The role of the gut flora in health and disease, and its modification as therapy

Aliment Pharmacol Ther 2002; 16: 1383–1393.
Review article: the role of the gut flora in health and disease,and its modification as therapy A . L . H A R T * ,   , A . J . S T A G G   , M . F R A M E à , H . G R A F F N E R à , H . G L I S E à , P . F A L K à & M . A . K A M M **St Mark’s Hospital, Harrow, Middlesex, UK;  Antigen Presentation Research Group, Imperial College School of Medicine,Northwick Park Institute of Medical Research, Harrow, Middlesex, UK; àAstraZeneca R & D Mo¨lndal, Mo¨lndal, Sweden Studies of single organism–epithelial interactions have revealed the large range of metabolic processes that The gut flora is a vast interior ecosystem whose nature gut bacteria may influence. In inflammatory bowel is only beginning to be unravelled, due to the diseases, bacteria drive the inflammatory process, and emergence of sophisticated molecular tools. Techniques genetic predisposition to disease identified to date, such such as 16S ribosomal RNA analysis, polymerase as the recently described NOD2/CARD15 gene vari- chain reaction amplification and the use of DNA ants, may relate to altered bacterial recognition. Extra- microarrays now facilitate rapid identification and intestinal disorders, such as atopy and arthritis, may characterization of species resistant to conventional also have an altered gut milieu as their basis. Clinical culture and possibly unknown species. Life-long cross- evidence is emerging that the modification of this talk between the host and the gut flora determines internal environment, using either antibiotics or pro- whether health is maintained or disease intervenes. An biotic bacteria, is beneficial in preventing and treating understanding of these bacteria–bacteria and bacteria– disease. This natural and apparently safe approach host immune and epithelial cell interactions is likely to lead to a greater insight into disease pathogenesis.
the disease areas in which bacterial modification has been shown to be of therapeutic benefit.
There has been a re-emergence of interest in therelationship between the gastrointestinal flora and gut function, fuelled in part by the recognition of thepotential value of probiotics and other means of The flora of the gastrointestinal tract is complex and modifying gut flora as therapeutic modalities.
should be considered as a functionally active organ, the This review aims to consider our existing knowledge of full potential of which remains to be elucidated.1–4 Each the nature of the gut flora and new techniques that of us has about 1012 viable bacteria per gram of large allow enhanced characterization of this flora. It high- bowel content, which is equivalent to more bacteria in lights emerging methods that allow bacteria–epithelial one person’s gut than there have ever been humans on interactions to be evaluated. Finally, it considers some of the planet. This ecosystem is vast, both quantitativelyand in terms of diversity, with the presence of at least400–500 different species.5 A single organism, for Correspondence to: Professor M. A. Kamm, St Mark’s Hospital, Watford example Escherichia coli, is known to have huge meta- Road, Harrow, Middlesex, HA1 3UJ, UK.
E-mail: [email protected] bolic capability. Extrapolating from this, the metabolic capability of the entire gastrointestinal flora is immense.
encouraged a renewed interest in the analysis of the The magnitude of this organ is underlined when we consider that the total number of genes contained in the The techniques that have found most use to date microflora is about 50–100 times that in our ÔownÕ employ analysis of the small ribosomal subunit RNA, genome.6 A further level of complexity is the temporal 16S rRNA in the case of bacteria.9 16S rRNA contains and spatial diversity of the flora. The bacterial distribu- hypervariable regions that contain signatures of phylo- tion varies greatly at different levels of the gastrointes- genetic groups or species. Polymerase chain reaction tinal tract,7 ranging from < 103 colony-forming units/ (PCR) amplification using universal primers allows the mL (CFU/mL) in the stomach, where the number of identification of bacterial species by sequencing of ingested bacteria is dramatically reduced by contact hypervariable regions or by the less laborious approach with gastric acid, to 1011)1012 CFU/mL within the of denaturing gradient gel electrophoresis (DGGE).10 colon, where anaerobes outnumber aerobes by a ratio of Alternatively, species-specific primers from the hyper- 1000 : 1. The species composition varies as we develop variable sequences can allow the direct identification of and age, and is influenced by our environment.
particular organisms. There are currently in excess of Methods of identifying the gastrointestinal flora are 16 000 16S rRNA sequences stored in databases and limited by this complexity, the inaccessibility of parts of the development of bioinformatic tools required to the ecosystem and the temporal and spatial diversity.
efficiently exploit such information is paralleling the Furthermore, the definition of ÔnormalityÕ of the gut advances in molecular biology and genetics.11 flora needs to be clarified. How should normality be These molecular techniques are very powerful but defined, in terms of different sites in the gut, different they do have limitations. For instance, DNA/RNA may conditions and the population under evaluation? Much not be extracted with equal efficiency from all bacteria; of the work has focused on Western subjects with some techniques, such as PCR/DGGE, may not be very known diagnoses, many of whom are on medications sensitive; and PCR amplification, particularly for large or have received anaesthetics or antibiotics. It is numbers of cycles, may introduce mutation artefacts.
difficult to extrapolate from this to healthy individuals Future developments of identification technologies world-wide and to encompass the extremes of age.
will probably include the use of high-density DNA Furthermore, most of the studies undertaken to define microarrays as an alternative to PCR/DGGE for high- the normal flora have used faecal samples, which may throughput analysis, and real-time PCR analysis for not be representative of the mucosal microenviron- quantitative assessment of mRNA levels. In addition, ment. There are technical problems of sampling, microarray analysis will be combined with matrix- transport and storage, in addition to cultivation, assisted laser desorption ionization time-of-flight mass enumeration and identification, all of which may cause spectrometry. Using such technology, it is possible to changes in the flora.8 Much of the work published determine the DNA sequence of an 80-base pair over the last 40 years needs to be interpreted with fragment present in an individual array spot in milliseconds and make the transition from genomicsto proteomics.
