When an asthma attack strikes, the body turns against one of its most essential functions. Breathing becomes difficult, creating panic and increased inflammation. For millions of people living with asthma, the sensation can feel like trying to breathe through a straw.

According to Dan Huh, Professor in Bioengineering, much of what scientists still don’t understand about asthma may come down to something surprisingly physical.

“Asthma changes the lung, and not for the better,” says Huh. “It damages and reshapes airways over time. We tend to blame inflammation for it, but our work shows that the physical force of an airway squeezing shut may be just as important in understanding what drives those changes.”

In a new paper published in Nature Biomedical Engineering, Huh, together with lead author Jungwook Paek, a former postdoctoral researcher in Huh’s lab who is now an Assistant Professor of Electrical and Computer Engineering at Binghamton University, and colleagues have developed a bioengineered “asthma-on-a-chip” system that recreates the mechanical stresses experienced by human airways during an asthma attack. The platform allows scientists to observe how diseased lung tissue responds to physical compression in ways previously impossible to study directly in patients.

Looking at Asthma Differently

In healthy lungs, the airways widen and narrow freely with every breath. In asthmatic lungs, that flexibility is gradually lost. Beyond the familiar episodes of sudden narrowing, the airways themselves undergo lasting structural changes. Most notably, their walls thicken and stiffen as collagen builds up within them through a process known as airway remodeling. The resulting stiffer, narrowed airways are harder to keep open and respond poorly to standard medications, becoming a hallmark of the most severe and persistent forms of the disease. For decades, this remodeling was attributed almost entirely to chronic inflammation.

“Inflammation has long been seen as the main driver of asthma, and most therapies are built around the idea of controlling it,” says Huh. “The fact that asthma remains a major clinical challenge tells us we are missing part of the picture. One thing we know for certain is that the airways in asthmatic lungs constrict frequently. Yet we understand surprisingly little about how this defining mechanical feature affects the pathophysiology of the disease.”

A landmark clinical study published in the New England Journal of Medicine 15 years ago showed that the constriction of airways, even with no inflammation present, was by itself enough to drive tissue stiffening in patients with asthma. Physical force, not inflammation alone, was shown to reshape the airway.

But studies in patients can only go so far. This work established that the force mattered, but how it inflicted its damage remained a question.

Huh and Paek, who brought expertise in soft robotics and microfabrication into his role as a postdoctoral researcher in Huh’s lab, set out to answer that question by engineering realistic human airway tissues using patient cells.

Their lab-grown tissue was built using patient-derived airway epithelial cells, those at the air-tissue interface, and fibroblasts, cells involved in tissue repair and remodeling, embedded within a hydrogel structure that mimics the architecture of the human airway.

“Beyond developing the lung tissue on a chip, the bigger challenge was to subject this tissue to the same physical forces that in vivo lung tissues experience as the airways constrict,” says Paek. “To tackle this challenge, we drew on an approach more familiar to soft robotics than to biology. We built living human airway tissue into a device containing microfabricated air chambers and thin flexible membranes that gently squeeze the engineered tissue the way an asthmatic airway constricts in the lung.”

The chip gave the team a way to connect mechanics, tissue remodeling and molecular signaling all within the same system, which is very difficult to study directly in patients.

When Force Changes Biology

When the team applied controlled compression, mimicking airway constriction during an asthma attack, they discovered something surprising. The asthmatic tissue underwent pronounced fibrotic remodeling and stiffening while healthy tissue exposed to the identical force barely responded.

“By isolating mechanical compression from every other variable, our model identified the compression itself as a direct instigator of tissue remodeling and, for the first time, allowed us to visualize and probe the process as it unfolded,” says Paek.

The system then revealed something entirely new. While an abnormally high density of blood vessels is a well-documented feature of asthmatic airways, little is understood about why that happens. Here, the team found that the same compressive force responsible for fibrotic remodeling also induced this excess vascular growth, with the fibrosis itself acting as a key trigger.

