Asking whether a disease (or a behavior) is caused by genetic or environmental factors has been likened to asking whether the area of a rectangle is determined by its length or its width.

Many smoking-related disorders illustrate the limits of the question nicely. A case in point: Smoking causes 80 percent to 90 percent of cases of chronic pulmonary obstructive disease (COPD) - roughly speaking, chronic bronchitis and emphysema - but COPD affects only 15 percent to 20 percent of all smokers (which is still enough for it to kill almost 120,000 U.S. citizens annually).

Researchers from the University of Pittsburgh Medical Center, Nanjing Normal University in China, University of British Columbia in Vancouver and Cornell University's College of Physicians and Surgeons in New York wanted to know the genes that might be responsible for determining which unlucky smokers go on to develop COPD. Their findings are reported in the Oct. 12, 2004, issue of the Proceedings of the National Academy of Sciences in a paper titled "Comprehensive gene expression profiles reveal pathways related to the pathogenesis of chronic obstructive pulmonary disease."

Low-Throughput Method + Patience = High Data Output

The researchers used both microarray analysis and a technique known as serial analysis gene expression, or SAGE, to profile genes that are differentially expressed in the lung tissue of smokers with and without COPD. "The major method people are using now is microarrays, which is popular because it's very high-throughput. SAGE is based on actual sequencing, so it gives more comprehensive and quantitative data. But it is not a high-throughput method," said Augustine Choi, chief of pulmonary, allergy and critical care medicine at the University of Pittsburgh Medical Center and senior author of the study.

For a low-throughput method, it must be said that the output was pretty impressive. The scientists sequenced almost 60,000 gene tags to generate comprehensive gene-expression profiles in the lung tissues of either healthy smokers or smokers with moderate COPD. They identified 327 genes that were significantly up- or down-regulated in the COPD patients.

The scientists picked eight of those genes for further functional study. Choi said that determining selection criteria for those genes that were further investigated was difficult, but "we decided to go with interesting genes that fit the schematic of emphysema." To be included, genes had to be identified by both SAGE and microarray methods, and represent "functional classes of genes with important roles in the pathogenesis of COPD."

Anatomical studies revealed that the expression patterns of the affected genes were "diffuse, but primarily in the epithelial cells," Choi told BioWorld Today. "This fit with what we knew already: Smoking is the real way to get emphysema, and the first cells to get exposed to smoke are epithelial cells, because the epithelium lines the airway."

One tantalizing clue as to the mechanism of COPD was offered by the fact that several of the genes the scientists identified are involved in fibrosis. That process is basically the excessive formation of scar tissue in response to injury. Choi explained that fibrosis "is not just bad, because you need it to repair injury. But if there is dysregulation, then you have a really serious problem."

Fibrosis occurs in a variety of organs and is common in lung disease. "Emphysema has not been known as a fibrosis disease, but what we saw is that there was a hint of fibrosis," Choi said.

That hint spurred the researchers to investigate the effects of smoke in fibroblasts. Most cell culture studies on the mechanisms of COPD are done in other cell types, but "we studied fibroblasts because our gene-expression data suggested that there is a fibrotic component to COPD." They focused on the transcription factor EGR-1, which is highly expressed in the lung tissue of emphysematic patients, and its effects on matrix metalloproteinases. That class of proteins "is the one that chews up your tissue," Choi said, meaning they are involved in the breakdown of matrix tissue.

In human lung fibroblasts, concentrated cigarette smoke induced the expression of EGR-1. In mouse fibroblasts with EGR-1 knocked out, the expression of two matrix metalloproteinases was affected by knocking out EGR-1. Collectively, the experiments point to the possibility that matrix metalloproteinases might "add to the destruction of lungs" in COPD.

The research being reported in PNAS was obtained using whole-lung tissue. Next, the researchers plan to use laser-captured micro-dissection to "zero in on specific areas," Choi said. In the long run, he hopes that the data gathered using the SAGE approach will allow the identification of new molecular targets and ultimately, new therapeutics to treat COPD. That ability, he said, is the strength of comprehensive gene-expression profiling. "Doing one gene at a time just doesn't give you the same kind of possibility."