The Human Microbiome—A Physiologic Perspective

Abstract

The human microbiome consists of the microorganisms associated with the body, such as bacteria, fungi, archaea, protozoa, and viruses, along with their gene content and products. These microbes are abundant in the digestive, respiratory, renal/urinary, and reproductive systems. While microbes found in other organs/tissues are often associated with diseases, some reports suggest their presence even in healthy individuals. Lack of microbial colonization does not indicate a lack of microbial influence, as their metabolites can affect distant locations through circulation. In a healthy state, these microbes maintain a mutualistic relationship and help shape the host’s physiological functions. Unlike the host’s genetic content, microbial gene content and expression are dynamic and influenced by factors such as ethnicity, genetic background, sex, age, lifestyle/diet, and psychological/physical conditions. Therefore, defining a healthy microbiome becomes challenging as it is context dependent and can vary over time for an individual. Although differences in microbial composition have been observed in various diseases, these changes may reflect host alterations rather than causing the disease itself. As the field is evolving, there is increased emphasis on understanding when changes in the microbiome are an important component of pathogenesis rather than the consequence of a disease state. This article focuses on the microbial component in the digestive and respiratory tracts—the primary sites colonized by microorganisms—and the physiological functions of microbial metabolites in these systems. It also discusses their physiological functions in the central nervous and cardiovascular systems, which have no microorganism colonization under healthy conditions based on human studies.

Introduction

According to the latest definition, microbiome refers to “the microbiota (community of microorganisms) and their ‘theatre of activity’ (structural elements, metabolites/signal molecules, and the surrounding environmental conditions)” (323). Among the 3.3 million nonredundant microbial genes (150 times more than human genome) in human fecal samples, about 99% are from bacteria (273). According to recent estimates, a 70-kg adult man has approximately 3.8 × 10¹³ total bacteria in the body, similar to the total human cell number (~3.0 × 10¹³). The total bacteria occupy approximately 0.2 kg, approximately 0.3% of body mass. Among all colonized sites, the colon holds the largest amount (>95%) of bacteria, which is calculated based on an estimation of 10¹¹ bacteria/mL at approximately 400 mL volume of an adult man. Among all other sites, dental plaque holds approximately 10¹² (10¹¹ bacteria/mL × <10 mL; ~2.6% of total bacteria); ileum, saliva, and skin hold approximately 10¹¹ (10⁸ × 400 mL, 10⁹ × <100 mL, and 10¹¹/m² × 1.8 m², respectively; ~0.3% of the total bacteria for each); and stomach, duodenum, and jejunum hold only approximately 10⁷ bacteria at each site, corresponding to approximately 2.6 × 10⁻⁵% of the total bacteria (290). In reproductive age females, the vagina holds approximately 10¹⁰ to 10¹¹ bacteria (52), whereas in adult males, bacteria load is estimated to be approximately 10⁶ to 10⁷ bacteria/mL semen (141). In infants, the number of bacterial cells associated with the body is largely reduced due to the small colon content mass (~4.4 × 10¹² for a 4.4-kg infant at around 4 weeks old and ~7 × 10¹² for a 9.6-kg infant at around 1 year old), whereas in obese people, this number is dramatically increased (~5.6 × 10¹³ at 140 kg) (290). As simply looking at the number of bacterial cells at a location is highly descriptive, most studies focus on the functional implications of bacterial colonization. The advent of next-generation sequencing technologies has also led to a better understanding of microorganisms other than bacteria.

Over 200 fungal species have been identified in human gut (151). Similar to bacteria, fungi are also distributed in various body sites, including conjunctiva, skin, oral cavity, respiratory tract (nasal cavity and lung), gut, and reproductive tract in healthy humans (68). In adults, the fungal diversity is low compared to bacteria, with the composition being highly variable among individuals, and is influenced significantly by diet (122) and medications such as antibiotics.

There are two types of viruses in the human body—bacteriophage or phage, which infects bacteria, and eukaryotic virus, which originates from diet (e.g., plant viruses) or directly affects host cells. It is estimated that the total number of viruses in the human body are similar to bacteria, with approximately 90% being bacteriophages/phages (203).

Archaeal diversity is also low. There are only 17 complete archaeal sequences associated with human/homo sapiens reported by National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/genome/browse#!/overview/). They are separated into two categories—methane-producing archaea, that is, Methanobacteriales, Methanomassiliicoccales, Methanomicrobiales, and Methanosarcinales, and nonmethanogens, that is, Halobacteriales, Thaumarchaeota, and DPANN (Woesearchaeota). Most archaea are present in the gastrointestinal (GI) tract, followed by skin and respiratory tract. Analogous to bacteria, abundance of archaea in the GI tract and the skin increases with age (35, 171).

In humans, there are two groups of nonpathogenic protozoa—amebae and flagellates. Commensal amebae include Entamoeba gingivalis in the oral cavity and Entamoeba coli, Entamoeba hartmanni, Entamoeba polecki, Entamoeba dispar, Endolimax nana, and Iodamoeba buetschlii in the large intestine. Commensal flagellates include Trichomonas tenax in the oral cavity; Trichomonas hominis, Enteromonas hominis, and Retortamonas intestinalis in both small and large intestines; and Chilomastix mesnili in the large intestine only (79).

While all components of the microbiome may be physiologically relevant, we primarily discuss bacteria in this article, given the limited number of functional studies with other microorganisms. We highlight the physiological effects in systems directly exposed to the microbiome (digestive and respiratory) as well as in systems that are largely influenced by circulating bacterial products (nervous and cardiovascular).

Digestive System

Oral cavity

The aerodigestive tract (oral cavity, pharynx, nasal passages, sinuses, and esophagus) harbors 775 different bacterial species according to the Human Oral Microbiome Database (http://www.homd.org/). Among them, Streptococci is the most abundant in mucosa; Simonsiella is unique to the hard palate; and Neisseria, Prevotella, and Haemophilus are abundant in other sites (tongue, saliva, and sub- and supragingival plaque). Anaerobic bacteria (Actinomyces, Veillonella, and Fusobacterium) are more likely to be present in subgingival plaque specimens. A high abundance of Streptococci (Streptococcus mitis is the leading species, ~9.5% of the total detected species) with low carbohydrate consumption is associated with the maintenance of dental health. When Streptococcus mutans is present at high levels, its fermentation of simple carbohydrates to lactic acid may result in the outgrowth of acid-tolerant bacteria, such as Cryptobacterium, which can cause caries (158). The main physiological functions of the oral cavity are ingestion, mechanical digestion, and chemical digestion. The enzymes responsible for chemical digestion such as amylase, lysozyme, and lipase put selective pressure on the oral microbiome and stimulate formation into biofilms. An alteration in the biofilm microbial community allows for colonization by opportunistic pathogens such as Corynebacterium diphtheriae, Pseudomonas aeruginosa, Staphylococcus aureus, Acinetobacter spp., Escherichia coli, and Porphyromonas gingivalis, which triggers an inflammatory response and can contribute to periodontitis (302, 333). In addition, higher-than-normal levels of oral-fecal transmission where oral microbes are found in the stool, are associated with conditions such as colorectal cancer and rheumatoid arthritis (284). This is especially true for the opportunistic pathogens, suggesting that oral microbiome may have implications for both GI and systemic conditions beyond the oral cavity. An analysis of oral rinse samples, which is representative of the whole oral microbiome, found antibiotic resistance genes related to macrolides, lincosamides, streptogramin, tetracycline, and quinolones potentially originating in response to food that has been treated with low-dose antibiotics (47). The vast majority of known viruses present in the oral cavity are bacteriophages that serve as a reservoir of bacterial genes such as antibiotic resistance genes, allowing for horizontal gene transfer among bacteria (268, 275). Over 20 fungal genera (Candida and Malassezia) have been reported in oral cavity of healthy humans (136), and their behavior may be influenced by bacteria. For instance, Aggregatibacter actinomycetemcomitans inhibits, while Streptococcus gordonii enhances biofilm formation by Candida albicans (13, 15). Archaea have also been isolated from subgingival plaque; pathogenesis of periodontitis may be attributable to hydrogen-consuming archaea such as Methanobrevibacter oralis, which can favor the growth of hydrogen-producing bacterial pathogens, such as Streptococcus, or serve as a syntrophic partner to a known periodontal pathogen Treponema denticola by acting as a hydrogen reservoir (190).

