CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2007;5:274 –284
EVIEWS
ut Microbiota: Mining for Therapeutic Potential
NN M. O’HARA* and FERGUS SHANAHAN*,‡
Alimentary Pharmabiotic Centre, and ‡Department of Medicine, University College Cork, National University of Ireland, Cork, Ireland
Wdhbemiecc
ofhseawacb
cdtn(furpast
riMddtt
he resident microbiota of the human intestine exerts aonditioning effect on intestinal homeostasis, deliveringegulatory signals to the epithelium and instructing muco-al immune responses. Pattern recognition receptors areey mediators of innate host defense, and in healthy indi-iduals, the mucosal immune system exhibits an exquisitelyegulated restrained response to the resident microbiota.owever, in genetically susceptible hosts, unrestrained mu-
osal immune activation in response to local bacterial sig-als can contribute to the pathogenesis of inflammatoryowel disease. Manipulation of the microbiota to enhance
ts beneficial components thus represents a potential ther-peutic strategy for inflammatory bowel disease. Moreover,he microbiota might be a rich repository of metaboliteshat can be exploited for therapeutic benefit. Modern mo-ecular techniques are facilitating improved understandingf host-microbe dialogue in health and in several diseaserocesses, including inflammatory bowel disease. It follows
hat elucidating the molecular mechanisms of host-micro-ial interactions is now a prerequisite for a “bugs to drugs”rogram of discovery.
he chronic inflammatory bowel diseases (IBDs), Crohn’sdisease and ulcerative colitis, are characterized by bouts of
ncontrolled, chronic mucosal inflammation, followed by re-odeling processes that occur during periods of remission.ogether, these result in tissue-damaging inflammatory re-
ponses in the intestine that might result in much personaluffering and impaired quality of life. Crohn’s disease andlcerative colitis have a combined prevalence of 250/100,000 inhe Western world, and they represent a substantial economicurden on health care resources. Although the precise etiologyf IBD remains to be elucidated, a complex interaction ofnvironmental, genetic, and immunoregulatory factors contrib-tes to the initiation and perpetuation of the disease.1
Under normal circumstances, the mucosal immune system isxquisitely regulated and exhibits a restrained response to theesident microbiota while retaining an ability to mount appro-riate immune responses to pathogenic bacteria. However, sev-ral lines of evidence indicate that genetically influenced dys-egulation of the mucosal immune response to antigens of thendigenous microbiota can contribute to the pathogenesis ofBD.2,3 In susceptible individuals, environmental triggers canmpact on the initiation or reactivation of disease, and tissueamage might result from immune cell misperception of dan-er within the indigenous microbiota or from failure of normalolerance to enteric bacteria.2,4 Within the gastrointestinal tract,he inflammatory capacity of commensal bacteria is varied.ome resident bacteria are proinflammatory, whereas others
ttenuate inflammatory responses.5–11
The 2005 Nobel prize awarded to Barry Marshall and Robinarren is a sobering reminder that the solution to some chronic
isorders cannot be unraveled by exclusive investigation of theost response. The host response must be considered in terms ofacterial residents and prokaryotic-eukaryotic interactions. Mod-rn molecular techniques are facilitating our understanding oficrobe-microbe and host-microbe communications at mucosal
nterfaces and are providing glimpses of tangible therapeutic strat-gies. Therefore, the intent here is to present an overview of theommensal microbiota, their interactions with the intestinal mu-osa, and clinical relevance to IBD.
Indigenous Gut MicrobiotaComposition of the Normal Gut MicrobiotaDialogue between commensal bacteria and the host
ccurs primarily along mucosal surfaces, and the largest inter-ace is the human gastrointestinal mucosa. The intestine isabitat to a dynamic and diverse bacterial community that iseparated from the internal milieu by only a single layer ofpithelial cells. Intestinal bacteria outnumber human somaticnd germ cells 10-fold12 and represent a combined microbiomeell in excess of the human genome. Co-evolution of the hostnd indigenous microbes has fostered mutually beneficial andooperative interactions mediated by bidirectional host-micro-iota exchange.
