Like inflammation, fibrosis - the laying down of collagen to form scars - is a response to injury that's absolutely necessary for survival. But when the process goes awry, that means trouble and can lead to diseases such as idiopathic pulmonary fibrosis.

What sets off idiopathic pulmonary fibrosis, or lung scarring, is unknown. But once it starts, the statistics are grim. "It is a bad disease," Harold Chapman told BioWorld Today. About one-half of patients die within three years of being diagnosed, because there is practically no treatment. Other than a lung transplant, "in virtually all of the published literature, there are only marginal effects of any treatment" for IPF, Chapman said.

For all its severity, little is known about the mechanisms of idiopathic pulmonary fibrosis. But in a paper to be published in the Aug. 29, 2006, issue of the Proceedings of the National Academy of Sciences, now available online, Chapman and his colleagues at the University of California at San Francisco reported that one feature of the disease is the induction of epithelial-mesenchymal transition, or EMT, transforming alveolar cells into fibroblasts.

During EMT, cells free themselves from the extracellular matrix and become more mobile. The process is important during development and also is thought by some to play a role in cancer metastasis, though Chapman said that both the cancer and the injury response literature on EMT has been "controversial." The PNAS paper supports the idea that EMT is indeed a major feature of lung fibrosis.

"We got into this serendipitously," said Chapman, who is chief of the division of pulmonary and critical care medicine at UCSF. When staining patient tissue, they noticed cells markers for both alveolar cells, which are epithelial cells, and mesenchymal-type cells, suggesting that the alveolar cells were transforming into fibroblasts.

But those experiments can only go so far. "In the patients, we could only describe what we saw," Chapman said. For that reason, the scientists turned to an animal model that allowed them to trace the fate of alveolar cells as they are changing: a transgenic mouse with epithelial cells that express a reporter gene. As epithelial cells and their descendants divide, the reporter gene keeps being expressed, which makes it possible to determine whether a fibroblast was once an epithelial cell.

The researchers induced lung fibrosis in the transgenics and compared the injured lungs to those of regular transgenics. They found that mesenchymal cells in control animals never expressed the reporter gene, meaning that they could not have formed via EMT. In contrast, in injured lungs they found mesenchymal cells that expressed the reporter gene, meaning that they or their ancesters had once been epithelial cells.

Simultaneous staining for the reporter gene and fibroblast markers revealed that some but not all of the EMT-derived mesenchymal cells in injured lungs were fibroblasts. The number of fibroblasts had roughly doubled three weeks after lung injury, and most, though not all, of the increase came from EMT fibroblasts.

In the injured lungs, many of the EMT-derived cells were located within patches of fibronectin; the researchers set out to determine whether fibronectin drives the EMT transition, or whether EMT-derived cells deposit fibronectin in injured lungs. In cell culture experiments, culturing alveolar cells on fibronectin promoted their transformation into mesenchymal fibroblasts. Indeed, after one week of cozying up to fibronectin in a test tube, nearly 40 percent of mesenchymal cells expressed the reporter gene and were thus derived from epithelial cells.

Delving deeper into just how the cells are prodded into transforming, Chapman and his team found that blocking TGF-beta signaling also blocked the transition, and TGF-beta itself was regulated by a specific matrix protein - the poetically named integrin alpha-v-beta-6. However, epithelial cells needed to interact with the matrix for more than just the integrin, as cells treated with active TGF-beta that were not of a fibronecting matrix died off in massive numbers.

The research suggested two targets worth pursuing for idiopathic pulmonary fibrosis. "TGF-beta is an obvious one," Chapman noted, and already is being pursued by several companies. "But more than that, it suggests that there are pathways regulating TGF-beta that could be targeted, including this integrin."

As for Chapman and his colleagues, they intend to return to the patients to see whether the same transition is at work in the clinic. "We've only shown this convincingly so far in the mouse," Chapman cautioned. The next questions they want to answer are, "Does the same thing occur in patients? And if it does, is it crucial?"

At a more basic level, the research also suggests a new way of thinking about wound healing: "There are a number of ways that the healing process could occur," Chapman said. "Our work suggests that after an injury, epithelial cells are not just activating tissue fibroblasts. They are differentiating into fibroblasts and laying down a matrix that develops into a scar," which is "a different way of thinking about how healing could occur."