Immune system protein is a cancer killer on its own, new study shows

For decades, scientists have known that a protein made by T cells and natural killer (NK) cells in the immune system, called perforin, can kill tumor cells by punching a hole in the cell wall and injecting proteases called granzymes, which kill the cell. 

New work led by researchers at the University of Michigan Health Rogel Cancer Center found that tumor cells could be killed even when major granzymes were missing.  

“If T cells and NK cells are the soldiers of the immune system, perforin is the weapon, and granzymes are the ammunition,” said co-senior study author Weiping Zou, M.D., Ph.D., Charles B. de Nancrede Professor of Pathology, Immunology, Biology, and Surgery at U-M. “At least, that’s what the classic understanding was.” 

No granzymes needed 

Their new study, published in Cell, reveals that perforin is a cancer killer on its own, no major granzymes needed. They also found that a loss of perforin causes a catastrophic failure of anti-tumor immunity. The discoveries will likely have implications for understanding how cancers are killed by immune cells and could have ripple effects throughout the field of immunology. 

“Our findings show, for the first time, that perforin can independently induce ferroptosis in tumor cells,” said co-senior author Ilona Kryczek, Ph.D., a researcher at U-M. “This work changes our understanding of one of the longest held mechanisms of T cell and NK cell mediated tumor clearance.” 

Zou and Kryczek collaborated closely with co-senior author Judy Lieberman, Ph.D., of Harvard Medical School, one of the international pioneers of cytotoxic lymphocyte research. 

Cells can face different forms of death. There’s apoptosis, which is what granzymes do. Pyroptosis is a form of cell death often caused by bacterial infection. And there’s ferroptosis, which is an iron dependent form of cell death driven by the accumulation of lipid peroxides in cellular and organelle membranes. Ferroptosis is a relatively new concept, formally introduced over a decade ago, so scientists know less about it than other, more well established forms of cell death. 

“It’s a very hot area that needs a lot of research,” Zou said. In 2019, his group was the first to report the immune system’s role in ferroptosis in tumors, highlighting its potential as a target pathway for immune checkpoint therapy. But questions remained: Could perforin on its own really kill cancer cells? And if so, how? 

If we can understand what makes a tumor cell susceptible to perforin, we may be able to increase the effectiveness of immunotherapy.”  

-- Weiping Zou, M.D., Ph.D.  

 

Three pronged approach 

To find out, Zou and collaborators took a three pronged research approach, pairing in vitro experiments of genetic and biochemical manipulations with in vivo mouse studies and human cancer patient data.  

The in vitro experiments revealed that perforin without granzymes is still deadly to cancer cells, with ferroptosis as the mechanism. Mice with the iron rich diet had slower tumor progression that was related to their perforin expression. And cancer patients with higher perforin expression saw improved survival and better responses to immune checkpoint blockade treatments. 

With those three lines of evidence, the answer to Zou’s first question was clear. 

“We wanted to know if perforin, by itself, can do the job,” he said. “The answer is yes.” 

“It is a new way of thinking about how immune cells kill tumors,” said Hannah Bell, Ph.D., the study’s first author and a recent postdoc in Zou’s lab who is now an assistant professor in Molecular and Integrative Physiology at U-M Medical School.  

Death of tumor cell 

Perforin is capable of not only punching a hole in the target cell membrane, but also of interfering with the target cell’s iron uptake so that it accumulates problematically inside the cell, including inside the mitochondria. There, the excess iron rapidly oxidizes lipids. The cell’s whole system goes haywire, and the tumor cell dies. 

The takeaway is not to give all cancer patients an iron rich diet. “We need to be very cautious here,” Zou said. “That’s because some cancer types enjoy an iron rich diet. For those tumors, and we don’t totally know which ones yet, an iron rich diet may accelerate progression.” 

It’s mostly too soon to say what the clinical applications of the finding will be, but previous work found that immune checkpoint blockade can enhance T cell mediated ferroptosis, so that’s likely an avenue of future research. “If we can understand what makes a tumor cell susceptible to perforin, we may be able to increase the effectiveness of immunotherapy,” Zou said. 

The team’s future research will delve deeper into the mechanism of cell death, test whether multiple cell death pathways could be working together and make inroads to clinical applications. 

“At the end of the day, the whole goal of different forms of cancer therapies — immunotherapy, chemotherapy, targeted therapy — is to kill tumor cells,” Zou said. “Improving our understanding of cell death, like we do in this study, is so fundamental and so important. That’s why we’re going to continue to push on this.” 

Additional authors: Kavya Nair, Kole Tison, Peng Liao, Hongwei Lei, Marwa O. El-Derany, Ayijiang Yisimayi, Haiwei Zhang, Ryan Rebernick, Weichao Wang, Yoshitaka Adachi, Cristina Castillo, Nupur K. Das, Jinsu An, Vivien R. Sutton, Xun Lin, Heng Lin, Xinghao Wang, Yunfei Li, Linda Vatan, Amr Aboud, Shuang Wei, Miguel Gijon, Samantha Goodwin, Yatrik M. Shah, Joseph A. Trapani, Margarida Barroso 

Funding: National Cancer Institute grants CA217648, CA123088, CA099985, CA193136, CA152470, CA46592, R01 CA233188, R21 CA274622, R01 CA240955, K99CA299410-01, and National Science Foundation grant NSFCBET 234722 

Paper cited: “Perforin triggers iron-induced mitochondrial damage to drive ferroptosis and anti-tumor immunity,” Cell. DOI: 10.1016/j.cell.2026.09.062