Traditionally, studies of indigenous bacteria focused on theharacterization of fecal diversity. Detailed analysis was hin-ered by conventional microbiology techniques, and most bac-erial species still cannot be cultured. Modern molecular tech-iques such as broad-range sequencing of 16S ribosomal RNA
rRNA) from amplified bacterial nucleic acid extracted fromeces or biopsies indicate evolutionary divergence and can besed to identify and classify bacteria. The availability of bacte-ial sequence data has facilitated the development of molecularrobes for DNA microarrays, fluorescent in situ hybridization,nd gene chips that identify and enumerate specific microbialpecies. These molecular approaches are being used to examinehe individuality and temporal stability of the microbiota and
March 2007 THERAPEUTIC POTENTIAL OF THE GUT MICROBIOTA 275
he impact of weaning, antibiotics, or dietary changes on itsomposition. It is now realized that the mucosa-associatedommunity is significantly different from the luminal and fecalommunities.13 Although every adult intestine harbors a par-icular combination of predominant species, the compositionf the intestinal microbiota might alter with lifestyle, diet, andge.14,15 Nonetheless, a comparative study of the microbiota ofuman adults with varying degrees of genetic relatedness, in-luding monozygotic twins, emphasized the prevailing influ-nce of host genotype over diet in determining the microbialomposition of the gut.16
Acid, bile, and pancreatic secretions hinder the colonizationf the stomach and proximal small intestine by most bacteria.owever, bacterial density dramatically increases in the distal
mall intestine and in the large intestine increases to approxi-ately 1011–1012 bacteria per gram of colonic content, which
ontribute to 60% of fecal mass.17,18 Although archaea andukarya are also represented,19 bacteria predominate, and co-on-residing bacteria achieve the highest cell densities recordedor any ecosystem.20 The most common anaerobic genera in-lude Bifidobacterium, Clostridium, Bacteroides, and Eubacterium,nd aerobic Escherichia, Enterococcus, Streptococcus, and Klebsiellare also found. However, sequencing of 16S rRNA gene cloneibraries has indicated that a significant fraction of the bacteriaepresent uncultivated species and novel microorganisms.13
Mammalian fetuses are born germ-free, but immediatelyfter birth, establishment of the resident microbiota is driven bynvironmental factors such as mode of delivery, type of infantiet, hygiene levels, and medication.21,22 Enterobacteria andifidobacteria species represent early colonizers, but differences
igure 1. Changes in intestinal structure and function in germ-free animeconstitution of germ-free mice with a microbiota restores the mucotructural, and metabolic effects on the intestinal epithelium.
n gut microbial composition and incidence of infection occur e
etween breast-fed and formula-fed infants.21 These pioneeracteria can modulate gene expression in the host to create auitable environment for themselves and can prevent growth ofther bacteria introduced later in the ecosystem.23,24 Increasedredence is being given to the hypothesis that the modernanitized environment of developed societies has altered theormal pattern of gut colonization during infancy. This mightesult in a lack of tolerance to otherwise harmless food proteinsnd other antigens, including those of the intestinal micro-iota.25
Functions of the Normal Gut MicrobiotaEnteric bacteria confer many benefits to intestinal phys-
ology including protective, structural, and metabolic effects.heir influence on intestinal structure and function has beenemonstrated in comparative studies of germ-free and colo-ized animals. Some of the differences between animals raisednder germ-free and conventional conditions are listed in Fig-re 1. Such comparisons, together with studies indicating thateconstitution of gnotobiotic mice with a microbiota is suffi-ient to restore the mucosal immune system,26 illustrate thathe microbiota provide regulatory signals that instruct intesti-al development and function. Indeed, colonization of germ-
ree mice with a single species, Bacteroides thetaiotaomicron, af-ects the expression of a variety of host genes influencingutrient uptake, metabolism, angiogenesis, mucosal barrier
unction, and the development of the enteric nervous system.23
urthermore, ligands from resident bacteria and commensal-erived symbiosis factors influence the normal developmentnd function of mucosal immunity.27,28 Indigenous bacteria
ompared with colonized animals raised under conventional conditions.mune system, and commensal bacteria exert numerous protective,
als csal im
ducate the mucosal immune system and modulate the fine
tcci
bitmlftfeti
octRsrdnrapagitbmeotp
mmsphsac
roccsmcaca
aiecrlitsmDs
ppioCwettabria
abcomlttsdiwisaipbc
mdm(pccapr
276 O’HARA AND SHANAHAN CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 3
uning of T-cell repertoires and T-helper cell type 1 (Th1)/Th2ytokine profiles.29 Taken together, the composition of theolonizing microbiota might influence individual variations inmmunity.
