A new study is drawing attention to a possible metabolic vulnerability in breast cancer stem cells: the mitochondrial DNA maintenance enzyme polymerase gamma, or POLG. Published in Aging on August 8, 2026, the pilot study reports that disrupting either component of the POLG system weakened mitochondrial activity and reduced stem-cell-like behaviors in several breast cancer models. The findings do not yet establish POLG inhibition as a treatment, but they suggest that the enzyme may help cancer cells preserve the specialized metabolic state required for self-renewal, tumor progression, and resistance to therapy.
Breast cancer stem cells, commonly called CSCs, represent a relatively small and adaptable population within a tumor. Unlike most rapidly dividing cancer cells, CSCs can self-renew, generate more differentiated tumor cells, and survive conditions that eliminate other malignant cells. These properties have been linked to recurrence, metastasis, and treatment failure. Although cancer metabolism has often been associated with increased glucose consumption and glycolysis, CSCs can also depend heavily on mitochondria, the organelles that generate ATP through oxidative phosphorylation. This reliance led the researchers to investigate POLG, the enzyme responsible for copying and repairing mitochondrial DNA.
POLG operates as a two-part molecular machine. The catalytic subunit is encoded by the POLG1 gene, while the accessory subunit is encoded by POLG2. Together, these proteins help replicate the small circular genome carried inside mitochondria. Mitochondrial DNA encodes essential components of the respiratory chain, the series of protein complexes that transfer electrons and use the released energy to produce ATP. If mitochondrial DNA is depleted or damaged, respiratory-chain assembly can fail, reducing energy production and altering the balance of reactive oxygen species inside the cell. The researchers therefore asked whether impairing POLG would interfere with the biology of breast cancer stem cells.
The experiments were initially performed in MCF-7 cells, an estrogen receptor-positive breast cancer model. Using genetic silencing, the investigators reduced expression of either POLG1 or POLG2. The effects on mitochondrial DNA were substantial: POLG1 silencing lowered mitochondrial DNA content by about 80%, while POLG2 silencing caused an approximately 70% reduction. These changes were accompanied by decreased levels of MTCO2, a protein encoded by mitochondrial DNA and required for respiratory-chain function. Measurements of mitochondrial membrane potential, respiration, and ATP production also indicated that the organelles were operating less efficiently after POLG disruption.
The mitochondrial defects were closely associated with a loss of stemness-related characteristics. In MCF-7 cells, silencing POLG1 reduced expression of OCT4, a transcription factor involved in maintaining cellular plasticity and self-renewal. The cells also formed roughly 70% fewer mammospheres, three-dimensional structures used as an in vitro indicator of mammary stem-cell activity. Their ability to generate colonies was markedly suppressed as well. Silencing POLG2 produced a similar pattern, reducing OCT4 expression, mammosphere formation, and clonogenic growth. Notably, these interventions did not substantially affect short-term growth in conventional two-dimensional cultures, suggesting that mitochondrial POLG activity may be particularly important for long-term self-renewal rather than immediate cell proliferation.
The study also examined how POLG disruption changed the chemical environment within mitochondria. Loss of mitochondrial membrane potential is a sign that the electrochemical gradient used to drive ATP synthesis has been weakened. At the same time, POLG1 and POLG2 silencing increased mitochondrial superoxide, a reactive oxygen species generated when electrons leak from the respiratory chain. This increase did not translate into a statistically significant rise in total cellular reactive oxygen species in every experiment, indicating that mitochondrial stress and whole-cell oxidative stress are not identical measurements. The results instead point to a localized disturbance in mitochondrial redox balance that may contribute to the loss of CSC-associated properties.
To test whether the genetic results could be reproduced with drugs, the researchers examined Alovudine, a nucleoside reverse transcriptase inhibitor developed for antiviral use. Alovudine can inhibit POLG as an off-target effect, although it is not a selective POLG-directed cancer drug. In the breast cancer cells, the compound reduced mammosphere and colony formation while lowering MTCO2 expression, mitochondrial respiration, and ATP generation. The authors emphasize that Alovudine caused hematological toxicities during antiviral development, making it unsuitable as an established cancer treatment on the basis of these experiments. Its role in the study was primarily to provide pharmacological support for the idea that POLG-dependent mitochondrial function is connected to cancer stemness.
