COMPOSITION AND MOLECULAR ANALYSIS OF THE ENTERIC MICROBIOTA

Most human enteric bacteria cannot be cultured, because of a lack of truly selective growth media. Nonetheless, molecular profiling has shown that whereas the microbiota appear distinct in different persons, the composition of each person’s microbiota is relatively stable after infant weaning and throughout adulthood. Evidence from studies of twins suggests that the individuality of human microflora may be genetically controlled,¹ but environmental variables including diet and sanitation appear to have profound effects on early intestinal colonization with bacteria.^2,3 In adulthood, dietary fluctuations appear to induce changes in bacterial enzymes and metabolic activity rather than changes in the relative populations of the microflora.^2,4,5

The composition of the microflora varies quantitatively and qualitatively over the longitudinal and the cross-sectional axes of the alimentary tract. Beyond the oral cavity, which harbors approximately 200 different bacterial species, the size and diversity of the microflora increase distally along the digestive tract (Fig. 102-1). Gastric acid restricts bacterial numbers within the stomach to fewer than 10³ colony-forming units (CFU)/mL. The gradient in bacterial density is greatest across the ileocecal valve, with approximately 10⁸ bacteria per gram of ileal contents and up to 10¹² bacteria per gram of colonic contents, comprising more than 1000 different bacterial species.^2,4–6 More than 99% of the culturable bacteria in the ileum and the colon are obligate anaerobes, but the composition of the flora at the mucosal surface differs from that within the lumen; ratios of anaerobes to aerobes are lower at mucosal surfaces. Culture-independent methods, such as the various molecular approaches described later, suggest that mucosa-associated bacteria differ from those recovered from feces, thus supporting the idea that host-related factors have a role in determining the enteric microflora⁷ and implying that bacterial aspirates from the lumen may be an incomplete reflection of mucosa-associated bacteria.

Figure 102-1. Composition and distribution of bacterial flora within the human alimentary tract.

The microbiota of the proximal small intestine consist predominantly of Gram-positive facultative bacteria—bacteria that can survive under aerobic or anaerobic conditions—although enterobacteria and Bacteroides species also may be present. Peristalsis is the principal factor restricting bacterial numbers in the small intestine. In the distal small intestine, the composition of the microflora resembles that of the colon, with a preponderance of Gram-negative anaerobes. The most prominently represented genera in the distal bowel include Bacteroides, Clostridium, Lactobacillus, Fusobacterium, Bifidobacterium, Eubacterium, Peptococcus, and Escherichia species.^2,6

The impact of diet on the composition of the enteric microflora has been studied by several investigators, with varying results. The consensus is that breast-fed babies have a greater proportion of bifidobacteria in their fecal flora than that observed in infants who are formula fed; high-carbohydrate and high-fiber diets are associated with increased bifidobacteria, whereas a high-fat diet is linked with an increase in Bacteroides species. It is the metabolic activity of the microbiota, however, rather than their composition, that exhibits the greatest response to dietary changes in a given person.

Detailed analysis of the enteric microflora has been confounded by the limitations of traditional culture-dependent microbiology.⁸ First, obtaining representative material from different niches within the intestine is problematic; because most of the indigenous microflora are obligate anaerobes, major methodologic difficulties are encountered with sampling, contamination, transport, and storage. Second, the lack of truly selective growth media precludes culture of most components of the microflora. In this respect, it is noteworthy that culture of H. pylori and C. difficile was accomplished only within the past two to three decades. This difficulty has led to a shift in emphasis from conventional bacterial phenotyping toward genotyping and molecular approaches to study the unculturables.^4,9,10

Bacterial nucleic acid extracted from feces or mucosal biopsy samples can provide a profile of the composition of the indigenous microflora. The small ribosomal subunit RNA (16S rRNA in bacteria) contains highly conserved regions of base sequences that reflect an absence of evolutionary change. These conserved sequences are interspersed with hypervariable regions, which contain mutational changes reflecting the evolutionary divergence of different species. Sequencing of 16S rRNA, therefore, represents a method for identification and phylogenetic classification of intestinal bacteria.

