Chimpanzee maxilla (upper jaw bone).

Looking a Hominid in the Mouth

Chimpanzee “Oral Microbiome” Shows Surprising Frequency of Species Linked to Human Disease

Oral health may have a surprising impact on overall health. But scientists still don’t fully understand which of the bacteria in our mouths cause disease. Scientists used data-intensive computation on PSC’s Bridges platform to identify the bacteria of the “oral microbiome.” The project sequenced bacterial DNA from the tooth surfaces of chimpanzees and compared it to those of humans and Neanderthals. The scientists discovered that relatively rare bacteria associated with disease in human mouths are common in our close evolutionary relatives.

Why It’s Important

Gum and other dental diseases have been linked to heart disease, pregnancy and birth complications, and pneumonia. We suspect that some bacteria living in human mouths help us stay healthy. Others may threaten our health. Scientists would like to better understand how the collective bacteria in our mouths—the “oral microbiome”—changed as humans diverged from the great apes, and then from ancient human relatives like the Neanderthals. A better understanding of the ecology of the mouth could help identify which bacteria cause disease and which are just bystanders.

“Differences in bacteria in the oral cavity affect health states in different populations … I’ve been in the microbiome field for seven or eight years now, and we’re still trying to explore and understand what species are there, especially in humans. A few bacteria are known to associate with [oral] disease states; we wanted to see whether these types of bacteria are present in our closest evolutionary relatives and ancestors.”—Andrew Ozga, Nova Southeastern University

Arizona State University’s Andrew Ozga, now at Nova Southeastern University, and colleagues at Arizona State, the University of Minnesota and Duke University decided to study the bacterial DNA of the oral microbiome. They collected DNA from the teeth of modern humans, ancient “anatomically modern” humans, Neanderthals, and the population of chimpanzees in Gombe National Park, Tanzania. Their mission required them to determine the sequences—the A, G, T, C alphabet that makes up DNA—of thousands of species of bacteria in each sample. They needed to use extremely large, memory-intensive computation to sort and assemble each bacterium’s genes. So they turned to the Bridges at PSC.

How PSC Helped

The Arizona State-led team collected samples of calculus—the hardened coating on the teeth that results from calcium reacting with dental plaque—from 19 Gombe chimpanzees that had died of various causes over five decades at Gombe National Park in Tanzania. Many of these chimps had been studied by the famous primatologist Jane Goodall. While the calculus doesn’t contain live bacteria, the scientists were able to isolate DNA from dead bacteria from these samples. The team then compared these to samples from 41 ancient and more recent humans, four Neanderthals, and one “historical” chimpanzee from outside the Gombe.

Each sample generated the DNA sequences of millions of genetic fragments. Some of the DNA belonged to the animal being studied. But much of it came from fungi, viruses and, mostly, bacteria. To sort it all out, researchers used a technique called MALT—MEGAN alignment tool. MALT compares hundreds of millions of short DNA sequences to the National Center for Biotechnology Information’s genetic database. It then outputs the best match between each fragment and known genes in the database. MALT works well, and it requires massive computer power. Alignment using MALT has to be performed for each of the over 500 million DNA fragments in the study, comparing hundreds of gigabytes of data, over and over.

The computers available to the Arizona State researchers didn’t have enough memory to run MALT. These computers would have had to read from their hard drives over and over to fetch the data, slowing the calculation enormously. So instead, the collaborators turned to the “large memory (LM) nodes” of Bridges. These computing nodes can handle up to 3,000 gigabytes of data at once. That’s up to hundreds of times more than in a typical personal computer’s memory. Better, as Bridges has 42 LM nodes, the team was able to analyze 20 to 30 samples simultaneously in only a few days.


Image: A “tree” showing how many microbial species in the oral microflora are shared between different host species. The closer the branches, the more closely related are the assortment of microbes in a given sample. Orange indicates Neanderthals, blue anatomically modern humans and red chimpanzees.  Duplicated under Creative Commons from Ozga, AT et al. Oral microbiome diversity in chimpanzees from Gombe National Park. Nature Scientific Reports (2019) 9:17354. See

“We’re creating very large datasets … we need a lot of computing power to compare the samples with all the known databases. Bridges has basically allowed me to do work that I could not have done otherwise.”—Andrew Ozga, Nova Southeastern University

The results of the Bridges DNA assemblies offered some expected results and some surprises. The chimpanzees had different bacteria in their oral microbiome than the humans. The Neanderthal oral microbiome, while distinct from most of the humans’, was closer to ours than to the chimpanzees’. To some extent, the Neanderthal microbiome overlaps with ours. On the other hand, one of the two most common bacteria in the chimpanzee oral microbiome was Porphyromonas. These bacteria associate with oral disease in humans, in whose mouths they’re present, but much less abundant. Porphyromonas may cause oral disease in chimpanzees. Or it may be a component of a healthy chimp microbiome. It isn’t yet clear which. The Arizona and Florida team published their findings in the journal Nature Scientific Reports in November.

The next step in the research will include a deeper dive into how and whether the exact mix of bacteria associates with disease in both humans and chimps, including captive chimpanzee populations. The scientists would also like to study how age affects both dental disease and the oral microbiome in the chimpanzees. Also, they’d like to include more distant human relatives such as gorillas and orangutans. The scientists’ hope is to gain enough insight into how bacterial populations and disease states evolved. Ultimately, the goal is to identify which bacteria are causing disease, which prevent it, and which are only bystanders.

Read the paper