholar] [CrossRef] [PubMed]
  5. Hanif, F.; Muzaffar, K.; Perveen, K.; Malhi, S.M.; Simjee, S.U. Glioblastoma Multiforme: A Review of Its Epidemiology and Pathogenesis through Clinical Presentation and Treatment. Asian Pac. J. Cancer Prev. 2017, 18, 3–9. [Google Scholar] [CrossRef]
  6. Reifenberger, G.; Wirsching, H.-G.; Knobbe-Thomsen, C.B.; Weller, M. Advances in the Molecular Genetics of Gliomas—Implications for Classification and Therapy. Nat. Rev. Clin. Oncol. 2017, 14, 434–452. [Google Scholar] [CrossRef]
  7. Bale, T.A.; Rosenblum, M.K. The 2021 WHO Classification of Tumors of the Central Nervous System: An Update on Pediatric Low-grade Gliomas and Glioneuronal Tumors. Brain Pathol. 2022, 32, e13060. [Google Scholar] [CrossRef]
  8. Horbinski, C.; Berger, T.; Packer, R.J.; Wen, P.Y. Clinical Implications of the 2021 Edition of the WHO Classification of Central Nervous System Tumours. Nat. Rev. Neurol. 2022, 18, 515–529. [Google Scholar] [CrossRef]
  9. Louis, D.N.; Perry, A.; Wesseling, P.; Brat, D.J.; Cree, I.A.; Figarella-Branger, D.; Hawkins, C.; Ng, H.K.; Pfister, S.M.; Reifenberger, G.; et al. The 2021 WHO Classification of Tumors of the Central Nervous System: A Summary. Neuro Oncol. 2021, 23, 1231–1251. [Google Scholar] [CrossRef]
  10. Sokolenko, A.P.; Imyanitov, E.N. Molecular Diagnostics in Clinical Oncology. Front. Mol. Biosci. 2018, 5, 76. [Google Scholar] [CrossRef]
  11. Ceccarelli, M.; Barthel, F.P.; Malta, T.M.; Sabedot, T.S.; Salama, S.R.; Murray, B.A.; Morozova, O.; Newton, Y.; Radenbaugh, A.; Pagnotta, S.M.; et al. Molecular Profiling Reveals Biologically Discrete Subsets and Pathways of Progression in Diffuse Glioma. Cell 2016, 164, 550–563. [Google Scholar] [CrossRef]
  12. Radke, J.; Koch, A.; Pritsch, F.; Schumann, E.; Misch, M.; Hempt, C.; Lenz, K.; Löbel, F.; Paschereit, F.; Heppner, F.L.; et al. Predictive MGMT Status in a Homogeneous Cohort of IDH Wildtype Glioblastoma Patients. Acta Neuropathol. Commun. 2019, 7, 89. [Google Scholar] [CrossRef]
  13. Labreche, K.; Kinnersley, B.; Berzero, G.; Di Stefano, A.L.; Rahimian, A.; Detrait, I.; Marie, Y.; Grenier-Boley, B.; Hoang-Xuan, K.; Delattre, J.-Y.; et al. Diffuse Gliomas Classified by 1p/19q Co-Deletion, TERT Promoter and IDH Mutation Status Are Associated with Specific Genetic Risk Loci. Acta Neuropathol. 2018, 135, 743–755. [Google Scholar] [CrossRef]
  14. Cekanaviciute, E.; Dietrich, H.K.; Axtell, R.C.; Williams, A.M.; Egusquiza, R.; Wai, K.M.; Koshy, A.A.; Buckwalter, M.S. Astrocytic TGF-β Signaling Limits Inflammation and Reduces Neuronal Damage during Central Nervous System Toxoplasma Infection. J. Immunol. 2014, 193, 139–149. [Google Scholar] [CrossRef]
  15. Yang, J.; Xu, J.; Wang, W.; Zhang, B.; Yu, X.; Shi, S. Epigenetic Regulation in the Tumor Microenvironment: Molecular Mechanisms and Therapeutic Targets. Signal Transduct. Target. Ther. 2023, 8, 210. [Google Scholar] [CrossRef]
  16. Wu, Q.; Berglund, A.E.; Etame, A.B. The Impact of Epigenetic Modifications on Adaptive Resistance Evolution in Glioblastoma. Int. J. Mol. Sci. 2021, 22, 8324. [Google Scholar] [CrossRef]
  17. Dexheimer, P.J.; Cochella, L. MicroRNAs: From Mechanism to Organism. Front. Cell Dev. Biol. 2020, 8, 409. [Google Scholar] [CrossRef] [PubMed]
  18. Saliminejad, K.; Khorram Khorshid, H.R.; Soleymani Fard, S.; Ghaffari, S.H. An Overview of microRNAs: Biology, Functions, Therapeutics, and Analysis Methods. J. Cell. Physiol. 2019, 234, 5451–5465. [Google Scholar] [CrossRef] [PubMed]
  19. Naik, A.; Thakur, N. Epigenetic Regulation of TGF-β and Vice Versa in Cancers—A Review on Recent Developments. Biochim. Biophys. Acta (BBA)—Rev. Cancer 2024, 1879, 189219. [Google Scholar] [CrossRef] [PubMed]
  20. Park, J.W.; Lee, K.; Kim, E.E.; Kim, S.-I.; Park, S.-H. Brain Tumor Classification by Methylation Profile. J. Korean Med. Sci. 2023, 38, e356. [Google Scholar] [CrossRef]