For rapid profiling of the dominant culturable and nonculturable organisms within a complex ecosystem such as that in the intestinal tract, 16S rRNA can be amplified by polymerase chain reaction (PCR) with universal primers spanning conserved and variable regions. The mixture of hypervariable RNA fragments can then be separated by a chemical denaturing gradient or a temperature gradient gel electrophoresis (DGGE and TGGE). Complete denaturation of the RNA fragments is prevented by incorporating a GC-rich 5′ end to one of the primers (a GC clamp).¹¹ Variations in migration distance through the denaturing gradients reflect the diversity of 16S species in the sample (Fig. 102-2). Theoretically, the technique is semiquantitative, because the more dominant the organism, the more abundant the specific PCR product. The specific PCR product can be cut from the gel and further amplified, cloned, and sequenced to identify individual bacterial strains without requiring a conventional culture step. Further refinements of the technique can be achieved by using species-specific PCR primers.

Figure 102-2. Representative examples of molecular fingerprinting profiles. Denaturing gradient gel electrophoresis analysis of the intestinal microbiota in two patients tested over time, at 0, 4, 8, and 12 months. Although the profile for each patient is unique, it also is stable.

Other molecular techniques for analysis of specific bacterial species now are possible because of the increasing availability of genomic sequence data for the major components of the bacterial flora. These techniques include fluorescence in situ hybridization (FISH) flow cytometry (FISH-flow) and bacterial DNA microarrays. In disorders such as inflammatory bowel disease, immunologic reactivity against components of the microflora has been used to identify microbes that may be etiologic in the disease. Marker antibodies generated by hybridoma or phage-display technology have been used as reagents to identify microbial antigens. For example, antineutrophil cytoplasmic antibody (pANCA), which is associated with ulcerative colitis, has been used to identify colonic bacteria expressing a pANCA-related epitope.^12,13 In addition, candidate microbes can be identified by the presence of unique bacterial nucleic acid sequences associated with a particular lesion or disease location, by subtractive cloning using genomic representational difference analysis. With this technique, a sequence representing a bacterial transcription factor from an apparent commensal organism, Pseudomonas fluorescens,^14–16 was found in lesions of Crohn’s disease but not in adjacent nonlesional mucosa. Serologic expression cloning also has been used to identify bacterial flagellin as a dominant antigen in Crohn’s disease.¹⁷

The new science of metagenomics—the sequencing of genes from whole microbial environments at once—promises to address many of the unresolved questions about the microbiota. The microbiota comprise a repository of genetic information (microbiome) that greatly exceeds that of the host genome. By combining metagenomics with bioinformatics, biochemistry, and traditional bioassays, new insights into the metabolic capacity of the human intestinal microbiota can be achieved.⁴ Major consortia (including the human microbiome project⁴) around the globe are under way using metagenomics as a tool for bioprospecting the intestinal microbiota in health and disease.

INTERACTIONS BETWEEN HOST AND MICROBES

The microbiota exert both positive and negative regulatory effects on the development and function of the intestine. These complex influences first were shown in comparative studies of germ-free and conventionally colonized animals. A sterile intestine is associated with reductions in mucosal cell turnover, digestive enzyme activity, cytokine production, lymphoid tissue, lamina propria cellularity, vascularity, muscle wall thickness, and motility, but with an increase in enterochromaffin cell area.¹⁸ The molecular events underpinning this regulatory signaling from the lumen currently are being explored using modern techniques such as laser capture microdissection and gene array analysis; such studies promise to reveal new molecular targets to be exploited for the design of novel therapeutics.^19,20 Thus, for example, when applied to animals colonized with only a single bacterial strain, Bacteroides thetaiotaomicron, this combined approach has illustrated the impact of bacteria-derived signaling on the expression of host genes controlling mucosal barrier function, nutrient absorption, angiogenesis, and development of the enteric nervous system.

