skip navigation
Vol. LVII, No. 24
December 2, 2005

previous story

next story
More Microbe Than Human
Relman's Investigations Prove We Are Never Alone

Three years ago, Harvard professor Edward O. Wilson gave an NIH Director's Lecture demonstrating that science is far from knowing how many kinds of microscopic critters are on the premises when a simple log decomposes in the woods. Stanford's Dr. David Relman took that argument an intimate step further Oct. 27 at NIAID's annual Kinyoun Lecture; we don't even know how many kinds of microbes colonize our mouths, guts, skin and other mucosa.

Most of us are not used to thinking of ourselves as "communities," but that is what we are. Humans have 1013 cells, but host 1014 microbes. "In other words, we are 10 parts microbe, and one part human," Relman noted. "We are clearly outnumbered."

  Dr. David Relman

The notion that we are not alone began some 350 years ago, when Leeuwenhoek trained his microscope on such samples as human hair and teeth. He found in dental plaque a variety of tiny motile creatures, noting that the little "animalcules" withered and died on contact with hot coffee.

Subsequent centuries of scientific inquiry discovered the world of bacteria, entrée to which was possible chiefly by cell culturing techniques — literally getting the bugs to multiply on a bed of food. That approach has limits though. What if the microbes in a given environment don't like what you've given them to eat? The problem with taking roll by cell culture is that you may be serving steak to an audience of vegans, and would therefore not know how many were at the table simply by counting cleaned plates.

Today, the tools for discovering what kind of microbial baggage we carry are far more sophisticated and sensitive — measuring ribosomal DNA (rDNA), for example — and the roles our fellow travelers play are only starting to be understood.

Relman and his colleagues want to know the purpose of our internal gardens, and their role in health and disease.

The conventional wisdom is that we begin life "sterile" (that is, culture-negative, although there could be flora we don't yet know about) until the rupture of the amniotic membranes. There follows a long roster of benefits conferred by microbes throughout human development, including vitamin production, food degradation and colonization resistance. Microbial flora promote the terminal differentiation of mucosa, the innate immune defenses and epithelial homeostasis in the gut.

Recently, in what Relman termed "a truly surprising and revolutionary" role for microbes, they appear to regulate fat storage. A recent article in the Proceedings of the National Academy of Sciences showed that "germ-free animals have significantly lower body fat than those raised in a microbial-infected environment," Relman said. "That's a result not previously suspected."

Relman and his colleagues want to know the purpose
of our internal gardens, and their role in health and

Because technology now permits a far broader view of microbes than did cell culture, scientists can investigate the microbial universe in more detail. Although there remain limits to detectability, there seem to be about 80 phyla in nature, based on current estimates. Focusing on bacteria that colonize the mouth, investigators have discovered a wide diversity of microbes that belong primarily to four phyla — the firmicutes, bacteroidetes, fusobacteria and proteobacteria. Yet if you could follow a bite of dinner past the mouth into the esophagus, stomach, small intestines, colon and beyond, you would encounter distinct microbial neighborhoods with very different characteristics. "The biota of the stomach is distinct from the colon, or esophagus, or mouth," Relman said.

In the large intestine, for example, at least seven phyla are represented. In one study, a look at three individuals found close to 500 bacterial phylotypes in the colon and feces, 62 percent of which were novel and 80 percent of which could not be cultivated. When these scientists estimated the completeness of this survey, they were humbled to find that their analysis of nearly 12,000 bacterial sequences had revealed no more than two-thirds of the strains and species that were predicted to be present.

"Individuals vary quite distinctly in the make-up of their flora," Relman said. "To answer the question, 'Who's there?' would require a very large study population."

Beyond the fascination of what's aboard the human raft, Relman and associates want to know the sources of variability in human microbial colonization, and what factors perturb them. Clearly, host genetics, diet and age play roles, but "there are not enough data to know what's truly important, and to what degree."

As if it isn't hard enough to tease out the role of dominant microbial species, the part "rare members" may play is even more of a mystery. Relman suggested there may be "keystone" or "founder" species without which the entire microbial ecosystem collapses.

A range of diseases known to have microbial elements is under review in Relman's laboratory, including Crohn's disease, irritable bowel syndrome, periodontitis, diarrhea, and even premature labor and delivery. All are tied to disturbances in microbial ecology. Thus far, periodontitis, a disease affecting about 40 percent of the adult population in the U.S., is linked in a subset of patients to an abundance of archaeal sequences during stages of severe disease.

To gain a better perspective on this largely hidden frontier, Relman suggested harnessing microarray technology to do more high-throughput analyses of microbial community structure, continue with rDNA screening, establish correlations with host genotypes and develop more sophisticated cultivation technologies.

He called for a "second Human Genome Project" that would survey the genomes of the organisms that populate our bodies. "We need to ask, 'Who's there?', 'What are their genes?', 'What are their functional capabilities? 'How do their patterns of gene expression and associated activities define health?'"

One of his collaborations, with the Institute for Genomic Research in Rockville, involves the "metagenomics" of both the human fecal and oral microbiomes. He concluded, "The diversity and variation of the human indigenous microbial flora is still poorly understood."

Relman's lecture is archived at

back to top of page