  21. Ahuja, S.; Zaheer, S. Multifaceted TGF-β Signaling, a Master Regulator: From Bench-to-Bedside, Intricacies, and Complexities. Cell Biol. Int. 2024, 48, 87–127. [Google Scholar] [CrossRef]
  22. Galassi, C.; Esteller, M.; Vitale, I.; Galluzzi, L. Epigenetic Control of Immunoevasion in Cancer Stem Cells. Trends Cancer 2024, 10, 1052–1071. [Google Scholar] [CrossRef] [PubMed]
  23. Ding, Y.; Zhou, G.; Hu, W. Epigenetic Regulation of TGF-β Pathway and Its Role in Radiation Response. Int. J. Radiat. Biol. 2024, 100, 834–848. [Google Scholar] [CrossRef] [PubMed]
  24. Kurowska, N.; Strzalka-Mrozik, B.; Madej, M.; Pająk, K.; Kruszniewska-Rajs, C.; Kaspera, W.; Gola, J.M. Differences in the Expression Patterns of TGFβ Isoforms and Associated Genes in Astrocytic Brain Tumors. Cancers 2022, 14, 1876. [Google Scholar] [CrossRef] [PubMed]
  25. Boroń, D.; Zmarzły, N.; Wierzbik-Strońska, M.; Rosińczuk, J.; Mieszczański, P.; Grabarek, B.O. Recent Multiomics Approaches in Endometrial Cancer. Int. J. Mol. Sci. 2022, 23, 1237. [Google Scholar] [CrossRef]
  26. Fu, Y.; Tao, J.; Liu, T.; Liu, Y.; Qiu, J.; Su, D.; Wang, R.; Luo, W.; Cao, Z.; Weng, G.; et al. Unbiasedly Decoding the Tumor Microenvironment with Single-Cell Multiomics Analysis in Pancreatic Cancer. Mol. Cancer 2024, 23, 140. [Google Scholar] [CrossRef]
  27. Li, L.; Bai, L.; Lin, H.; Dong, L.; Zhang, R.; Cheng, X.; Liu, Z.; Ouyang, Y.; Ding, K. Multiomics Analysis of Tumor Mutational Burden across Cancer Types. Comput. Struct. Biotechnol. J. 2021, 19, 5637–5646. [Google Scholar] [CrossRef]
  28. Galozzi, P.; Bindoli, S.; Doria, A.; Sfriso, P. The Revisited Role of Interleukin-1 Alpha and Beta in Autoimmune and Inflammatory Disorders and in Comorbidities. Autoimmun. Rev. 2021, 20, 102785. [Google Scholar] [CrossRef]
  29. Nigam, M.; Mishra, A.P.; Deb, V.K.; Dimri, D.B.; Tiwari, V.; Bungau, S.G.; Bungau, A.F.; Radu, A.-F. Evaluation of the Association of Chronic Inflammation and Cancer: Insights and Implications. Biomed. Pharmacother. 2023, 164, 115015. [Google Scholar] [CrossRef]
  30. Michels, N.; van Aart, C.; Morisse, J.; Mullee, A.; Huybrechts, I. Chronic Inflammation towards Cancer Incidence: A Systematic Review and Meta-Analysis of Epidemiological Studies. Crit. Rev. Oncol./Hematol. 2021, 157, 103177. [Google Scholar] [CrossRef]
  31. Basheer, A.S.; Abas, F.; Othman, I.; Naidu, R. Role of Inflammatory Mediators, Macrophages, and Neutrophils in Glioma Maintenance and Progression: Mechanistic Understanding and Potential Therapeutic Applications. Cancers 2021, 13, 4226. [Google Scholar] [CrossRef]
  32. Skora, K.; Strojny, D.; Sobanski, D.; Staszkiewicz, R.; Brys, K.; Gogol, P.; Bereza, K.; Masternak, M.; Grabarek, B.O. Multi-Omics Profiling of TGF-β Isoforms and Regulatory miRNAs in Astrocytic Tumors Reveals TGF-β-3 as a Prognostic Biomarker. Front. Oncol. 2025, 15, 1592685. [Google Scholar] [CrossRef] [PubMed]
  33. Boye, A. A Cytokine in Turmoil: Transforming Growth Factor Beta in Cancer. Biomed. Pharmacother. 2021, 139, 111657. [Google Scholar] [CrossRef] [PubMed]
  34. Golán-Cancela, I.; Caja, L. The TGF-β Family in Glioblastoma. Int. J. Mol. Sci. 2024, 25, 1067. [Google Scholar] [CrossRef] [PubMed]
  35. Lee, J.H.; Massagué, J. TGF-β in Developmental and Fibrogenic EMTs. Semin. Cancer Biol. 2022, 86, 136–145. [Google Scholar] [CrossRef]
  36. Wang, X.; Eichhorn, P.J.A.; Thiery, J.P. TGF-β, EMT, and Resistance to Anti-Cancer Treatment. Semin. Cancer Biol. 2023, 97, 1–11. [Google Scholar] [CrossRef]
  37. Zhou, K.; Zhao, Z.; Li, S.; Liu, Y.; Li, G.; Jiang, T. A New Glioma Grading Model Based on Histopathology and Bone Morphogenetic Protein 2 mRNA Expression. Sci. Rep. 2020, 10, 18420. [Google Scholar] [CrossRef]