Esophagus

Esophageal bacteria are largely associated with mucosal surfaces reaching up to approximately 10⁴ bacteria/mm² in distal esophagus (258), with Streptococcus, Prevotella, and Veillonella being the most common genera. The resident microbial community is highly similar to that of the oral cavity, but the mechanisms that promote colonization and selection of specific oral bacteria (e.g., exclusion of Spirochaetes) remain unclear (258, 259). Changes in the resident esophageal microbial community and their physiological consequences have been explored in a few studies. The main physiological function of the esophagus is propelling swallowed food or fluid into the stomach while preventing or clearing gastroesophageal reflux. Improper functioning of these processes is linked to multiple conditions, the most common being gastroesophageal reflux disease (GERD). In some people, GERD triggers a change in the structure of the epithelium that lines the lower esophagus, leading to Barrett’s esophagus (BE). BE is in turn associated with an increased risk of developing esophageal cancer. However, most esophageal cancers are associated with environmental triggers such as smoking, obesity, and alcohol. These environmental factors are also known to affect the microbiome. Eosinophilic esophagitis (EoE) is a chronic immune-mediated disease of the esophagus in response to T helper type 2 (Th2) cells and is often related to GERD. Local Th1/Th2 ratios are known to be affected by the gut microbiome (271), and this may also be the case for the esophagus. There is an increase in the abundance of Fusobacteria in the saliva of patients with GERD and BE compared to healthy subjects (211, 259), a decrease in the ratio of Streptococcus to Prevotella in patients with BE and esophageal adenocarcinoma (EAC) (108), and an increase in the abundance of Streptococcus and Prevotella in esophageal squamous cell carcinoma (215). Biopsies from patients with EoE show increased abundance of Haemophilus, which is associated with the persistent inflammation in EoE, likely due to its ability to perturb the epithelial barrier function (126). Treatment with proton pump inhibitors (PPIs), swallowed topical corticosteroids, and food-elimination diets changed the abundance of metabolic pathways related to the biosynthesis of arginine and ornithine, degradation of 4-aminobutanoate, and bacterial cell-wall peptidoglycan synthesis, but the consequence of these findings is unclear (184). A shift toward Gram-negative bacteria increases the abundance of lipopolysaccharides (LPS)—which can activate the inducible nitric oxide synthase (iNOS) pathway to cause gastric reflux, which in turn promotes the selection against acid-sensitive bacteria in the esophagus (357). LPS can also bind directly to cell surface receptors such as Toll-like receptor 4 to activate NF-κβ pathway, which together with the activation of iNOS may induce neoplastic progression (371).

Stomach

Although previously considered to be an inhospitable environment for bacteria, given the low pH, recent studies have described a resident stomach microbiome of approximately 10² to 10⁴ colony-forming units (CFU)/g cultivatable acid-resistant/acidophilic bacteria (76), with Firmicutes, Bacteroidetes, and Actinobacteria being the predominant phyla in gastric fluid, and Proteobacteria being abundant in mucosa (307). Lactobacillus, Streptococcus, Prevotella, Rothia, and Veillonella are the predominant genera in the healthy mucosa, while Helicobacter pylori is a common pathobiont (26, 149) associated with peptic ulcer disease and gastric cancer. Resident bacteria such as Lactobacillus compete against H. pylori for the same gastric niche. Lactobacillus decreases motility and colonization of H. pylori by masking their binding sites on gastric epithelium (139). In addition, Lactobacillus can prevent an increase in gastric pH by neutralizing the ammonia produced by H. pylori.

The main physiological functions of the stomach are formation of acid chyme to aid digestion; synthesis of proteins necessary for motility, absorption, and microbial defense (e.g., gastrin, haptocorrin, and mucus); and propagating the peristaltic reflex. Regulation of gastric acid production is essential as acid overflow can induce pain and inflammatory oversensitivity (hyperalgesia), which is seen in gastric diseases such as gastritis, gastroduodenal ulceration, and dyspepsia (103, 140, 338). The acidic environment in the stomach largely restricts bacterial growth and migration into the intestine, and bacteria that colonize the stomach facilitate the maintenance of gastric pH. For example, Lactobacillus converts lactose to lactic acid, contributing to the low pH in the mucus layer, which in turn inhibits the production of gastrin and gastric acid secretion. Increased gastrin and gastric acid secretion within normal gastric pH range is often seen in patients with duodenal ulcer (103, 338). On the other hand, an increase in gastric pH is unfavorable for the resident gastric microbial community. Changes in gastric pH also have downstream consequences as evidenced by the association of PPI use (which reduce stomach pH) in spontaneous bacterial peritonitis with bacterial translocation and colonization of mesenteric lymph nodes (5), overgrowth of bacteria in the small intestine (SI), and a change in the composition of stool microbiota (155, 168), with increased susceptibility to opportunistic pathogens such as Clostridioides difficile (3).

Intestines

Microbial distribution in the gastrointestinal tract

The SI is divided into duodenum, jejunum, and ileum, while the large intestine includes the cecum and colon (ascending, transverse, and descending). Histologically, both small and large intestines are composed of four layers—mucosa (epithelium, laminal propria, and muscularis mucosa), submucosa, muscularis propria (inner circular muscular layer and longitudinal outer muscular layer with myenteric plexus in between), and serosa. Physiologically, the SI has a thin barrier with little mucus and is mainly responsible for digestion and absorption of nutrients. Its epithelium is composed of enterocytes (absorption), enteroendocrine cells (EECs; secretion of hormones), goblet cells (secretion of mucin), Paneth cells (secretion of antimicrobial substances and lysozyme), tuft cells (defense against parasites), M cells (antigen presentation), and cup cells (limited in ileum with unclear functions). The large intestine is responsible for the absorption of water and electrolytes. Except for Paneth cells, M cells, and cup cells, all other cell types found in the SI are also present in the large intestine, although in different proportions.

The SI is a harsh environment for most bacteria, given limited nutrient availability, rapid transit, higher oxygen availability, and secretion of antimicrobials [bile acids (BAs), antimicrobial peptides (AMPs), and secretory immunoglobin A (sIgA)]. As a result, the bacterial load and alpha diversity is much lower in the SI (10³–10⁶ CFU/g) but steadily increases in density and diversity as we move toward the colon (10¹¹ CFU/g) (88, 306). In healthy human subjects, phylum TM7 is mainly found in duodenum (189). Bacilli, Clostridia, Gammaproteobacteria, Actinobacteria, and Fusobacteria are the leading classes in all of the segments. Lactobacillales is the leading order of Firmicutes in the SI, with Carnobacteriaceae being uniquely found in the SI. A decrease in Carnobacteriaceae is associated with autoimmune diseases (209), and increased abundance is associated with fecal zonulin (measure of intestinal permeability) and obesity (1, 9, 163, 169, 181, 238, 263). Disorders associated with slowing of transit through the SI can result in small intestinal bacterial overgrowth (SIBO) resulting in early deconjugation of BAs, excessive carbohydrate fermentation resulting in increased osmotic load, and competition for nutrients resulting in diarrhea, weight loss, and malnutrition. The same factors that govern longitudinal distribution, such as oxygen gradient, mucus thickness and composition, and AMPs, are also important determinants of the cross-sectional distribution of microbes resulting in distinct populations close to mucosa and in the lumen of the intestine. This also results in a high density of anaerobes in the lumen of the colon compared to other parts of the intestine (128, 189, 336). Apart from bacteria, fungi such as C. albicans are present in jejunal aspirates from healthy subjects (61), and methanogenic archaea such as Methanobrevibacter smithii are found in duodenal luminal fluid (110).

In the colon, facultative anaerobes seen in the meconium and early life (e.g., Bifidobacterium) are slowly replaced by obligate anaerobes as influenced by weaning and introduction of solid foods, and largely belong to two main phyla Firmicutes and Bacteroidetes (243). Microbial composition is indistinguishable from adults at 3 to 5 years of age. Among fungi, Saccharomyces, Malassezia, and Candida are the most abundant (245) and are influenced by diet and age. In humans, consumption of carbohydrates and saturated fat is positively and negatively correlated with the abundance of Candida spp., respectively, and short-chain fatty acid (SCFA) production is negatively correlated with the abundance of Aspergillus (138). The fungal composition may be a determinant of GI diseases. In IBD patients, Malassezia restricta stimulates stronger immune response in patients with polymorphism in CARD9 which can potentially exacerbate colitis (206). Bacteriophages are abundant in feces, approximately 10¹⁰ to 10¹² virus-like particles/g, with crAssphage (associated with Bacteroides) being the most common (136, 142). The peak abundance of bacteriophages is seen in the infant gut and is inversely related to bacterial diversity (292). An increased abundance of Caudovirales is reported in fecal and mucosal samples from IBD patients (102, 204, 382). An increased fecal viral diversity is seen in colorectal cancer (244), while diversity is decreased in malnutrition (279) and type I diabetes (373). Phages can also migrate through the lymphatic system into the blood (134), and patients with cardiovascular disease (CVD) have been found to have higher circulating phage levels (85). Though at very low abundance, at least 15 families of eukaryotic viruses have been reported in the human gut, with Circoviridae being the most consistently presented (275).

Only approximately 50% of stool samples showed detectable archaeal organisms (172). Methanobrevibacter and Methanosphaera are the leading genera of archaea (138, 234). The abundance of Methanobrevibacter smithii is increased in patients with constipation, which could be a result of constipation or may be due to inhibitory effect of methane on GI transit (170). Methanosphaera stadtmanae has been implicated in IBD, given its ability to activate both innate and adaptive immune responses through the modulation of cell-surface receptors on monocyte-derived dendritic cells (17, 31).

Structural parts of the gastrointestinal tract

Enteric nervous system

The ENS is the largest component of the autonomic nervous system. The ENS innervates the whole digestive tract, and these intrinsic neural circuits can control motor function, local blood flow, mucosal transport, and secretions, without input from the brain. The ENS consists of two major network hubs of interlacing nerves, the myenteric plexus and the submucosal plexus. The myenteric plexus is found between the circular and longitudinal muscles in the gut wall, whereas the submucosal plexus can be found in the submucosa of the gut wall. The myenteric plexus regulates motor patterns (e.g., peristalsis) through the interplay between cholinergic excitatory and nitrergic inhibitory neurons, whereas the submucosal plexus is involved in secretion (e.g., fluid or electrolyte secretion). The ENS undergoes maturation after birth, and this process depends on the presence of a microbiome. In the absence of microbial colonization in germ-free (GF) mice, the ENS of an adult GF mouse resembles the ENS of a newborn. However, the immature ENS of GF mice retains its plasticity, with a high expression of the neuronal stem cell marker nestin and remains capable of maturation once microbes are introduced (74, 118). Microbial cell wall components such as LPS as well as microbial metabolites can alter GI transit, secretion, and sensation via their effect on signaling pathways in the epithelium, ENS, and glia. At the same time, ENS functions that determine the GI transit time and stool consistency are important determinants of the gut microbiome (350).