The functional definition of bacterial populations has primarily involved immunochemical strategies. Anti- New methods of identification have enabled a renewed body responses to local immunogenic microbial cells assessment to be made of the gut microflora. The can be useful to identify microorganisms participating traditional microbiological techniques of culture and in the disordered environment. Expression cloning using antibody targets and subtractive cloning using genomic analysis. This has been driven by the fact that representational difference analysis, cDNA suppression traditional methods are unsuitable for high-throughput analysis and there is the added problem that a large hybridization offer functional methods of identifying proportion of the gut microbial residents are non- culturable by traditional techniques. Techniques that In a period of about 30 years, the identification of can detect and semi-quantify both culturable and non- intestinal bacteria has developed a long way from culturable gut bacteria are now available and have Ôculture and phenotypeÕ to Ôno culture, just genotypeÕ.
Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 R E V IE W : G U T F L O R A IN H E A L T H A N D D I S E AS E There is a great deal of evidence supporting the view Translocation or penetration of intestinal bacteria into that what is generally considered as normal gut the underlying tissue or mesenteric lymph nodes can be structure and function is the end-point of a complex demonstrated in animals and humans,13 but occurs set of interactions between the host and microorgan- without serious consequence in most healthy individ- isms colonizing the gut. Such normal features include uals. The acquired immune response appears to play only a minor role in limiting bacterial penetration.
mitotic activity, villous length and crypt depth. This Although translocation is increased in T-cell-deficient has been known for decades from observations in mice,14, 15 these animals remain healthy. Similarly, germ-free animals.3 The molecular basis for these there is evidence for increased translocation in immu- host–flora interactions is now being studied. For noglobulin A (IgA)-deficient mice,16 but commensal- example, a prototypic commensal, Bacteroides thetaio- related sepsis is not observed in IgA ÔknockoutÕ mice or taomicron, has been used to colonize germ-free mice, in IgA-deficient humans. In contrast, gp91phox–/–/ and the effect on epithelial cell gene expression has been assessed by DNA microarrays and laser capture impaired antimicrobial activity due to the inability to microdissection. This commensal has been shown to make both reactive oxygen and reactive nitrogen influence many genes dictating nutrient absorption, intermediates, develop massive abdominal abscesses containing enteric commensals.17 On the C57BL/6 angiogenesis, the enteric nervous system and postna- genetic background, all such mice die by the age of tal intestinal maturation. Furthermore, different spe- about 4 weeks despite the use of prophylactic antibiot- cies have been shown to elicit different changes in ics. Mice with a deficiency in only one enzyme pathway gene expression and to participate in different phy- (gp91phox–/– mice or NOS2–/– mice) do not develop siological functions.12 This is a reductionist approach abscesses. These results demonstrate the importance of and it will be the cumulative summation of the entire phagocyte killing for limiting the consequence of flora that will determine the end result. Nonetheless, bacterial translocation from the gut, and illustrate it makes it possible to use in vivo systems to quantify redundancy in the cellular pathways involved. They the impact of a microbial population on host cell gene also indicate a constant exposure of host tissues to bacteria and their products, which occurs despite afully functional acquired immune system.
Adaptation of the immune system to the normal flora A relationship between the normal flora and the host Whilst not playing the dominant role in the control of immune system exists with a mutual dependency bacterial penetration from the intestine, it is clear that between the two. Successful coexistence with a com- those parts of the host’s immune system that mediate plex microflora presents a particular challenge to the acquired responses are not ignorant of the gut immune system of the host. On the one hand, the host microflora. Comparison of germ-free animals with needs to avoid an overly aggressive response to this those with a normal flora clearly demonstrates adap- microbial population that would lead to the elimin- tation of the immune system to the presence of ation of beneficial organisms and would almost bacteria. Germ-free animals have reduced numbers of certainly result in inflammation and extensive tissue lamina propria or intra-epithelial T cells,18 and these damage. On the other hand, the capability to limit the are increased following the restoration of a normal spread of bacteria from the lumen into underlying microflora.19, 20 The turnover of lamina propria T tissues and to mount an effective response to intestinal cells, but not intra-epithelial T cells, is estimated to be pathogens needs to be maintained. A generalized rapid,21 and this process of lymphocyte activation, hyporesponsiveness in the intestine would leave the homing and death may be flora driven. Peyer’s patches host highly vulnerable to a wide range of intestinal are small in germ-free animals, but cryptopatches, tiny sites of lymphopoiesis in the crypt lamina propria of Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 the intestine,22 do not appear to require a flora for their development.23 Levels of secretory IgA andnumbers of mucosal IgA-producing plasma cells are The presence of large amounts of immunogenic and both low in germ-free mice,12, 16 but they increase pro-inflammatory molecules in the intestinal lumen and rapidly upon introduction of even a single commensal the evidence for the interaction of the host’s immune bacterial species, such as Morganella morganii24 or system with this material have led to the concept that the gut mucosa is in a state of restrained immune Some of the prodigious IgA production that occurs in reactivity driven by the intestinal flora. This has been the intestine of non-germ-free mice has no obvious termed ÔcontrolledÕ or ÔphysiologicÕ inflammation. It is specificity for microbial (or food) antigens. This ÔnaturalÕ clear that such a restrained response must be very antibody may result from polyclonal activation by tightly regulated. In support of this concept, many microbial products, such as lipopolysaccharide. How- knockout mouse strains, in which the genes involved in ever, some of the IgA can be shown to have specificity immune regulation are targeted (e.g. interleukin-10, for components of the bacterial flora. Recent work T-cell receptor a or b chain, interleukin-2, interleukin-2 suggests that at least some of this bacteria-specific IgA is receptor), develop intestinal inflammation.28 Similarly, produced, in a T-cell-independent fashion, by the B1 perturbations in T-cell development or lymphocyte lineage cells in mice, derived from the pleuroperitoneal subpopulations can also result in inflammatory gut cavity.16 B1 cells, characteristically expressing CD5 and disease. Where tested, this inflammation was shown to high levels of IgM, arise early in ontogeny and reside originally in the peritoneal and pleural cavities. Cur- Control of this highly regulated response is probably rently, it is not clear whether intestinal B1 cells are achieved by multiple mechanisms acting at different activated directly, perhaps by multimeric bacterial levels; a network of interactions between bacteria, the antigens, or whether other cell types, such as antigen- immune system, epithelial cells, stromal elements and presenting cells, are involved. The contribution of B1 components of the nervous system is likely to be involved.