  38. Rampazzo, E.; Dettin, M.; Maule, F.; Scabello, A.; Calvanese, L.; D’Auria, G.; Falcigno, L.; Porcù, E.; Zamuner, A.; Della Puppa, A.; et al. A Synthetic BMP-2 Mimicking Peptide Induces Glioblastoma Stem Cell Differentiation. Biochim. Biophys. Acta Gen. Subj. 2017, 1861, 2282–2292. [Google Scholar] [CrossRef]
  39. Zou, M.-L.; Chen, Z.-H.; Teng, Y.-Y.; Liu, S.-Y.; Jia, Y.; Zhang, K.-W.; Sun, Z.-L.; Wu, J.-J.; Yuan, Z.-D.; Feng, Y.; et al. The Smad Dependent TGF-β and BMP Signaling Pathway in Bone Remodeling and Therapies. Front. Mol. Biosci. 2021, 8, 593310. [Google Scholar] [CrossRef]
  40. Miao, X.; Yuan, J.; Wu, J.; Zheng, J.; Zheng, W.; Wang, F.; Wang, C.; Li, X.; Liu, S.; Shi, Z.; et al. Bone Morphogenetic Protein-2 Promotes Osteoclasts-Mediated Osteolysis via Smad1 and P65 Signaling Pathways. Spine 2021, 46, E234–E242. [Google Scholar] [CrossRef]
  41. Jing, N.; Gao, W.-Q.; Fang, Y.-X. Regulation of Formation, Stemness and Therapeutic Resistance of Cancer Stem Cells. Front. Cell Dev. Biol. 2021, 9, 641498. [Google Scholar] [CrossRef]
  42. Zhou, W.; Yan, K.; Xi, Q. BMP Signaling in Cancer Stemness and Differentiation. Cell Regen. 2023, 12, 37. [Google Scholar] [CrossRef]
  43. Ganjoo, S.; Puebla-Osorio, N.; Nanez, S.; Hsu, E.; Voss, T.; Barsoumian, H.; Duong, L.K.; Welsh, J.W.; Cortez, M.A. Bone Morphogenetic Proteins, Activins, and Growth and Differentiation Factors in Tumor Immunology and Immunotherapy Resistance. Front. Immunol. 2022, 13, 1033642. [Google Scholar] [CrossRef]
  44. Abdulmawjood, B.; Costa, B.; Roma-Rodrigues, C.; Baptista, P.V.; Fernandes, A.R. Genetic Biomarkers in Chronic Myeloid Leukemia: What Have We Learned So Far? Int. J. Mol. Sci. 2021, 22, 12516. [Google Scholar] [CrossRef]
  45. Heider, T.R.; Lyman, S.; Schoonhoven, R.; Behrns, K.E. Ski Promotes Tumor Growth through Abrogation of Transforming Growth Factor-Beta Signaling in Pancreatic Cancer. Ann. Surg. 2007, 246, 61–68. [Google Scholar] [CrossRef]
  46. Kawamura, E.; Matsubara, T.; Daikoku, A.; Deguchi, S.; Kinoshita, M.; Yuasa, H.; Urushima, H.; Odagiri, N.; Motoyama, H.; Kotani, K.; et al. Suppression of Intrahepatic Cholangiocarcinoma Cell Growth by SKI via Upregulation of the CDK Inhibitor P21. FEBS Open Bio 2022, 12, 2122–2135. [Google Scholar] [CrossRef]
  47. Liao, H.-Y.; Da, C.-M.; Wu, Z.-L.; Zhang, H.-H. Ski: Double Roles in Cancers. Clin. Biochem. 2021, 87, 1–12. [Google Scholar] [CrossRef]
  48. Tavleeva, M.M.; Belykh, E.S.; Rybak, A.V.; Rasova, E.E.; Chernykh, A.A.; Ismailov, Z.B.; Velegzhaninov, I.O. Effects of Antioxidant Gene Overexpression on Stress Resistance and Malignization In Vitro and In Vivo: A Review. Antioxidants 2022, 11, 2316. [Google Scholar] [CrossRef]
  49. Liu, H.; Tang, T. MAPK Signaling Pathway-Based Glioma Subtypes, Machine-Learning Risk Model, and Key Hub Proteins Identification. Sci. Rep. 2023, 13, 19055. [Google Scholar] [CrossRef]
  50. Aashaq, S.; Batool, A.; Mir, S.A.; Beigh, M.A.; Andrabi, K.I.; Shah, Z.A. TGF-β Signaling: A Recap of SMAD-Independent and SMAD-Dependent Pathways. J. Cell Physiol. 2022, 237, 59–85. [Google Scholar] [CrossRef]
  51. Turati, M.; Mousset, A.; Issa, N.; Turtoi, A.; Ronca, R. TGF-β Mediated Drug Resistance in Solid Cancer. Cytokine Growth Factor. Rev. 2023, 71–72, 54–65. [Google Scholar] [CrossRef]
  52. Papa, S.; Choy, P.M.; Bubici, C. The ERK and JNK Pathways in the Regulation of Metabolic Reprogramming. Oncogene 2019, 38, 2223–2240. [Google Scholar] [CrossRef]
  53. Strickland, M.; Stoll, E.A. Metabolic Reprogramming in Glioma. Front. Cell Dev. Biol. 2017, 5, 43. [Google Scholar] [CrossRef]
  54. Ouellette, M.M.; Zhou, S.; Yan, Y. Cell Signaling Pathways That Promote Radioresistance of Cancer Cells. Diagnostics 2022, 12, 656. [Google Scholar] [CrossRef]
  55. Hamosh, A.; Scott, A.F.; Amberger, J.; Valle, D.; McKusick, V.A. Online Mendelian Inheritance in Man (OMIM). Hum. Mutat. 2000, 15, 57–61. [Google Scholar] [CrossRef]