Enteric immune system

The EIS protects the host from potentially harmful microbes, while inducing tolerogenic responses to innocuous food, commensals, and self-antigens. The EIS can also be referred to as the mucosal immune system and consists of immune cell populations and enterocytes that secrete defensins (AMPs and host defense proteins), sIgA, and cytokines to maintain barrier function (6). Levels of AMPs and sIgA are high in the inner mucous layer, which result in a largely sterile zone with few of any microbes, although some microbial products can translocate through this layer. Some of these microbial products such as propionate and butyrate can affect the colonic regulatory T-cell (Treg) differentiation and accumulation which is important in the pathogenesis of inflammatory conditions such as IBD (10, 107).

Physiological functions

Digestion and absorption of nutrients

In the proximal SI, gut microbes can utilize dietary components that escape uptake or cannot be digested by host enzymes. The resulting microbial end products serve as the energy source for the host. It is estimated that around 10% of the host energy requirements are met by the products of the gut microbiome (269). Here, we briefly discuss polysaccharide, protein, and fat utilization as well as the production of vitamins by microbes and the effect of diet-driven microbial metabolites on host physiology (Figure 1). There is emerging data that suggests a trade-off between saccharolytic and proteolytic metabolism among gut microbes (e.g., low saccharolytic fiber intake can increase protein degradation) driven by diet and related factors such as transit time. Diet shapes the microbiome by supporting the networks of cross-feeding bacteria that can efficiently utilize the dietary nutrients (365).

Figure 1.

Summary of microbial metabolite modulation of physiology in different systems. Gut microbes produce the key metabolites including short-chain fatty acids (SCFAs), secondary bile acids (BAs), trimethylamine (TMA), tryptamine, and lipopolysaccharide (LPS), which can have effects in the gastrointestinal (GI) tract and other sites, such as the cardiovascular system through the circulation and the central nervous system through circulation and the blood-brain barrier (BBB). Tryptamine, produced by the microbial decarboxylation of dietary amino acid tryptophan, can facilitate GI transit by activating serotonin receptor 4 (5-HT4R) on intestinal epithelial cells and goblet cells, resulting in increased colonic secretion and mucus release (24, 25). SCFAs, fermented from dietary fiber, alter GI transit indirectly by increasing the synthesis of serotonin (5-HT) (276, 360) or directly by affecting the cholinergic neurons in the myenteric plexus. Primary BAs are converted into secondary BAs by microbes, which can be influenced by dietary components, to increase colonic secretion and contractility and influence GI transit via the potential effect on enteric glia (45, 80, 90). LPS, a cell-wall component of Gram-negative bacteria, can affect enteric neuronal survival, increase mast cell degranulation, and influence GI motility by disrupting cross-talk between muscularis macrophages and neurons via the activation of the TLR4/NF-κB pathway (7, 241, 296). TMA comes from the biotransformation of choline from diet and further oxidized into trimethylamine N-oxide (TMAO) in the liver catalyzed by the enzyme flavin-like monooxygenase 3 (FMO3), which enters the circulation. High level of circulating TMAO is associated with platelet hyperresponsiveness, increased thrombotic event risk, and aortic stiffness (40, 376, 378). It also increases the expression of CD36, which is a scavenger reporter on the surface of macrophages, to engulf oxidized low-density lipoprotein—an autoantigen involved in atherosclerosis—and thus overload the macrophages to become foam cells (358). Meanwhile, TMAO also stimulates cardiomyocyte hypertrophy and impairs endothelial function via inducing superoxide-associated oxidative stress and reducing the availability of nitric oxide (NO), which protects against oxidative stress (41). By reducing the pool of BAs and altering the BA profiles, TMAO disrupts lipid metabolism and limits the reverse cholesterol transport, which together promotes the plaque formation and atherosclerosis (87, 340). On the contrary, BAs and SCFAs prevent oxidative stress and inflammation and promote vasodilatation, lipid metabolism, and cardiomyocyte contraction (2, 324, 367). In the central nervous system, tryptamine induces autophagy and the formation of helical fibrillary tangles in the neurons and glial cells, leading to neurodegeneration (131, 249). Meanwhile, it exerts anti-depressant-like effect through the activation of serotonin 1A/2A receptors (5-HT1AR/2AR) (32, 44, 178). Another tryptophan metabolite, indole-3-propionic acid (IPA), plays a protective role against astrocyte inflammation, DNA damage, oxidative stress, and β-amyloid accumulation (57, 111, 153). In contrast, indole decreases motor activity and induced anxiety-like emotional disorder with unknown mechanisms (159). While SCFAs, such as butyrate and propionate, as well as TMAO promote endothelial tight junction integrity, certain Bas, such as chenodeoxycholic acid (CDCA) and deoxycholic acid (DCA), as well as TMA interrupt tight junction integrity (143, 174, 274). Similarly, butyrate, TMAO, taurochenodeoxycholic acid (TCDCA), tauroursodeoxycholic acid (TUDCA), glycoursodeoxycholic acid (GUDCA), and taurolithocholic acid (TLCA) exert anti-inflammatory effect on microglia cells, whereas taurocholic acid (TCA) is pro-inflammatory (152, 192, 232, 250, 359, 370). Moreover, TMAO suppresses the activation of astrocyte, while it impairs synaptic plasticity by inhibiting the mTOR signaling pathway (192). Except for the GI tract, the respiratory tract is also colonized by specific microbes, which produce metabolites that modulate the local physiological effects. Lipophilic bacteria Propionibacterium convert sebum into SCFAs to lower the pH of the nasal cavity, creating a hospitable environment for Corynebacterium and Staphylococcus, which thrive in enriched oxygen and humidity (349). SCFAs further suppress immune response by reducing the production of pro-inflammatory cytokines through histone deacetylase 9 (HDAC9) inhibition (313, 315, 316). While chronic low-dose or single high-dose exposure to LPS before home dust mite challenge prevented the development of asthma symptoms by suppressing the activation of epithelial cells and dendritic cells via induction of the ubiquitin-modifying enzyme A20 in adult mice, neonates under chronic exposure of LPS had interrupted the lung development which is more prone to inflammation (286, 295). Interestingly, LPS also reduces mucus secretion in both pretreated postnatal pups before asthma induction by ovalbumin and adult mice before home dust mite challenge-induced allergic lung inflammation (70, 86). Under normal physiological conditions, commensal Streptococcus species proliferate by catabolizing carbohydrates into organic acids to feed other commensal bacteria such as Veillonella. Meanwhile, it facilitates biofilm formation and regulates cell-cell interaction through autoinducer 2 signaling to generate a balanced community (67, 218, 242, 260, 332).

Polysaccharide utilization

Complex polysaccharides (also referred to glycans) are only partially digested by endogenous digestive enzymes; the remaining undigested fiber components serve as nutrient sources for colonic bacteria. There are a multitude of different fiber types such as resistant starch, inulin, pectin, and cellulose. Gut microbial members express dedicated uptake systems and enzymatic machinery (e.g., glycoside hydrolases, polysaccharide lyases, and carbohydrate esterases) required for breaking specific linkages in these fibers. These enzymes have flexibility in substrates, targeting dietary fiber when present and switching to host mucus glycans with the same linkages in the absence of dietary fiber (225). As individual bacteria often lack enzymes for complete degradation of complex polysaccharides, there is cooperation among different members in a community, and this complex food web allows establishment of a stable and resilient microbial community. Methanogenic archaea, such as Methanobrevibacter smithii, can interact with bacteria to promote utilization of polysaccharides such as fructans and drive increased host adiposity (19).

SCFAs are the main products of carbohydrate fermentation, and their concentration increases from less than 1 in the jejunum to greater than 100 mmol/kg in the cecum and colon (69), reflecting the significantly higher microbial density and strictly anerobic environment in the colon. The fecal concentration of the total SCFAs is estimated to be approximately 120 mM (221) (or ~200 μmol/g wet feces) (148) with estimated ratios of 60%, 20%, and 20% for acetate, propionate, and butyrate, respectively. The abundance and ratio of these SCFAs are relatively consistent among healthy adults in different ethnic groups (251, 253), but they may be altered by age and diet (78).

Absorption of SCFAs is largely facilitated by H⁺-coupled monocarboxylate-transporter 1 (MCT1) and the Na⁺-coupled monocarboxylate-transporter 1 (SMCT1), and they can exert biological effect via G-protein-coupled receptors GPR41/FFAR3, GPR43/FFAR2, and GPR109A (251) or driving epigenetic changes. GPR43 binds all three SCFAs with similar affinity, whereas GPR41 favors propionate and butyrate over acetate (38). GPR109A is specific for butyrate (312). SCFAs have been shown to influence diverse physiological processes such as GI transit, secretion, epithelial barrier function, immune modulation, and hormone secretion, but this is not surprising, given the different signaling pathways that can be activated via SCFAs.

