Microbiome Research: Bacteria in Human Health Explained

A 70-kilogram adult carries roughly 38 trillion bacterial cells, a number close enough to the body's own 30 trillion human cells that the old "ten bacteria for every human cell" claim had to be retired (Sender, Fuchs & Milo, 2016, PLOS Biology). Most of those microbes are not invaders. They digest fiber humans cannot break down, manufacture vitamins, and train the immune system from birth. This piece explains what the microbiome actually is, how researchers learned to study it without growing it in a dish, and why disrupting it is now linked to conditions from Clostridioides difficile infection to depression.

Educational content, not medical advice. Talk to your doctor before changing any medication, treatment, or routine. Information here reflects current research but every person's situation differs.

Key Takeaways

• The body carries roughly as many bacterial cells as human cells — about 38 trillion, not the old "10-to-1" figure (Sender et al., 2016)
• Microbiome composition is linked with metabolism, immune regulation, and — through the gut-brain axis — mood and behavior
• Dysbiosis, an imbalanced microbial community, is associated with obesity, inflammatory bowel disease, and neuropsychiatric conditions, though association is not the same as causation
• DNA sequencing let researchers catalog bacteria that can't be grown in a lab, which is what made large-scale microbiome research possible
• Two FDA-approved fecal microbiota products (Rebyota, 2022; Vowst, 2023) now treat recurrent C. difficile infection — the field's first mainstream therapeutic wins

Why bacteria were "the enemy" for a century

Germ theory, formalized by Louis Pasteur and Robert Koch in the late 1800s, gave medicine an enormously useful frame: identify the microbe, kill the microbe, cure the disease. Antibiotics extended that logic into a golden age of infection control starting with penicillin's wartime rollout in the 1940s. The working assumption was simple. Bacteria in the body were either dangerous or irrelevant.

That assumption started cracking earlier than most people realize. Researchers raising "germ-free" animals — born and kept in sterile isolators with no microbial exposure at all — noticed something odd. These animals developed abnormal gut linings, underdeveloped immune tissue, and skewed behavior compared to normally colonized littermates. Something about ordinary bacterial exposure was necessary, not just tolerated.

Clinical medicine kept bumping into the same contradiction. Patients on broad-spectrum antibiotics who cleared their original infection often came down with something new: diarrhea, yeast overgrowth, or a dangerous C. difficile infection that hadn't been there before. Wiping out bacteria wholesale, it turned out, could open the door to worse problems. The logical conclusion, that the resident microbes had been doing protective work all along, took decades to become mainstream.

The oldest hint that gut bacteria matter for host development predates DNA sequencing by decades: germ-free rodents raised without any microbial exposure develop visibly underdeveloped gut-associated lymphoid tissue, a finding first documented in gnotobiotic research programs of the 1960s and still replicated today.

Why couldn't scientists just look at the bacteria under a microscope?

They could see them, but not identify most of them. Roughly 80% of gut bacterial species resist standard lab cultivation because they need oxygen-free conditions, specific nutrients, or cross-feeding relationships with other microbes that a petri dish can't replicate. Species biologists never managed to grow simply didn't show up in catalogs of "what lives in the gut."

How researchers learned to read a microbial community without growing it

The breakthrough was genetic, not microbiological. In the 1970s and 1980s, Carl Woese and colleagues showed that a single gene, the one coding for 16S ribosomal RNA, varies enough between bacterial species to work as a universal identification tag, while staying similar enough within a species to be recognizable. Every bacterium has it. Sequence that one gene from a stool or skin sample, compare it against a reference database, and you get a census of who's there, no culturing required.

This is the microbiome research equivalent of a wildlife census done by DNA left in footprints instead of counting animals by eye. You don't have to trap and identify each creature; you read the genetic trace it leaves behind and match it to a known catalog. Just as camera traps revealed how much wildlife biologists had been missing by relying only on direct sightings, 16S sequencing revealed how much of the human microbial world direct culturing had been missing.

Early culture-independent surveys, notably a 2005 study using this approach on intestinal samples, found hundreds of bacterial species per person, most never previously described in the medical literature (Eckburg et al., 2005, Science). The field needed a bigger, systematic effort to make sense of that diversity across a real population, which is exactly what came next.

What was the Human Microbiome Project?

Launched by the U.S. National Institutes of Health in 2007 as a 10-year, roughly $215 million initiative, the Human Microbiome Project sequenced microbial DNA from over 300 healthy adults across multiple body sites and produced around 3,000 reference bacterial genomes, creating the baseline catalog that later disease-comparison studies rely on (NIH Common Fund).

