Although neurons are the excitable and firing cells of the brain driving the nervous system signaling, glial cells in the brain parenchyma are key players for the correct functioning and the homeostasis of the central nervous system. Among glial cells, “star-like” astrocytes are the cells whose relative number, size, number of ramified processes and volume increased with phylogeny and brain complexity (Nedergaard et al., 2003). Depending on the different regions, astrocytes represent 20−40% of all brain cells (Herculano-Houzel, 2014) and show different morphology, ranging from protoplasmic to spherical shape (Emsley and Macklis, 2006; Oberheim et al., 2006). In the brain, each astrocyte occupies a specific territory, with less of 5% of overlap with neighboring astrocytes (Ogata and Kosaka, 2002). Within their specific competence territory, astrocytes contact blood vessels and up to hundred-thousands of different synapses (Halassa et al., 2007); moreover, due to the presence of connexin gap junctions between different astrocytes, these cells are organized in networks (D’Ambrosio et al., 1998; Giaume et al., 2010) that appear to be organized in functional domain (Giaume et al., 2010). Generally believed to mainly have a supportive function (Kettenmann and Ransom, 2005), astrocytic cells play many active roles. During development, astrocytes play a role in guiding the migration of neuronal axons and neuroblast (Powell and Geller, 1999), and the formation of developing synapses (Ullian et al., 2001; Christopherson et al., 2005); moreover, they can drive microglial synapse engulfment, or actively engulf synapses and sculpt neuronal circuits (Chung et al., 2013; Vainchtein et al., 2018). With their terminal processes (end-feet), astrocytes contribute to the formation and maintenance of brain-blood integrity (Abbott, 2002); thank to the presence of several plasma membrane transporters, during neuronal activity they can buffer extracellular K⁺ concentration and water content (Simard and Nedergaard, 2004), regulate the extracellular pH and remove excessive glutamate from the synapses (Rose et al., 2018). Astrocytes sense neuronal activity via metabotropic neurotransmitter receptors, and are able to provide energy substrate to neurons through the so call “astrocyte-neuron lactate shuttle” (Magistretti and Pellerin, 1999); in addition, astrocytic networks can support the high energy demand of neuronal activity, also at site distant from blood vessels (Rouach et al., 2008), thus ensuring glia-neurons metabolic coupling necessary for memory formation (Suzuki et al., 2011; Gao et al., 2016). Also, astroglial endfeets that enwrap blood vessels are characterized by high levels of connexins expression (Rouach et al., 2008) and Ca²⁺ signaling within astrocytes can trigger the release of vasoactive molecules that modulate local or regional cerebral blood flow (Koehler et al., 2009; Institoris et al., 2022). Being part of the “tripartite” synapse (Araque et al., 1999), astrocytes respond to neurotransmitter release by presynaptic terminals with an increase in intracellular Ca²⁺, and consequent release of “gliotransmitters” that can act regulating synaptic plasticity at local synapse (Fellin et al., 2006; Di Castro et al., 2011). Moreover, intracellular calcium increase can be spread to other connected astrocytes (Bazargani and Attwell, 2016; Goenaga et al., 2023) resulting in neurotransmitter release and modulation of synapses at the level of network activity (Fellin, 2009; Miguel-Quesada et al., 2023).

In non-physiological conditions, such as CNS (central nervous system) injuries, disease or brain tumor, astrocytes lose their “quiescent” state, become “reactive” and undergo changes in molecular expression, progressive cellular hypertrophy and in some cases also proliferation and scar formation (Lukaszevicz et al., 2002; Sofroniew, 2009; Faideau et al., 2010; Acevedo-Arozena et al., 2011; Cuevas-Diaz Duran et al., 2019; Makarava et al., 2019). These changes are regulated in a context-specific manner, and lead to altered astrocytic activities, either loss or gain of functions, that can be either detrimental or beneficial to the brain (Sofroniew, 2005).

CNS primary tumors are the most frequent in children between 0 and 14 years of age and are the eighth most frequent in adults (van den Bent et al., 2023). These tumors are extremely heterogeneous, depending not only on the tissue of origin but also on the genetic and/or molecular modifications that characterize them and on the ethnicity of the affected population; all these aspects define the average outcome of patients (Louis et al., 2021). A first distinction is between malignant and non-malignant tumors; the first ones are able to invade the surrounding tissue and have a terrible outcome; the others are classified based on histological and molecular characteristics. Among the malignant CNS tumors, gliomas are the most common, and glioblastoma (GBM) is the most aggressive and frequent primary malignant CNS tumor, with the prognosis of an overall survival of 7−17 months after surgical removal (Molinaro et al., 2020). Among the non-malignant tumors, the most common is meningioma (Louis et al., 2016).