Proteins

Undigested dietary protein (~5%) (96, 224), mucus, exfoliated epithelial cells, and pancreatic enzymes are all substrates for microbial fermentation in the colon (352). For instance, Akkermansia muciniphila predominantly consumes mucus glycans from which it produces AMPs and helps maintain barrier function (97). The end products of microbial protein fermentation include branched-chain fatty acids (BCFAs; 2-methylbutyrate, isobutyrate, and isovalerate) in addition to SCFAs, ammonia, amines, phenols, indoles, gases (H₂S, NO, CO₂, CH₄, and H₂), and organic acids (29, 352). Unlike carbohydrate fermentation which largely occurs in the proximal colon, protein fermentation predominantly occurs in the distal colon, and this activity increases with pH (280). pH is lower in the proximal colon, with pH ~5.5, compared to the descending colon, with pH ~6.9 (297).

Fat

In adults, the absorption of dietary fat is largely completed in the mid-third of the jejunum, and less than 10% of the daily intake is estimated to escape absorption. This is in contrast to infants where 15% to 35% may escape absorption (18). Fat digestibility and absorption is further reduced in the presence of dietary fiber (77, 132). Bacterial fatty acid metabolites such as reuterin or polyunsaturated fatty acids (PUFAs) have antimicrobial properties. Trimethylamine N-oxide (TMAO) produced by the interplay of bacterial and host metabolism of choline is associated with CVD (87).

Vitamin production

Bacteria predominately in the phyla Bacteroidetes and Fusobacteria can synthesize some or all of the different B vitamins (B1/thiamin, B2/riboflavin, B3/niacin, B5/pantothenic acid, B6/pyridoxine, B7/biotin, B9/folic acid, and B12/cobalamin) (223), among which the microbial production of B6, B9, B12, and B3 can be as high as 86%, 37%, 31%, and 27%, respectively, of the human daily reference intake (280). There are two forms of vitamin K—K1/phylloquinone, which comes directly from diet, and K2/menaquinone (MK), which is largely dependent on microbial synthesis (Bacteroides, Enterobacteria, Eubacterium lentum, and Veillonella) (310, 339).

Inorganic nitrate

Inorganic nitrate (NO₃⁻) is acquired largely from dietary vegetables, such as spinach, radish, beetroot, and celery. Commensal bacteria such as Rothia, Neisseria, and Lactobacilli are highly effective in converting NO₃⁻ to NO (327). Under normoxic conditions, the conversion of NO₃⁻ to NO is dependent on the oxide synthase enzymes [nitric oxide synthase (NOS)], whereas under hypoxia, this is independent of NOS and favored under acidic environment (pH <5.5) (298, 317). In the stomach, NO exerts antimicrobial effects and inhibits the pathobiont H. pylori (92). NO also protects the stomach from stress-induced mucosal injury via suppressing neutrophil infiltration into gastric mucosa (236). In the intestine, NO facilitates fluid absorption under physiological conditions, whereas under pathophysiological conditions, high concentrations of NO can lead to hypersecretion and diarrhea (240). Upon oral administration of nitrate, increased plasma NO₂⁻ with decreased blood pressure was seen in older normotensive subjects, which is correlated with increased abundance of Rothia and Neisseria and decreased abundance of Prevotella and Veillonella in the oral microbiome (327).

Barrier function

The epithelial lining of the digestive tract with the overlaying mucus layer is the first line of defense from luminal contents. The barrier is generally permeable to allow for absorption of nutrients, exchange of electrolytes and excretion of end products, and sampling of luminal contents allowing for development of tolerance within the gut immune system. The barrier is composed of epithelial cells connected by desmosomes, adherent junctions, and tight junctions, which together form epithelial junctional complexes (119). Depletion of the mucus layer or disruption of the epithelial junctional complexes can result in increased permeability of the barrier, resulting in translocation of harmful luminal contents (e.g., LPS) and activation of the ENS and EIS, etc. Microbial metabolites including BAs, tryptophan metabolites, and SCFAs can influence epithelial barrier function, while availability of dietary fiber can affect the thickness of the overlying mucus layer (113, 119, 248, 254, 277, 288, 361).

BAs can upregulate expression of Claudin-1 via the activation of G-protein-coupled receptor TGR5 and regulate barrier integrity and iNOS expression which has antibacterial effects via the farnesoid X receptor (FXR) in the enterocytes (156). Dietary tryptophan is extensively metabolized by both the host and the microbes. The tryptophan side-chain indole is input for a range of microbially derived indoles that can engage the arylhydrocarbon receptor (AhR). These molecules include indole, indoxyl sulfate, indoleacetic acid, indolecarboxaldehyde, indoleacetaldehyde, 3-methylindole, indole-3-ethanol, indole-3-pyruvate, indole-3-aldehyde, and tryptamine (288, 326). AhR signaling promotes tissue repair through induction of interleukin 22 production, which is impaired in a subset of IBD patients. In contrast, indolepropionic acid (IPA) is not a strong AhR agonist (147) but seems to strengthen the intestinal barrier through engaging the pregnane X receptor (PXR) and upregulating genes regulating intestinal permeability (331). Microbial metabolites also modulate barrier function. Acetate activates NLRP3 inflammasome via GPR43 (222), butyrate upregulates actin-binding protein synaptopodin via histone deacetylase (HDAC) inhibition (345), and butyrate oxidation stabilizes hypoxia-inducible factor 1 (269). The purine derivative hypoxanthine can augment the epithelial barrier via its effect on goblet cells (186, 187).

Propulsion of gastrointestinal contents: gastrointestinal motility and secretion

The propulsion of contents through the GI tract is facilitated by GI motility and secretion. GI motility involves complex interplay between smooth muscle cells, ENS, glia, and immune cells. The secretion of water, electrolytes, and mucin facilitates transit through the GI tract; dilutes intestinal contents; and helps maintain body fluid and electrolyte balance. Gut microbial products influence GI motility by modulating enteric neurons, enteric muscularis macrophages, and enteric glia. For instance, LPS derived from gut microbiota enhances enteric neuronal survival and promotes GI motility by acting through the TLR4 and NF-κB pathway (7). Gut microbial products also affect the communication between enteric neurons and muscularis macrophages, which is essential for maintaining normal GI motility (241). Additionally, gut microbes modulate enterochromaffin (EC) cell biosynthesis of serotonin [5-hydroxytryptamine (5-HT)], an important neurotransmitter involved in regulating GI motility and secretion (276). SCFAs and BAs can increase serotonin biosynthesis and release, thereby accelerating GI transit by increasing GI motility and secretion in a diet-dependent manner (165). Butyrate also promotes colonic transit and circular muscle contraction by increasing the proportion of choline acetyltransferase (ChAT)-immunoreactive myenteric neurons and their expression of MCT2 (301). Butyrate enhances the absorption of NaCl while inhibiting Cl⁻ secretion, which could be leveraged to treat antibiotic-associated diarrhea (46). Butyrate also stimulates mucin production (46). On the contrary, propionate reduces both short (<50% movement along the segment) and complete gut propagations in guinea pigs (150).

BAs increase 5-HT release in the duodenum by stimulating motilin secretion at low pH (5.0) from M cells via TGR5 receptor (4). Ileocolonic delivery of chenodeoxycholic acid (CDCA), the most potent FXR agonist, stimulates propagation pressure waves in the proximal colon (16). CDCA increases colonic secretion by activating CFTR channels via the cAMP-PKA pathway. Indeed, elevated levels of CDCA are seen in the stool samples from diarrhea-predominated irritable bowel syndrome (IBS-D) patients, and they correlate with stool consistency and frequency (90, 228). This may be linked to differences in microbial capacity for conversion of CDCA to LCA in a subset of patients with IBS-D. We have previously shown that bacteria-derived tryptamine promotes fluid secretion and accelerated GI transit by activating 5-HT4 receptor (25). Fecal tryptamine levels are elevated in patients with diarrhea-predominant IBS, further supporting its effect on intestinal secretion (228).

H₂S, produced by sulfate-reducing bacteria via a dissimilatory sulfate reduction pathway, may inhibit motility by the relaxation of smooth muscles via three potential pathways: (i) activation of K_ATP channels, (ii) inactivation of L-type calcium channels, and (iii) suppression of cholinergic and tachykinergic excitatory pathways (229). H₂S stimulates the secretion of anion via K⁺ channels on both apical and basolateral sides of the gut (130). It also enhances acid-induced HCO₃₋ secretion via production of prostaglandin E2, release of luminal NO, and activation of capsaicin-sensitive afferent neurons in the duodenum (157). H₂S also evokes chloride secretion from mucosa by activating cholinergic secretomotor neurons (180).

Gastrointestinal sensation

Sensation in the GI tract relates to how the nervous system assesses the volume and chemical composition of the luminal contents along the entire length of the GI tract (98). The information is conveyed following input from mechanoreceptors, chemoreceptors, osmoreceptors, thermoreceptors, and nociceptors (98). Aberrant sensory input results in abdominal pain, which is a hallmark of disorders of the gut brain axis such as IBS (89, 99, 100). A few studies have reported potential links between microbial products and abdominal pain (228, 270, 281). Similar to GI transit, 5-HT signaling via 5-HT4 and 5-HT3 receptors is also important in visceral sensation (60, 137, 231, 304). There are some indications from animal studies that females are more sensitive to disrupted 5-HT signaling, which is consistent with higher female prevalence of IBS and other diseases linked to abdominal pain/hypersensitivity (109). A role for BAs has been proposed based on reduced sensory (pain) thresholds in healthy people following rectal infusion of the physiological levels of CDCA. In animal models, colonic infusion of CDCA or DCA induces visceral hypersensitivity via FXR-induced expression of transient receptor potential vanilloid 1 (TRPV1) in the dorsal root ganglia—a known driver of visceral hypersensitivity (199). SCFAs such as butyrate when given at physiological levels reduce visceral pain in a dose-dependent manner in healthy humans (328). This is dependent on the activation of AMP-activated protein kinase (AMPK) and PPARγ (247), whereas supraphysiological level of butyrate could induce visceral hypersensitivity in rats via enteric glial cell-derived nerve growth factor (217).