Along the epithelium, enteric bacteria form a natural defensearrier against exogenous microbes. Colonization resistance
nvolves several mechanisms including displacement, competi-ion for nutrients and epithelial binding sites by the resident
icrobiota, and production of antimicrobial factors such asactic acids and bacteriocins.30 Commensal bacteria also helportify the epithelial barrier by various mechanisms that includehe induction of the complement inhibitor decay-acceleratingactor or increasing barrier function.31,32 Exposure of colonicpithelial cell lines to bacterial ligands also results in apicalightening and sealing of the tight junctional protein ZO-1 andncreased transepithelial resistance.33
The metabolic activity of the microbiota is equivalent to thatf an organ within an organ.34 Through the production of shorthain fatty acids, resident microorganisms positively influencehe differentiation and proliferation of intestinal enterocytes.35
esident bacteria can break down dietary carcinogens; synthe-ize biotin, folate, and vitamin K; ferment nondigestible dietaryesidue, particularly carbohydrates and endogenous epithelial-erived mucus; and assist in the absorption of calcium, mag-esium, and iron.36 –38 Together, this complex metabolic activityecovers valuable energy and absorbable substrates for the hostnd provides energy and nutrients for bacterial growth androliferation. Colonization increases glucose uptake in the gut,nd compared with colonized mice, germ-free mice require areater caloric intake to sustain a normal body weight.39 Thismplicates intestinal bacteria as modulators of fat deposition inhe host. It has been proposed that an individual’s gut micro-iota has a specific metabolic efficiency, and differences in guticrobial composition between individuals might regulate en-
rgy storage and predispose to obesity.19 Therefore, explorationf the metabolic activity of the microbiota offers strong poten-ial for exploitation of the microbiota as biomarkers for im-ending disease and the development of novel therapies.
Host-Microbiota Dialogue at the MucosalSurfaceHost defense demands precise interpretation of the
icroenvironment to distinguish commensal from pathogenicicroorganism and must entail exquisite regulation of re-
ponses. Disruption of these processes might lead to inappro-riate responses. Thus, although the microbiota are generally aealth asset, they might become a liability in certain circum-tances. These include syndromes of bacterial overgrowthnd/or translocation, metabolite-mediated conversion of pro-arcinogens to carcinogens, and IBD.29,40,41
Along the epithelium, active sampling of commensal bacte-ia, pathogens, and other antigens is mediated by various typesf immunosensory cells. These include surface enterocytes, Mells, and dendritic cells (DCs). Surface enterocytes are inter-onnected with tight junctions and overlain with mucus. Theyerve as afferent sensors of danger signals within the luminal
icroenvironment by secreting defensins, immunoglobulin A,hemokines, and cytokines that alert and direct innate anddaptive immune responses.42– 45 M cells, specialized epithelialells, that overlie lymphoid follicles sample the environment
nd transport luminal antigens to subadjacent DCs and other s
ntigen-presenting cells. Intestinal DCs themselves participaten immune surveillance and can directly sample gut contents byither entering or extending dendrites between surface entero-ytes without disrupting tight junctions.46 DCs can ingest andetain live commensal bacteria and transit to the mesentericymph node where immune responses to these bacteria arenduced locally. This prevents access of commensal bacteria tohe internal milieu.47 In addition, enterocyte-derived mediatorsuch as thymic stromal lymphopoietin can induce noninflam-
atory DCs, suggesting that physiologic interactions betweenCs and surface enterocytes contribute to intestinal homeosta-
is.48
Mucosal Recognition of BacteriaThe ability of immunosensory cells to discriminate
athogen from commensal is mediated, in part, by 2 major hostattern recognition receptor (PRR) systems. These are the fam-
ly of toll-like receptors (TLRs) and the nucleotide-bindingligomerization domain/caspase recruitment domain (NOD/ARD) molecules.49 Several of the signal transduction path-ays induced by these receptors have been elucidated and their
ffector responses characterized.49 –53 Both TLRs and NOD pro-eins trigger innate and adaptive immune responses, includinghe synthesis of proinflammatory cytokines and chemokines,nd TLR signaling also impacts on subsequent T-cell responsesy activating DCs that overcome the suppressive effects of Tegulatory cells.54 Together, these PRRs play a fundamental rolen immune cell activation in response to specific microbial-ssociated molecular patterns (Table 1).