Despite malignant brain tumors can originate from neuronal stem cells or oligodendrocyte precursor cells, astrocytes represent the cellular origin at least for a defined number of cases (Zong et al., 2015). In fact, a specific mouse model carrying co-deletion of the tumor suppressor Tp53, Pten and Rb1 genes was created using the site-specific recombinase technology (Cre-Lox/GFAP) in adult mice (Chow et al., 2011). This model shows deletion of the three tumor suppressor genes widespread in mature astrocytes but also in a subpopulation of GFAP-expressing neuronal stem cells (NSCs) in the brain proliferative niches (subventricular zone and subgranular layer). Most tumors grew in these niches but more than 20% of tumors appeared in the non-proliferative areas (specifically in cortex, brainstem, cerebellum, and spinal cord). These data demonstrate that mature astrocytes are a cell type from which malignant CNS tumors arise even if the majority originates from stem cells.

Astrocytes can originate malignant CNS tumors not only by alteration of their proliferation and/or cell survival process, but also in case of alterations in the differentiation state maintenance. Indeed, it has been shown that astrocytes dedifferentiated into NSCs - after stimulation with TNFα - become susceptible to the process of cancerization by irradiation; in contrast mature astrocytes do not undergo transformation upon the same oncogenic stress (Dufour et al., 2009).

Metastatic brain tumors are secondary tumors that develop from cells of a primary systemic tumor that invade the brain and are often the main cause of mortality (Achrol et al., 2019). The most common primary tumors that develop brain metastasis are the lung (40−50%), breast (15−20%), skin (5−10%), and gastrointestinal (4−6%) tumors, and most patients develop more than one brain lesion. Similar to glioma, standard of care to treat brain metastasis is the surgical removal, followed by radiation- and chemo- therapies; nevertheless, these treatments have a reduced efficacy, with most patients developing local recurrence in less than one year (Brastianos et al., 2013). Brain metastasis formation requires several steps such as detachment from the primary tumor, invasion of surrounding tissue, intravasation in blood vessel, dissemination and arrest in brain capillary, extravasation through non-fenestrated capillaries, colonization of surrounding tissue and local proliferation and neo-angiogenesis (Svokos et al., 2014). The interaction between circulating tumor cells and blood–brain barrier (BBB) components is mainly mediated by cytokines and chemokines (Seike et al., 2011); tumor cell proliferation and angiogenesis in the brain depend on the release of local growth factors (Hoshide and Jandial, 2017).

It is well established that the crosstalk between brain tumor cells and the surrounding microenvironment is determinant for tumor progression. In particular 50% of glioma tumor mass is made by the infiltration of brain-resident microglia, and peripheral macrophages (Hambardzumyan et al., 2016) that actively contribute to tumor proliferation and invasion, but also to the formation of an immune suppressive environment (Catalano et al., 2020). Indeed, other infiltrating immune cells are present, these are primarily T lymphocytes, and also rare NK, dendritic and B cells (Gieryng et al., 2017). In particular the infiltrating T cells display an exhausted phenotype and undergo programmed cell death (Woroniecka et al., 2018), resulting in the inability to contrast the tumor.

In this context, astrocytes in the proximity of the tumor became reactive (Nagashima et al., 2002; O’Brien et al., 2013) as a consequence of their ability to sense the tumor and to engage a crosstalk with the tumor microenvironment (TME), becoming part of it. Reactive astrocytes appear to play an important role in supporting glioma growth since, as it has been recently found, genetic ablation of tumor-associated astrocytes in glioma GL261 bearing mice, not only stalls GBM progression but drives the tumors into regression and prolongs animal survival (Perelroizen et al., 2022). The ability of astrocytes to modulate tumor growth seems to depend on the phenotype of glioma cells. In fact, if the glioma is minimally invasive (i.e., U87MG glioma cell line), astrocytes are able to totally block its migration and probably to stimulate a robust immune response. If the glioma is highly invasive (i.e., LN229 glioma cell line), astrocytes increase the migratory capacity of tumor cells by inducing over-expression of genes related to migratory signaling pathways, such as STAT3 (signal transducer and activator of transcription) and HGF/MET (hepatocyte growth factor/mesenchymal-epithelial transition factor) (Cui et al., 2023).