  56. Jean-Quartier, C.; Jeanquartier, F.; Holzinger, A. Open Data for Differential Network Analysis in Glioma. Int. J. Mol. Sci. 2020, 21, 547. [Google Scholar] [CrossRef]
  57. Hao, J.; Cao, Y.; Yu, H.; Zong, L.; An, R.; Xue, Y. Effect of MAP3K8 on Prognosis and Tumor-Related Inflammation in Renal Clear Cell Carcinoma. Front. Genet. 2021, 12, 674613. [Google Scholar] [CrossRef]
  58. Chiu, C.-Y.; Willis-Owen, S.A.G.; Wong, K.C.C.; Farrow, S.N.; Cookson, W.O.C.; Moffatt, M.F.; Zhang, Y. MAP3K8 Is a Potential Therapeutic Target in Airway Epithelial Inflammation. J. Inflamm. 2024, 21, 27. [Google Scholar] [CrossRef]
  59. Gianì, F.; Russo, G.; Pennisi, M.; Sciacca, L.; Frasca, F.; Pappalardo, F. Computational Modeling Reveals MAP3K8 as Mediator of Resistance to Vemurafenib in Thyroid Cancer Stem Cells. Bioinformatics 2019, 35, 2267–2275. [Google Scholar] [CrossRef]
  60. Weng, J.Y.; Salazar, N. DNA Methylation Analysis Identifies Patterns in Progressive Glioma Grades to Predict Patient Survival. Int. J. Mol. Sci. 2021, 22, 1020. [Google Scholar] [CrossRef]
  61. Ferreyra Vega, S.; Olsson Bontell, T.; Corell, A.; Smits, A.; Jakola, A.S.; Carén, H. DNA Methylation Profiling for Molecular Classification of Adult Diffuse Lower-Grade Gliomas. Clin. Epigenetics 2021, 13, 102. [Google Scholar] [CrossRef]
  62. Waha, A.; Rodrigues, F.J.; Waha, A.; Meyer-Puttlitz, B.; Cavenee, W.K.; Huang, T.H.-M.; Wiestler, O.D.; Yan, P.S. Methylation Profiling Identifies Epigenetic Markers for High-Grade Gliomas. Cancer Genom. Proteom. 2004, 1, 209–214. [Google Scholar]
  63. de Souza, C.F.; Sabedot, T.S.; Malta, T.M.; Stetson, L.; Morozova, O.; Sokolov, A.; Laird, P.W.; Wiznerowicz, M.; Iavarone, A.; Snyder, J.; et al. A Distinct DNA Methylation Shift in a Subset of Glioma CpG Island Methylator Phenotypes during Tumor Recurrence. Cell Rep. 2018, 23, 637–651. [Google Scholar] [CrossRef]
  64. Lebrun, L.; Bizet, M.; Melendez, B.; Alexiou, B.; Absil, L.; Van Campenhout, C.; D’Haene, N.; Rorive, S.; Fuks, F.; Decaestecker, C.; et al. Analyses of DNA Methylation Profiling in the Diagnosis of Intramedullary Astrocytomas. J. Neuropathol. Exp. Neurol. 2021, 80, 663–673. [Google Scholar] [CrossRef]
  65. Pratt, D.; Sahm, F.; Aldape, K. DNA Methylation Profiling as a Model for Discovery and Precision Diagnostics in Neuro-Oncology. Neuro Oncol. 2021, 23, S16–S29. [Google Scholar] [CrossRef]
  66. Singh, J.; Sahu, S.; Mohan, T.; Mahajan, S.; Sharma, M.C.; Sarkar, C.; Suri, V. Current Status of DNA Methylation Profiling in Neuro-Oncology as a Diagnostic Support Tool: A Review. Neurooncol Pract. 2023, 10, 518–526. [Google Scholar] [CrossRef]
  67. White, C.L.; Kinross, K.M.; Moore, M.K.; Rasouli, E.; Strong, R.; Jones, J.M.; Cain, J.E.; Sturm, D.; Sahm, F.; Jones, D.T.W.; et al. Implementation of DNA Methylation Array Profiling in Pediatric Central Nervous System Tumors: The AIM BRAIN Project: An Australian and New Zealand Children’s Haematology/Oncology Group Study. J. Mol. Diagn. 2023, 25, 709–728. [Google Scholar] [CrossRef]
  68. Zmarzły, N.; Wojdas, E.; Skubis, A.; Sikora, B.; Mazurek, U. DNA Methylation: Gene Expression Regulation. Acta Univ. Lodziensis. Folia Biol. Et Oecologica 2017, 12, 1–10. [Google Scholar] [CrossRef]
  69. Guo, L.; Zhang, Y.; Zhang, L.; Huang, F.; Li, J.; Wang, S. MicroRNAs, TGF-β Signaling, and the Inflammatory Microenvironment in Cancer. Tumour Biol. 2016, 37, 115–125. [Google Scholar] [CrossRef]
  70. Saadh, M.J.; Allela, O.Q.B.; Sattay, Z.J.; Al Zuhairi, R.A.H.; Ahmad, H.; Eldesoky, G.E.; Adil, M.; Ali, M.S. Deciphering the Functional Landscape and Therapeutic Implications of Noncoding RNAs in the TGF-β Signaling Pathway in Colorectal Cancer: A Comprehensive Review. Pathol. Res. Pract. 2024, 255, 155158. [Google Scholar] [CrossRef]
  71. Gong, Y.; Gong, D.; Liu, S.; Gong, X.; Xiong, J.; Zhang, J.; Jiang, L.; Liu, J.; Zhu, L.; Luo, H.; et al. Deciphering the Role of NcRNAs in Pancreatic Cancer Immune Evasion and Drug Resistance: A New Perspective for Targeted Therapy. Front. Immunol. 2024, 15, 1480572. [Google Scholar] [CrossRef]
