Researchers identify new treatment targets for lung diseases using big data

Mar 20 2018

Every year, approximately 12 million adults in the U.S. are diagnosed with Chronic Obstructive Pulmonary Disease (COPD), and 120,000 die from it. For people with COPD, Haemophilus influenzae, a bacterium, can be particularly dangerous.

The microbe can reside in their lungs and wreak havoc within already weakened organs. Identifying the genetic variations of this microbe is vital to treating these patients effectively.

Now, a University of Maryland School of Medicine (UMSOM) researcher teamed up with researchers at the University of Buffalo and Yale University to better understand how the bacterium adapts quickly, which may open new avenues for therapy for COPD patients. The new findings, which relied on genomic analysis, may also be useful for people who have other diseases such as ear infections or pneumonia, since this microbe can cause these diseases as well.

The research was published today in the Proceedings of the National Academy of Sciences.

"The question we asked was why are certain strains of the bacterium are so much more dangerous than others. We discovered a genetic pattern, which helps explain why certain strains are so virulent," said Hervé Tettelin, PhD, associate professor of microbiology and immunology at the Institute for Genome Sciences (IGS) at UMSOM. "This offers key clues about what this microbe does to evolve in the lungs of people with COPD, and it may open exciting new avenues for treatments and vaccines for the future." Dr. Tettelin oversaw the genomic data mining with the isolates.

He collaborated with Timothy Murphy, MD, senior associate dean for clinical and translational research and professor at the University at Buffalo Jacobs School of Medicine and Melinda Pettigrew, PhD, senior associate dean for academic affairs and a professor of epidemiology at the Yale School of Public Health. Over 20 years, Dr. Murphy has collected thousands of strains from COPD patients.

Some strains of the bacteria are more dangerous than others. The level of lethality depends in part on which genes get turned on, and which are turned off. Certain patterns allow the bacteria to adapt more efficiently to the lungs, allowing it to cause more damage.

With data mining, the researchers detected certain patterns of genetic activation or inactivation. With this information, the researchers say they may be able to develop new treatments and new vaccines. "We now have a much better understanding of how certain strains of the bacterium adapt to the lungs," said Dr. Murphy.

The team studied the genomic isolates from distinct time periods: what the isolates look like when Haemophilus is acquired by a patient, and how they look when they're about to be cleared out of the lungs. The researchers learned that there is evidence of adaptation. H. influenzae uses mechanisms to vary its genome and the proteins it uses to interact with the host.

Interestingly, much of the variation appears to be random. In the bacterium, a subset of genes are randomly turning off and on constantly. Some of these mutations are useful to the microbe, while some are not. Those that work are conserved, while those that fail do not survive. Essentially, the bacterium undergoes a constant state of evolution.

These findings are important for vaccine researchers. This kind of data mining helps researchers more precisely identify better vaccine candidates, which can lead to better treatments for COPD patients.