Genetic difference in staph infects some heart devices, not others

Now researchers from Duke University Medical Center and Ohio State University (OSU) have discovered how and why certain strains of staphylococcus aureus (SA) bacteria, the leading cause of these device infections, have infected thousands of implanted cardiac devices. About 4 percent of the one million annually implanted devices become infected.

The researchers examined SA's ability to bind to a sticky human (mammalian) substance called fibronectin that circulates in blood and sticks to the surfaces of implanted devices, like pacemakers.

Staph bacteria have fibronectin-binding molecules and bind to the human protein to establish an infection on the implanted medical device. Once established, these infections are difficult or impossible to eradicate without removing the device itself.

"This the first step in biofilm-based disease work," said Vance Fowler, MD, MHS, an associate professor of infectious diseases in the Duke Department of Medicine and co-corresponding author of the study. "I would expect the findings would be relevant for most implanted devices. The difference is that the cardiac devices are in direct contact with the bloodstream, and thus with fibronectin, so we need to do further work to clarify."

The study appears online in the Proceedings of the National Academy of Sciences.

"The question was, 'are all SA created equal when binding with fibronectin?' and the answer is no," Fowler said. "We identified differing SA isolates from the blood of patients. All of the patients had SA, but some of the cardiac devices were infected and some were not, and we wanted to learn why. Most people had the infection but a lucky few didn't."

Working with the lab of Steven K. Lower, PhD, at OSU, which specializes in atomic force microscopy, the team sequenced the binding regions of the gene that coded for fibronectin-binding protein in the bacteria.

They found that SA with three specific one-letter differences in their DNA were significantly more common in the infected cardiac-device group. The infectious bacteria had one to three of these changes. The research team also verified that the ability to bind was stronger in the three SA strains found in the infected group.

"We often hear that nanoscience will make the world a better place, and our study demonstrates a direct correlation between something that occurs at the scale of a nanometer (i.e. a bond between a bacterium and implant) and the health of human patients with cardiovascular implants," said Steven K. Lower, co-corresponding author and associate professor in the OSU School of Earth Sciences.

"Some practical implications of this research could be a new protocol to determine risk of serious biofilm-related infections for patients with prostheses or patients who are considering surgical implants. For example, we could obtain a culture of S. aureus from the skin of a patient, and determine the risk of a biofilm-based infection, using the methods we described."

Roberto D. Lins, a computer engineer with structural biochemistry and modeling expertise at the Universidade Federal de Pernambuco in Recife, Brazil, showed through dynamic modeling the interaction of the protein and the SA polymorphs of interest.

Using a powerful computer, the team saved about two-and-a-half years' time and learned that the three DNA differences were associated with SA's ability to form more chemical bonds with fibronectin. These SA strains had an increased number of hydrogen bonds between the fibronectin (in people) and the fibronectin-binding protein (in the SA).

"Getting to the fundamental answers of common, serious infections that plague our patients is why I stay in research," Fowler said. "Now we have a plausible biological explanation of why these particular SNP mutations matter. We have a basis for working on prevention strategies."

Lower noted, "I like to think that one day we will discover a fundamental force law that we can exploit so that S. aureus never forms a bond with the surface of an implanted device."

The two other co-corresponding authors are Pao Lamlertthon of the Fowler lab, and Nadia Casillas-Ituarte of the Lower lab. Other authors include L. Barth Reller of Duke University Medical Center; Ruchirej Yongsunthon, Eric S. Taylor, Alex C. DiBartola, and Brian H. Lower of OSU; Catherine Edmonson and Lauren M. McIntyre of University of Florida, Gainesville; Yok-Ai Que of University of Lausanne in Switzerland; and Robert Ros of Arizona State University in Tempe.

Funding came from grants from the National Institutes of Health and the National Science Foundation, as well as the Swiss National Science Foundation/Swiss Medical Association and SICPA.