  72. Wu, Y.; Xu, J.; Xu, J.; Cheng, J.; Jiao, D.; Zhou, C.; Dai, Y.; Chen, Q. Lower Serum Levels of miR-29c-3p and miR-19b-3p as Biomarkers for Alzheimer’s Disease. Tohoku J. Exp. Med. 2017, 242, 129–136. [Google Scholar] [CrossRef]
  73. Hashemi, M.; Hajimazdarany, S.; Mohan, C.D.; Mohammadi, M.; Rezaei, S.; Olyaee, Y.; Goldoost, Y.; Ghorbani, A.; Mirmazloomi, S.R.; Gholinia, N.; et al. Long Non-Coding RNA/Epithelial-Mesenchymal Transition Axis in Human Cancers: Tumorigenesis, Chemoresistance, and Radioresistance. Pharmacol. Res. 2022, 186, 106535. [Google Scholar] [CrossRef] [PubMed]
  74. González-Cantó, E.; Monteiro, M.; Aghababyan, C.; Ferrero-Micó, A.; Navarro-Serna, S.; Mellado-López, M.; Tomás-Pérez, S.; Sandoval, J.; Llueca, A.; Herreros-Pomares, A.; et al. Reduced Levels of miR-145-3p Drive Cell Cycle Progression in Advanced High-Grade Serous Ovarian Cancer. Cells 2024, 13, 1904. [Google Scholar] [CrossRef] [PubMed]
  75. Zhao, M.; Mishra, L.; Deng, C.-X. The Role of TGF-β/SMAD4 Signaling in Cancer. Int. J. Biol. Sci. 2018, 14, 111–123. [Google Scholar] [CrossRef] [PubMed]
  76. Lucarelli, P.; Schilling, M.; Kreutz, C.; Vlasov, A.; Boehm, M.E.; Iwamoto, N.; Steiert, B.; Lattermann, S.; Wäsch, M.; Stepath, M.; et al. Resolving the Combinatorial Complexity of Smad Protein Complex Formation and Its Link to Gene Expression. Cell Syst. 2018, 6, 75–89.e11. [Google Scholar] [CrossRef]
  77. Yue, J.; López, J.M. Understanding MAPK Signaling Pathways in Apoptosis. Int. J. Mol. Sci. 2020, 21, 2346. [Google Scholar] [CrossRef]
  78. Wang, Z. Regulation of Cell Cycle Progression by Growth Factor-Induced Cell Signaling. Cells 2021, 10, 3327. [Google Scholar] [CrossRef]
  79. Mansisidor, A.R.; Risca, V.I. Chromatin Accessibility: Methods, Mechanisms, and Biological Insights. Nucleus 2022, 13, 236–276. [Google Scholar] [CrossRef]
  80. Li, B.; Liu, Y.; Sun, S. Proton Pump Inhibitors Display Anti-Tumour Potential in Glioma. Cell Prolif. 2023, 56, e13321. [Google Scholar] [CrossRef]
  81. Wang, C.; Song, W.; Zhang, Y.; Deng, H.; Zhou, Z.; Zhu, J.; Wang, X. SKIL Promotes Pancreatic Cancer Metastasis by Inhibiting TSPYL2 to Activate the TGF-β Pathway. Cancer Innov. 2025, 4, e70011. [Google Scholar] [CrossRef]
  82. Luo, K. Signaling Cross Talk between TGF-β/Smad and Other Signaling Pathways. Cold Spring Harb. Perspect. Biol. 2017, 9, a022137. [Google Scholar] [CrossRef]
  83. Yan, C.; Wang, J.; Yang, Y.; Ma, W.; Chen, X. Molecular Biomarker-Guided Anti-Angiogenic Targeted Therapy for Malignant Glioma. J. Cell Mol. Med. 2019, 23, 4876–4882. [Google Scholar] [CrossRef]
  84. Renovanz, M.; Kurz, S.C.; Rieger, J.; Walter, B.; Becker, H.; Hille, H.; Bombach, P.; Rieger, D.; Grosse, L.; Häusser, L.; et al. Clinical Outcome of Biomarker-Guided Therapies in Adult Patients with Tumors of the Nervous System. Neurooncol Adv. 2023, 5, vdad012. [Google Scholar] [CrossRef]
  85. Shah, S.; Nag, A.; Sachithanandam, S.V.; Lucke-Wold, B. Predictive and Prognostic Significance of Molecular Biomarkers in Glioblastoma. Biomedicines 2024, 12, 2664. [Google Scholar] [CrossRef]
  86. Śledzińska, P.; Bebyn, M.G.; Furtak, J.; Kowalewski, J.; Lewandowska, M.A. Prognostic and Predictive Biomarkers in Gliomas. Int. J. Mol. Sci. 2021, 22, 10373. [Google Scholar] [CrossRef]
  87. Skóra, K.; Strojny, D.; Sobański, D.; Staszkiewicz, R.; Gogol, P.; Miller, M.; Grabarek, B.O. Analysis of the Expression Patterns of Tumor Necrosis Factor Alpha Signaling Pathways and Regulatory MicroRNAs in Astrocytic Tumors. Int. J. Mol. Sci. 2025, 26, 5892. [Google Scholar] [CrossRef]
  88. Kanehisa, M.; Furumichi, M.; Sato, Y.; Kawashima, M.; Ishiguro-Watanabe, M. KEGG for Taxonomy-Based Analysis of Pathways and Genomes. Nucleic Acids Res. 2023, 51, D587–D592. [Google Scholar] [CrossRef] [PubMed]