TLRs and NOD proteins are expressed by surface enterocytesnd DCs,51 and in the gut, TLRs and NOD proteins appear toe crucial for bacterial-host communication. Decreased entero-yte proliferation and levels of cytoprotective factors have beenbserved in TLR-defective mice compared with wild-typeice.27 TLR signals mediated by commensal bacteria or their
igands are essential for intestinal barrier function and repair ofhe gut.55 Many PRR ligands are expressed by commensal bac-eria, yet the healthy gut does not evoke inflammatory re-ponses to these bacteria.11 PRRs participate in several recentlyescribed host molecular immune mechanisms that mediate
ntestinal homeostasis (Table 2), and these are reviewed else-here.56 In the healthy gut, regulatory T cells and tolerance-
nducing DCs contribute to the control of excessive Th1 re-ponses to the indigenous microbiota also.57,58 Phenotypiclterations in regulatory T-cell populations have been observedn IBD patients,59 and several lines of evidence indicate thateripheral tolerance, as well as local tolerance to commensalacteria, is mediated by the suppressive effects of regulatory Tells.59 – 61
Commensal bacteria have been shown to suppress inflam-atory responses and inhibit specific intracellular signal trans-
uction pathways, thereby contributing to the maintenance ofucosal homeostasis. The transcription factor, nuclear factor
NF)-�B, is a master coordinator of immune responses toathogenic bacteria and other stress signals. However, mostommensal bacteria do not activate NF-�B8,11; instead, someommensal strains antagonize NF-�B within epithelial cells byvariety of mechanisms. These include inhibition of epithelialroteasome function or degradation of the NF-�B counter-egulatory factor I�B-� or by the nuclear export of the p65
ubunit of NF-�B in a peroxisome proliferator activated recep-
trvubsmdt
c
fmtrIe1Cao
ipe
T
TT
TTTT
TTTTTNNN
T
D
L
I
E
CL
a
t
March 2007 THERAPEUTIC POTENTIAL OF THE GUT MICROBIOTA 277
or (PPAR)–�-dependent manner.7,8,62 Some commensal bacte-ia can attenuate colonic inflammation through TLR4 by ele-ating PPAR-� expression in intestinal enterocytes andncoupling NF-�B– dependent target genes in a negative feed-ack loop.8,63 It is clear that many of these mechanisms aretrain-specific, and not all commensal bacteria with immuno-
odulatory effects use these mechanisms; other signal trans-uction pathways are likely to account for their anti-inflamma-ory effects.
Pathogenesis of Inflammatory BowelDiseaseBoth forms of IBD represent the clinical outcome of a
omplex interaction of environmental, genetic, and immune
able 2. Host Systems That Limit PRR Signaling and Inflamm
Mechanism
ecreased ligand recognition due to TLRexpression profiles
Low expressionTLR localizationState of cell diff
expression)imited repertoire of TLR co-receptorexpression
Low expressionLack of express
nhibitors of TLR signaling High expressioninteracting pro
Sequestration oMyD88 and Mmember)a
nzyme-mediated inhibition of TLRsignaling
Proteolytic degraa ubiquitin lig
Deubiquitinationfinger protein
ross-regulation by other PRRs Negative regulaigand neutralization Secretion of imm
enterocytes
The regulatory effects of ST2, TRIAD3A, and A20 on TLR signals have
hese suppressors mediate TLR signaling in the gut.
actors. The normal physiologic response to the indigenousicrobiota is one of immunologic quiescence. Deviations from
his and, in particular, genetically influenced aberrant immuneesponses to luminal antigens can lead to the development ofBD. Crohn’s disease bears the immunologic signature of anxaggerated Th1 response, with excess interleukin (IL)-12, IL-8, interferon (IFN)-�, and tumor necrosis factor (TNF)–�.64
onversely, ulcerative colitis is associated with a dominanttypical Th2 response that is probably driven by the productionf IL-13.65
An environmental contribution to the pathogenesis of IBDs clear from studies of monozygotic twins in which an incom-lete accordance rate has been observed for both Crohn’s dis-ase (�50%) and ulcerative colitis (�10%).66 Smoking, dietary
TLR signaling adaptor proteinsST-2 (toll IL-1 receptor family
138
n of TLR4 and TLR9 by TRIAD3A, 139
LR adaptor proteins by the zinc 140, 141
f TLR signaling by NOD2 103, 104, 142, 143globulin A by intestinal 47
identified in other systems, but it remains to be determined whether
NA
elic a
atio
of fu(surferent
of TLion oof TLtein/
f theal by
datioasea
of TA20a
tion ouno
been
fctmsmaE
bwoiammddsoiwmMstbMfi
brdrcfdrgchidhbcgarHarCsaiod
bimIclcmlbtrdibfdtpcf
wnpdclwctFaTtIaTtcdmurao
ttbcso
278 O’HARA AND SHANAHAN CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 3