Also in the case of brain metastasis derived from a primary systemic tumor, the interaction between metastatic cells and the surrounding brain TME plays a fundamental role in the establishment of brain metastasis (Gwak, 2023). Among cells of the TME, reactive astrocytes are the most active host cell population, that immediately localizes to individual invading tumor cells and continuously associates with growing metastatic lesions (Lorger and Felding-Habermann, 2010). Moreover, in a model of spontaneous brain metastasis, in immunocompetent mice, Schwartz et al. (2016) demonstrated that astrocytes acquire a proinflammatory phenotype in the brain metastatic niche before the formation of macrometastasis.

The crosstalk between astrocytes and tumoral cells within the brain is mediated by the release of soluble factors, the release of extracellular vesicles (EVs) and through the direct contact between cells, due to gap junctions or to tunneling nanotubes (Figure 1).

Although non-neoplastic astrocytes engage direct and indirect dialogue with glioma cells contributing to tumor progression, recent data suggest that these cells can also represent key elements to contrast tumor progression (Fletcher-Sananikone et al., 2021; Serpe et al., 2022).

One strategy to counteract glioma might be represented by EVs released by non-neoplastic astrocytes. Recently (Serpe et al., 2022) we have found that EVs derived from glioma-stimulated astrocytes increase glioma proliferation and in vivo tumor volume. In contrast, EVs derived from normal astrocytes are able, in vitro, to reduce glioma cell proliferation, migration and invasion capability. Moreover in vivo, administration of these EVs reduces glioma tumor volume, proliferation rate, and in addition, impairs cell volume regulation. We found that among molecules transported by these EVs there is miR124. Such molecules are able to reduce the expression in glioma cells of LRCC8C protein, a subunit of the volume regulated anion channels (VRACs) that play a role in the modification of cell volume, necessary for cell migration and proliferation. Glioma cells kill surrounding neurons by releasing glutamate in order to create the necessary space to grow (Ye and Sontheimer, 1999). VRACs are also permeable to excitatory amino acids including glutamate (Feustel et al., 2004); thus, we speculate that the reduction of VRAC expression might contribute to reduction in glioma release of glutamate.

Another strategy could be the use of non-neoplastic astrocytes as therapeutic targets. After surgical resection of glioma, the standard of care is patients’ treatment with up to 60 Gy of fractionated ionized irradiation (IR) with concurrent adjuvant chemotherapy, such as Temozolomide (Stupp et al., 2005). However, despite the positive response following IR therapy, later on the recurrence of a more invasive and resistant glioma lead to a fatal prognosis (Osuka and Van Meir, 2017, Scoccianti et al., 2021). Recently, it has been demonstrated in vivo, using syngeneic GL261 mouse model of glioma, and in vitro, using coculture of normal astrocytes and GL261 glioma cells, that upon irradiation normal astrocytes became senescent and release factors, including HGF. HGF then activates the Met receptor on glioma, promoting tumor invasiveness. In vivo, blocking Met activation by pharmacological approach results in attenuation of tumor growth and increased mice survival. Further, the elimination of senescent astrocytes using a senolytic drug results in delayed tumor growth in pre-irradiated brains (Fletcher-Sananikone et al., 2021).

In this Mini Review different strategies of the crosstalk between surrounding non-neoplastic astrocytes and brain primary or metastatic tumor cells have been reported. Astrocytes react to brain tumor cells engaging in undirect and direct dialogues mainly to support the tumor growth. Astrocytic release of soluble factors fosters migration, invasion, and growth of primary brain tumor cells. In brain metastasis these factors, released by astrocytes close to the extravasated cancer cells, support the brain entering.

Extracellular vesicles released by primary tumor cells promote astrocytic polarization toward an immunosuppressive phenotype and astrocytic release of soluble factors that contribute to the tumor progression. EVs released by astrocytes transfer miRs and growth factors into metastatic cancer cells that support metastatization into the brain.

Direct exchange of molecules occurs through gap junctions or tunneling nanotubes between astrocytes and brain tumor cells also support their crosstalk.

However, recent data suggest that EVs from non-neoplastic astrocytes or selective elimination of non-neoplastic astrocytes may be used to counteract brain tumors. Considering these findings, it is possible to speculate that in the future, administration of EVs obtained from non-neoplastic astrocytes (for example derived from patient differentiated IPSCs), and/or targeting non-neoplastic astrocytes by using senolytic therapy, could represent an alternative or coadiuvant therapeutic approach to limit brain tumor progression and to contrast glioma recurrence.

MC: Conceptualization, Data curation, Supervision, Writing – original draft, Writing – review & editing, Funding acquisition. CL: Funding acquisition, Supervision, Writing – review & editing. FT: Conceptualization, Data curation, Supervision, Writing – original draft, Writing – review & editing.