  89. Staszkiewicz, R.; Sobański, D.; Pulka, W.; Gładysz, D.; Gadzieliński, M.; Strojny, D.; Grabarek, B.O. Variances in the Expression Profile of Circadian Clock-Related Genes in Astrocytic Brain Tumors. Cancers 2024, 16, 2335. [Google Scholar] [CrossRef]
  90. Agarwal, V.; Bell, G.W.; Nam, J.-W.; Bartel, D.P. Predicting Effective microRNA Target Sites in Mammalian mRNAs. eLife 2015, 4, e05005. [Google Scholar] [CrossRef] [PubMed]
  91. Chen, Y.; Wang, X. miRDB: An Online Database for Prediction of Functional microRNA Targets. Nucleic Acids Res. 2020, 48, D127–D131. [Google Scholar] [CrossRef] [PubMed]
  92. Liu, W.; Wang, X. Prediction of Functional microRNA Targets by Integrative Modeling of microRNA Binding and Target Expression Data. Genome Biol. 2019, 20, 18. [Google Scholar] [CrossRef]
  93. Li, L.-C.; Dahiya, R. MethPrimer: Designing Primers for Methylation PCRs. Bioinformatics 2002, 18, 1427–1431. [Google Scholar] [CrossRef]
  94. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2024. [Google Scholar]
  95. Therneau, T.M.; Grambsch, P.M. Modeling Survival Data: Extending the Cox Model; Springer: New York, NY, USA, 2000; ISBN 0-387-98784-3. [Google Scholar]
  96. Therneau, T.M. A Package for Survival Analysis in R. Available online: https://ftp2.uib.no/cran/web/packages/survival/vignettes/survival.pdf (accessed on 15 July 2025).
  97. Kassambara, A.; Kosinski, M.; Biecek, P. Survminer: Drawing Survival Curves Using “Ggplot2”. Available online: https://cir.nii.ac.jp/crid/1360300776601012224 (accessed on 15 July 2025).
  98. Raporty|Krajowy Rejestr Nowotworów. Available online: http://onkologia.org.pl/pl/raporty (accessed on 5 May 2025).
Ijms 26 07798 g001
Figure 1. Quantitative expression analysis of six TGF-β-associated mRNAs using reverse transcription quantitative polymerase chain reaction (RT-qPCR). Data represent mean ± standard deviation (SD) from three independent experiments (n = 3). Statistical significance was evaluated using one-way ANOVA with post hoc testing. Asterisk indicates statistically significant differences compared to G2 (p < 0.05). SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2).
Figure 1. Quantitative expression analysis of six TGF-β-associated mRNAs using reverse transcription quantitative polymerase chain reaction (RT-qPCR). Data represent mean ± standard deviation (SD) from three independent experiments (n = 3). Statistical significance was evaluated using one-way ANOVA with post hoc testing. Asterisk indicates statistically significant differences compared to G2 (p < 0.05). SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2).
Ijms 26 07798 g001
Ijms 26 07798 g002
Figure 2. Frequency of promoter methylation in TGF-β signaling-related genes (SKIL, SMAD1, SMAD3, SMAD4, BMP2, and MAPK1) across astrocytic tumor grades (G2, G3, and G4). Methylation status was assessed using methylation-specific PCR (MSP) and interpreted in a binary manner (methylated vs. unmethylated). The histogram-like representation reflects the number of unmethylated cases per gene and grade, indicating a trend toward promoter hypomethylation with increasing tumor grade.
Figure 2. Frequency of promoter methylation in TGF-β signaling-related genes (SKIL, SMAD1, SMAD3, SMAD4, BMP2, and MAPK1) across astrocytic tumor grades (G2, G3, and G4). Methylation status was assessed using methylation-specific PCR (MSP) and interpreted in a binary manner (methylated vs. unmethylated). The histogram-like representation reflects the number of unmethylated cases per gene and grade, indicating a trend toward promoter hypomethylation with increasing tumor grade.
Ijms 26 07798 g002
Ijms 26 07798 g003
Figure 3. Interaction network of selected proteins generated using the STRING database.
Figure 3. Interaction network of selected proteins generated using the STRING database.