By the early 2010s, a second wave of technology, shotgun metagenomics, pushed further. Instead of sequencing one identifying gene, metagenomics sequences all the DNA in a sample at once, which means researchers can see not just which species are present but which genes and metabolic pathways those species carry. That distinction matters: two gut communities can have similar species lists but wildly different functional capacities, and functional capacity is what actually touches host physiology.

What the microbiome does for the body it lives in

Once researchers could reliably catalog gut bacteria, the next question was what those bacteria were actually doing. The answer turned out to be more than digestion.

Function	Mechanism	Example
Nutrient production	Fermentation of dietary fiber the human gut cannot digest alone	Bacteria synthesize vitamin K and several B vitamins
Energy for gut cells	Short-chain fatty acid (SCFA) production, especially butyrate	Butyrate is the primary fuel source for colon lining cells (colonocytes)
Immune training	Continuous low-level signaling shapes regulatory immune cells	Germ-free mice show underdeveloped gut-associated lymphoid tissue
Pathogen resistance	Competitive exclusion — beneficial bacteria occupy niches and nutrients	Reduced colonization space for opportunistic species like C. difficile
Gut-brain signaling	Bacterial metabolites and vagus nerve activation	SCFAs and bile acids activate vagal neurons that project to the brainstem

Table 1: Core functions of a healthy gut microbiome and the mechanisms behind them.

Butyrate deserves particular attention because it's one of the best-characterized links between a specific microbe and a specific health outcome. Faecalibacterium prausnitzii, one of the most abundant butyrate producers in a healthy colon, supplies energy directly to colonocytes and produces anti-inflammatory metabolites. In both ulcerative colitis and Crohn's disease, this species and other butyrate producers are consistently depleted, and fecal butyrate concentrations drop along with them (Frontiers in Immunology, 2019). That's an association, not proof that low F. prausnitzii causes IBD, but the mechanistic plausibility (less butyrate, less epithelial fuel, less anti-inflammatory signaling) is strong enough that it now guides drug development.

What is dysbiosis, exactly?

Dysbiosis describes a gut microbial community that has shifted away from the range typically seen in healthy individuals — usually some combination of reduced species diversity, loss of beneficial taxa such as butyrate producers, and overgrowth of opportunistic species. It's a pattern, not a single measurable threshold, and researchers usually define it relative to a matched healthy comparison group in a given study.

The gut-brain axis: how bacteria reach the brain without leaving the gut

Of all the microbiome's effects, the connection to mood and behavior draws the most public attention and the most scientific caution. The mechanism runs through the vagus nerve, the long cranial nerve that carries signals between the gut and the brainstem. Germ-free mice, raised with no gut bacteria at all, show measurably lower baseline vagal nerve activity than normally colonized mice, and that activity returns to typical levels once bacteria are reintroduced (Neuron, 2019).

Bacterial metabolites appear to be the messengers. Short-chain fatty acids and bile acids produced by gut bacteria directly activate vagal sensory neurons, giving the microbiome a real-time channel to signal the brain about the state of the gut. Separately, germ-free mice show an exaggerated hormonal stress response through the hypothalamic-pituitary-adrenal (HPA) axis, along with altered anxiety-like behavior compared to conventionally raised mice (International Journal of Neuropsychopharmacology).

Think of the vagus nerve as a phone line the gut and brain have always had open, and the microbiome as one of the callers who gets to use it. Remove that caller entirely, as germ-free mouse studies do, and the line's baseline activity changes measurably. That doesn't mean any single probiotic strain can reliably "call the brain" and change a person's mood on command. Mouse studies establish a plausible mechanism; extrapolating a specific behavioral outcome to humans requires human trials, which are far fewer and much harder to control.

How does the gut microbiome affect the brain?

Through three main channels: bacterial metabolites (short-chain fatty acids, bile acids) that directly activate vagal nerve endings, immune signaling molecules that cross into circulation, and indirect effects on neurotransmitter precursor availability. Dysbiosis is associated with depression, anxiety, and autism spectrum disorder in observational studies, but causation in humans remains unproven.

The single most counterintuitive fact in microbiome research: germ-free mice aren't healthier without bacteria, they're developmentally behind. No microbes means underdeveloped gut immune tissue, altered stress hormones, and measurably different brain chemistry. The absence of bacteria is itself a pathology.