Respiratory System

The respiratory tract is composed of the nose, pharynx, larynx, trachea, bronchi, and the lungs. Current studies frequently separate the respiratory tract into upper and lower parts, where the nasal cavity, oropharynx, nasopharynx, and the larynx above the vocal cords are counted as the upper respiratory tract (URT), and the larynx below the vocal cords, trachea, bronchi, and the lungs are counted as the lower respiratory tract (LRT). Under normal conditions, an adult human inhales approximately 7000 liter air/day, along with which approximately 10⁴ to 10⁶ bacteria/m³ of air are inhaled (182). Large particles (>10 μm) are trapped by the nasal hair, and particles of 3 to 10 μm are trapped in the URT. Small particles (0.4–3 μm) which are normally bacterial or viral particles can either be held by mucus covering the URT or directly enter the lungs (182, 226).

URT microbiome and its physiological functions

Nasal cavity

The nares and the nasal cavity have distinct structures and characteristics. The nares are adjacent to the nasal cavity and have a keratinized stratified squamous epithelium with hairs and sebaceous glands. In contrast, the nasal cavity has a nonkeratinized stratified squamous epithelium that changes from without microvilli to short microvilli and eventually transitions into ciliated pseudostratified columnar epithelium with mucus-producing goblet cells (21, 182). While the nares are relatively dry and inhospitable for microorganisms, the mucus layer in the nasal cavity traps and clears the invading particles with the help of antimicrobial components such as lysozyme, lactoferrin, defensins, and H₂O₂ (21, 182). Bitter receptors located on solitary chemosensory cells sense molecules, such as acyl-homoserine lactone, released by bacterial pathogens. Activation of the bitter receptors stimulates the production of NO, which facilitates the mucociliary clearance of the pathogens (188).

In healthy adults, the predominant bacterial species in the nasal cavity are from Corynebacterium and Staphylococcus. These bacteria benefit from lower pH resulting from fermentation of lipids from sebum into SCFAs by commensal Propionibacterium (349). Certain species of Corynebacterium inhibit the transcription of virulence genes and hemolysin activity in Staphylococcus to keep it in a commensal state (21, 53, 282). However, under disease conditions such as chronic rhinosinusitis, there are alterations in the species and abundance of bacteria (Table 1). While the abundance of Corynebacterium varies, S. aureus abundance is consistently higher (28, 84). This shift in the microenvironment and reduced clearance by the host innate immune response lead to the production of more polysaccharide inter-cellular adhesin (PIA), a major component of S. aureus, and other surface-associated proteins that facilitate the formation of biofilms. Secreted proteins from S. aureus promote the maturation of biofilms, and extracellular DNA (eDNA) from lysed cells interacts with bacterial proteins to form insoluble oligomers that keep the biofilm matrix intact. This matrix protects microorganisms from the host immune defense and provides better tolerance to antibiotics, ultimately promoting the dispersion of bacteria (208).

Table 1.