The investigators then used a second compound, zalcitabine, also known as ddC, to determine whether the observations depended on a single drug. In MCF-7 cells, ddC impaired mitochondrial respiration, reduced MTCO2 levels, and suppressed both mammosphere and colony formation. The researchers extended the analysis to T47D, MDA-MB-231, MDA-MB-436, and MDA-MB-453 breast cancer cells, representing different biological subtypes and molecular characteristics. Across these models, ddC consistently reduced mammosphere formation, although its effects on ordinary monolayer growth varied. Non-tumoral MCF10A mammary epithelial cells showed limited viability changes under the tested conditions, a result that may indicate some degree of selectivity but cannot substitute for formal toxicity and therapeutic-index studies.
The molecular response to ddC also varied between cancer models. In MCF-7 cells, the compound markedly reduced the stemness-associated transcription factors SOX2 and NANOG. In MDA-MB-453 cells, SOX2 declined, whereas the effect on NANOG was more modest. Such differences are important because breast cancer is not a single disease: tumors differ in receptor status, genomic alterations, mitochondrial activity, and dependence on particular metabolic pathways. The researchers additionally analyzed clinical data from 458 patients with high-risk, estrogen receptor-positive, lymph node-positive luminal A breast cancer. Higher POLG1 expression was associated with poorer overall survival, with a hazard ratio of 1.34, and with shorter relapse-free, distant metastasis-free, and post-progression survival. These preliminary associations do not prove that POLG drives poor outcomes, and they require validation in independent patient cohorts using multivariable analyses.
The authors describe the work as a pilot study and stress that important questions remain unanswered. Much of the genetic evidence came from MCF-7 cells, and shRNA-based experiments can produce off-target effects that complicate interpretation. Neither Alovudine nor ddC is a fully selective POLG inhibitor, so their effects may involve additional molecular targets. The experiments were performed in cultured cells rather than animals or patients, meaning that the study does not yet demonstrate tumor suppression in a living organism or establish whether POLG inhibition can spare normal tissues. Future research will need to use independent genetic strategies, selective chemical probes, patient-derived models, xenografts, and orthotopic tumors. Nevertheless, the study identifies a compelling connection between mitochondrial DNA maintenance and breast cancer stem-cell behavior. By showing that POLG disruption can weaken mitochondrial respiration, alter redox balance, and suppress self-renewal across multiple models, the findings position POLG as a candidate biomarker and a possible target for therapies designed to attack the metabolically resilient cells that help breast tumors persist and return.
Subject of Research: Cells
Article Title: Investigating POLG-driven modulation of cancer stemness: a pilot study in breast cancer cells
News Publication Date: 24 August 2026
Web References: Aging, Volume 18; https://doi.org/10.18632/aging.206406; Aging-US
References: Chinigò et al., “Investigating POLG-driven modulation of cancer stemness: a pilot study in breast cancer cells,” Aging, DOI: 10.18632/aging.206406
Image Credits: Copyright © 2026 Chinigò et al.; Figure 2, distributed under the Creative Commons Attribution License (CC BY 4.0).
Keywords: breast cancer, cancer stem cells, cancer metabolism, POLG, POLG1, POLG2, mitochondrial DNA, mitochondrial function, oxidative phosphorylation, cancer stemness
Tags: breast cancer stem cellscancer metabolismmetabolic vulnerabilities in tumorsmitochondrial DNA maintenancemitochondrial DNA repairmitochondrial dysfunction in cancer treatmentmitochondrial function in canceroxidative phosphorylation in cancerPOLG enzyme inhibitiontargeting cancer stemnesstherapy resistance in breast cancertumor cell self-renewal