From lab finding to approved treatment: what's actually working in clinic

Most microbiome science remains observational, but one application has cleared the highest bar in U.S. medicine: full FDA approval as a drug.

Fecal microbiota transplantation (FMT), transferring processed stool from a screened healthy donor into a patient with a depleted microbiome, treats recurrent Clostridioides difficile infection, an antibiotic-resistant gut infection that kills tens of thousands of people annually when standard antibiotics fail to prevent relapse. The FDA approved Rebyota (fecal microbiota, live-jslm) in November 2022 and Vowst (fecal microbiota spores, live-brpk), an oral capsule version, in April 2023 — the first two microbiome-based biologic drugs ever approved in the United States (FDA, Fecal Microbiota Products; American Gastroenterological Association). Clinical trials supporting these approvals showed FMT substantially reduces recurrence rates compared to standard antibiotic-only management, which is why it's now recommended after a second recurrence in current treatment guidelines.

Beyond C. difficile, the picture gets more mixed:

• Obesity and metabolism. A landmark 2006 study found that transferring gut microbiota from obese mice into germ-free lean mice increased the recipients' capacity to harvest energy from food, gaining more fat than mice that received lean-donor microbiota on an identical diet (Turnbaugh et al., 2006, Nature). This showed causation in mice; equivalent causal evidence in humans is still limited.
• Inflammatory bowel disease. FMT trials for ulcerative colitis show modest, inconsistent remission rates, well below the C. difficile results, and it is not FDA-approved for this use.
• Probiotics and prebiotics. Evidence is strongest for specific strains in specific contexts, such as reducing antibiotic-associated diarrhea, and considerably weaker for general "gut health" claims made on yogurt labels and supplement bottles.
• Neuropsychiatric conditions. No microbiome-targeted treatment for depression, anxiety, or autism has cleared regulatory approval; research here is earlier-stage than the metabolic and infectious-disease work.

What people get wrong about the microbiome

The biggest misconception is treating "microbiome" as a single dial that can be turned toward health with the right supplement. In reality, an individual's gut community is shaped by genetics, birth mode, early diet, geography, and years of dietary history, and it settles into a relatively personalized, stable configuration by adulthood. A probiotic capsule introducing a handful of strains rarely restocks or restructures an entire ecosystem that took years to establish, which is one reason probiotic effects in trials are often modest and strain-specific rather than universal.

A second common error is reading correlation as proof. Studies routinely find that a certain bacterial signature is more common in people with a given condition, obesity, depression, IBD, and so on. That's a real, useful finding for generating hypotheses. It is not evidence that changing the bacteria changes the disease, because the causal arrow could run the other way (the disease state changes the gut environment, which then changes the bacteria) or a third factor (diet, medication, inflammation) could be driving both. The Turnbaugh mouse-transfer study is compelling precisely because it demonstrates directional causation, which most human microbiome studies cannot.

A third mistake is assuming "diverse" always means "healthy." Diversity correlates with health in many contexts, but a community can be diverse and still dysfunctional, or less diverse and perfectly healthy for a given individual's baseline. Diversity metrics are a useful signal, not a verdict.

Where the field is headed

Several directions are moving from early research toward plausible clinical tools over the next five to ten years.

Live biotherapeutic products beyond C. difficile. Rebyota and Vowst prove the regulatory pathway works. Several companies are now running trials of defined bacterial consortia (rather than whole, unprocessed stool) for IBD and other gastrointestinal conditions, aiming for more standardized, reproducible products than first-generation FMT.

Precision dosing based on baseline microbiome. Because response to fiber, probiotics, and even some drugs varies with an individual's existing microbial community, researchers are exploring whether a baseline microbiome profile can predict who will respond to a given dietary or therapeutic intervention before they start it.

Postbiotics. Rather than delivering live bacteria, some researchers are isolating and administering the beneficial metabolites bacteria produce, such as purified short-chain fatty acids, sidestepping questions about whether transplanted organisms will actually colonize.

Engineered and phage-based editing. Early-stage work uses bacteriophages (viruses that infect specific bacteria) to selectively remove harmful strains while leaving beneficial ones intact, offering a more targeted alternative to broad-spectrum antibiotics that also wipe out protective species.

The takeaway

Microbiome research replaced a century-old assumption, that bacteria in the body are threats to be managed, with a more accurate and more useful one: most resident bacteria are partners the body depends on for digestion, immune development, and even nervous system signaling. The field's biggest wins so far are narrow and well-earned, particularly FMT's FDA approval for recurrent C. difficile infection, while broader claims about diet, mood, and disease remain mostly associational and still being tested. The honest state of the science is genuinely useful: it explains why wiping out bacteria indiscriminately backfires, and it gives a clear, evidence-based reason to be skeptical of any product that promises to "fix your gut" without a controlled trial behind it.