actors, nonsteroidal anti-inflammatory drugs, socioeconomiconditions, and the more hygienic lifestyles of developed coun-ries have all been associated with IBD.67 Contributing environ-
ental factors are probably multiple and complex. Numeroustudies have pursued a specific infectious cause for IBD. The
ost notable inconclusively incriminate either Mycobacteriumvium subspecies paratuberculosis (MAP) or adherent-invasivescherichia coli (AIEC) in the development of Crohn’s disease.68,69
Viable MAP have occasionally been detected in peripherallood and intestinal tissue in a higher proportion of individualsith Crohn’s disease than in controls, and antibodies to therganism have been disease-associated.68,70,71 A recent study
dentified 2 sets of cross-reactive MAP and human peptides thatre specific for Crohn’s disease.72 Although these peptide pairsight represent structural mimics whereby an anti-MAP im-une response could trigger a host autoimmune response,
efinite proof for a role of molecular mimicry in Crohn’sisease–specific autoimmunity has not been established. Thistudy did not test the existence of cross-reactive autoantibodiesr T-cell immunity with pathogenic activity. Nonetheless, there
s limited evidence for T-cell immunity to MAP in associationith Crohn’s disease, and a cause and effect relationship re-ains unproved.73 Among several arguments against persistentAP infection as a cause of Crohn’s disease, the most persua-
ive is the clinical experience with anti-TNF therapy. Therapeu-ic blockade of TNF-� greatly impairs host resistance to myco-acterial infection, but this is not associated with disseminatedAP in patients with IBD.73 Furthermore, controlled trials have
ailed to show a therapeutic effect of anti-tuberculosis therapyn Crohn’s disease.74
A number of studies have reported abnormally high num-ers of virulent AIEC and hemagglutinin-expressing Esche-ichia coli adjacent to the ileal or colonic mucosa in Crohn’sisease.69,75 AIEC can adhere to epithelial cells, invade mac-ophages without inducing apoptosis, and persist and repli-ate within phagocytes.76 Nonetheless, these reports do notulfil Koch’s postulates, and several lines of available evi-ence that are reviewed elsewhere stand against an etiologicole for any single pathogenic microorganism in the patho-enesis of IBD.67 Moreover, it is difficult to reconcile theoncept of an infectious etiology with the beneficial effects ofost immunosuppression; any infection putatively stimulat-
ng such an intense inflammatory reaction as seen in Crohn’sisease is unlikely to respond to long-term suppression ofost immune defenses. However, after the lesson of Helico-acter pylori and peptic ulcer disease, it would be unwise toompletely dismiss an infectious contribution to the patho-enesis of IBD. It is interesting to note that both MAP DNAnd E coli DNA have been detected in the granulomas ofesected tissues from some Crohn’s disease patients.77,78
owever, other forms of bacterial DNA were also detected,67
nd rather than reflect a specific causal relationship, it mighteflect impaired innate handling of luminal bacteria inrohn’s disease. It is possible that in a subset of genetically
usceptible Crohn’s disease patients, particularly those withdefect in their ability to sense bacterial ligands or clear
ntracellular infections, persistent infection with MAP, AIEC,r an as yet unidentified pathogen might underlie their
isease. p
Microbial Contribution to InflammatoryBowel DiseaseAlthough a specific infectious cause for IBDs has not
een established, current clinical and experimental data incrim-nate a loss of tolerance to commensal bacteria in the develop-
ent of chronic intestinal inflammation.67 Animal models ofBD provide persuasive evidence implicating the resident mi-robiota as a causative factor of IBD. Irrespective of the under-ying genetic defect(s) in various animal models of the disease,olonization with commensal bacteria is required for an inflam-atory phenotype, and germ-free animals do not develop IBD-
ike lesions.29,79 – 82 Of note, antibiotics are beneficial in a num-er of animal models. It has been shown in some models thathe disease can be adoptively transferred to immune-deficientecipients when reconstituted with T cells from a diseasedonor that was sensitized to enteric bacteria.83 Moreover, mod-
fication of the microbiota by the administration of probioticacteria, which are commensal organisms that can be harnessedor therapeutic benefits, has been shown to delay the onset ofisease and attenuate the inflammatory process.84 – 86 Althoughhese animal models do not reveal whether commensal bacteriaassively or directly elicit the inflammatory response, theylearly demonstrate that exposure to the microbiota is essentialor expression of the disease.