“It is the first evidence that the mechanical force of constriction drives vascular remodeling in the asthmatic airway, linking a physical event to two of the disease’s defining structural changes and pointing to how repeated constriction could progressively worsen the disease over time,” says Huh.

Reading the Molecular Signals of Stress

Having shown that mechanical force could drive both fibrosis and abnormal vascular remodeling, the team set out to investigate the molecular basis of those changes and whether such mechanistic insight could point toward new treatments.

As the airway tissue was compressed and remodeled, its cells released a shifting mixture of proteins into the surrounding fluid. By collecting this fluid from the chip’s microchannels and running proteomic analysis, the team could read those signals directly.

“We looked at which proteins were turned up or down in response to mechanical stress,” says Paek. “This kind of analysis tells us how the tissue is responding at the molecular level and can also reveal disrupted pathways we might be able to rescue with drugs.”

The result cut both ways. Some of the disrupted proteins were already known players in asthma, confirming that the model was reproducing the right biology. Others had not been connected to the disease before, pointing toward potential new therapeutic targets.

The team then went a step further, treating the tissue with compounds aimed at some of these pathways and tracking how the tissue response changed. The experiments demonstrated how the platform could eventually support drug discovery and biomarker development.

“This work shows how bioengineered human tissue models can be used to advance our mechanistic understanding of disease pathophysiology and to discover and test new drugs,” says Huh. “It allows researchers to come full circle from asking the initial question, to conducting the study and then to offering potential solutions for patients living with the very condition we rebuilt from their own cells.”

Building the Future of Preclinical Medicine

The asthma chip is part of a broader scientific movement known as New Approach Methodologies (NAMs): human-relevant experimental systems designed to complement or reduce reliance on animal testing.

When Huh started his lab at Penn nearly 14 years ago, organ-on-a-chip technology was still emerging. Today, the field is rapidly evolving toward more realistic and complex living models.

“We’re raising the bar,” Huh says. “Real tissues don’t exist and function in isolation, so we don’t study them that way. We are developing new ways to recreate how different tissues live and work together, and then to watch what happens when we put them under the kinds of stress that drive disease.”

His lab has previously engineered chip-based models of the lung, intestine, liver, placenta and even the blinking human eye. But for Huh, sophisticated devices have never been the point.

“The fun part, as an engineer, is building the gadget,” he says. “The important part is making sure it helps us answer questions that matter.”

That conviction runs through his lab’s culture, where researchers are encouraged to think simultaneously as engineers, biologists and problem solvers. The goal is to pursue questions that are clinically grounded and produce data of real value, not only for fundamental science but for patients living with the conditions they study.

For Paek, who is now applying similar approaches to neurodegenerative diseases like Alzheimer’s and Parkinson’s, the lesson extends beyond any one technology.

“The tools, the way we collect and assess data and the scientific challenges we are trying to address will continue changing,” he says. “But the mindset — curiosity, asking important questions, approaching problems creatively — stays the same.”

Learn more about the work being done in Dan Huh’s lab here.

Lakshminarayan Reddy Teegala, Charles Thodeti and Sailaja M. Paruchuri of the Department of Physiology and Pharmacology in the College of Medicine and Life Sciences at the University of Toledo; Farid Alisafaei and Vivek Shenoy, both in the Department of Material Science and Engineering at Penn Engineering and the NSF Science and Technology Center for Engineering Mechanobiology at the University of Pennsylvania; Anika Alim in the Department of Electrical and Computer Engineering at Binghamton University; Joseph W. Song, Sunghee E. Park, Jeehan Chang and Haijiao Liu in the Department of Bioengineering at the University of Pennsylvania; Bang-Jin Kim and Sandra Ryeom in the Department of Cancer Biology at the Perelman School of Medicine at the University of Pennsylvania; and Geremy C. Clair in the Biological Sciences Division at the Pacific Northwest National Laboratory also contributed to this work.