The supercomputing power of UT Austin Texas Advanced Computing Center’s Stampede and Lonestar systems was deployed in service of micro biome research of gum disease, diabetes, and Crohn’s disease for a study published in April 2014 in the American Society for Microbiology (ASM)’s journal mBio.
The study, led by Dr. Marvin Whiteley, professor of molecular biosciences and director of the Center for Infectious Disease at The University of Texas at Austin, employed Stampede and Lonestar to compare gene expression of 160,000 genes in healthy and diseased periodontal communities, and paves the way for biomarkers to predict illness from wide-ranging diseases like periodontitis, diabetes, and Crohn’s.
Entitled “Metatranscriptomics of the Human Oral Microbiome during Health and Disease” (doi: 10.1128/mBio.01012-14 1 April 2014 mBio vol. 5 no. 2 e01012-14), the paper is coauthored by Peter Jorth, Keith H. Turner, Nejat Nizama, and Marvin Whiteley of the Department of Molecular Biosciences, Institute for Cellular and Molecular Biology, Center for Infectious Disease, University of Texas, Austin; with Pinar Gumus and Nurcan Buduneli of the Department of Periodontology, School of Dentistry, Ege University, Izmir, Turkey; and edited by Roberto Kolter of Harvard Medical School.
The researchers note that the human microbiome plays important roles in health, but when disrupted, these same indigenous microbes can cause disease. And while the composition of the microbiome changes during transition from health to disease, these changes are often not conserved among patients. Since microbiome-associated diseases like periodontitis cause similar patient symptoms despite interpatient variability in microbial community composition, the coauthors hypothesize that human-associated microbial communities undergo conserved changes in metabolism during disease.
Using patient-matched healthy and diseased samples to compare gene expression of 160,000 genes in healthy and diseased periodontal communities, the scientists show that health-and disease-associated communities exhibit defined differences in metabolism that are conserved between patients. In contrast, they say metabolic gene expression of individual species was highly variable between patients, and that these results demonstrate that despite high interpatient variability in microbial composition, disease-associated communities display conserved metabolic profiles that are generally accomplished by a patient-specific cohort of microbes.
The coauthors note further that the human microbiome project has shown that shifts in human hosted microbiota are associated with many diseases, including obesity, Crohn’s disease, diabetes, and periodontitis. However, they observe that while changes in microbial populations are apparent during the curse of these diseases, species associated with particular diseases can vary from patient to patient.
The researchers observe that among the best-characterized human-associated microbial communities are extremely diverse gut and oral microbiota and it is now appreciated that our indigenous microbiota are tightly linked to the state of a person’s health. Several human diseases, including diabetes, Crohns disease, and periodontitis, are linked to disruptions in the gut and oral microbial populations, and accordingly, microbiota-associated diseases such as periodontitis are increasingly examined through an ecological lens.
Researchers at The Whiteley Lab found that bacteria inside the mouth drastically alter their behavior in diseased individuals, according to data processed by the TACC supercomputers, and say these surprising findings could potentially lead to better ways to prevent, treat, or even reverse development of these diseases.
With periodontal disease, a huge microbial shift is associated with massive reorganization of the microbiota residing in the subgingival crevice, the region between the tooth surface and the gingival epithelium (gum). While marked changes in microbial population structure are observed during periodontitis, the actual community members can differ greatly from person to person. In fact, both healthy and disease-associated oral microbial communities vary significantly among people, among locations in the mouth, and even on a daily basis at the same site within the mouth.
“Stampede allows us to use 6,400 desktop computers, all at the same time. There are a lot of problems in biology that can benefit from the supercomputing approach,” comments study coauthor Keith Turner, a postdoctoral fellow with the Whiteley Lab, in a TACC release by Jorge Salazar.
“What we we’re trying to figure out,” says Dr. Whiteley, “is how do these bacteria act when you’re healthy, and how do they act when they’re in a diseased state. The really big finding is that they do act very differently.”
Bacteria share nutrients, and one species will even feed on another as they constantly interact. “The thing that we found in this paper,” Dr. Whiteley notes, “is that this sharing, and how they interact with each other changes quite drastically in disease than it does in health.”
The UT Austin researchers used “shotgun metagenomic sequencing,” described as a non-targeted method of studying all the genetic material of the bacterial communities. Dr. Whiteley and his colleagues analyzed the RNA collected with the Lonestar and Stampede supercomputers at TACC, having been awarded computing allocations through the University of Texas System Research Cyberinfrastructure initiative. The research was funded by grants from the National Institutes of Health, administered by the National Institute of Dental and Craniofacial Research.
Most people are probably unaware that according to a five-year, $115 million research effort that began in 2008 by the National Institutes of Health (NIH) called the Human Microbiome Project, microbes — mainly bacteria — outnumber human cells in the human body by a ratio of roughly 10 to 1, and that scientists have identified 10,000 different species of bacteria that live inside each individuals. These microbial communities are collectively known as the human microbiome.