Microbial Compositions Under Healthy and Disease Conditions

System	Source of microbes/microbial metabolites	Main microbial components under healthy state	Microbial shifts in diseases (adults only; comparisons made to healthy controls or nondisease controls)
Gastrointestinal (GI) tract	Oral cavity	Bacteria: Streptococci, Simonsiella, Neisseria, Prevotella, Haemophilus, Actinomyces, Veillonella, Fusobacterium (47) Fungi: Candida, Malassezia (136) Viruses: bacteriophages of oral commensals (268) Archaea: Methanobrevibacter oralis (190)	(Data from stool samples with active diseases unless specified) Irritable bowel syndrome (IBS): Bacteria ↑Shigella ↓Eubacterium rectale, Faecalibacterium prausnitzii, Paraprevotella clara, Prevotella corporis, Roseburia intestinalis, Ruminococcus lactaris (all IBS subtypes) (266) ↑Streptococcus (Constipation-predominated IBS (IBS-C) and IBS-D), Lactobacillus spp. (severe IBS-D) (228); ↑Enterobacteriaceae ↓Bifidobacteria (IBS-C) (66) Fungi ↑Saccharomycetes ↓Aspergillus, Rhodotorula, Penicillium, Pichia (mixed IBS-C and IBS-D) (287) Viruses ↑Pandoravirus salinus ↓Choristoneura biennis entomopoxvirus, Aureococcus anophagefferens virus, Phaeocystis globosa virus, Pandoravirus inopinatum, unclassified Adenoviridae, Rudiviridae, Orthopoxvirus, Centapoxvirus, Ligamenvirales, Chlorovirus, Prymnesiovirus, Capripoxvirus (mixed IBS subtypes) (8) ↑Cronobacter virus CR3, unclassified Microviridae sp. Cat1991, crAss-like viruses sp. Cat2025 (IBS-D) ↑Lactobacillus virus LBR48, Poophage MBI-2016a, Spiromicrovirus undefined species (IBS-C) ↓Norovirus (IBS-C and IBS-D) (235) Archaea ↑Methanobrevibacter smithii (IBS-C) (170) Inflammatory bowel disease (IBD): Bacteria ↑Ruminococcus gnavus and Eggerthella lenta (CD) ↑E. lenta, Holdemania filiformis, Clostridium innocuum (UC) ↓Eubacterium eligens, Eubacterium rectale, Faecalibacterium prausnitzii (CD) ↓E. eligens and Clostridium aminobutyricum (UC) (58) ↑Unclassified Roseburia (CD and UC) ↑Bifidobacterium breve and Clostridium symbiosum (UC) ↑Ruminococcus gnavus, Escherichia coli, Clostridium clostridioforme (CD) (104) Fungi ↑Candida [UC mucosa (200); CD stool (303)] ↓Saccharomyces [UC mucosa (200); mixed subtypes stool (299)] Viruses ↑Caudovirales (UC and CD) (59, 102, 246, 382) Archaea ↑Methanosphaera stadtmanae (17, 31) Colorectal cancer (CRC): Bacteria ↑Fusobacterium, Peptostreptococcus, Porphyromonas, Bacteroides, Parvimonas, Prevotella, Gemella, Streptococcus, Solobacterium, Clostridium, Campylobacter, Faecalibacterium, Eubacterium, Roseburia, Bifidobacterium, Anaerostipes (133, 176, 314) Fungi ↑Aspergillus flavus, Kwoniella mangrovensis, Pseudogymnoascus sp. VKM F-4518, Debaryomyces fabryi, Aspergillus sydowii, Moniliophthora perniciosa, Kwoniella heavenensis, Aspergillus ochraceoroseus, Talaromyces islandicus, Malassezia globosa, Pseudogymnoascus sp. VKM F-4520, Aspergillus rambellii, Pneumocystis murina, Nosemia apis, Sistotremastrum suecicum, Tilletia controversa, Erysiphe pulchra (62, 212) ↓(class) Saccharomycetes and Pneumocystidomycetes (62), (species) Aspergillus niger and Talaromyces islandicus (212) Viruses ↑(family) Myoviridae, Podoviridae, Siphoviridae, Drexlerviridae, Inoviridae, Microviridae, (genus) Orthobunyavirus, (species) Vibrio phage ICP1, Perigonia lusca nucleopolyhedrovirus, Chrysodeixis includens nucleopolyhedrovirus ↓ (family) Herelleviridae, (species) Citrobacter virus Moon, Bacillus virus Mater (54, 212, 244, 383) Archaea ↑Halophiles (e.g., Haloplanus CBA1113, Halopelagius longus, Halorubrum tropicale, Halococcus morrhuae, Halococcus salifodinae, Halovenus aranensis, Natrinema species J7–2), ↑methanogens Methanothrix soehngenii and Methanoculleus marisnigri (63) ↑Pyrobaculum (212) ↓Methanogens (e.g., Methanosphaera, Methanococcoides, Methanocorpusculum, Methanocaldococcus, Methanobacterium) (63) Gastroesophageal reflux disease (GERD): Bacteria (saliva sample) ↑Actinomyces, Atopobium, Stomatobaculum, Ruminococcaceae [G-2], Veillonella, Leptotrichia ↓Porphyromonas, Gemella, Peptostreptococcus, Neisseria (379) ↓Prevotella melaninogenica, Prevotella pallens, Leptotrichia, Solobacterium moorei (166) (biopsy above esophagogastric junction) ↑Campylobacter (30), Fusobacterium, Veilonella, Neisseria, Bacillus (211) ↓Prevotella, Helicobacter, Moraxella (362) (stool) ↑Unclassified Ruminococcaceae, unclassified Enterobacteriaceae, Blautia, Parabacteroides, Streptococcus, Dorea, Clostridium, Collinsella, unclassified Mogibacteriaceae, Eubacterium, unclassified TM73, Eggerthella, Adlercreutzia, Escherichia, Actinomyces ↓Sutterella, Butyricimonas (356) Celiac disease: Bacteria (stool) ↑Barnesiellaceae, Odoricateriacecae, Lactobacillus ↓Erysipelotrichaceae, Lachnospiraceae, Prevotella, Dorea, Akkermansia (33, 291) (duodenal biopsy) ↑Helicobacter, Megasphaera, Prevotella, Lactobacillus, Catenibacterium ↓Intestinibacter, Eubacterium, Morexella, Ruminococcus, Turicibacter (33) (oropharyngeal swab) ↑Neisseria, Actinomyces ↓Streptococcus, Veilonella, Gemella, Staphylococcus, Rothia, Prevotella, Leptotrichia (154) Fungi No difference in duodenal biopsy (72) ↑Candida and Saccharomyces in stool (125) Archaea Methanomassiliicoccus ↑in duodenal biopsy and ↓ in stool (33)
	Esophagus	Bacteria: Streptococcus, Prevotella, Veilonella, Haemophilus, Neisseria (252, 258)
	Stomach	Bacteria: Lactobacillus, Streptococcus, Prevotella, Rothia, Veillonella, Clostridum (26)
	Intestines	Bacteria: Enterococcus faecalis, Staphylococcus epidermidis, Escherichia coli, Enterobacter spp. (162) Fungi: Candida spp., Saccharomyces, Malassezia (61, 245) Viruses: crAssphage, Circoviridae, Anelloviruses (136, 142, 205, 275) Archaea: Methanobrevibacter, Methanosphaera (138, 234)
Respiratory tract	URT (nasal cavity, pharynx, and larynx)	Bacteria: Corynebacterium, Propionibacterium, Streptococcus, Fusobacterium, Staphylococcus, Veillonella, Prevotella (12, 115, 116, 335) Fungi: Candida, Saccharomyces, Meyerozyma (325)	Asthma: Bacteria (Nasal lavage fluid) ↑Streptococcus, Pelomonas ↓Faecalibacterium, Lactobacillus, Clostridium_XlVa, Blautia, Escherichia, Butyricicoccus (55) (Nasal/Nasopharyngeal swabs) ↓ (genus) Corynebacteriales (young adults), Moraxella (elderly) (185), (species) Prevotella buccalis, Alkanindiges hongkongensis, Gardnerella vaginalis, Dialister invisus (101) (sputum) ↑Streptococcus, Gemella, Neisseria (145) (bronchial biopsy) ↑(genus) Klebsiella, Mycobacterium, Streptomyces, Cellulomonas, Leclercia, Kitasatospora (severe asthma), (species) Mycoplasma pneumoniae, Chlamydophila pneumoniae (146, 179) (stool) ↑Ruminococcus gnavus, Clostridium clostridioforme, Bifidobacterium pseudocatenulatum, Eggerthella lenta, Clostridium bolteae, Clostridium ramosum, Clostridium spiroforme ↓Roseburia intestinalis, Roseburia inulinivorans, Faecalibacterium prausnitzii, Sutterella wadsworthensis, Bacteroides stercoris (344, 380) Fungi (sputum) ↑Wallemia, Mortierella, Fusarium, unclassified Chaetomiaceae, Phialophora, Metarhizium, unclassified Sporormiaceae, Irpex, Schizophyllum, Rhodotorula ↓unclassified Sclerotiniaceae, Mycosphaerella, Sporobolomyces, Trametes, Naganishia, Aspergillus (145) (bronchial samples) ↑Trichoderma, Alternaria, Cladosporium, Fusarium, Penicillium (293) Chronic obstructive pulmonary disease (COPD): Bacteria (bronchoalveolar lavage fluid) ↑Afipia, Brevundimonas, Curvibacter, Moraxella, Neisseria, Undibacterium, Corynebacterium, Capnocytophaga, Leptolyngbia, Haemophilus, Pseudomonas ↓Haemophilus parainfluenzae, Propionibacterium acnes, Micrococcus luteus, Terrahaemophilus, Eubacterium saburreum, Selenomonas, Centipeda periodontii, Lactobacillus sakei, Prevotella (93, 364) (stool) ↑Prevotellaceae (severity GOLD I-II), Bifidobacteriaceae, Eubacteriaceae, Lactobacillaceae, Micrococcaceae, Streptococcaceae, Veillonellaceae (mixed severity) ↓Bacteroidaceae, Fusobacteriaceae (severity GOLD III-IV), Desulfovibrionaceae, Gastranaerophilaceae, Selenomonadaceae, uncharacterized Bacilli, Clostridia (mixed severity) (36, 197) Fungi (sputum) ↑Aspergillus, Penicillium (14) Acute respiratory distress syndrome (ARDS): Bacteria (bronchoalveolar lavage fluid) ↑Enterobacteriaceae, Pasteurellaceae, Bacteroides (82, 83) Cystic fibrosis (idiopathic): Bacteria (bronchoalveolar lavage fluid) ↑Haemophilus, Streptococcus, Neisseria, Veillonella, Pseudobutyrivibrio, Anaerorhabdus (237, 319) (stool) ↑Fusobacteria, Lactobacillus, Streptococcus, Veillonella, Enterococcus ↓Bifidobacterium, Clostridium, Akkermansia, Faecalibacterium, Roseburia (43, 91, 283) Viruses (lung specimen) ↑human herpes viruses (HHV) (309)
	LRT (trachea, bronchi, and lungs)	Bacteria: Streptococcus, Haemophilus, Corynebacterium, Neisseria, Moraxella, Veillonella, Stenotrophomonas, Prevotella, Pseudomonas, Staphylococcus, Acinetobacter (261, 262) Fungi: Aspergillus, Penicillium, Blumeria, Mycosphaerella (293)
CNS	GI (through blood-brain barrier)	None	Alzheimer’s disease: Bacteria (saliva) ↑(genus) Moraxella, Leptotrichia, Sphaerochaeta (214), (species) Capnocytophaga sp. ora clone DZ074, Eubacterium infirmum, Prevotella buccae, Selenomonas artemidis (106) ↓(genus) Rothia (214), (species) Streptococcus mutans (106) (stool) ↑Prevotella, Escherichia/Shigella, Phascolarctobacterium, Gemella, Alistipes ↓(genus) Lachnospira, Ruminiclostridium, Ruminococcus, SMB53 (family Clostridiaceae), Dialister, Clostridium, Turicibacter, cc115 (family Erysipelotrichaceae), Bifidobacterium, Adlercreutzia, (species) Eubacterium rectale (48, 121, 213, 334) Parkinson’s disease: Bacteria (oral swab) ↑ Streptococcus mitis ↓Neisseria subflava (202) (colon mucosa) ↑Ralstonia ↓Faecalibacterium, Dorea (167) (stool) ↑Oscillospira, Akkermansia, Bifidobacterium, unidentified Ruminococcaceae, Christensenella, Catabacter, Lactococcus, Klebsiella, unclassified Enterobacteriaceae, Proteus, Escherichia/Shigella, Streptococcus, Enterococcus ↓Blautia, Coprococcus, Dorea, Roseburia, Clostradium, Eubacterium, Faecalibacterium, Ruminococcus Contradictory results: Bacteroides, Lactobacillus, Prevotella (22, 127, 167, 191, 198, 202, 264, 267, 322) Fungi (stool) ↑Hanseniaspora, Kazachstania, uncultured Tremellaceae, Penicillium ↓Saccharomyces (73) Anxiety and depression: Bacteria (stool) (generalized anxiety disorder) ↑ (genus) Fusobacterium, (species) Ruminococcus gnavus ↓ (genus) Faecalibacterium, Sutterella, Lachnospira, Butyricicoccus, (species) Eubacterium rectale (161) (social anxiety disorder) ↑(genus) Gordonibacter, (species) Anaeromassilibacillus sp. An250 ↓Parasutterella excrementihominis (42) (major depression disorder)
Cardiovascular	GI (through circulation)	None	↑Enterobacteriaceae, Bacteroides, Escherichia, Ruminococcus, Parabacteroides, Collinsella, Olsenella, Parvimonas, Anaerostipes, Blautia, Adlercreutzia, Bifidobacterium, Clostridium XI, Eggerthella, Holdemania, Streptococcus ↓Faecalibacterium, Prevotella, Subdoligranulum, Clostridium XlVa, Roseburia, Megamonas, Sutterella Contradictory results: Alistipes, Phascolarctobacterium (56, 160, 369, 374) Coronary artery disease: Bacteria ↑(genus) Klebsiella, Clostridium IV, Collinsella, Escherichia/Shigella, Enterococcus, Parabacteroides, Bifidobacterium, Butyricimonas, (species) Odoribacter splanchnicus, Escherichia Coli, Ruminococcus Gnavus ↓(family) Clostridiaceae, (genus) Roseburia, Eubacterium, Faecalibacterium, Subdoligranulum, Clostridium, Blautia, Anaerostipes hadrus, Streptococcus, Butyricicoccus, (species) Lachnospiraceae Anaerosporobacter, Lachnospiraceae K4B4 group, Ruminococcus gauvreauii (164, 216, 308, 320, 375) Hypertension: Bacteria (saliva) ↑Prevotella, Veillonella ↓Fusobacterium (stool) ↑Prevotella, Klebsiella, Desulfovibrio, Parabacteroides, Porphyromonas, Actinomyces, Streptococcus ↓Faecalibacterium, Oscillibacter, Roseburia, Bifidobacterium, Coprococcus, Butyrivibrio, Clostridium, Enterococcus, Blautia (195, 355) Fungi (saliva) ↑Malassezia, Kluyveromyces, Tetrapisispora, Agaricus ↓Metarhisium, Sugiyamaella, unclassified Hypocreomycetidae, Zymoseptoria, Melampsora (subgingival) ↑Nannizzia, Blastomyces, Wallemia ↓Pestalotiopsis, Leptosphaeria, Saccharomycopsis, Gaeumannomyces, Scheffersomyces, unclassified Hypocreales, Torulaspora (stool) ↑Aspergillus, Cryptococcus, Colletotrichum, Spizellomyces, Torulaspora, Komagataella, unclassified Dothideomycetes, Malassezia ↓Saccharomyces, Coniophora, Mortierella (51, 381) Viruses (stool) ↑Cnaphalocrocis medinalis granulovirus, Simbu orthobunyavirus, Tevenviriae T4virus, Siphoviridae, WW nAnB ↓ Unclassified Siphoviridae, Streptococcus virus phiAbc2, Thermus virus P23.77, Mycobacterium phage Toto, Morganella phage vB MmoM MP1, contagiosum virus subtype1, Lambda-like viruses, Pandoravirus inopinatum (123) Ischemic stroke: Bacteria (oral) ↑Streptococcus, Prevotella, Veillonella, Fusobacterium, Treponema (305) (stool) ↑(post-stroke cognitive vs. non-cognitive impairment) Streptococcus, Klebsiella, Lactobacillus, Prevotella, Veillonella ↓Roseburia, Fusicatenibacter (207) ↑(family) Enterobacteriaceae, Ruminococcaceae, Veillonellaceae, Lachnospiraceae, Neisseriaceae, Aerococcaceae, (genus) Pseudomonas, Flavobacterium, Staphylococcus, Prevotella, Micrococcus, Corynebacterium, Enhydrobacter, Collinsella, Faecalibacterium, Blautia, Anaerococcus, Finegoldia, Dialister ↓(family) Bacteroidaceae, Prevotellaceae, Lachnospiraceae, (genus) Bacteroides, Akkermansia, Lactobacillus, Ruminococcus, Parabacteroides, Oscillospira, Mucispirillum, Actinomyces, Klebsiella, Stenotrophomonas Contradictory result: Ruminococcaceae (49, 351)