cells to human IgA responses is unclear. Another From the perspective of the immune system, it is often unanswered question is whether the intestinal flora argued that the gut displays an inherent bias towards influences the primary repertoire of B cells — that is the anti-inflammatory (Th2) or regulatory (Th3/Tr) T-cell constellation of binding specificities available in a naive responses, with the production of ÔregulatoryÕ cytokines such as interleukin-10 and transforming growth factor- One of the special features of immunoregulation at b. This concept has its origins in studies on the massive mucosal sites is the ability of orally administered IgA production associated with the gut and on the soluble antigens to induce a profound state of systemic propensity of T cells from gut-associated lymphoid tissue to support IgA responses. It has been adapted to include non-responsiveness, also termed oral tolerance, is concepts of Th1 vs. Th2/Th3/Tr responses, as our presumably related to physiological responses invoked understanding of functional T-cell subsets has developed.
by dietary proteins and perhaps commensal bacteria.26 However, a simple model of a default pathway away from The mechanisms that underlie oral tolerance remain Th1 responses is clearly an oversimplification. Inter- somewhat controversial, but could include active feron-c is produced in response to fed antigens, even cytokine regulation, anergy or clonal deletion; differ- when they are administered at tolerogenic doses,29 and ent mechanisms are probably favoured at different human Peyer’s patch cells display a T-cell type 1 cytokine antigen doses. Some aspects of oral tolerance are profile in response to dietary antigens.30 Indeed, a bias deficient in germ-free mice, and this deficiency can be away from Th1 responses may be a feature of mice, but reversed by reconstitution with Bifidobacterium infantis not humans, and it could be a consequence of a reduced in the neonatal period;27 reconstitution of adult mice flora (quantitatively and qualitatively) in young, specific is not effective. These data suggest that the presence of bacterial flora during the neonatal period is Despite the complexities of Th cell differentiation in required for the development of a properly regulated the intestine, animal models provide convincing evi- immune system in which oral tolerance can be fully dence that regulatory T-cell populations are involved in restraining the response to the intestinal flora.32, 33 Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 R E V IE W : G U T F L O R A IN H E A L T H A N D D I S E AS E Local immunosuppression by these cells is mediated by genicÕ dendritic cell subpopulation. The dendritic cells interleukin-10 and transforming growth factor-b and bearing apoptotic epithelial cells could be a candidate is dependent on signalling through the cytolytic T for such a population. Additional mechanisms or lymphocyte-associated antigen-4 receptor for members combinations thereof are also possible.
of the B7 family of co-stimulatory molecules. The It is important to emphasize that dendritic cells do not conditions that favour the induction of these regula- function in isolation and will be influenced by signals tory T cells are poorly understood, but at least some other than the bacteria themselves. Prime amongst regulatory T cells are present in germ-free animals, such signals will be those provided by epithelial cells, suggesting that they are not specific for, or dependent which can produce a range of cytokine and chemokine on the presence of, components of the flora.
signals from the basal surface in response to interac- Dendritic cells are one of the first cell types of the tions with bacteria at their apical surface. Enteric immune system to come into contact with components epithelial cells act as sensors, particularly for the of the flora, and are likely to play a crucial role in presence of pathogenic organisms, and signal the onset shaping the type of response that develops. Gut dendritic of mucosal inflammation. Epithelial signalling by a cells can acquire antigens such as ovalbumin when number of different pathogens appears to converge on these are fed to an animal and, upon isolation, these activation of the transcription factor NF-jB, and some dendritic cells can stimulate ovalbumin-specific T non-pathogenic components of the flora appear to cells.34 Sampling of the luminal contents by dendritic attenuate these pro-inflammatory responses by block- cells could occur by several routes. Firstly, dendritic cells ing the degradation of the counter-regulatory factor in the subepithelial dome of Peyer’s patches could IjB.42 Thus there may be fundamental differences in acquire materials transported across M cells. Secondly, the ways in which pathogens and non-pathogens recent evidence suggests that dendritic cells in the interact with enterocytes. Chemokine production by lamina propria are able to interdigitate between epithe- gut epithelial cells is, at least in part, likely to lial cells and directly sample luminal contents;35 orchestrate the recruitment and spatial distribution of expression of tight junction proteins by the dendritic cells of the immune system. For instance, constitutive cells ensures that mucosal integrity is preserved.
production of thymus-expressed chemokine (TECK)/ Thirdly, some gut dendritic cells appear to contain CCL25 by small intestinal epithelium preferentially apoptotic epithelial cells,36 and this may represent an recruits a4b7+ T cells, expressing its cognate receptor indirect pathway by which dendritic cells sample CCR9, to the small bowel.43 In contrast, colonic materials derived from the lumen that have been taken epithelial cells do not express TECK/CCL25, but up by the epithelial cells. Dendritic cells may also take up materials that have been transported across epithe- CCL28, whose ligand is CCR10. Thus, immune cells lial cells or gained access through transient breaches in expressing CCR10 may be preferentially recruited to the mucosal integrity. Recognition and response to bacterial products by dendritic cells will be determined by It is evident from the above discussion that the molecules such as pattern recognition molecules and selective recruitment of cells and their spatial organiza- toll-like receptors that are expressed on the surface of tion, as well as their function upon contact with dendritic cells.37 It is conceivable that dendritic cells bacterial products, will contribute to the tightly regu- may also express the bacteria-sensing molecule NOD2, lated response to the intestinal flora.
genetic variants of which have recently been reported tobe associated with Crohn’s disease.38, 39 T H E R O L E O F T H E F L O R A IN D I S E A S E In the steady state, dendritic cells may play a part in mediating non-responsiveness to intestinal flora, but We need only consider the impact that the eradication can be induced to stimulate a response by cytokines that of Helicobacter pylori has had in the management of signal tissue damage or the presence of a pathogen.40 peptic ulcer disease to see the potential for the enteric Non-responsiveness could be induced by constitutive antigen presentation by immature dendritic cells, With regard to inflammatory bowel disease, the resulting in the induction of anergic or regulatory aetiology appears to involve three overlapping elements: populations,41 or by activation of T cells by a Ôtolero- genetic susceptibility, priming by an environmental Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 factor, which appears to be part of the normal enteric inflammation,55 but are influenced by the host genetic flora, and immune-mediated tissue injury. The genetic susceptibility may determine the dysregulated immune Secondly, evidence comes from studies of patients response, a leaky mucosal barrier or even an imbalance with defects in phagocyte killing of bacteria, such as in the enteric flora.45 It is not clear whether the barrier children with chronic granulomatous disease or gly- function is primarily compromised by intrinsic defects in cogen storage disease type 1b. A subgroup of these epithelial integrity, by infection with pathogens, or by children develops a disease similar to Crohn’s disease.
the loss of or changes in commensal-dependent signals When the phagocyte defect is treated, the gut lesions The search for a single organism as the causative Thirdly, it has been demonstrated that T cells from the agent has yielded inconsistent results. Entero-adherent lamina propria of Crohn’s disease patients respond strains of Escherichia coli in the ileal mucosa have been in vitro to the antigens of their own flora with a Th1 found in patients with Crohn’s disease.46, 47 More polarized response,58 indicating a loss of tolerance to recently, a novel bacterial sequence has been reported their normal flora. This is reflected by a serologic in association with intestinal lesions in patients with Crohn’s disease. This sequence represents a novel between autoantibodies, such as antineutrophil cyto- bacterial transcription factor which, through genomic plasmic antibodies and enteric bacterial antigens.59, 60 homologue evaluation, appears to originate from Pseu- In addition, diversion of the faecal stream after surgery domonas fluorescens.48 However, whether these changes are primary or secondary to the disease process is not re-anastomosis results in inflammation.61–64 Further- more, this effect has been demonstrated directly by Instead, it may be that environmental factors are infusion of luminal contents into excluded ileum.65, 66 exerting their effects via the immune system. Indeed, The lesions occur in areas of the bowel with the highest the increasing prevalence of inflammatory bowel bacterial counts, the terminal ileum and colon, and in disease parallels an increase in atopy, asthma and ileal pouches formed after colectomy in patients with insulin-dependent diabetes mellitus, all of which are ulcerative colitis. Modulating the luminal bacteria with immunologically mediated.49 Differences in the neona- antibiotics has been beneficial in pouchitis67 and tal gut microflora precede the development of atopy, suggesting a role for the balance of intestinal bacteria in Finally, the recent identification of defective variants of a gene, NOD2/CARD15 on chromosome 16, that There is increasing evidence that the antigens that predisposes to Crohn’s disease is an encouraging drive the tissue-damaging response are derived from the development.38, 39 The function of the protein encoded normal bacterial flora. Proof of this concept has come is partly known, detecting bacteria and playing a role in from studies of at least 11 models of inflammatory the inflammatory response to them. NOD2/CARD15 bowel disease. In these models, inflammation is depend- seems to function as an intracellular receptor for ent on the presence of a normal flora and in its absence lipopolysaccharide via a leucine-rich repeat domain there is no disease. The phenomenon is seen in different and is involved in the regulation of the NF-jB pathway, species (mice, rats and guinea pigs) and occurs in which is the master modulator of the genes involved in manipulated systems, such as knockout or transgenic inflammation. It is suggested that the NOD2/CARD15 animals, as well as in induced models, such as gene product confers susceptibility to Crohn’s disease by indometacin- or carrageenan-induced colitis. Moreover, altering the recognition of bacterial components and/or in some models, colonization with normal flora rapidly by altering the activation of NF-jB in monocytes. This results in T-cell-mediated gut inflammation,51 and clearly points to a link between microorganisms that disease can be transferred with effector T cells against could be recruited from the intestinal microflora and the enteric bacteria.52, 53 It appears that a constant bac- innate immune system in triggering episodes of Crohn’s terial stimulus is needed for the induction and perpetu- ation of inflammation,54 and that all commensal Taken together, the evidence for bacterial flora driving bacteria are not equal in their ability to induce an inflammatory process is compelling.
Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 R E V IE W : G U T F L O R A IN H E A L T H A N D D I S E AS E A different approach to altering the balance of the M OD I F I C AT I O N O F G U T M I C R O F L O R A — beneficial and harmful bacteria in the gut is achieved through dietary supplementation. A prebiotic is a non- The role of the flora in disease has led to an interest in digestible food ingredient that beneficially affects the the use of probiotics as a therapeutic modality. Probi- host by stimulating the growth or activity of a genus of otics are defined as Ôliving organisms, which upon bacteria, such as lactobacilli or bifidobacteria, that can ingestion in certain numbers, exert health benefits improve the host’s health. Fructo-oligosaccharides have beyond inherent basic nutritionÕ. An added stimulus in a specific colonic fermentation directed towards bifido- this area has been the relentless use of antibiotics and bacteria and their activity has been confirmed in the associated spread of antibiotic resistance.
laboratory and human trials.87–89 As prebiotics exploit The most extensively studied gastrointestinal condition non-viable food ingredients, their applicability in diets is treated with probiotics is acute infantile diarrhoea.
wide ranging. The challenge will be to achieve finer Lactobacillus GG decreases the duration of diarrhoea in control over the microflora and modulate at the level of infants with acute rotavirus diarrhoea,72 a result that the species rather than the genus. A further approach is has been confirmed in similar73 and different74 popu- synbiotics,90 where probiotics and prebiotics are com- lation groups and in multicentre studies.75 In addition, bined and, indeed, probiotics could be encapsulated Lactobacillus GG is efficacious in the prevention of within a prebiotic coating to enhance targeted distal There is increasing evidence from animal models and The ingestion of large numbers of viable bacteria human studies that probiotics may be an important requires an assurance of safety. If bacteria are used treatment in inflammatory bowel disease. In particular, which are derived from the normal flora, their natural the administration of Lactobacillus reuterii was effective presence attests to their safety. Such bacteria have in ameliorating acetic acid-induced colitis77 and met- rarely caused disease through translocation, and their hotrexate-induced colitis78 in rats. More recently, safety record through use in fermented milk, vegeta- Lactobacillus sp. were able to prevent the development bles and cereals is excellent. In the case of genetically modified organisms, such as the Lactococcus lactis mice,79 and continuous feeding with Lactobacillus engineered to locally produce interleukin-10, the most plantarum attenuated an established colitis in the same important safety factor is the content of antibiotic resistance markers in the genetic constituents of the In humans, definitive evidence indicating the efficacy modified organisms. Another issue to consider is that a of probiotics is emerging. A non-pathogenic strain of genetically modified component of the flora may Escherichia coli has been reported to have efficacy acquire a growth advantage that will disturb the equivalent to that of mesalazine in preventing relapse homeostasis in the gut ecosystem. It is already possible in patients with ulcerative colitis.81–83 Perhaps the most to assess these risks, for example, by using suicide impressive evidence for the efficacy of probiotics in vectors to limit the proliferation of these microbial inflammatory bowel disease has been the maintenance of remission and prevention of onset of pouchitis with amixture of eight bacterial strains.84, 85 The high concentrations and multiple species of bacteria usedmay be more effective in dealing with the heterogeneity Life-long cross-talk occurs between the host and the of disease than single strains. Nonetheless, the import- intestinal flora and the outcome of this can determine ance of the careful selection of single strains with whether health is maintained or disease intervenes. Our particular physical, biochemical, genetic and immuno- knowledge of the intestinal flora in health is limited, but logical properties is clear, and their role in treating a combination of sophisticated molecular techniques, in conjunction with traditional microbiological methods Recently, the introduction of genetically engineered and an appreciation of their limitations, will pave the organisms to produce and deliver cytokines or other way to a greater understanding of this complex organ, biologically relevant molecules to the mucosa offers which can then be explored in disease states. An under- standing of the bacteria–bacteria and bacteria–host cell Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 interactions in the gastrointestinal tract will provide us 10 Muyzer G, Smalla K. Application of denaturing gradient gel with potential ways of modulating the flora.
electrophoresis (DGGE) and temperature gradient gel electro-phoresis (TGGE) in microbial ecology. Antonie van Leeuwen- The intestinal flora appears to have a role in driving inflammation. It remains to be seen which components 11 Thornton JM. From genome to function. Science 2001; of the flora are important in this process. Indeed, it may be that unique bacterial antigens are associated with 12 Hooper LV, Wong MH, Thelin A, Hansson L, Falk PG, Gordon disease in hosts with different genetic backgrounds.
JI. Molecular analysis of commensal host–microbial relation- Manipulating such a complex ecosystem is challen- ships in the intestine. Science 2001; 291(5505): 881–4.
13 O’Boyle CJ, MacFie J, Mitchell CJ, Johnstone D, Sagar PM, ging. Therapeutic interventions are superimposed on a Sedman PC. Microbiology of bacterial translocation in complex background with a high noise to signal ratio.
Nonetheless, there is tantalizing evidence from animal 14 Gautreaux MD, Gelder FB, Deitch EA, Berg RD. Adoptive and human studies that such manipulation may be transfer of T lymphocytes to T-cell-depleted mice inhibits effective in treating disease. More evidence-based medi- Escherichia coli translocation from the gastrointestinal tract.
Infect Immun 1995; 63(10): 3827–34.
cine is needed, but the proof of principle has led to a 15 Owens WE, Berg RD. Bacterial translocation from the gas- requirement to establish the mechanism of action, so trointestinal tract of athymic (nu/nu) mice. Infect Immun that treatments can be targeted and novel innovative 16 Macpherson AJ, Gatto D, Sainsbury E, Harriman GR, Hen- gartner H, Zinkernagel RM. A primitive T cell-independentmechanism of intestinal mucosal IgA responses to commensal bacteria. Science 2000; 288(5474): 2222–6.
17 Shiloh MU, MacMicking JD, Nicholson S, et al. Phenotype of This review includes highlights of a meeting held in mice and macrophages deficient in both phagocyte oxidase April 2001, entitled ÔGut Ecology — To Explore the Role and inducible nitric oxide synthase. Immunity 1999; 10(1): of the Gut Flora in Maintaining Health and in Disease and its Modification as a TherapyÕ. The meeting was 18 Guy-Grand D, Griscelli C, Vassalli P. The mouse gut T lym- phocyte, a novel type of T cell. Nature, origin, and traffic in mice in normal and graft-versus-host conditions. J Exp Med1978; 148(6): 1661–77.
19 Imaoka A, Matsumoto S, Setoyama H, Okada Y, Umesaki Y.
Proliferative recruitment of intestinal intraepithelial lympho- 1 Berg RD. The indigenous gastrointestinal microflora. Trends cytes after microbial colonization of germ-free mice. Eur J 2 Mackowiak PA. The normal microbial flora. N Engl J Med 20 Umesaki Y, Setoyama H, Matsumoto S, Okada Y. Expansion of alpha beta T-cell receptor-bearing intestinal intraepithelial 3 Midtvedt T. Microbial functional activities. In: Hanson LA, lymphocytes after microbial colonization in germ-free mice Yolken RH, eds. Intestinal Microflora, Nestle Nutrition Work- and its independence from thymus. Immunology 1993; 79(1): shop Series. Philadelphia: Lippincott-Raven, 1999: 79–96.
4 Falk PG, Hooper LV, Midtvedt T, Gordon JI. Creating and 21 Poussier P, Edouard P, Lee C, Binnie M, Julius M. Thymus- maintaining the gastrointestinal ecosystem: what we know independent development and negative selection of T cells and need to know from gnotobiology. Microbiol Mol Biol Rev expressing T cell receptor alpha/beta in the intestinal epithe- lium: evidence for distinct circulation patterns of gut- and 5 Simon GL, Gorbach SL. Intestinal flora in health and disease.
thymus-derived T lymphocytes. J Exp Med 1992; 176(1): Gastroenterology 1984; 86(1): 174–93.
6 Hooper LV, Gordon JI. Commensal host–bacterial relationships 22 Kanamori Y, Ishimaru K, Nanno M, et al. Identification of in the gut. Science 2001; 292(5519): 1115–8.
novel lymphoid tissues in murine intestinal mucosa where 7 Borriello SP. Microbial flora of the gastrointestinal tract. In: Hill MJ, ed. Microbial Metabolisms in the Digestive Tract.
progenitors develop. J Exp Med 1996; 184(4): 1449–59.
23 Saito H, Kanamori Y, Takemori T, et al. Generation of 8 Borriello SP, Hudson M, Hill M. Investigation of the gastro- intestinal T cells from progenitors residing in gut crypto- intestinal bacteria flora. In: Russell RI, ed. Clinics in Gastro- patches. Science 1998; 280(5361): 275–8.
enterology. Philadelphia: W. B. Saunders, 1978: 329.
24 Shroff KE, Meslin K, Cebra JJ. Commensal enteric bacteria 9 Wilson KH, Blitchington RB. Human colonic biota studied by engender a self-limiting humoral mucosal immune response ribosomal DNA sequence analysis. Appl Environ Microbiol while permanently colonizing the gut. Infect Immun 1995; Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 R E V IE W : G U T F L O R A IN H E A L T H A N D D I S E AS E 25 Mowat AM, Viney JL. The anatomical basis of intestinal 42 Neish AS, Gewirtz AT, Zeng H, et al. Prokaryotic regulation of immunity. Immunol Rev 1997; 156: 145–66.
epithelial responses by inhibition of IkappaB-alpha ubiquiti- 26 Strobel S, Mowat AM. Immune responses to dietary antigens: nation. Science 2000; 289(5484): 1560–3.
oral tolerance. Immunol Today 1998; 19(4): 173–81.
43 Kunkel EJ, Campbell JJ, Haraldsen G, et al. Lymphocyte CC 27 Sudo N, Sawamura S, Tanaka K, Aiba Y, Kubo C, Koga Y. The requirement of intestinal bacterial flora for the development of chemokine (TECK) expression distinguish the small intestinal an IgE production system fully susceptible to oral tolerance immune compartment: Epithelial expression of tissue-specific induction. J Immunol 1997; 159(4): 1739–45.
chemokines as an organizing principle in regional immunity.
28 Blumberg RS, Saubermann LJ, Strober W. Animal models of mucosal inflammation and their relation to human inflam- 44 Pan J, Kunkel EJ, Gosslar U, et al. A novel chemokine ligand matory bowel disease [published erratum appears in Curr for CCR10 and CCR3 expressed by epithelial cells in mucosal Opin Immunol 2000; 12(2): 226]. Curr Opin Immunol 1999; tissues. J Immunol 2000; 165(6): 2943–9.
45 Van de Merwe JP, Stegeman JH, Hazenberg MP. The resident 29 Mowat AM, Steel M, Leishman AJ, Garside P. Normal faecal flora is determined by genetic characteristics of the host.
induction of oral tolerance in the absence of a functional Implications for Crohn’s disease? Antonie van Leeuwenhoek IL-12-dependent IFN-gamma signaling pathway. J Immunol 46 Darfeuille-Michaud A, Neut C, Barnich N, et al. Presence of 30 Nagata S, McKenzie C, Pender SL, et al. Human Peyer’s patch adherent Escherichia coli strains in ileal mucosa of patients T cells are sensitized to dietary antigen and display a Th cell with Crohn’s disease. Gastroenterology 1998; 115(6): 1405– type 1 cytokine profile. J Immunol 2000; 165(9): 5315–21.
31 MacDonald TT, Monteleone G. IL-12 and Th1 immune 47 Masseret E, Boudeau J, Colombel JF, et al. Genetically related responses in human Peyer’s patches. Trends Immunol 2001; Escherichia coli strains associated with Crohn’s disease. Gut 32 Groux H, O’Garra A, Bigler M, et al. A CD4+ T-cell subset 48 Sutton CL, Kim J, Yamane A, et al. Identification of a novel inhibits antigen-specific T-cell responses and prevents colitis.
bacterial sequence associated with Crohn’s disease. Gastro- 33 Powrie F, Leach MW, Mauze S, Menon S, Caddle LB, Coffman 49 Rook GA, Stanford JL. Give us this day our daily germs.
RL. Inhibition of Th1 responses prevents inflammatory bowel disease in scid mice reconstituted with CD45RBhi CD4+ 50 Kalliomaki M, Kirjavainen P, Eerola E, Kero P, Salminen S, T cells. Immunity 1994; 1(7): 553–62.
Isolauri E. Distinct patterns of neonatal gut microflora in 34 Liu LM, MacPherson GG. Lymph-borne (veiled) dendritic cells infants in whom atopy was and was not developing. J Allergy can acquire and present intestinally administered antigens.
Clin Immunol 2001; 107(1): 129–34.
51 Sartor RB. Colitis in HLA-B27/beta 2 microglobulin trans- 35 Rescigno M, Urbano M, Valzasina B, et al. Dendritic cells ex- genic rats. Int Rev Immunol 2000; 19(1): 39–50.
press tight junction proteins and penetrate gut epithelial 52 Aranda R, Sydora BC, McAllister PL, et al. Analysis of monolayers to sample bacteria. Nat Immunol 2001; 2(4): intestinal lymphocytes in mouse colitis mediated by transfer of CD4+, CD45RBhigh T cells to SCID recipients. J Immunol 36 Huang FP, Platt N, Wykes M, et al. A discrete subpopulation of dendritic cells transports apoptotic intestinal epithelial cells to 53 Cong Y, Brandwein SL, McCabe RP, et al. CD4+ T cells reactive T cell areas of mesenteric lymph nodes. J Exp Med 2000; to enteric bacterial antigens in spontaneously colitic C3H/ HeJBir mice: increased T helper cell type 1 response and ability 37 Muzio M, Bosisio D, Polentarutti N, et al. Differential to transfer disease. J Exp Med 1998; 187(6): 855–64.
expression and regulation of toll-like receptors (TLR) in 54 Veltkamp C, Tonkonogy SL, De Jong YP, et al. Continuous human leukocytes: selective expression of TLR3 in dendritic stimulation by normal luminal bacteria is essential for the cells. J Immunol 2000; 164(11): 5998–6004.
development and perpetuation of colitis in Tg(epsilon26) mice.
38 Hugot JP, Chamaillard M, Zouali H, et al. Association of NOD2 Gastroenterology 2001; 120(4): 900–13.
leucine-rich repeat variants with susceptibility to Crohn’s 55 Rath HC, Wilson KH, Sartor RB. Differential induction of disease. Nature 2001; 411(6837): 599–603.
colitis and gastritis in HLA-B27 transgenic rats selectively 39 Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation in colonized with Bacteroides vulgatus or Escherichia coli. Infect NOD2 associated with susceptibility to Crohn’s disease. Nature 56 Winkelstein JA, Marino MC, Johnston RB Jr, et al. Chronic 40 Williamson E, Westrich GM, Viney JL. Modulating dendritic granulomatous disease. Report on a national registry of 368 cells to optimize mucosal immunization protocols. J Immunol patients. Medicine (Baltimore) 2000; 79(3): 155–69.
57 Roe TF, Coates TD, Thomas DW, Miller JH, Gilsanz V. Brief 41 Viney JL, Mowat AM, O’Malley JM, Williamson E, Fanger NA.
report: treatment of chronic inflammatory bowel disease in Expanding dendritic cells in vivo enhances the induction of glycogen storage disease type Ib with colony-stimulating oral tolerance. J Immunol 1998; 160(12): 5815–25.
factors. N Engl J Med 1992; 326(25): 1666–9.
Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 58 Duchmann R, Kaiser I, Hermann E, Mayet W, Ewe K, Meyer 75 Guandalini S, Pensabene L, Zikri MA, et al. Lactobacillus GG zum Buschenfelde KH. Tolerance exists towards resident administered in oral rehydration solution to children with intestinal flora but is broken in active inflammatory bowel acute diarrhea: a multicenter European trial. J Pediatr Gast- disease (IBD). Clin Exp Immunol 1995; 102(3): 448–55.
roenterol Nutr 2000; 30(1): 54–60.
59 Macpherson A, Khoo UY, Forgacs I, Philpott-Howard J, 76 Szajewska H, Kotowska M, Mrukowicz JZ, Armanska M, Bjarnason I. Mucosal antibodies in inflammatory bowel dis- Mikolajczyk W. Efficacy of Lactobacillus GG in prevention of ease are directed against intestinal bacteria. Gut 1996; 38(3): nosocomial diarrhea in infants. J Pediatr 2001; 138(3): 60 Shanahan F. Immunological and genetic links in Crohn’s 77 Fabia R, Ar’Rajab A, Johansson ML, et al. The effect of exogenous administration of Lactobacillus reuteri R2LC and 61 Janowitz HD, Croen EC, Sachar DB. The role of the fecal oat fiber on acetic acid-induced colitis in the rat. Scand J stream in Crohn’s disease: an historical and analytic review.
Gastroenterol 1993; 28(2): 155–62.
Inflamm Bowel Dis 1998; 4(1): 29–39.