Incoming bacterial signals include secreted chemoattractants, such as the formylated peptide f-Met-Leu-Phe, cellular constituents such as lipopolysaccharide (LPS) and peptidoglycans, flagellin, and bacterial nucleic acids (i.e., CpG DNA). Detection of bacterial stimuli by the host and discrimination of pathogens from commensals are mediated in part by pattern recognition receptors such as Toll-like receptors (TLRs) that are present on epithelial and immune (dendritic) cells. In health, engagement of TLRs by ligands from the commensal microbiota appears to be required for mucosal homeostasis.^21,22 Thus, not only are bacterial signals required for optimal mucosal and immune development, but they also actually are required to maintain and condition the mucosa for responses to injury.²²

The immune system mediates the sense of microbial danger and responses to injury. Although the primary lymphoid organs are developed at birth, mucosal immune functions require continual education and fine-tuning of cytokine balances and T-cell responses; this process is achieved by microbial colonization and sporadic mucosal infections. Without the microbiota, mucosal lymphoid tissue is rudimentary, and induction of mucosal immune responses and tolerance is suboptimal.^23,24

With a surface area similar to that of a tennis court (approximately 400 m²) and only one cell layer separating the internal milieu from the lumen, the enteric mucosa is well adapted to immunologic sampling of the intraluminal microbial community. Sampling of the microbiota across the epithelial barrier is mediated by M cells, which deliver particulate and microbial antigens to underlying immune cells, and by mucosal dendritic cells, which appear to extend processes into the lumen between the surface enterocytes without disrupting tight junctions.²⁵ It appears that intestinal dendritic cells can ingest and retain intact live commensal bacteria and then transit to the mesenteric lymph node, where immune responses to commensals are induced locally.²⁶ Thus, the mesenteric lymph node acts as a gatekeeper, preventing access of commensal bacteria to the internal milieu and protecting the host from harmful systemic immune reactivity. The immunosensory function of dendritic cells is facilitated by their plasticity and versatility of responses,²⁷ depending on whether they are presented with commensals or pathogens; moreover, they appear to exhibit tissue-specific specialization in the intestine.^23,24

In addition to specific immune responses to enteric bacteria, the surface epithelial cells serve a sensory function to detect microbial danger by producing chemokines that activate the host immune response and recruit it to any breach in the mucosal barrier caused by pathogenic infection.²⁸

Transduction of bacterial signals into host immune responses after engagement of TLRs may proceed along more than one molecular pathway. The transcription factor nuclear factor-κB (NF-κB) is the pivotal regulator of epithelial responses to invasive pathogens, but nonpathogenic bacteria can attenuate inflammatory responses by delaying the degradation of IκB, which is counter-regulatory to NF-κB.²⁹ Other signal transduction pathways are likely to emerge to account for the anti-inflammatory effects of probiotics and other commensal organisms such as Bacteroides thetaiotaomicron. This anaerobic commensal can antagonize the proinflammatory effects of NF-κB within the epithelial cell by enhancing the nuclear export of its transcriptionally active subunit (RelA) in a peroxisome proliferator-activated receptor-γ (PPAR-γ)–dependent manner.³⁰

METABOLIC ACTIVITY OF THE MICROBIOTA

The enteric microbiota are tantamount to a hidden metabolic organ (Table 102-1). Although our understanding of indigenous bacterial metabolites is still superficial, coevolution with this living inner mass of bacteria has several apparent benefits for the host. In addition to the production of regulatory signals for mucosal homeostasis as discussed earlier, the microbiota exhibit important metabolic properties not possessed by the host. These include biotransformation of bile acids; degradation of oxalate; breakdown of otherwise indigestible dietary components, such as plant polysaccharides; and production of short-chain fatty acids, a major energy source for colonic epithelium, from fermentable carbohydrates. Other activities include synthesis of biotin, folate, and vitamin K.^2,20 Clinicians also have exploited enteric bacterial enzymes such as azoreductase to convert prodrugs such as sulfasalazine to active drug metabolites (e.g., aminosalicylate). Other examples of bacterial action on drug bioavailability include the metabolism of l-dopa to dopamine and degradation of digoxin. Not all of the metabolic changes induced by the enteric microbiota are beneficial to the host, however, and although bacteria probably degrade some carcinogens, they also might promote the production of carcinogens from dietary procarcinogens.³¹

Table 102-1 Examples of Metabolic Activities of Intestinal Microbiota

A striking example of the importance of bacterial metabolism is exemplified by the regulatory effect that the enteric microbiota exert on fat storage.³² It has long been known that germ-free animals need a significantly greater caloric intake to sustain a body weight similar to that of normal colonized animals. Thus, the normal host-microbiota relationship has nutritional benefit, in contrast with the negative nutritional effect associated with bacterial overgrowth syndromes. Elegant studies with germ-free mice have shown that upon colonization, body weight increases despite a reduced caloric intake. The bacteria in the microbiota colonizing the intestine promote storage of dietary calories in fat by increasing absorption of monosaccharides and suppressing epithelial-derived fasting-induced adipocyte factor (FIAF).³² Thus, the composition and activity of the intestinal microbiota should be considered as a diet-influenced variable that can influence susceptibility to obesity.

One of the outcomes of bacterial metabolic activity is gas production. Of the five gases—N₂, O₂, CO₂, H₂, CH₄—that constitute 99% of flatus, the latter three are produced by the enteric bacteria, and bacteria are the sole source of hydrogen and methane in the intestine. Hydrogen production by bacterial action on carbohydrates, and to lesser extent on protein, normally occurs in the colon. In patients with small intestinal bacterial overgrowth, however, the small intestine also becomes a site of H₂ production. Bacterial methanogens occur in the colon and produce methane from H₂ and CO₂, with significantly detectable excretion in approximately 30% of humans.^33–37 The principal gases produced are odorless, but bacterial metabolism also is responsible for producing various trace and odiferous gases in flatus such as hydrogen disulfide.^38,39 Qualitative and quantitative variability in gas production with diet illustrates the fluctuations in bacterial metabolic activity despite the apparent stability of the microbiota in adulthood.

SMALL INTESTINAL BACTERIAL OVERGROWTH

Small intestinal bacterial overgrowth (SIBO) is characterized by malabsorption and overgrowth of bacteria in the small intestine. The syndrome often is referred to as blind loop syndrome because of recognition of the disorder in patients with predisposing anatomic abnormalities. Other terms that have been used to describe the disorder include stagnant loop syndrome, contaminated small bowel, small intestinal colonization, and small bowel stasis. In 1939, Barker and Hummel⁴⁰ reported macrocytic anemia in association with intestinal strictures and anastomoses and postulated that the anemia was secondary to bacterial overgrowth, or “putrefaction.” SIBO is not confined to humans and is well recognized in dogs.⁴¹ The syndrome is associated with a variety of anatomic disturbances, such as blind loops,⁴² and motility disorders, such as scleroderma,⁴³ but it can occur in the absence of any specific predisposing factor. It is likely that the condition is underdiagnosed, particularly in the elderly.⁴⁴

Patients with SIBO do not necessarily present with a florid malabsorption syndrome, and symptoms may be minor and nonspecific. Considerable debate has concerned the relationship between irritable bowel syndrome (IBS) and SIBO (see Chapter 118).⁴⁵ Bacterial overgrowth has been documented in asymptomatic elderly persons in the community,^46,47 in whom it is debatable whether the phenomenon is of any significance⁴⁸; in the absence of malabsorption or related symptoms, such overgrowth probably should not be considered to represent true SIBO. Asymptomatic SIBO may be termed simple colonization and probably results from achlorhydria and abnormal fasting intestinal motility (see later).

The diagnosis of SIBO usually is made by noninvasive breath testing,⁴⁹ even though studies on the accuracy of these tests report very variable results. For this reason, culture of a small intestinal aspirate must be regarded as the diagnostic gold standard. Unfortunately, much of the published literature on SIBO is based on breath tests, rather than on culture, and the findings must be interpreted with caution.