Humans host a profusion of resident microorganisms, and a huge genomics-based effort is revealing how they affect health and disease
● The NIH began planning the Human Microbiome Project (HMP) in 2007, committing $140 million to an initial five-year effort.
● During its first phase, five academic centers were charged with creating a reference catalogue of microbial DNA, recruiting healthy adults for microbial sampling, and performing 16S rRNA gene analyses on bacteria from several anatomic sites.
● The second phase of HMP, which includes demonstration projects, is designed to test the initiative's ability to answer key questions, including how an individual's personal microbial signature relates to health and disease.
● Microbiologists are encouraged to collaborate with HMP by contributing microbial isolates from human anatomic sites for genomic sequencing.
The human genome proved surprisingly skimpy, encoding somewhere in the range of a mere 20,000 proteins, on the same order as the fruit fly genome- a psychologically humbling number. However, humans provide a scaffold upon which microbes build elaborate ecosystems and, by young adulthood, each of us carries many more microbial genes than our own. This endogenous microbial life includes 100 trillion bacteria in the distal gut alone. "Perhaps the genes supplied by our microbes are part of what make us human," says microbiome researcher Peter Turnbaugh from the Center for Genome Sciences at Washington University School of Medicine in St. Louis, Mo. Together, the genomes of our resident microbes, collectively defined as the microbiome, provide traits that human genomes do not carry or were, perhaps, discarded as our species evolved, he suggests.
"The Human Microbiome Project (HMP) is a logical extension of the Human Genome Project (HGP)," Turnbaugh says. Viewed from amicrobiological perspective, the HGP can be considered the "warm-up exercise" for the HMP, an expanding effort to decipher the associate microorganisms that help extend our genetic diversity. Multiple projects are already under way and others are being launched worldwide, including in the United States, Canada, Europe, and Asia, according to officials from National Institutes of Health (NIH) in Bethesda, Md.
"It's the biggest life sciences project of all times," says microbiologist Julian E. Davies of the University of British Columbia in Vancouver, British Columbia, Canada, who chaired the first HMP Research Network Meeting, held this past June in Gaithersburg, Md. "The HMP is a great example of the power of collaboration, and we're all privileged to be involved. But we've got a long way to go before we understand all the good and essential things that our microbes do for us."
In 2007, an NIH project team with staff from 22 NIH Institutes and Centers and coordinated by the National Human Genome Research Institute (NHGRI) unveiled their own HMP plan-a five-year, $140 million effort building on the HGP. In its first phase, the initiative granted four DNA-sequencing centers-Baylor College of Medicine in Houston, Tex., the Broad Institute of the Massachusetts Institute of Technology and Harvard University in Cambridge, Mass., the J. Craig Venter Institute in Rockville, Md., and Washington University-$8.2 million to develop a reference catalogue of microbial DNA and, among other things, characterize core human microbiomes and determine how they are acquired and transmitted. Another set of goals is to learn how stable individual microbiota remain during the lifetime of each human host, how alike family microbiomes are, and whether communities are colonized with similar microbes. (See http://nih roadmap.nih.gov/hmp/)
An Early Focus on Analytic Tool Development, Data Analysis and Access
During the initial, or "jumpstart," phase, the HMP sequencing centers began recruiting healthy adults for microbial sampling as well as sequencing 600 bacterial and several nonbacterial genomes. The centers also determined the16S RNA ribosomal (rRNA) gene sequences from several specific body sites-the gastrointestinal, naso-pharyngeal, and female urogenital tracts, the oral cavity, and the skin-and characterized their bacterial biodiversity for eventual human community profiling. To complete the profiling, 500 new bacterial reference genomes were added to the sequencing list. Combined with current and contemplated efforts, the total HMP reference collection soon will exceed 1,000 microbial genomes, all of which are being rapidly released into publicly accessible databases.
Ethical, Legal, Social Safeguards
The NIHHMP Working Group (WG) reported this past May that the first phase of volunteer recruitment and sampling is proceeding rapidly, all the targeted microbial genomes are now either available or in the sequencing pipeline, and the 16S rRNA work is well under way. "The jumpstart investigators have also developed a common set of sampling and sequencing protocols plus a set of rigorous standards and quality control guidelines to ensure that data from the various centers and different sequencing platforms are comparable and reliable," according to aWG draft report. The centers have been funded for another four years to continue building a genome reference set expanded to include viruses and eukaryotic microorganisms.