78 Mao Y, Nobaek S, Kasravi B, et al. The effects of Lactobacillus 62 Rutgeerts P, Geboes K, Vantrappen G, Beyls J, Kerremans R, strains and oat fiber on methotrexate-induced enterocolitis in Hiele M. Predictability of the postoperative course of Crohn’s rats. Gastroenterology 1996; 111(2): 334–44.
disease. Gastroenterology 1990; 99(4): 956–63.
79 Madsen KL, Doyle JS, Jewell LD, Tavernini MM, Fedorak RN.
63 Rutgeerts P, Goboes K, Peeters M, et al. Effect of faecal stream Lactobacillus species prevents colitis in interleukin 10 gene- diversion on recurrence of Crohn’s disease in the neoterminal deficient mice. Gastroenterology 1999; 116(5): 1107–14.
ileum. Lancet 1991; 338(8770): 771–4.
80 Schultz M, Sartor RB. Probiotics and inflammatory bowel dis- 64 Tytgat GN, Mulder CJ, Brummelkamp WH. Endoscopic lesions eases. Am J Gastroenterol 2000; 95(1 Suppl.): S19–S21.
in Crohn’s disease early after ileocecal resection. Endoscopy 81 Rembacken BJ, Snelling AM, Hawkey PM, Chalmers DM, Axon AT. Non-pathogenic Escherichia coli versus mesalazine 65 D’Haens GR, Geboes K, Peeters M, Baert F, Penninckx F, for the treatment of ulcerative colitis: a randomised trial.
Rutgeerts P. Early lesions of recurrent Crohn’s disease caused by infusion of intestinal contents in excluded ileum. Gastro- 82 Kruis W, Schutz E, Fric P, Fixa B, Judmaier G, Stolte M.
66 Harper PH, Lee EC, Kettlewell MG, Bennett MK, Jewell DP.
preparation and mesalazine in maintaining remission of Role of the faecal stream in the maintenance of Crohn’s colitis.
ulcerative colitis. Aliment Pharmacol Ther 1997; 11(5): 67 Gionchetti P, Rizzello F, Venturi A, et al. Antibiotic combina- 83 Kruis W, Fric P, Stolte M. Maintenence of remission in tion therapy in patients with chronic, treatment-resistant ulcerative colitis is equally effective with Escherichia coli pouchitis. Aliment Pharmacol Ther 1999; 13(6): 713–8.
Nissle 1917 and with standard mesalamine. Gastroenterology 68 Moss AA, Carbone JV, Kressel HY. Radiologic and clinical 2001; 120(5): A127–A127(Abstract).
assessment of broad-spectrum antibiotic therapy in Crohn’s 84 Gionchetti P, Rizzello F, Venturi A, et al. Oral bacteriotherapy disease. Am J Roentgenol 1978; 131(5): 787–90.
as maintenance treatment in patients with chronic pouchitis: 69 Rosen A, Ursing B, Alm T, et al. A comparative study of A double-blind, placebo-controlled trial. Gastroenterology metronidazole and sulfasalazine for active Crohn’s disease: the cooperative Crohn’s disease study in Sweden. I. Design and 85 Gionchetti P, Rizzello F, Venturi A, et al. Prophylaxis of methodologic considerations. Gastroenterology 1982; 83(3): pouchitis onset with probiotic therapy: a double-blind, placebo 70 Sutherland L, Singleton J, Sessions J, et al. Double blind, placebo controlled trial of metronidazole in Crohn’s disease.
86 Steidler L, Hans W, Schotte L, et al. Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science 2000; 71 Ursing B, Alm T, Barany F, et al. A comparative study of metronidazole and sulfasalazine for active Crohn’s disease: the 87 Gibson GR, Beatty ER, Wang X, Cummings JH. Selective cooperative Crohn’s disease study in Sweden. II. Result.
stimulation of bifidobacteria in the human colon by oligofruc- Gastroenterology 1982; 83(3): 550–62.
tose and inulin. Gastroenterology 1995; 108(4): 975–82.
72 Isolauri E, Juntunen M, Rautanen T, Sillanaukee P, Koivula T.
88 Kleessen B, Sykura B, Zunft HJ, Blaut M. Effects of inulin and A human Lactobacillus strain (Lactobacillus casei sp strain lactose on fecal microflora, microbial activity, and bowel habit GG) promotes recovery from acute diarrhea in children.
in elderly constipated persons. Am J Clin Nutr 1997; 65(5): 73 Majamaa H, Isolauri E, Saxelin M, Vesikari T. Lactic acid 89 Wang X, Gibson GR. Effects of the in vitro fermentation of bacteria in the treatment of acute rotavirus gastroenteritis.
oligofructose and inulin by bacteria growing in the human J Pediatr Gastroenterol Nutr 1995; 20(3): 333–8.
large intestine. J Appl Bacteriol 1993; 75(4): 373–80.
74 Pant AR, Graham SM, Allen SJ, et al. Lactobacillus GG and 90 Gibson GR, Roberfroid MB. Dietary modulation of the human acute diarrhoea in young children in the tropics. J Trop colonic microbiota: introducing the concept of prebiotics.
Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393 R E V IE W : G U T F L O R A IN H E A L T H A N D D I S E AS E 91 Palmeros B, Wild J, Szybalski W, et al. A family of removable 92 Recorbet G, Robert C, Givaudan A, Kudla B, Normand P, cassettes designed to obtain antibiotic-resistance-free genomic Faurie G. Conditional suicide system of Escherichia coli modifications of Escherichia coli and other bacteria. Gene released into soil that uses the Bacillus subtilis sacB gene. Appl Environ Microbiol 1993; 59(5): 1361–6.
Ó 2002 Blackwell Science Ltd, Aliment Pharmacol Ther 16, 1383–1393

Source: http://s.bio119.co.kr/Shop_Image/Board/shopboard_5/298a0f5af935ed75b4743e294c39911c_13_.pdf