Could a 70-year-old high blood pressure medicine hold the secret to stopping one of the deadliest brain cancers? New research indicates that hydralazine, a vasodilator introduced in the 1950s, directly acts on a molecular oxygen sensor that glioblastoma cells rely on for their survival when oxygen is scarce.
For decades, the exact mechanism of hydralazine remained unknown. It was developed in what Kyosuke Shishikura of the University of Pennsylvania calls a “pre-target era of drug discovery,” when efficacy was observed in patients long before the underlying biology was understood. That mystery has now been resolved: hydralazine binds to and disables 2‑aminoethanethiol dioxygenase (ADO), an iron-dependent enzyme that functions as a rapid-response oxygen sensor in cells.
ADO’s function is to sense falling oxygen levels and initiate the proteolytic degradation of certain regulatory proteins, including the regulators of G‑protein signaling RGS4 and RGS5. In vascular smooth muscle, this degradation maintains calciummediated constriction of blood vessels. Hydralazine exerts its effects by chelating the metal cofactor of ADO and covalently modifying a key histidine (His112) in its active site, as visualized by highresolution X‑ray crystallography. This prevents the degradation, leading to accumulation of RGS proteins, suppression of G‑protein coupled receptor signaling, a decrease in intracellular calcium, and vasodilation.
In the case of glioblastoma, this same ADO pathway underlines tumor adaptation to hypoxia. These tumors grow so aggressively that their blood supply cannot keep pace, creating oxygen-poor niches. Normal cells falter under such conditions, while glioblastoma cells activate hypoxia-adaptive programs, including ADO-mediated protein turnover, to continue dividing. Hydralazine silences ADO and thereby disrupts this adaptation, forcing tumor cells into senescence-a state of permanent growth arrest. Three days of hydralazine treatment reduced the proliferation of cultured human glioblastoma cells, which became enlarged and flattened, exhibiting hallmark senescence phenotypes in the absence of outright cell death.
This cytostatic effect parallels other therapy-induced senescence pathways in glioblastoma, such as those induced by temozolomide, but with a different upstream target. Senescent tumor cells can still affect their environment through the senescence-associated secretory phenotype (SASP), which may facilitate inflammation or recurrence if not removed. Investigators point out that the combination of ADO inhibition with senolytic agents-small molecules that selectively kill senescent cells-could further improve therapeutic response.
Selectivity studies showed that hydralazine is more than 100‑fold more potent against ADO compared to PHD enzymes, a class of oxygen sensors that regulate the stability of hypoxia-inducible factor. This specificity distinguishes hydralazine from broad-spectrum iron chelators, which disrupt multiple oxygen-sensing pathways and cause off-target effects.
This also speaks to the inherent value of drug repurposing in oncology. As has been discussed within the broader research in cancer pharmacology, the repositioning of FDA-approved drugs avoids many years of safety testing, accelerating translation into clinical trials. Given hydralazine’s established safety profile-despite known side effects such as lupus-like autoimmune reactions-it would be well-suited for rapid testing in glioblastoma patients. Its relatively low molecular weight and chemical tractability open a window to designing next-generation ADO inhibitors with improved brain penetration while reducing systemic exposure.
The work also connects to the molecular biology of oxygen sensing illuminated by the 2019 Nobel Prize in Physiology or Medicine. Whereas much attention has focused on the HIF pathway, ADO represents a HIF-independent oxygen sensor with unusually low oxygen affinity, making it particularly relevant in the extreme hypoxia of solid tumors. By targeting this “first responder” enzyme, hydralazine exploits a vulnerability in the tumor’s metabolic infrastructure.
Future steps will be in vivo testing to see whether ADO inhibition can be achieved safely in the brain and whether senescent glioblastoma cells remain arrested long-term or can escape dormancy. If effective, this approach could be used in conjunction with current treatments, which include surgery, radiotherapy, and temozolomide, to prevent recurrence, which is otherwise rarely halted. As co-lead author Megan Matthews explains, “Understanding how hydralazine works at the molecular level offers a path toward safer, more selective treatments.” For a cancer type where median survival remains under 15 months, the prospect of disabling a core survival mechanism with a well-known cardiovascular drug is a rare and promising development.
