
Artist's image depicting a group of cancer cells (© fotoyou - stock.adobe.com)
In A Nutshell
- Polyamines, small molecules found in all cells, were shown to boost cancer growth by stimulating the protein eIF5A2.
- Silencing eIF5A2 in cancer cell lines quickly halted tumor growth, while its twin protein eIF5A1 had little effect.
- Analysis of over 2,400 breast cancer patients linked higher eIF5A2 levels with poorer survival.
- Existing drugs like DFMO already disrupt the polyamine pathway — pairing them with eIF5A2-targeting drugs could create more precise treatments.
TOKYO — Scientists have discovered that small molecules naturally present in every cell may hold the key to understanding why cancer spreads so aggressively. Researchers have identified a specific protein that could become the target for more precise treatments.
A new study reveals that polyamines, which are small molecules that help maintain normal cell functions, actually hijack cancer cells in a surprising way. Instead of supporting healthy growth, these molecules boost production of a protein called eIF5A2, which appears essential for cancer survival. Even more intriguing: eIF5A2’s nearly identical twin protein, eIF5A1, doesn’t have the same effect on cancer cells.
Published in the Journal of Biological Chemistry, the research shows that targeting eIF5A2 could offer a way to attack cancer cells while leaving healthy tissue largely unharmed.
Cancer Cells Switch Energy Systems for Rapid Growth
When researchers from Tokyo University of Science, led by associate professor Kyohei Higashi, removed polyamines from cancer cells, something unexpected happened. The cells’ entire energy production system shifted gears.
Normal cells prefer using their mitochondria, the cellular powerhouses that efficiently convert nutrients into energy. Cancer cells, however, switched to a faster but less efficient process called glycolysis, even when plenty of oxygen was available. Cancer cells choose this approach because it supports their rapid, uncontrolled growth.
The study found that polyamines orchestrate this metabolic switch by increasing levels of two key proteins: PDK1 and PKM2. Both proteins are crucial players in the glycolysis pathway that fuels cancer cell multiplication.
Using advanced protein analysis techniques on over 6,700 proteins, researchers discovered that polyamine removal affected 300 proteins, about 5.3% of the cell’s entire protein machinery.

Twin Proteins With Different Jobs in Cancer
eIF5A1 and eIF5A2 share 84% of their genetic code and scientists long assumed they worked similarly. But the new research reveals these molecular twins have drastically different jobs in cancer.
When researchers silenced eIF5A2 in multiple cancer cell lines including cervical and breast cancer, cell growth stopped within three days. Silencing eIF5A1 barely affected cancer growth until day five, and even then the impact was minimal. Based on their experimental results, the researchers concluded that eIF5A2 plays a more critical role in cancer cell proliferation than eIF5A1.
The difference comes down to a cellular brake system. A microRNA called miR-6514-5p normally prevents eIF5A2 production. Polyamines essentially cut those brake lines, allowing eIF5A2 to be produced freely. Meanwhile, eIF5A1 production remains under normal cellular control.
Computer simulations revealed why eIF5A2 might be more effective in cancer cells. Though the proteins differ by just 28 building blocks, those differences occur in the region that interacts with ribosomes, or the cellular factories that make proteins.
Patient Survival Data Confirms Laboratory Findings
The laboratory discoveries translate to real patient outcomes. Analysis of breast cancer data from over 2,400 patients showed that higher eIF5A2 levels correlated with worse survival rates. eIF5A1 levels showed no such correlation.
Cancer cells also modify their protein-making factories by increasing production of specific ribosomal proteins, particularly RPS27A, RPL36A, and RPL22L1. All three have been previously linked to aggressive cancer behavior and poor patient outcomes.
Several existing drugs already target parts of the polyamine pathway. DFMO, the compound used in this study to deplete polyamines, is currently in clinical trials for various cancers. The new research suggests that combining such treatments with drugs specifically targeting eIF5A2 might prove more effective than current approaches.
The discovery changes how scientists understand cancer cell behavior. Rather than being passive victims of genetic mutations, cancer cells actively recruit natural molecules like polyamines to support their growth. By identifying eIF5A2 as a critical player in this process — and one that healthy cells don’t depend on — researchers have uncovered a potential vulnerability that could lead to more precise treatments for cancer patients.
Paper Summary
Methodology
Researchers used multiple cancer cell lines including HeLa S3 cervical cancer cells and breast cancer lines (MCF7, MDA-MB-231, MDA-MB-468) to study polyamine effects on cancer growth. They depleted polyamines using DFMO and analyzed changes in over 6,700 proteins using mass spectrometry. The team used gene silencing techniques to knock down specific proteins, measured cellular energy production rates, and conducted computer simulations to understand protein interactions. Clinical relevance was assessed using breast cancer patient data from over 2,400 patients in the METABRIC database.
Results
Polyamine depletion changed expression of 300 proteins, shifting cancer cells toward glycolysis rather than normal energy production. eIF5A2 silencing stopped cancer cell growth within 3 days, while eIF5A1 silencing had minimal early effects. Polyamines stimulated eIF5A2 production by blocking a regulatory microRNA (miR-6514-5p). The proteins showed distinct roles despite genetic similarity, with eIF5A2 promoting cancer-associated ribosomal proteins. Higher eIF5A2 levels correlated with worse survival outcomes in breast cancer patients.
Limitations
The study relied primarily on established cancer cell lines rather than tumor samples from patients, which may not fully represent real cancer behavior. Experiments focused mainly on cervical and breast cancers, so results may not apply to all cancer types. Computer simulations provide helpful models but may not perfectly reflect actual cellular conditions. The patient data analysis was observational, limiting conclusions about direct causation.
Funding and Disclosures
Research was supported by Grant-in-Aid for Scientific Research (C) No. 18K06652 from the Japan Society for the Promotion of Science, the Hamaguchi Foundation for the Advancement of Biochemistry, and an Extramural Collaborative Research Grant from the Cancer Research Institute at Kanazawa University. Authors declared no conflicts of interest.
Publication Information
“Polyamines stimulate the protein synthesis of the translation initiation factor eIF5A2, participating in mRNA decoding, distinct from eIF5A1” was published in the Journal of Biological Chemistry, Volume 301, Issue 8, Article 110453, in 2025. DOI: 10.1016/j.jbc.2025.110453.