Pharynx and larynx

The pharynx can be divided into three regions: nasopharynx, oropharynx, and laryngopharynx. The bacterial community composition is relatively similar among the three regions, and due to the anatomical connection of the nasal cavity and the lower respiratory tract, the microbial structure is also like that of the nasal cavity, and in the case of reflux diseases, the lower respiratory tract. The most abundant genera in these regions are Streptococcus, Fusobacterium, Staphylococcus, Veillonella, and Prevotella (Table 1) (12, 115, 116).

Under normal physiological conditions, commensal Streptococcus species proliferate by catabolizing carbohydrates into organic acids to feed other commensal bacteria such as Veillonella. Meanwhile, they facilitate biofilm formation and regulate cell-cell interaction through autoinducer 2 signaling to generate a balanced community (67, 218, 242, 260, 332). Commensal S. mitis also cross-interacts with Th17 and produces hydrogen peroxide and bacteriocins to protect against pathogenic Streptococcus pneumoniae (294). A disruption in such a community may trigger an immune response in the host, contributing to inflammation and potentially tumorigenesis. Indeed, in the mucosa and swab samples from patients with laryngeal squamous cell carcinoma, there is an increase in Fusobacterium and Prevotella, challenging the predominance of Streptococcus (115, 116). Fusobacterium is involved in biofilm formation and triggers the production of pro-inflammatory cytokines, such as IL-1α, IL-6, and IL-8 (265). Similarly, Prevotella induces the activation of NF-κB and production of IL-8 and TNFα, preferably under oxygen-poor anaerobic conditions (117).

LRT microbiome and its physiological functions

Trachea

The tracheal region is anatomically located at the intersection of the URT and LRT. Studies focusing on the tracheal aspirates of healthy children, adolescents, and adults have identified several predominant genera including Streptococcus, Haemophilus, Corynebacterium, Neisseria, Moraxella, Stenotrophomonas, Prevotella, Pseudomonas, Staphylococcus, and Acinetobacter (135, 210, 261, 262). The composition of the tracheal microbiome is influenced by the environment and the physiological status of the host and shares similarities with the URT and LRT microbiomes.

In children with lower respiratory tract infections, Haemophilus abundance is increased, while Pseudomonas, Corynebacterium, and Acinetobacter are decreased (261). In patients with tracheostomy, Haemophilus and Moraxella are increased (262). Haemophilus may interfere with ciliary function and cause death of ciliary mucosal cells due to the presence of lipooligosaccharide (LOS) in its cell wall. The pili on its surface may also cause agglutination of red blood cells. Haemophilus proteases, such as endopeptidases, break down IgA and secretory factors such as lactoferrin and lysozymes, which may contribute to an environment sensitive to infection (173).

Patients with malignant tracheal tumors have a higher abundance of Prevotella and Alloprevotella compared to those with nonmalignant tumors (210). Prevotella induces the activation of NF-κB and production of pro-inflammatory cytokines such as IL-8 and TNFα, which could lead to immune dysfunction. Alloprevotella is positively associated with the severity of inflammation potentially driven by IgA dysfunction (23).

Bronchi and lungs

The lung microbiome is believed to be derived from the oral cavity due to the similarities between genera found in both sites (20). Oral microbes can migrate into the lungs through microaspiration as well as inhalation of air and direct dispersion along the mucosal surfaces (330). The homeostasis of the lung microbiome is maintained through a balance of migration, elimination, and the constant turnover of the community (81). In adults, the predominant bacterial genera of the lung microbiome are Prevotella, Veillonella, and Streptococcus, which are also commonly found in the upper respiratory tract (20, 50).

It is hypothesized that maternal microbial colonization may affect the fetal respiratory system and prime the development of their innate immune system. Indeed, infants born to women who had higher serum acetate levels during pregnancy had fewer respiratory issues. SCFAs such as acetate are present in the amniotic fluid and suppress the activity of HDAC9, which in turn increases the acetylation of Foxp3 to activate Tregs.

The “hygiene hypothesis” proposes that early-childhood exposure to environmental bacteria and certain infections helps to develop the immune system. Although specific mechanisms have not been studied in human fetal tissue samples, studies in animal models have demonstrated beneficial effects of early exposure to environmental bacteria. In mice experiments, chronic low-dose or single high-dose exposure to LPS or farm dust extract before home dust mite challenge prevented the development of asthma symptoms by suppressing the activation of epithelial cells and dendritic cells via induction of the ubiquitin-modifying enzyme A20 (286). There is an increase in bacterial load and a shift in microbial composition soon after birth, which is associated with weakened response to aeroallergens, and programmed death ligand 1 (PD-L1)-driven appearance of Helios-Treg cell subset (114, 227). Apart from shaping the immune response, bacteria also facilitate the shaping of lung structure and its sequential physiological activities. In rodent studies, alveolar morphology and mucus production was associated with the abundance of bacteria in the lung, and colonizing germ-free mice with two Lactobacillus strains improved mucus production and altered the alveoli structure (363).

Lung diseases have been associated with changes in the microbiome (Table 1). For example, in sepsis and acute respiratory distress syndrome (ARDS), lung microbiome has a higher representation of gut-associated bacteria (83), and Lachnospiraceae and Enterobacteriaceae have been associated with the use of ventilation systems (82). Chrysiogenales is prevalent in patients with severe cystic fibrosis and is associated with the abundance of metabolites such as pyruvate, lactate, and putrescine (321). Patients with chronic obstructive pulmonary disease (COPD) have a high abundance of Pseudomonas (94, 112, 289), which can trigger IL-17A production to activate an immune response (353). Patients with a functional FUT2 gene (secretor), which encodes fucosyltransferase-2 to determine blood group secretor status, or heterozygous secretors are more susceptible to P. aeruginosa-induced infection compared to those with a nonfunctional FUT2 gene (311). Fungi are rare in bronchioalveolar lavage samples from healthy individuals, and those present in the oropharyngeal wash samples are often of environmental origin.

Physiologic Systems Not Colonized with Bacteria

Central nervous system

Formation, maturation, and maintenance of the nervous system are influenced by genetic determinates, environmental cues, and experiential influences. On the cellular and anatomical levels, intricate physiological processes that regulate neurodevelopment occur at dedicated time points to regulate the development of maturation of the brain. Disturbances of these processes can result in abnormal neurodevelopment or abnormal integration of neural networks. These abnormalities may eventually accumulate into (observable) deviations of behavior. One environmental cue for neurodevelopment may be the gut microbiome as colonization of the gut in early life coincides with a time of high neural remodeling in the brain. Indeed, early-life disturbances in microbial colonization have been associated with the development of behavioral alterations later in life (144, 220, 348).