The HMP program and the researchers it supports are making their sequencing data and technical information freely accessible. To this end, the HMP has funded the Institute for Genome Sciences (IGS) at the University of Maryland School of Medicine in Baltimore to establish a Data Analysis and Coordination Center (DACC) to track, store, analyze, and distribute the data. In addition, DACC will develop datamining and retrieval tools and serve as the project's international portal. Plans also call for the DACC to hold training workshops on singlegenome and metagenome analysis as well as on the use of its tools and resources.
The Center, directed by Owen White, includes researchers from other institutions, including Lawrence Berkeley National Laboratory and the University of Colorado. Jennifer Wortman and Michelle Giglio from IGS introduced the program to ASM members during the 2009 General Meeting in Philadelphia, describing its comprehensive web resource in detail (http://www.hmpdacc.org). They encourage microbiologists who are not directly involved in HMP to provide feedback to DACC, especially on reference strains, and urge anyone with unusual single-microorganism isolates from a human body site to collaborate with HMP.
HMP Begins Phase Two, Probing the Microbiome's Impact on Health
The second phase of HMP has just begun and includes a diverse set of projects addressing the project's central question-how the microbial complement of an individual relates to his or her state of health. Researchers involved in the first set of 15 specific projects will study bacterial, fungal, and viral changes in microbiomes from individuals with gastrointestinal, skin, nasal tract, oral cavity, genital, urinary tract, and blood disorders.
"The real gains of the HMP begin as the initiative shifts its focus to biology and starts applying the technologies developed during jumpstart," says microbiologist B. Brett Finlay, from the University of British Columbia in Vancouver, Canada, who calls the HMP "extremely important" and notes that he and many colleagues plan to use its findings in their own research. "However, I don't think anyone realized it was going to be so complex and different between sites and individuals."
The Vast Majority of Endogenous Microbes Inhabit the GI Tract
HMP research is only at its beginning stage, but much is already known about the microbiota associated with the gastrointestinal (GI) tract and the skin. Early genomic analyses of the microbes from these sites are already providing provocative insights.
Groundwork for the second phase of HMP comes from a series of genomic analyses of microbes from the human GI tract. For example, NIH-funded California scientists recently challenged the long-held notion that the human immune system is set to destroy any microbial pathogen that comes its way, the "take-noprisoner" view of immunity. Instead, the immune system interacts with the vast and complex microbial mix within the GI tract and, more often than not, the host and the GI microbes peacefully coexist, according to June L. Round and Sarkis K. Mazmanian from the California Institute of Technology in Pasadena. "Recent evidence suggests that a beneficial partnership has coevolved between symbiotic bacteria and the immune system," they note.
More generally, GI-tract bacteria influence development of adaptive immunity, while the immune system helps to shape the GI microbiota. Moreover, gut microbes exert both pro- and antiinflammatory activity, and a well-balanced microbiome is necessary for proper immune function. Altering that microbial balance can deregulate the immune system, and lead to chronic inflammatory disorders such as Crohn's disease and ulcerative colitis. Moreover, because the GI tract is a primary meeting place for microorganisms and immune cells, what happens there could affect other immune functions such as allergy and maybe even the response to some cancers.
The question of what happens when GI microbes are decimated by diarrhea or disease is another issue for HMP to explore. One possibility is that a backup GI microbiota is kept intact within the appendix, which might be a "safe house for beneficial bacteria," says William Parker at Duke University Medical Center in Durham, N.C. Not only is the appendix separated from the fecal stream and difficult to breach, it is well configured to harbor bacteria and to support biofilms. "We and our bacteria have evolved together," he stresses. "Humans are the ones messing with the system-in the brief time since the Stone Age, we humans have profoundly changed our diet, exercise habits, and hygiene, leaving the poor microbes to cope with all the new things we're throwing at them. Yet, we continue to point fingers at bacteria when something goes amiss. It's an outdated human- centric mindset that the HMP will help change."
GI Microbiota Might Play a Role in Obesity
Two groups of beneficial bacteria, the Bacteroides and Firmicutes, dominate the human GI and early environmental exposures contribute greatly to the content and structure of gut microbial communities in human adults, according to Turnbaugh, Jeffrey I. Gordon, and their colleagues at Washington University School of Medicine. "Yet despite a remarkable variety in GI species and composition from one individual to another, metagenomic methods are helping to identify a core gut microbiome."
"Furthermore, in humans there is a relationship between the structure of the gut microbiota, diet, and obesity," Gordon adds. "The relative proportions of key bacterial phyla, including the Bacteroidetes, were altered in obese compared to lean individuals and certain sets of genes in their gut microbiomes were significantly overor underrepresented."
Gut microbial communities can be transplanted from obese mice, which typically carry decreased proportions of Bacteroidetes but increased Firmicutes and Actinobacteria, into germfree lean animals. The recipients develop an "increased fat phenotype," Gordon says. "Such findings suggest that obesity may have a microbial component, and this could have potential therapeutic implications." Notably, figuring out ways to counter obesity is no mean accomplishment considering its estimated aggregate cost of "$60 billion a year in the U.S. alone, with a large proportion of this attributable to obesityassociated type 2 diabetes," according to Jeffrey M. Friedman of Rockefeller University in New York, N.Y.
If the GI microbiota influence the nutritional and energetic availability of foods, perhaps affecting how individuals regulate weight, understanding how those bacteria behave toward one another could prove important for designing effective anti-obesity therapies. Integrating genome sequencing with in vivo studies, Gordon's students Michael A. Mahowald and Federico Rey and their teammates showed a complex and nuanced relationship between two human gut bacterial species, Bacteroides thetaiotaomicron and Eubacterium rectale, representing the Bacteroidetes and Firmicutes, respectively. Both bacterial species colonize the gut to similar levels when carbohydrates are readily available. However, when host food supplies shift, B. thetaiotaomicron signals the host to produce mucosal glycans that it but not E. rectale can access. In response, E. rectale lowers the levels of its glycan- degrading enzymes and increases expression of selected amino acids and sugar transporters to help it harvest the nutrients supplied by B. thetaiotaomicron.
This example is one among many of frequent and complicated microbial readjustments within the gut ecosystem, according to Gordon. However, he stresses that it simply illustrates basic principles of nutrient interchange and metabolic reciprocity. Medical microbiology, it appears, is finally catching up with microbial ecology.
Meanwhile, Claire Fraser-Liggett and Alan R. Shuldiner from the University of Maryland Medical Center in Baltimore are investigating the GI microbiota of individuals from a nearby Old Order Amish community, whose members are more genetically homogenous than the general U.S. population. This HMP-funded demonstration study will help to test the "thrifty microbiome hypothesis" which, in simple terms, claims that a microbiota biased toward weight build-up evolved to help the host survive nutrient- scarce periods. However, the results of these researchers' analyses of microbial genomes from overweight and thin human donors is unlikely to quell the controversy over whether an imbalance among Bacteroides, Firmicutes, and other gut microbes triggers obesity-amatter that continues to spur skepticism from some experts, including Friedman from Rockefeller University.
"Beliefs are just beliefs, however expert," says Deinococcus specialist Michael J. Daly from the Uniformed Services University of the Health Sciences in Bethesda, Md. "What we need is proof, and now that we have the tools to get that proof, deciphering the human microbiome is absolutely essential."
Daly is convinced that bacteria affect host health in both good and bad ways. "Deinococcus, for example, which have been found in human stomachs, synthesize and accumulate potent manganese antioxidants that are eventually released into the environment," he says. "Other radiationresistant bacteria such as lactobacilli also accumulate very high manganese concentrations, which could explain their gastrointestinal benefits. These bacteria are little antioxidant factories." The research being done by HMP scientists, he adds, "will give us a better idea which bacteria should be nurtured and which avoided."
The Complex Landscape of the Skin Also Supports a Wide Variety of Microbes
"The skin microbiome is vastly different from, but just as complex as, the gut," says Julia A. Segre, a senior investigator at the Genetics and Molecular Biology Branch of the National Human Genome Research Institute in Bethesda, Md. "The hairy, moist underarms are but a short distance from the smooth, dry forearms, and the two niches are as ecologically dissimilar as rainforests are to deserts." Whereas culture-based testing reveals limited numbers of microbes, such as Staphylococcus species that readily grow under standard conditions, molecular technologies uncover a much greater diversity of skin microbiota within and between distinct topographical sites. "Culture-independent molecular approaches give us a way to systematically survey multiple skin sites; especially important are the areas characteristically affected by dermatologic disorders," she says.
The Human Microbiome Project: Centers and Investigators
Segre, Elizabeth Grice, and their collaborators are analyzing human skin microbiota. They began a pilot project by swabbing, scraping, and doing punch biopsies on five healthy human volunteers and analyzing 5,373 16S ribosomal RNA gene sequences, which were used to infer phylogenetic relationships among sampled bacteria. The 16S sequences derived mainly from Proteobacteria, predominantly Pseudomonas and Janthinobacterium. Both these pseudomonads are not typically considered skin microbes and, when clinically isolated, have historically been labeled as secondary invaders, particularly of wounds. Moreover, fewer than 5% of the microorganisms identified in this study were Staphylococcus epidermidis and Propionibacterium acnes, the bacteria usually thought to predominate on skin. Indeed, of the 113 kinds of resident dermal bacteria, almost 60% were pseudomonads. Especially surprising was the large number of microbes living under the skin, according to Segre.
"Dermatologic disorders commonly affect specific skin sites, and characterizing the microbiota that inhabit these niches will provide insight into the delicate balance between skin health and disease," Segre says. In a follow-up study, samples of 20 distinct skin sites from 10 new and 5 resampled volunteers were analyzed and 112,283 near-full-length bacterial DNA sequences generated. One finding was that physiologically comparable sites harbor similar bacterial communities. "In all, 19 bacterial phyla were detected in the disease-prone sites; however, 4 phyla dominated: Actinobacteria (51.8%), Firmicutes (24.4%), Proteobacteria (16.5%), and Bacteroides (6.3%)," note Segre and her collaborators.
The forearm hosts the most diverse microbial communities, while the least diverse are found behind the ear (44 versus 15 bacterial species, respectively). Eczema preferentially affects highly diverse and richly inhabited skin sites typified by the inner bend of the elbow; in contrast, common characteristics between bacterial communities at psoriasis-associated locations such as the outer elbows, knees, and navel could not be identified. Four of the five re-sampled volunteers were significantly more like themselves over time than they were like other volunteers, according to Segre. "However, it was the skin sites themselves that seemed to determine both intra- and interpersonal variation," she says.
"The fact that the locations of bacterial species are relatively consistent from person to person suggests that they are beneficial to the host," comments microbiologist Martin J. Blaser, at New York University Langone Medical Center, whose own studies have identified 183 different kinds of bacteria on the human arm. He and his colleagues are investigating psoriatic lesions in an effort to determine if they have a distinctive microbial colonization pattern. "The overall ecological diversity of the microbial population in and around the lesions appears greater than that of normal skin, a finding that suggests a substantial ecological disturbance in the diseased tissue," he says. Moreover, there are significant differences in the bacterial distribution on psoriatic tissue, with Actinobacteria and Proteobacteria underrepresented and Firmicutes overrepresented.
Knowing more about the specific microbial components of psoriatic lesions should enable development of therapies aimed at re-establishing normal populations of microbes. "But doing so requires much more information about the skin's usual bacterial constituents," Blaser says. Th
us, his HMP demonstration project "Evaluation of the Cutaneous Microbiome in Psoriasis," is intended to characterize the resident microbes of 75 subjects with and without psoriasis. The effect of commonly used immunosuppressive treatment on the microbiomes of patients with psoriasis will also be investigated.
Bollinger, R. R., A. S. Barbas, E. L. Bush, S. S. Lin, and W. Parker. 2009. Biofilms in the large bowel suggest an apparent function of the human veriform appendix. Gut 56:1481-1482.
Gao, Z., C. Tseng, B. E. Strober, P. Zhiheng, and M. J. Blaser. 2008. Substantial alterations of the cutaneous bacterial biota in psoriatic lesions. PLoS ONE 3:e2719.
Grice, E. A., H. H. Kong, S. Contan, C. B Deming, J. Davis, A. C. Young, NISC Comparative Sequencing Program, G. G. Bouffard, R. W. Blakesley, P. R. Murray, E. D. Green, M. L. Turner, and J. A. Segre. 2009. Topographical and temporal diversity of the human skin microbiome. Science 324:1190-1192.
Grice, E. A., H. H. Kong, G. Renaud, A. C. Young, NISC Comparative Sequencing Program, G. G. Bouffard, R. W. Blakesley, T. G. Wolfsberg, M. L. Turner, and J. A. Segre. 2008. A diversity profile of the human skin microbiota. Genome Res. 18:1043-1050.
Ley, R., P. J. Turnbaugh, S. Klein, and J. I. Gordon. 2006. Human gut microbes associated with obesity. Nature 444:1023.
McGuire, A. L., J. Colgrove, S. N. Whitney, C. M. Diaz, D. Bustillos, and J. Versalovic. 2008. Ethical, legal, and social considerations in conducting the Human Microbiome Project. Genome Res. 18:1861-1864.
Round, J. L., and S. K. Mazmanian. 2009. The gut microbiota shapes intestinal and immune responses during health and disease. Nature Rev. Immunol. 9:313-323.
Turnbaugh, P. J., R. E. Ley, M. Hamady, C. M. Fraser-Liggett, R. Knight, and J. I. Gordon. 2007. The Human Microbiome Project. Nature 449:804-810.