Astrocytes and glioblastoma: how tumor cells wake from dormancy

Astrocytes and glioblastoma may be connected by much more than physical proximity in the brain. In experimental models, these normally supportive brain cells transferred mitochondria to tumor cells, helping some dormant glioblastoma cells resume proliferation while becoming more resistant to immune attack.

Glioblastoma is dangerous not only because it grows rapidly. Some tumor cells can enter a quiescent state in which they divide very little, survive adverse conditions and potentially resume growth later. Such cells are thought to contribute to the persistence and recurrence of the disease.

The Nature Communications study focused on astrocytes, abundant brain cells that support neurons, regulate the local environment and respond strongly to tissue injury. At the invasive edge of glioblastoma, tumor cells and reactive astrocytes are often found close together. The researchers asked whether this interaction could actively change the state of dormant cancer cells.

Mitochondria moving between cells

The experiments combined the human U87 glioblastoma cell line, mouse GL261 and CT2A glioma cells, primary astrocytes and intracranial mouse tumor models. Reporter systems allowed the researchers to distinguish quiescent cells from cells that had re-entered the cell cycle.

When astrocytes were experimentally depleted in tumor-bearing mice, a larger fraction of glioma cells remained quiescent. Conversely, exposure to astrocytes encouraged cells to leave dormancy.

Mitochondrial transfer emerged as a central part of this effect. Mitochondria are best known for their role in cellular energy production, but they also carry out metabolic reactions that generate building blocks and signaling molecules needed for cell growth.

By labeling astrocyte mitochondria, the researchers could trace them inside glioma cells. Astrocyte-derived mitochondrial signals were particularly abundant in proliferating tumor cells. Disrupting the intercellular structures involved in this transfer slowed the transition of glioma cells out of quiescence.

Cells that acquired astrocyte mitochondria also became more difficult for NK cells and CD8+ T cells to kill. When transplanted into mice, mitochondria-receiving glioma cells produced faster-growing tumors than cells that had not received them.

Analysis of human glioma datasets provided additional support for the model. Tumor cells computationally identified as likely recipients of astrocyte mitochondria showed stronger cell-cycle and metabolic programs, and higher inferred mitochondrial-transfer scores were associated with poorer prognosis. This part of the study is observational, however, and does not establish mitochondrial transfer itself as a cause of clinical outcome in patients.

The metabolic cargo matters

The effect was not simply a matter of giving tumor cells more cellular “power plants.” Mitochondria from reactive astrocytes were enriched in SHMT2, an enzyme involved in mitochondrial one-carbon metabolism.

One-carbon metabolism moves small carbon units through a network of reactions used to make nucleotides, amino acids and molecules required for methylation. After glioma cells acquired astrocyte mitochondria, several metabolites in this pathway increased, including S-adenosylmethionine, or SAM.

SAM is the cell's major methyl-group donor. Its increased availability was associated with higher levels of m6A, a common chemical modification of RNA.

This created a link between metabolism and gene regulation. The nuclear protein YTHDC1 recognizes m6A-modified RNA and, in the experimental system, promoted the release of RNA polymerase II from transcriptional pausing at selected genes. The resulting transcriptional program favored ribosome production, protein synthesis, focal-adhesion pathways and genes associated with immune evasion.

The proposed sequence is therefore unusually direct: an astrocyte transfers mitochondria; mitochondrial SHMT2 strengthens one-carbon metabolism; SAM availability rises; RNA methylation changes; and transcriptional programs supporting proliferation and immune resistance become more active.

Removing Shmt2 specifically from astrocytes weakened this chain. Tumors in these mice contained fewer cycling glioma cells, grew more slowly and were associated with longer survival. Disrupting YTHDC1 in glioma cells likewise kept more cells dormant and increased their susceptibility to NK-cell killing.

A mechanism, not yet a treatment

The researchers also used a pharmacological approach to interfere with intercellular mitochondrial transfer. In cultured cells, the intervention suppressed the transition from quiescence to proliferation. In tumor-bearing mice, it increased the proportion of quiescent glioma cells and slowed tumor growth.

That does not make mitochondrial transfer an established therapeutic target in patients.

Most of the mechanistic evidence comes from established cell lines and mouse models. The authors note that these models form relatively focal tumors rather than the diffusely infiltrative disease typical of human glioblastoma. The GL261 model is also relatively immunogenic, which may influence experiments involving immune-cell killing. Therapeutic inhibition of SHMT2 or mitochondrial transfer has so far been tested only in preclinical models.

The study therefore does not show that an existing drug can prevent glioblastoma recurrence in people. What it provides is a mechanistic framework linking the tumor microenvironment to the awakening of quiescent cancer cells.

In that framework, an astrocyte is not merely a passive neighbor. It acts as a metabolic donor. A transferred mitochondrion supplies not only additional metabolic capacity but also biochemical machinery capable of changing RNA regulation and, ultimately, tumor-cell behavior.

That makes the most interesting finding broader than mitochondrial transfer itself. The movement of a single organelle between two cells can, at least in these experimental systems, connect metabolism, RNA regulation, proliferation and immune escape in one coordinated change of cellular state.