Ijms 26 07798 g003
Ijms 26 07798 g004
Figure 4. Kaplan–Meier survival curves for overall survival stratified by protein expression levels of (A) SKIL, (B) SMAD1, (C) SMAD3, (D) SMAD4, (E) BMP2, and (F) MAPK1 in patients with astrocytic tumors. X-axis: survival time (months). Y-axis: cumulative survival probability. Red lines indicate high expression; black lines indicate low expression. Number of patients at risk is displayed for all time points below each survival curve to improve transparency and interpretability.
Figure 4. Kaplan–Meier survival curves for overall survival stratified by protein expression levels of (A) SKIL, (B) SMAD1, (C) SMAD3, (D) SMAD4, (E) BMP2, and (F) MAPK1 in patients with astrocytic tumors. X-axis: survival time (months). Y-axis: cumulative survival probability. Red lines indicate high expression; black lines indicate low expression. Number of patients at risk is displayed for all time points below each survival curve to improve transparency and interpretability.
Ijms 26 07798 g004
Table 1. Differentially expressed genes between high-grade (G3/G4) and low-grade (G2) astrocytomas identified by microarray profiling (FC > 5.0 or <–5.0; p < 0.05).
Table 1. Differentially expressed genes between high-grade (G3/G4) and low-grade (G2) astrocytomas identified by microarray profiling (FC > 5.0 or <–5.0; p < 0.05).
IDmRNAlog2FC G3 Vs. G2log2FC G4 Vs. G2
206675_s_atSKIL7.43 ± 0.549.18 ± 0.43
215889_at7.42 ± 0.439.54 ± 0.76
217591_at7.18 ± 0.879.71 ± 0.43
225227_at7.21 ± 0.329.81 ± 0.51
210993_s_atSMAD15.76 ± 0.878.18 ± 0.98
227798_at5.87 ± 0.717.34 ± 1.02
205396_atSMAD36.13 ± 0.777.81 ± 1.03
205397_x_at6.21 ± 0.548.12 ± 0.76
205398_s_at6.43 ± 0.918.43 ± 0.76
202526_atSMAD48.16 ± 0.789.12 ± 0.65
202527_s_at8.23 ± 0.609.87 ± 0.17
235725_at8.43 ± 0.869.12 ± 1.09
1565702_at8.11 ± 0.819.54 ± 1.02
1565703_at8.71 ± 0.449.88 ± 1.04
205289_atBMP25.11 ± 0.317.65 ± 0.87
205290_s_at6.09 ± 0.617.53 ± 0.56
208351_s_atMAPK110.11 ± 0.9814.13 ± 1.23
212271_at9.98 ± 1.0714.55 ± 1.24
224620_at9.73 ± 0.4316.12 ± 1.32
224621_at10.16 ± 0.7615.65 ± 0.98
229847_at10.54 ± 0.1114.76 ± 1.19
1552263_at10.17 ± 0.7614.61 ± 1.25
1552264_a_at10.78 ± 0.1914.19 ± 1.28
Data are presented as mean ± standard deviation. SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2).
Table 2. miRNAs predicted to target TGF-β signaling components and their expression profiles in G3/G4 vs. G2 tumors.
Table 2. miRNAs predicted to target TGF-β signaling components and their expression profiles in G3/G4 vs. G2 tumors.
miRNATarget ScoreG3 Vs. G2 (log2FC)G4 Vs. G2 (log2FC)Predicted mRNA
Target
hsa-miR-1277-3p100−3.18 ± 0.98−3.81 ± 0.12SKIL
hsa-miR-30a-5p96−4.91 ± 0.71−4.78 ± 0.19SMAD1
hsa-miR-miR-145-5p98−4.18 ± 0.73−3.91 ± 0.98SMAD3
hsa-miR- miR-155-3p83−2.11 ± 0.41−2.13 ± 0.12SMAD4
hsa-miR-58794−3.65 ± 0.17−4.71 ± 0.19BMP2
has-miR-302c-5p93−2.18 ± 0.51−3.18 ± 0.61BMP2
hsa-miR-130b-3p96+2.87 ± 0.37+3.16 ± 0.12MAPK1
Data are presented as mean ± standard deviation. SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2).
Table 3. Concentration of SKIL, SMAD1, SMAD3, SMAD4, BMP2, and MAPK1 proteins in astrocytic tumor samples of different malignancy grades (G2, G3, G4), measured by ELISA.
Table 3. Concentration of SKIL, SMAD1, SMAD3, SMAD4, BMP2, and MAPK1 proteins in astrocytic tumor samples of different malignancy grades (G2, G3, G4), measured by ELISA.
ProteinG2G3G4
SKIL [ng/mL]456.98 ± 43.911234.13 ± 76.12 *1345.19 ± 51.36 *
SMAD1 [ng/mL]2.17 ± 0.185.16 ± 0.61 *8.17 ± 0.17 *
SMAD3 [ng/mL]5.81 ± 0.349.18 ± 0.12 *13.34 ± 0.13 *
SMAD4 [ng/mL]2.15 ± 0.192.87 ± 0.297.87 ± 0.91 *
BMP2 [pg/mL]121.12 ± 7.12541.90 ± 19.33 *601.19 ± 12.81 *
MAPK1 [ng/mL]1.19 ± 0.815.91 ± 0.76 *5.21 ± 2.09 *
Protein concentrations were normalized to total protein content in each sample to enable inter-sample comparison. Data are presented as mean ± standard deviation. SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2); * p < 0.05 vs. G2.
Table 4. RT-qPCR primers.
