The same protein that shields tumors from immunity helps mice survive the flu

The Jackson Laboratory

In cancer, PD-L1 is the enemy. Tumors plaster it on their surfaces to shut down immune cells, and some of the most successful cancer drugs of the past decade work by blocking it. The logic is straightforward: remove the shield, let the immune system attack.

But in the lungs of mice fighting influenza, PD-L1 does something entirely different. It helps.

A new study from The Jackson Laboratory (JAX), published in Cell Reports, found that Programmed Death-Ligand 1 (PD-L1) - the very protein that cancer immunotherapies are designed to neutralize - actually strengthened antiviral defenses in immunocompromised mice. When researchers treated infected mice with an antibody that activated PD-L1, the animals survived longer. Their natural killer cells became more effective at destroying virus-infected lung cells, and critically, this improved viral clearance did not increase lung damage.

"This discovery suggests that a pathway targeted in cancer could also be useful in infectious disease, but in the opposite way," said Silke Paust, a JAX professor and immunologist who led the research. "Cancer therapies block PD-L1, whereas in flu, enhancing it may strengthen host defense."

A protein caught doing two jobs

PD-L1 has been understood primarily as an immune suppressor. It sits on the surface of tumor cells and interacts with PD-1, a receptor on T cells, to dampen immune activity. That interaction is the basis for checkpoint inhibitor drugs like atezolizumab and durvalumab, which have transformed treatment for cancers including lung, bladder, and triple-negative breast cancer.

For years, scientists assumed that immune suppression was essentially PD-L1's only function. But Paust's team stumbled onto a different story after unexpectedly finding PD-L1 on immune cells in human and mouse lungs. That observation prompted a question: if PD-L1 is present on lung immune cells, is it doing something beyond shielding tumors?

To find out, the researchers used mice that lacked T and B cells - the adaptive immune system's heavy hitters - and were therefore dependent on innate immunity. In these mice, the primary antiviral defense comes from natural killer (NK) cells, which can rapidly identify and destroy infected or abnormal cells without prior exposure to a pathogen.

What happened in the lungs

After infecting the mice with influenza, the team observed that NK cells in the lungs produced high levels of PD-L1. When they treated the mice with an antibody that activated PD-L1 signaling, the NK cells became substantially better at killing virus-infected cells. The treated mice survived longer than untreated controls.

One detail stood out: the improved viral clearance did not come at the cost of collateral tissue damage. In many respiratory infections, the immune response itself causes as much harm as the virus - a phenomenon familiar from severe COVID-19 cases. Here, activating PD-L1 appeared to enhance precision rather than brute-force destruction.

The mechanism involved a downstream molecule called TRAIL (tumor necrosis factor-related apoptosis-inducing ligand). PD-L1 signaling in NK cells increased TRAIL expression, and TRAIL is specifically designed to trigger programmed cell death in infected or damaged cells. The result was more targeted killing of flu-infected cells without the inflammatory collateral damage that often accompanies aggressive immune responses.

And there was another surprise. In cancer, PD-L1 works by interacting with PD-1. But in this flu model, the enhanced NK cell function operated independently of PD-1. PD-L1 was sending signals inside the NK cells themselves, a form of reverse signaling that had not been well characterized in viral infection.

Human signals in COVID-19 patients

The researchers also examined human lung tissue and blood samples from COVID-19 patients. They found similarly increased expression of both PD-L1 and TRAIL in lung NK cells, suggesting that the mechanism observed in mice may extend to human respiratory viral infections. This does not prove the same protective function operates in people, but it establishes that the molecular players are present and active during severe viral lung disease.

"Most healthy people don't die from the flu when they get sick," Paust said. "It's really the very young, the very old, and certainly the immunocompromised who are at risk of dying. Immunocompromised patients with flu usually die from severe complications, especially viral or secondary bacterial pneumonia leading to respiratory failure, sometimes with sepsis and multi-organ failure."

What this might mean for cancer therapy

The findings carry an intriguing implication for oncology. If PD-L1 can send functional signals inside cells during viral infection, it may do the same in tumors. That could help explain a puzzle that has nagged the immunotherapy field: why drugs that block PD-L1 and drugs that block PD-1 - two sides of the same checkpoint pathway - do not always produce identical results in cancer patients.

"If PD-L1 is sending signals inside cells during flu infection, it may do the same in cancer and other diseases," Paust said. "That could reshape how we think about immune checkpoints, because this molecule's role may depend on which cells carry it. In the lung, PD-L1 may be doing two jobs at once."

Mouse model, open questions

The study's most significant limitation is also its most obvious: the mice used lacked T and B cells entirely. That is a useful model for studying immunocompromised patients, but it does not replicate the immune environment of healthy individuals, where T cells, B cells, and NK cells all interact. The next step, which Paust's team is already pursuing, is to study how PD-L1 behaves in lungs with a full, intact immune system.

"We still don't understand how PD-L1 behaves in healthy lungs with a full immune system," Paust acknowledged. "Figuring that out could help us understand how the lung regulates immune responses to viral infection."

The study also involved only influenza. Whether PD-L1 plays the same protective role in other respiratory viruses - RSV, parainfluenza, or future pandemic strains - remains unknown. And translating a finding from mice to a viable human therapy would require years of additional preclinical and clinical work.

But the core finding reframes a well-known molecule. PD-L1 is not simply a brake on immunity. In the right context, on the right cells, it can be an accelerator - and that distinction may matter for both infectious disease and the next generation of cancer treatments.