FAQ

Q: How many bacteria live in the human body? A: Roughly 38 trillion, a number close to the body's approximately 30 trillion human cells. This overturned the older "10 bacteria for every human cell" estimate (Sender et al., 2016, PLOS Biology).

Q: Are all bacteria harmful? A: No. Most bacteria that live in and on the human body are neutral or beneficial. Only a small subset are pathogenic, and even some of those only cause disease when the surrounding microbial community is disrupted, as with antibiotic-triggered C. difficile overgrowth.

Q: What is dysbiosis? A: An imbalance in the gut microbial community, typically involving reduced species diversity and loss of beneficial bacteria such as butyrate producers, defined relative to a matched healthy comparison population rather than a fixed universal threshold.

Q: How do gut bacteria influence health? A: They ferment dietary fiber into short-chain fatty acids that fuel colon cells and regulate inflammation, produce vitamins the body cannot synthesize on its own, train immune cells from early life onward, and occupy space and nutrients that would otherwise be available to pathogens.

Q: What is fecal microbiota transplantation and does it actually work? A: FMT transfers processed stool from a screened healthy donor to a patient, most often to treat recurrent C. difficile infection. Two FDA-approved products, Rebyota (2022) and Vowst (2023), now deliver this therapy in standardized form (FDA). It works well for this specific use; evidence for other conditions like IBD is weaker.

Q: Can probiotics fix an unhealthy microbiome? A: Evidence supports specific probiotic strains for specific problems, such as reducing antibiotic-associated diarrhea, but a handful of introduced strains rarely restructures an entire established gut community. Effects are usually modest and strain-specific rather than a general "reset."

Q: How does the gut microbiome affect the brain? A: Primarily through the vagus nerve, which bacterial metabolites like short-chain fatty acids and bile acids can directly activate, plus immune signaling that reaches the brain via circulation. Germ-free mice show measurably altered vagal activity, stress hormone response, and anxiety-like behavior, but human causal evidence is still limited.

Q: Why did the "10 bacteria for every human cell" statistic turn out to be wrong? A: It traced back to rough, unverified estimates repeated for decades without updated measurement. A 2016 reanalysis using better cell-counting methods found the true ratio is closer to 1-to-1 (Sender, Fuchs & Milo, 2016).

Q: What made large-scale microbiome research possible in the first place? A: 16S ribosomal RNA gene sequencing, developed from Carl Woese's work in the 1970s-1980s, let researchers identify bacterial species from a DNA sample without needing to grow them in a lab. Around 80% of gut bacteria resist standard cultivation, so this was the technology that unlocked the field.

Q: Is a more diverse microbiome always healthier? A: Not necessarily. Diversity correlates with health in many population-level studies, but an individual can have a less diverse, perfectly healthy baseline community, or a diverse one that's still functionally imbalanced. Diversity is a useful signal, not a standalone verdict.

References

Sender, R., Fuchs, S., & Milo, R. (2016). "Revised Estimates for the Number of Human and Bacteria Cells in the Body." PLOS Biology, 14(8), e1002533.

Eckburg, P.B., et al. (2005). "Diversity of the Human Intestinal Microbial Flora." Science, 308(5728), 1635-1638.

Turnbaugh, P.J., et al. (2006). "An Obesity-Associated Gut Microbiome with Increased Capacity for Energy Harvest." Nature, 444(7122), 1027-1031.

Lynch, S.V., & Pedersen, O. (2016). "The Human Microbiome and Immunity." Nature Reviews Immunology, 16(3), 213-226.

NIH Common Fund. "Human Microbiome Project (HMP)." commonfund.nih.gov/hmp.

U.S. Food and Drug Administration. "Fecal Microbiota Products." fda.gov/vaccines-blood-biologics/fecal-microbiota-products.

Frontiers in Immunology (2019). "Short Chain Fatty Acids (SCFAs)-Mediated Gut Epithelial and Immune Regulation and Its Relevance for Inflammatory Bowel Diseases."

Neuron (2019). "Gut Microbe to Brain Signaling: What Happens in Vagus."

---

Read next:

• How the Immune System Recognizes Threats
• How Antibiotic Resistance Spreads
• Why Fevers Help Fight Infections