Gut-brain axis

The microbiota-gut-brain axis encompasses several direct and indirect signaling routes. Certain microbial metabolites can cross the blood-brain barrier (BBB) and interact directly with neurons and/or glial cells. One of the most important bacterial metabolite classes to cross the BBB are SCFAs (Figure 1). However, harmful microbial substances, such as LPS, can also cross the BBB under certain conditions (329). Indirect pathways through which the gut microbiota can send signals toward the brain involve the vagus nerve, a component of the parasympathetic nervous system that sends information from the GI tract to the brain via afferent fibers. Although the luminal microbiota is physically separated from the vagal afferent terminals by the gut epithelium, microbial products can still directly or indirectly stimulate the vagal nerve activity (341, 346). For instance, SCFAs or neurotransmitters synthesized by gut microbes can diffuse across the gut epithelium and directly stimulate their respective receptors on vagal afferents. Alternatively, these microbial products can stimulate epithelial cells or immune cells in the gut wall to release mediators that will activate the vagal afferents (39). Upon activation, vagal afferent signals are relayed to the nucleus tractus solitarius, from where the signal is dispersed to the thalamus, hypothalamus, locus coeruleus, amygdala, and periaqueductal gray (11). Vagal stimulation is an important regulator of the 5-HT, gama-aminobutyric acid (GABA), and glutamate levels in the brain (278). However, vagal stimulation can also lead to the activation of the hypothalamic-pituitary-adrenal (HPA) axis and the release of stress hormones in the blood circulation. Certain brain regions, such as the hippocampus, are extremely sensitive to circulating glucocorticoids, and abnormal levels may impede normal functioning.

Alterations of gut microbiome in neurodevelopment and functions

Gut microbiota plays a crucial role in neurodevelopment, as demonstrated by preclinical studies on rodents. Alterations to the gut microbiome during pregnancy, neonatal stage, or adulthood of the rodents can lead to behavioral abnormalities in their offspring. These abnormalities include changes in social behavior, anxiety-, or depressive-like behavior. The maternal gut microbiota composition can also affect fetal neurodevelopment, with evidence showing that fetal neurodevelopment can be influenced by maternal immune activation, high-fat diet, or exposure to chronic stress during pregnancy (75, 318, 372). Because elevated pro-inflammatory cytokines and glucocorticoids can also influence fetal neurodevelopment, to further separate out the effect of microbes, researchers found that antibiotic-induced perturbations of the maternal gut microbiota can also lead to similar alterations in offspring behavior (75, 318, 372). The exact mechanisms that lead to these alterations are not yet fully understood but may involve changes in maternal 5-HT levels—which is regulated by gut microbiota and vital for fetal neuronal cell division, differentiation, and synaptogenesis (65) or increased BBB permeability—which can result in a pro-inflammatory phenotype in glial cells to interrupt neurodevelopment (37, 219). Depletion of maternal gut microbiota can also result in deficient axon outgrowth and altered sensory behavior in offspring (337).

Microglia play a critical role in early postnatal neurodevelopment. GF mice showed morphological and molecular abnormalities in microglia, which contributed to increased synapse density (due to lack of pruning) and decreased neural activity (219). Administration of butyrate or colonization with butyrate-producing bacteria can reverse microglial abnormalities, likely by shifting microglia toward an anti-inflammatory phenotype through inhibition of HDACs (300, 343). In addition, butyrate ensures proper BBB permeability via regulating tight junction expression, which promotes optimal neuronal growth and development (37).

While the neural network becomes less plastic in adulthood, gut microbiota becomes stable and diverse. Preclinical studies showed that changes in diet, chronic stress, or immune challenges could still induce anxiety- and depressive-like behaviors (120, 342). For instance, antibiotics-induced gut dysbiosis caused a change in neuronal activity in brain regions associated with processing of emotional responses (e.g., the anterior cingulate cortex, prefrontal cortex, and insular cortex) (Figure 1). Such abnormalities could be rescued by administration of specific probiotics (342). Gut dysbiosis in mice was linked to decreased expression of brain-derived neurotrophic factor (BDNF) and altered microglial activity in hippocampus (64), while butyrate supplementation effectively reversed cognitive impairment, likely via reducing the number of active pro-inflammatory microglia (256, 354). Indeed, in mice fecal microbiota, transplantation from young donors to aged recipients ameliorated age-associated cognitive impairment and reversed age-associated differences in brain immunity and hippocampal gene expression pattern (34). The opposite transplantation impaired spatial learning and memory, reduced the expression of proteins involved in synaptic plasticity and neurotransmission in the hippocampus, and shifted the microglia toward a pro-inflammatory phenotype in the young recipients (71).

Cardiovascular system

The main physiological functions of the cardiovascular system are to provide adequate circulation of blood through the body. Concomitantly, it must regulate blood pressure, vasodilation, and health of the blood vessels and arteries. CVD is the leading cause of death in the United States and continues to rise on a global level. The links between the gut microbiome and CVD are (i) production of specific metabolic toxic products, (ii) affecting blood pressure, (iii) modulating cholesterol levels, and (iv) inciting systemic chronic low-grade inflammation through regulating intestinal permeability and endotoxemia.

A series of large CVD-microbiome cohort studies have been reported. Metabolic aberrations linked to microbiome differences in acute coronary syndrome (ACS) were also observed in control individuals with metabolic impairment, suggesting that these aberrations may proceed the development of ACS (308). Similarly, for ischemic heart disease (IHD), it was found that about 75% of microbiome and metabolome features that distinguish individuals with IHD from healthy individuals are present in individuals exhibiting dysmetabolism (105). Besides dyslipidemia, another consistent metabolic aberration is elevated circulating branched-chain amino acid (BCAA) levels (257, 272). Interestingly, a gut symbiont that degrades BCAA (Parabacteroides merdae) was depleted in individuals with atherosclerosis (272).

Another metabolite consistently implicated in CVD is TMAO (193). TMAO can directly modulate cell inflammation by enhancing platelet reactivity and thrombosis potential to promote the development of atherosclerosis (377). TMAO is converted from trimethylamine (TMA) by flavin monooxidases 3 (FMO3), and FMO3 is regulated by FXR (285). Therefore, one link between TMAO and CVD is lipid metabolism through BA-induced activation of FXR (129). Indeed, mice lacking FXR and apolipoprotein E (APOE) had increased atherosclerotic plaque formation (accumulation of modified lipids) (124), while activation of FXR reduced plaques, enhanced lipid metabolism, and improved immune response upon LPS-induced inflammation (233). However, effects of TMAO and BA seem bidirectional as adding TMAO to a mouse diet notably changed BA profiles, inducing abnormal lipid metabolism, foam cell formation, and eventually atherosclerosis (87, 175, 347) (Figure 1).

BA signaling through FXR is also directly involved in CVD physiology. FXR stimulation promotes the transcription of endothelial nitric oxide synthase (eNOS) and sequential production of NO, which is a vasodilator essential for hemodynamics (194). In smooth muscle cells, FXR upregulates the expression of Ang II type 2 receptor (AT2R) to modulate vasodilatation in the resistance arteries (368) and suppresses inflammation via interlukin-1β (IL-1β)-induced NFκB, iNOS, and cyclooxygenase-2 as well as cell migration (201). Through FXR or other mechanisms, BAs also modulate cardiomyocyte contraction and electrophysiology (27, 177).

Functionally lower levels of SCFA-producing bacteria have been observed in patients with CVD (95). SCFAs such as butyrate have anti-inflammatory and antioxidative stress effects and suppress the hyperproliferation of vascular smooth muscle cells which drives the development of atherosclerosis (230). Butyrate also alleviates aorta constriction-induced cardiac hypertrophy (255) and suppresses the progression of hypertrophy (366). Beyond the roles of butyrate, propionate salvages mitochondrial membrane potential and sequentially normalizes cardiomyocyte contractile function (196), while acetate is a potent vasodilator of human colonic resistance arteries (239) and enhances mucosal blood flow (183).

Conclusions

In this article, we comprehensively summarized the colonization of various microorganisms along the GI and respiratory tracts, and their physiological functions in these two systems, as well as in the central nervous system and cardiovascular system, under health and diseases. In recent decades, culture-independent next-generation DNA sequencing technologies have transformed the field by facilitating the characterization of microbial communities in different niches in the body. This led to myriad studies describing differences in the microbial composition among different disease states and healthy individuals. Over the past few years, the field has slowly moved away from microbiome associations toward more mechanistic studies that are focused on understanding how microbes contribute to the pathogenesis of different disease states.

As with any new field, there are still significant challenges. As technology evolved, studies used different methods for sequencing and analyzing microbiome data, leading to significant heterogeneity and difficulties combining data across studies. Most of the human studies are cross-sectional, have small cohort sizes, and do not account for confounding factors in addition to dramatic variability in terms of methodologies, sampling sites, broad representation of demographics, etc. The field has been very gut microbiome centric, given the microbial diversity in the gut and the ease of sample collection, but as outlined in this article, there are increasing investigations into both the direct and indirect effects of the microbiome in other organ systems. There is also an increasing interest in other microbes such as fungi, viruses, and archaea, but this is currently hindered by the availability of robust databases. Despite these challenges, a vast amount of knowledge has been generated over the past decade, and this is fueling more mechanistic and translational research. The outlook for the future is optimistic, and over the next decade, we can expect significant advances in our understanding of specific microbial pathways that contribute to disease pathogenesis as well as treatment responses.

Didactic Synopsis.

Major teaching points

• There are different types of microorganisms colonizing the human body, including bacteria, viruses, fungi, protozoa, and archaea.

• Microbial metabolites, such as short-chain fatty acids, tryptamine, secondary bile acids, and trimethylamine N-oxide, produced from indigestible or residual carbohydrates, proteins, fats, etc., modulate physiological processes not only in the systems colonized with microbes but also the systems without microbial colonization, such as the central nervous system and the cardiovascular system.

• Individual microbes/metabolites can have distinct effects in different physiological systems.

• Different microbes/metabolites can have synergistic or antagonistic effects on individual physiological processes, and hence, the entire microbiome/metabolome should be considered.

• The heterogeneity in sequencing and analytical methods makes it difficult to compare and combine the study results, complicating the development of microbial-targeted therapies.