Mechanisms of Genomic Instability, Centrosome Dysfunction, and Cancer Stemness in Glioblastoma Pathogenesis and Therapeutic Implications

Proteins such as Intraflagellar Transport (IFT) interact with HSET to stabilize centrosome clusters [90], while Aurora Kinase A (AURKA)‐driven phosphorylated Nuclear Mitotic Apparatus (NuMA) protein anchors centrosomes at the spindle poles, ensuring bipolar spindle formation [91, 92]. Mitotic kinases that control centrosome maturation—notably AURKA—additionally couple centrosome function to actin/vimentin reorganization and Rho‐GTPases signaling, enabling chemokine‐directed migration and periventricular invasion [93]. AURKA ensures mitotic fidelity by recruiting its substrate, Transforming Acidic Coiled‐Coil Containing Protein 3 (TACC3), for stabilizing kinetochore fibers [94]. However, overexpression of TACC3 in glioblastoma drives multiple pro‐tumorigenic programs including enhanced proliferation, migration/invasion, stem-like properties, and survival—in part through its roles at centrosomes, spindles, and in mitotic centrosome‐clustering networks, making TACC3 both a marker of aggressive disease and a therapeutic vulnerability [95]. Interestingly, in its fusion form with Fibroblast growth factor receptor (FGFR), notably identified as FGFR3–TACC3 fusion, it exerts a dual pathogenic effect in GBM. Constitutive FGFR kinase signaling promotes growth and survival, while structural disruption of mitotic machinery via the TACC3 moiety—sequestering endogenous TACC3 away from the spindle—results in destabilizing microtubule dynamics, increasing chromosome misalignment/missegregation, and promoting aneuploidy and chromosomal instability [95, 96]. Mechanistic studies show that CA alone is sufficient to perturb cell–cell junctions, remodel the extracellular matrix, and engage Rap1/adhesion signaling to enable collective and single‐cell invasion, providing a unifying link between centrosome deregulation and the highly infiltrative phenotype of GBM [97]. Hence, centrosomal clustering promoting proteins could be potential therapeutic targets to selectively induce cell death in tumor cells [98].

The differential cellular responses to centrosome amplification are significantly influenced by p53 status. The fate of cells harboring supernumerary centrosomes is determined by whether they possess wild‐type or mutant p53. In cells with wild‐type p53, centrosome amplification is sensed through the PIDDosome–caspase‐2–MDM2–p53 signaling axis, leading to stabilization and nuclear accumulation of p53. Activated p53 either induces cell cycle arrest (via p21) or apoptosis (via BAX, PUMA), thereby preventing proliferation of aneuploid cells and maintaining genomic integrity. In contrast, in p53‐mutant or p53‐deficient cells, this checkpoint fails to activate, allowing continued cell division despite centrosome amplification. These cells evade mitotic catastrophe by engaging centrosome clustering mechanisms involving Aurora A kinase, HSET/KIFC1, Eg5, and dynein, which bundle multiple centrosomes into two spindle poles to form a pseudo‐bipolar spindle. This results in cell survival and permits chromosome segregation errors, ultimately leading to chromosomal instability (CIN).