Table 4. RT-qPCR primers.
mRNART-qPCR Amplification Primers (5′-3′)
SKILForward: AGAGACTCTGTTTGCCCCAA
Reverse: CAGGATGGGGCATTGAATGG
SMAD1Forward: CACTCAACGCCACTTTTCCA
Reverse: TCTTCAGGAGGCAGGTAAGC
SMAD3Forward: CTACCAGAGAGTAGAGACAC
Reverse: TCTCTGGAATATTGCTCTGG
SMAD4Forward: AAAGGTCTTTGATTTGCGTC
Reverse: CTATTCCACCTACTGATCCTG
BMP2Forward: AATGCAAGCAGGTGGGAAAG
Reverse: GCTGTGTTCATCTTGGTGCA
MAPK1Forward: TGGAATAGGTTGTTTTTAAATGTTG
Reverse: AAACTTTTCCTTAAACAAATCATCC
ACTBForward: TCACCCACACTGTGCCCATCTACGA
Reverse: CAGCGGAACCGCTCATTGCCAATGG
GAPDHForward: GGTGAAGGTCGGAGTCAACGGA
Reverse: GAGGGATCTCGCTCCTGGAAGA
SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2); ACTB, beta actin; GAPDH, Glyceraldehyde-3-phosphate dehydrogenase.
Table 5. Characteristics of primers designed for the MSP.
Table 5. Characteristics of primers designed for the MSP.
mRNAM/UNCBI Reference Sequence Database (RefSeq)Primers (5′-3′)
SKILMNM_005414Forward: ATAAGGAGAATTAAAATTAAGTCGT
Reverse: AATAAATAAATACAAATACCTATCGTA
UForward: ATAAGGAGAATTAAAATTAAGTTGT
Reverse: TAATAAATAAATACAAATACCTATCATA
SMAD1MNM_001003688.1Forward: AGGTTTTGAGTTGTTTAGGGTAATC
Reverse: ATAACATAAAACAATCCCTTCCGA
UForward: AGGTTTTGAGTTGTTTAGGGTAATT
Reverse: C ATAACATAAAACAATCCCTTCCAAT
SMAD3MNM_005902Forward: TTTTTAAATTTATTTTCGAATTCGA
Reverse: TAAAAAACAACCCTAAACAAAAACG
UForward: TTAGATGGGTTTTTTAAGTATTTGT
Reverse: ACATCCACCTCTAAATTTACTCATA
SMAD4MNM_005359Forward: TTGGGTTAGGTGTTTTAGTGATTAC
Reverse: A ACCGCCTACTACTACATCTATCGAT
UForward: TTTGGGTTAGGTGTTTTAGTGATTAT
Reverse: ACCACCTACTACTACATCTATCAAT
BMP2MNM_001200.4Forward: ATTTACGAGGAAGGTTGTAATAGTC
Reverse: TAAATTTAACTTAATCCAAATCGAT
UForward: TTATGAGGAAGGTTGTAATAGTTGT
Reverse: AAATTTAACTTAATCCAAATCAAT
MAPK1MNM_002745.5Forward: TATCGTCGAAGTATTATTTAAGTTCGA
Reverse: AAAAAAACACCGATATCTAAACACG
UForward: TTGTTGAAGTATTATTTAAGTTTGA
Reverse: AAAAACACCAATATCTAAACACATC
M, primers designed for methylated sequences; U, primers designed for non-methylated sequence; SKIL, SKI-like proto-oncogene; BMP2, bone morphogenetic protein 2; SMAD1, SMAD family member 1; SMAD3, SMAD family member 3; SMAD4, SMAD family member 4; MAPK1, mitogen-activated protein kinase 1 (ERK2).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Skóra, K.; Strojny, D.; Sobański, D.; Staszkiewicz, R.; Gogol, P.; Miller, M.; Rogoziński, P.; Zmarzły, N.; Grabarek, B.O. Multi-Layered Analysis of TGF-β Signaling and Regulation via DNA Methylation and microRNAs in Astrocytic Tumors. Int. J. Mol. Sci. 2025, 26, 7798. https://doi.org/10.3390/ijms26167798

AMA Style

Skóra K, Strojny D, Sobański D, Staszkiewicz R, Gogol P, Miller M, Rogoziński P, Zmarzły N, Grabarek BO. Multi-Layered Analysis of TGF-β Signaling and Regulation via DNA Methylation and microRNAs in Astrocytic Tumors. International Journal of Molecular Sciences. 2025; 26(16):7798. https://doi.org/10.3390/ijms26167798

Chicago/Turabian Style

Skóra, Klaudia, Damian Strojny, Dawid Sobański, Rafał Staszkiewicz, Paweł Gogol, Mateusz Miller, Przemysław Rogoziński, Nikola Zmarzły, and Beniamin Oskar Grabarek. 2025. "Multi-Layered Analysis of TGF-β Signaling and Regulation via DNA Methylation and microRNAs in Astrocytic Tumors" International Journal of Molecular Sciences 26, no. 16: 7798. https://doi.org/10.3390/ijms26167798

APA Style

Skóra, K., Strojny, D., Sobański, D., Staszkiewicz, R., Gogol, P., Miller, M., Rogoziński, P., Zmarzły, N., & Grabarek, B. O. (2025). Multi-Layered Analysis of TGF-β Signaling and Regulation via DNA Methylation and microRNAs in Astrocytic Tumors. International Journal of Molecular Sciences, 26(16), 7798. https://doi.org/10.3390/ijms26167798

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop