The Body’s Ecosystem

Altogether, the members of the human body’s microbial ecosystem make up anywhere from two to six pounds of a 200-pound adult’s total body weight, according to estimates from the Human Microbiome Project, launched in 2007 by the National Institutes of Health (NIH). The gastrointestinal tract is home to an overwhelming majority of these microbes, and, correspondingly, has attracted the most interest from the research community. But scientists are learning ever more about the microbiomes that inhabit parts of the body outside the gut, and they’re finding that these communities are likely just as important. Strong patterns, along with high diversity and variation across and within individuals, are recurring themes in microbiome research. While surveys of the body’s microbial communities continue, the field is also entering a second stage of inquiry: a quest to understand how the human microbiome promotes health or permits disease.

“None of us in the field—and this is true for the gut, this is true for the skin—none of us can actually tell how our experimental observations really relate to human disease, but we’re getting closer to mechanistic insights,” says immunologist Yasmine Belkaid, chief of mucosal immunology at the National Institute of Allergy and Infectious Disease.

Mouth Microbes

© CAROL DEL ANGEL/IKON/GETTY IMAGESThe late zoologist Charles Atwood Kofoid couldn’t possibly have known that he and his University of California, Berkeley, colleagues had begun to chip away at the human oral microbiome when, in 1929, they described in the Journal of the American Dental Association “animal parasites of the mouth and their relation to dental disease.” Scientists studying periodontal diseases have for decades realized that certain pathogenic bacteria contribute to inflammation and the eventual destruction of tissues within the oral cavity. But it’s now recognized that the mouth is populated with commensal microbes, too, and that these typically benign bacteria can contribute to a person’s health beyond their gums, tongue, and teeth.

Exploring the composition of the mouth microbiome is not without its challenges, however. “The whole world passes through the oral cavity,” says Purnima Kumar, an assistant professor of periodontology at Ohio State University College of Dentistry. “When we collect a sample, we don’t know if it’s just something that’s passing by, or if it’s truly a member of the community,” she says.

Some microbes that dwell in the mouth readily move on from the oral cavity, passing with saliva and food farther through the gastrointestinal tract, for example, or becoming aerosolized and spreading into the lungs. (See “A Lungful of Microbes,” below.) A recent study showed that the placental microbiome more closely resembles that of the mouth than of any other body site, suggesting the oral cavity, by way of the maternal bloodstream, might also supply the organ with commensal microbes. (See “The Maternal Microbiome,” below.)

The mouth may also pass along not-so-friendly bacteria to other body sites. “There’s a lot of evidence linking oral bacteria to distal infections,” says Kumar. To date, oral bacteria have been implicated in cardiovascular disease, pancreatic cancer, colorectal cancer, rheumatoid arthritis, and preterm birth, among other things.

The first step in understanding how mouth microbes affect human health and disease is to determine which species inhabit the human oral cavity. In 2010, microbiologist Floyd Dewhirst from the Forsyth Institute in Cambridge, Massachusetts, and his colleagues published the first comprehensive examination of mouth-dwelling microbes, finding distinct communities on the tongue, on the roof of the mouth, within the biofilms that coat the teeth and gums, and elsewhere in the oral cavity (J Bacteriol, 192:5002-17, 2010). Researchers have identified some 700 microbial species that inhabit the human mouth. “We’re doing really well in terms of who’s there,” Dewhirst says.

Scientists are also starting to get a better understanding of how microbes are organized within the oral cavity. Accumulating evidence suggests that the structure of this microbiome “is not haphazard or random,” says Boston University’s Salomon Amar. “We don’t have the full picture yet, but we understand that there is [a] first layer of microorganisms that allows for the attachment of the second-comers, the third-comers, the fourth, and so on, in a very hierarchical type of organization.”

Such a diversity of species makes for varied cellular interactions. At any one time, “there might be 200 or 300 species interacting with one another and the host,” says William Wade, a professor of oral microbiology at Barts and The London School of Medicine and Dentistry’s Blizard Institute. “Trying to model these interactions is extremely difficult.” By better understanding the dynamics of how these communities promote health or thwart pathogenesis, however, researchers may one day be able to disrupt the oral microbiome in targeted ways to prevent harmful growth.—Tracy Vence

A Lungful of Microbes

© ROMAN YA/SHUTTERSTOCK.COMIf the human digestive tract were a river extending from the mouth through the stomach and intestines, the lungs would be adjacent pools that are fed by the current, according to Gary Huffnagle of the University of Michigan who began studying the bacterial communities that inhabit these organs nearly a decade ago. “There’s a constant flow into [the] lungs of aspirated bacteria from the mouth,” he says. But through the action of cilia, the cough reflex, and other cleansing responses, there’s also an outward flow of microbes, making the lung microbiome a dynamic community.

Like many other body sites now known to harbor commensal bacteria, the disease-free lung was long thought by researchers and clinicians to be largely sterile. Over the last 10 years, however, evidence has been building that the lungs are also populated by a persistent community of microbial residents—albeit a small one. The lung microbiome is about 1,000 times less dense than the oral microbiome, and about 1 million to 1 billion times sparser than the microbial community of the gut, says Huffnagle. That is in part because the lung lacks the microbe-friendly mucosal lining found in the mouth and gastrointestinal tract, instead harboring a thin layer of much-less-inviting surfactant to keep the respiratory organs from drying out, as well as ciliated cells that beat rhythmically to move debris and invading microbes.

In a review article published this March (The Lancet Respiratory Medicine, 2:238-46, 2014), Huffnagle and his colleagues argued that the lungs are like the South Pacific, with small islands of clustered bacteria and wide stretches of unpopulated regions between them. It appears that the lung microbiome is populated from the oral microbiome and the air, and among this population exists a small subset of bacteria that can survive the unique environment of these organs. The most common bacteria found in healthy lungs are Streptococcus, Prevotella, and Veillonella species.

The lung microbiome is about 1,000 times less dense than the oral microbiome, and about 1 million to 1 billion times sparser than the microbial community of the gut.

Recent studies have linked shifts in the lung microbiome to chronic diseases, such as cystic fibrosis (CF) or chronic obstructive pulmonary disease (COPD). In a 2012 study led by epidemiologist John LiPuma of the University of Michigan, the researchers collected specimens from the lungs of CF patients for more than a decade and found that, as the disease progressed, the lung microbiome became less diverse, although overall microbe density stayed the same (PNAS, 10.1073/pnas.1120577109, 2012). They ascribed this shift in the microbiome to the use of antibiotics, which are typically administered to those with CF. “Are antibiotics bad? We’re not saying that at all,” LiPuma says. “The question this paper raises is: Is there a tipping point where antibiotics start to turn against us in CF?”

Leopoldo Segal of the New York University Langone Medical Center who studies small-airway disorders with an eye toward early detection of COPD, has found in a series of studies that inflammation of the lungs is often accompanied by a shift in their bacterial makeup. But the mechanisms behind these changes, and their consequences, are still not well understood. Other studies have uncovered associations between shifts in the lung microbiome and HIV or asthma, but again, causality has been difficult to show.

According to Yvonne Huang of the University of California, San Francisco, Medical Center, characterization of lung microbiomes in relation to health and disease progression is just starting to yield meaningful results. “This field is where studies of gut microbiome were 10 to 15 years ago.” —Rina Shaikh-Lesko

Microbes of the Penis and Vagina

© UIG GETTY IMAGESMicrobiologist David Nelson of Indiana University in Bloomington was investigating Chlamydia infections when he and his colleagues found evidence to suggest that the sexually transmitted pathogens in the male urogenital tract were mingling with other microbes (PLOS ONE, 5:e14116, 2010). Specifically, Nelson learned that the Chlamydia strains of the urogenital tract encode an enzyme that allows them to make tryptophan from an organic compound called indole, which is produced by other bacteria inhabiting the penis. “There was a signature in the chlamydial genome that suggested this organism might be interacting with other microorganisms,” says Nelson. “That’s what initially piqued our interest. And when we went in and started to look, we found that there were a lot more [microbes] than we would have anticipated being there.”

Nelson and his team were among those discovering that the penis harbors its own unique microbiomes, inside and out. Some men pass urine containing a variety of lactobacilli and streptococci species, likely washed from the urethra, whereas others’ urine has more anaerobes, such as Prevotella and Fusobacterium. In terms of overall composition, “we see a lot of parallels to the gut,” says Nelson, noting that there doesn’t seem to be a standout formula for a “healthy” urogenital tract. Commensal microbes within the urethra could make a man more susceptible to infection by supporting colonization by pathogens such as Chlamydia, whereas bacteria that consume the environment’s nutrients could help prevent infection. “We just don’t know at this point,” says Nelson.

On the outside of the penis, circumcision has the largest known influence on the composition of the microbiome. “Men who are uncircumcised have significantly more bacteria on their penis, and the types of bacteria are also very different,” explained Cindy Liu, now a research pathologist at Johns Hopkins Medicine in Baltimore.

In 2010, Lance Price of the Flagstaff, Arizona, office of Translational Genomics Research Institute and his colleagues, including Liu, showed that the base of the penis’s head, or glans, harbored fewer anaerobic bacteria within six months after the men in a study were circumcised (PLOS ONE, 10.1371/journal.pone.0008422, 2010). Last year, the team confirmed its finding in a larger cohort (mBio, 4:e00076-13, 2013). “It really appears that [the penis microbiome] depends on whether you’re circumcised or uncircumcised—different organisms dominate,” says Price.

Some of the anaerobes commonly found on the uncircumcised penis and on occasion inside the male urogenital tract are the same species associated with bacterial vaginosis (BV) in women, says Liu. Deborah Anderson, an OB-GYN and microbiologist at Boston University School of Medicine, and her colleagues have found similar results. “One hypothesis is that the male microbiome might reflect or be related to [his] partner’s microbiome,” says Anderson.

Researchers studying the vagina have for years characterized its microbial community as being dominated by Lactobacillus bacteria, which ferment carbohydrates to lactic acid, yielding a low pH that is toxic to many pathogenic microbes. When levels of Lactobacillus drop, the pH becomes more neutral, and the risk of infections such as BV rises. But with research revealing notable variation among women’s vaginal microbiomes, as well as some interesting dynamics of the microbial communities within a single organ, “that dogma is changing a little bit,” says Gregory Buck of the Vaginal Microbiome Consortium at Virginia Commonwealth University (VCU).

A few years ago, Larry Forney of the University of Idaho, Jacques Ravel of the University of Maryland School of Medicine, and their collaborators published a survey of the vaginal microbiomes of nearly 400 women and found that the majority harbored bacterial communities dominated by one of four Lactobacillus strains (PNAS, 108:4680-87, 2011). More than a quarter of the women studied, however, did not follow this pattern. Instead, their vaginas had fewer lactobacilli and greater numbers of other anaerobic bacteria, although the bacterial communities always included members of genera known to produce lactic acid.

The researchers also found that the composition of a woman’s vaginal microbiome was linked to her ethnicity. Eighty percent of Asian women and nearly 90 percent of white women harbored vaginal microbiomes that were dominated by Lactobacillus, while only about 60 percent of Hispanic and black women did. Vaginal pH varied with ethnicity as well, with Hispanic and black women averaging 5.0 and 4.7, respectively, and Asian and white women averaging 4.4 and 4.2. “There is a racial difference in the vaginal environment and the microbial [community] in parallel,” says Buck.

To complicate matters even further, it is now recognized that the vaginal microbiome is not stable. After menopause, the vagina harbors fewer lactobacilli than during the reproductive stage of women’s lives, with the notable exception of individuals on hormone-replacement therapies. More recent research from Forney, Ravel, and their colleagues has also revealed that the composition of the vaginal microbiome can change in as little as 24 hours. And once again, there are differences among individuals, with some women’s microbiomes appearing to be more stable than others’ (Sci Transl Med, 4:132ra52, 2012).

“In the past we’ve made some generalizations about what kinds of bacteria are found in the vagina, what kinds of bacteria are good or healthy or protective,” says Forney. “What the research is showing is there are tremendous differences between women in terms of the kinds of bacteria that are present and the changes in the communities that occur over time.”—Tracy Vence and Jef Akst

The Maternal Microbiome

© MAN_HALF-TUBE/ISTOCKPHOTO.COMThroughout her training in obstetrics, Kjersti Aagaard was taught that the womb is a sterile sanctuary for baby to develop, and “the only time it’s not is when we have a pathogenic infection,” says Aagaard, who studies the in utero environment of humans and animal models at Baylor College of Medicine and Texas Children’s Hospital. But recent evidence doesn’t seem to support such an idea.

In 2012, Aagaard and her colleagues found that while the vaginal microbiome did change during pregnancy, it didn’t resemble the microbial makeup of newborn babies: the vagina harbored bacterial communities of about 80 percent Lactobacillus, while newborn humans have a relatively greater abundance of other taxa, such as Actinobacteria, Proteobacteria, and Bacteroides (PLOS ONE, 7:e36466, 2012). This suggested that babies aren’t merely painted with vaginal microbes during childbirth, but that bacterial exposure might happen sooner.

A few years earlier, Juan Miguel Rodríguez’s group at the Complutense University of Madrid in Spain had inoculated pregnant mice with labeled bacteria and identified the strain in the meconium (the feces that develop in a fetus) of pups delivered by C-section, similarly implying that an infant’s first meeting with microbes is not at birth (Res Microbiol, 159:187-93, 2008). And in her latest study, Aagaard and colleagues collected placental tissue from 320 mothers immediately after they gave birth and documented a diverse community of microbes that resembled the mother’s oral microbial community more than any other site on the body (Sci Transl Med, 6:237ra65, 2014). “Based on the sum of evidence, it is time to overturn the sterile-womb paradigm and recognize the unborn child is first colonized in the womb,” says Seth Bordenstein of Vanderbilt University.

And then there’s breast milk, which for many decades was also considered sterile, but which is, in fact, a creamy bacterial soup.

The birthing process, then, would be the second stop on a tour of the maternal microbiome. Once on the outside, a baby’s first embrace with his mother is really a group hug with her skin microbiome. And then there’s breast milk, which for many decades was also considered sterile, but which is, in fact, a creamy bacterial soup.

When Rodríguez first began examining breast milk in the 1990s and found evidence that it served as a potential source of microbes in infant feces, many people didn’t believe him. They assumed that his samples were contaminated, “maybe from the mother’s skin or maybe the mouth of baby,” he says, but the bacterial strains he found in breast milk didn’t exist in the mouth or on the skin. And later, his group confirmed that these breast-milk bacteria were finding their way into the infant gut.

In 2011, Mark McGuire of the University of Idaho and his colleagues characterized the microbiome of human breast milk from 16 women and found a diverse community of microbes (PLOS ONE, 6:e21313, 2011). The most abundant bacteria were Streptococcus, Staphylococcus, Serratia, and Corynebacteria, although each woman’s sample was different. “It was very personalized,” says McGuire. “Part of that personalization means she’s sampling her environment and providing that environment to her offspring, and maybe that’s a way to train the immune system and help the infant expand what it’s going to be exposed to early in life.”

In addition to introducing microbes to populate her infant’s gut, a mother’s microbiome during pregnancy and lactation appears to affect her own health. Changes in the gut microbiome during pregnancy correlate with gains in body fat and dips in insulin sensitivity in mice, for example (Cell, 150:470-80, 2012). And several years ago, Rodríguez discovered that the breast-milk microbiomes of women with mastitis, a painful infection of the breast tissue, are characterized by what he calls a “huge dysbiosis”: a single strain of pathogenic bacteria dominating the sample. Providing oral supplements of the missing bacteria helped the women clear the infection. “For the first time we said, ‘Maybe this is important for the treatment of mastitis or painful breast-feeding,’” says Rodríguez, whose team is now wrapping up subsequent trials testing the ability of therapeutic bacteria, rather than antibiotics, to treat mastitis during breastfeeding.—Kerry Grens

Microbial Suit of Armor

© KOWALSKA-ART/ISTOCKPHOTO.COMThe skin is characterized by a multiplicity of habitats, including invaginations, appendages, glands, and follicles. Such environmental heterogeneity not surprisingly breeds diversity at the level of the microbiome. The skin is in constant contact with the outside world, making the bacterial communities that populate the skin some of the most varied of human microbiomes. “Between humidity and hygiene approaches and clothing and everything else, [the environment that skin microbes are exposed to] has infinitely more variation,” says Richard Gallo, chief of dermatology at the University of California, San Diego, School of Medicine.

Nevertheless, the skin is not simply covered with a random suite of bacterial species from the environment. Surveys of the bacterial communities that live in and on the skin of healthy adults have revealed three distinct skin microbiomes, each with fairly consistent patterns of microbial composition. The oily, or sebaceous, glands of the head, neck, and trunk—which secrete a mixture of lipids called sebum—are dominated by Propionibacterium species, including P. acnes, which is associated with blemishes. Moist sites, such as the crease of the elbow, below a woman’s breasts, or between the toes, are frequented by genus Corynebacterium. And the dry surfaces of the body, the uncreased expanses of skin such as the forearm or leg, are home to Staphylococcus species, in particular S. epidermidis.

While causal links between the skin’s commensal microbes and health or disease remain to be demonstrated, the evidence that has accumulated in the past few years paints a suggestive picture. Recent research has begun to document how skin commensals interact with one another, with pathogenic microbes, and with human cells. Staphylococcus epidermidis secretes antimicrobial substances that help fight pathogenic invaders, and P. acnes uses the skin’s lipids to generate short-chain fatty acids that can similarly ward off microbial threats. Meanwhile, these and other skin microbes can impact the local molecular environment, and may be able to alter the behavior of human immune cells.

“The field is exploding in terms of the types of observations that have been made,” says Gallo, “and they’re reaching into every aspect of immunology.”

Recently, molecular microbiologist Gitte Julie Christensen of Aarhus University in Denmark and her colleagues found that the P. acnes strains associated with healthy skin carry genes for thiopeptides, antimicrobial compounds that inhibit the growth of gram-positive species. Strains associated with acne, on the other hand, don’t appear to produce such compounds. In culture, Christensen says, “we can see that these health-associated strains are much better at killing other bacteria than the other strains.”

S. epidermidis itself plays a notable role in host immunity. In 2009, Gallo and colleagues showed that the species secreted lipoteichoic acid, which prevents inflammatory cytokine release from keratinocytes of human skin (Nat Med, 15:1377-82, 2009), and in 2012, Yasmine Belkaid of the National Institute of Allergy and Infectious Disease and her colleagues demonstrated that the addition of S. epidermidis to the skin of germ-free mice altered T-cell function to boost host immunity (Science, 337:1115-19, 2012). “We were able to show that these microbes were sufficient to make the immune system of the skin capable to control infection,” says Belkaid.

Another hint of the skin microbiome’s involvement in immunity came last October, when National Cancer Institute dermatologist Heidi Kong and her colleagues found that immunodeficient patients tended to have more “permissive” skin (Genome Res, doi: 10.1101/gr.159467.113, 2013). That is, people with primary immunodeficiencies harbored more bacterial and fungal communities, including species not normally found on healthy adults. “It’s possible that the focal defects in the immune system allow or permit these otherwise uncommon bacteria and fungi to be present on these patients,” says Kong.

As the importance of the skin microbiome in health and disease is further investigated, researchers are also looking into the possibility of manipulating it. With all our modern changes in lifestyle, diet, hygiene practices, and more, “we have dramatically altered our skin microbiota,” says Belkaid. “I think anything we can do to restore more balance or more appropriate microbe composition in the skin, as in all the different tissues, is extremely important.” —Jef Akst

Portions of this article were first run online as part of a Web series, “Beyond the Gut,” published on www.the-scientist.com between May 12, 2014 and June 18, 2014.