Defects in fundamental DNA‐repair pathways—such as homologous recombination (HR), non‐homologous end joining (NHEJ), base‐excision repair (BER), and mismatch repair (MMR)—lead to replication stress and accumulation of DNA double‐strand breaks (DSBs), producing aneuploidy and CIN through mis‐segregation, micronucleus formation, and chromosome bridge breakage [34, 56]. Notably, rare gliomas harboring germline or somatic mutations in the DNA polymerase genes POLE and POLD1—key enzymes responsible for DNA replication fidelity and proofreading—exhibit profound replication errors that drive chromosomal instability and give rise to extreme hypermutated tumor phenotypes [5]. Once established, persistent aneuploidy exacerbates DNA‐repair dysfunction by inducing haploinsufficiency of key repair genes (e.g., Breast Cancer gene 1/2 (BRCA1/2), Radiation sensitive protein 51 (RAD51), Meiotic Recombination 11 (MRE11), ATRX, Ataxia‐telangiectasia mutated (ATM)) and proteotoxic stress, which destabilizes repair protein networks and impairs DNA Damage Response (DDR) signaling fidelity [99, 100]. Aneuploid cells also exhibit micronuclear entrapment of chromosomes, where aberrant replication and nuclear envelope rupture cause defective recruitment of repair proteins (e.g., Mediator of DNA damage checkpoint 1 (MDC1), 53BP1, BRCA1), leading to chromothripsis and large‐scale structural rearrangements highlighted from the accumulation of Phosphorylated H2A Histone Family Member X (γH2AX) in the micronuclei (MN), a marker for earliest cellular responses to the formation of DSBs [101, 102]. These self‐reinforcing cycles of repair deficiency and karyotypic chaos are particularly pronounced in GBM, where loss of ATM, ATRX, MutS homolog 6 (MSH6), or O6‐methylguanine-DNA methyltransferase (MGMT) promoter methylation‐mediated silencing increases both mutation burden and chromosomal mis‐segregation [34]. In parallel, aneuploidy‐associated replication stress activates Ataxia Telangiectasia and Rad3‐related protein – Checkpoint kinase 1 (ATR–CHK1) and WEE1 G2 checkpoint kinase (WEE1) checkpoints as compensatory survival pathways, which have emerged as therapeutic vulnerabilities in GBM with DNA‐repair deficiencies [56, 103]. Targeting these compensatory DDR nodes—through PARP inhibitors in HR‐deficient tumors, or ATR/CHK1/WEE1 inhibitors to induce mitotic catastrophe in CIN‐positive GBM—represents a promising synthetic‐lethal approach [101, 104]. Therefore, the intersection of DNA‐repair impairment, aneuploidy, and CIN defines a key evolutionary axis in GBM, where genomic chaos both drives malignancy and exposes actionable molecular liabilities that can be therapeutically exploited. These ongoing efforts in medicinal chemistry to develop precise anti-cancer agents offer new hope for patients.

Cancer stem cells (CSCs) drive intra‐tumor heterogeneity through their dual ability to self‐renew and differentiate into diverse tumor cell types. Their plasticity enables transitions between stem‐like and non‐stem‐like states in response to environmental cues, allowing tumors to adapt to stresses, including therapeutic interventions, and contributing to treatment resistance [105]. The detrimental effect of stemness is consistent in brain cancers, where enrichment of CSCs known as GSCs, expressing nestin, CD133, and CD163 cell surface markers [14] in the tumor niche, exerts poor prognosis in patients [106] despite targeting signaling axes involved in GBM pathogenesis [107].

CIN contributes to tumor heterogeneity [108] through generating Copy Number Variations (CNVs) determined by single‐cell RNA sequencing (scRNA‐seq) in a tumor [109, 110], which promotes the emergence of genetically distinct subpopulations called CSCs [34] once the degree of genetic change reaches a tolerable threshold [111]. Since RNA expression correlates with gene copy numbers, scRNA‐seq enables the detection of chromosomal gains or losses by comparing transcriptional profiles from individual tumor cells to those of normal diploid references. Upregulated or downregulated gene expression across chromosomal regions reflects CNVs, which are then used to compute a chromosomal instability index (CII), quantitatively capturing CIN levels, distinguishing CIN ^high from CIN ^low in each cell across a tumor [74, 109]. Importantly, these CNV‐driven transcriptomic alterations are not merely by‐products of instability but serve as functional mediators of tumor cell reprogramming. CIN‐induced CNVs alter the expression of key oncogenic and stemness‐related genes, thereby enhancing genetic variability and promoting the emergence of CSCs. This occurs through chromosomal imbalance–driven stress responses that rewire transcriptional programs toward self‐renewal and pluripotency. Mechanistically, CIN can trigger de‐differentiation of non‐stem tumor cells via suppression of p53, amplification of cellular myelocytomatosis oncogene (c‐MYC), and aberrant activation of Wnt/β‐catenin signaling, all of which converge to sustain self‐renewal and pluripotency programs [58, 112]. Interestingly, the heterogeneous distribution of stem cell–associated markers across diploid and aneuploid tumor clones, as summarized in Table 1, indicates that stem‐like phenotypes are maintained across genetically distinct karyotypic states rather than being restricted to a single ploidy class [113, 114, 115, 116].

Table 1: Distribution of Stem Cell Markers Across Different Ploidy States in Tumor Clones