“The easiest way to think of it is just the collection of bacteria that are in or on your body,” Dr. Whiteley explains. “We think of it as not only the bacteria, but the genetic composition. What’s their DNA? And from that we can infer what these bacteria might be doing for us.”
Dr. Whiteley’s lab started by isolating RNA from oral plaque samples collected. Dr. Turner explains in the UTCC release, noting that “RNA, for those who know about computers, is kind of like the RAM [random access memory] — the working memory of the cell.” The RNA sample acts like a memory image or ‘core dump’ to reveal the processes of the as-yet unknown bacterium it came from. And unfortunately, Dr. Turner says, you can’t get a full picture of the activity because there are so many molecules in the sample.
“But what you do,” he explains, “is get what you can and profile it by sequencing, using some recent technological advances. Then it’s essentially a search problem.”
Dr. Turner searched a metagenomic database, essentially a vast genetic clearing house sampled from the environment instead of lab-grown, looking for matches at the NIH’s Human Microbiome Project. A match told him what bacterium a gene came from in the sample, and he tallied each match. “The more it’s thinking about a certain process, the more it seems to be important to it,” he says. “The shotgun approach, as you might imagine, is very computationally intensive, which is why we turned to TACC for some of these problems.”
So how big were these problems? The release notes that Dr. Turner and colleagues chose 60 different species of bacteria to represent the total community. More than 160,000 genes were analyzed, yielding 28 to 85 million reads of RNA snippets, including about 17 million mRNA reads for each sample.
“What we were trying to figure out is how do these bacteria act when you’re healthy, and how do they act when they’re in a diseased state. The really big finding is that they do act very differently,” says Dr. Whiteley, whose main findings show that bacteria act differently when the human host is healthy compared to when diseased. “The main thing that they change when they go from health to disease is that they change their metabolism,” Dr. Whiteley observes. In other words, a species of bacteria that ate one thing, fructose for example, can switch to a different kind of sugar to feed on if diseased.
“The kind of thing that might have taken a desktop computer a week, two weeks to run we can run at TACC in just a couple of hours,” Dr. Turner says. “Stampede allows us to use 6,400 desktop computers, all at the same time. There are a lot of problems in biology that can benefit from the supercomputing approach.”
Dr. Whiteley is particularly focused on periodontitis because it’s one of the most prevalent diseases globally. “It’s an interesting disease, because the same bacteria that are in your mouth when you’re healthy are the same ones, more or less when you’re sick,” he notes.
“What our study says is that it doesn’t really matter what bacteria you have, because the communities are acting very similarly. So a healthy community has this metabolism, no matter what the members are. And a diseased community has a very different metabolism, no matter what the members are. It’s this conservation of a metabolic community. ”
Dr. Whiteley draws an analogy between what goes on under our gums to an ecosystem in the African Savannah — interactions among the respective ‘animals’ being key. “You have lions, and you have leopards, and wildebeest, and all of these animals that are there. If you look at it as a whole community, it kind of makes sense. But if you were to only take a one-acre plot out of the African Savannah and look at it, it may not make sense because there may not be a lion in that one acre. So trying to understand interactions, you need to take a much larger, bigger context. And that’s what this study did,” He explains.
According to science results from the Human Microbiome Project, a shift to more harmful bacteria in the community is linked to wide-ranging diseases such as periodontitis, diabetes, and Crohn’s disease, and Dr. Whiteley says his research can help people by helping develop biomarkers that predict if a person is going to get sick. The question is: “Can you actually come up with a very quick way to assess the behavior of the community quickly and say, are you on the progression of moving from health to disease, and then provide some sort of preventative measure when you get there?”
Could pathogenic bacterial communities that rewired themselves to be harmful also be rewired for health? It’s possible in theory, anyway, Dr. Whiteley maintains. “You can manipulate bacterial populations numerically very easily. You feed them something else. So you might be able to shift them back. These are some of the ideas that we’ve been thinking about in our lab that might be more pervasive as we move forward. Medicine is going to change a lot in the next 10 to 50 years. We’re going to be thinking about these sort of questions a lot more, questions like what is your microbiome actually doing, and is that impacting why you’re in the doctor’s office,” he concludes.
This story follows a recent report covered on IBD News Today that oral bacteria might aggravate IBD, according to a new study.
Founded in June 2001, TACC has expanded from a staff of only a dozen and one mid-level Cray supercomputer to more than 110 staff and students who operate several of the most powerful supercomputers and visualization systems in the world, with the network and data storage infrastructure to support them. TACC has become one of the leading academic advanced computing centers in the US, deploying and operating since its inception 10 National Science Foundation (NSF)-funded supercomputers and advanced visualization systems for national programs.
Texas Advanced Computing Center
University of Texas at Austin
Texas Advanced Computing Center
University of Texas at Austin
Dr. Keith Turner