In patient-related studies, increased numbers of bacteriaithin the mucosa of IBD patients compared with that ofoninflamed and inflammatory disease controls have been re-orted.87,88 Furthermore, the distribution of lesions and inci-ence of inflammation in either Crohn’s disease or ulcerativeolitis is greatest in the area with the highest concentrations ofuminal bacteria.29 Diversion of the fecal stream is associatedith an improvement in disease severity in the distal bowel, and
ellular and humoral immune reactivity against components ofhe bacterial microbiota is evident in patients with IBD.2,29,89
lagellin derived from commensal bacteria has been identifieds a dominant antigen in Crohn’s disease, suggesting thatLR5-dependent recognition of flagellin might play a role in
he immune responses to the commensal bacteria observed inBD.90 In addition, therapeutic efficacy from the administration ofntibiotics and probiotics in IBD patients has been reported.3,91–93
ogether, these and other circumstantial observations in humanshat are reviewed in more detail elsewhere29,94 substantiate theurrent dogma that IBD is a heterogeneous syndrome caused byysregulated immunity to normal microbiota. However, the modelust allow for heterogeneity within both Crohn’s disease and
lcerative colitis. It remains unclear whether the inflammatoryesponses in IBD, both within the gut and at extraintestinal sites,re elicited in response to a specific subset of intestinal microbes,r whether tolerance to commensal bacteria in general is affected.
Genetic Influences on Microbial Perception inInflammatory Bowel DiseaseThe pathophysiologic impact of bacterial-mucosal in-
eractions has been emphasized by an increased awareness ofhe association of genetic aberrations with increased suscepti-ility to IBD. In particular, mutations of CARD15, which en-odes NOD2, and certain TLR polymorphisms have beentrongly associated with Crohn’s disease. This implicates a rolef PRR dysfunction and impaired bacterial sensing in the
athogenesis of IBD.
mCtdnenfe
baITdswvaddm2tmaotdsstpiffNnC
apctlCtIfHpfdpfltbien
Tta
atdmnianb
cVSppabattRalsbea
sdddrrheitpLiPaciUmtmbab
March 2007 THERAPEUTIC POTENTIAL OF THE GUT MICROBIOTA 279
Three major polymorphisms in the CARD15 gene on chro-osome 16q12 have been specifically associated with �15% ofrohn’s disease patients, and individuals who carry 2 copies of
he risk alleles have a 20-fold to 40-fold increase in their risk ofeveloping Crohn’s disease.95–97 However, this risk factor iseither sufficient nor necessary for development of the dis-ase.98 Moreover, NOD2-deficient mice do not develop sponta-eous gut inflammation, and NOD2 knockout mice that lack
ull-length NOD2 protein do not demonstrate significantlynhanced susceptibility to experiment-induced colitis.99,100
The ligands for NOD2 are muramyl dipeptides (MDP) ofacterial peptidoglycan (Table 1). All 3 Crohn’s disease–associ-ted mutations occur in the ligand-binding domain of NOD2.ntracellular triggering of NOD2 leads to NF-�B activation.hus, mutations in the NOD2 protein compromise MDP-in-uced signaling and result in aberrant or defective NF-�B re-ponses to gut bacteria.101,102 However, stimulation of NOD2ith MDP has been shown to inhibit TLR2-driven NF-�B acti-
ation and Th1 cytokine responses. This suggests that in thebsence of NOD2, stimulation with MDP results in NF-�Bysregulation and imbalanced TLR2-mediated cytokine pro-uction (elevated IL-12, decreased IL-10).103,104 Together theseight contribute to the pathogenesis of Crohn’s disease (Figure
A). On the other hand, in a study of knock-in mice expressinghe most common Crohn’s disease–associated NOD2/CARD15
utation, it was shown that MDP triggers increased NF-�Bctivation, IL-1� secretion, and apoptosis105 (Figure 2B). More-ver, NOD2-deficient mice are unusually susceptible to infec-ion with Listeria monocytogenes and have reduced production ofefensin-like antimicrobial peptides termed cryptidins.100 Con-idering that reduced expression of �-defensin has been ob-erved in Crohn’s disease patients with NOD2 mutations,106
his has led to the proposal that NOD2 mutations mightredispose to Crohn’s disease by reducing �-defensin–mediated
nnate antimicrobial activity100 (Figure 2C). To date, these dif-erent experimental models have implicated conflicting loss ofunction, loss of regulation, or gain of function phenotypes forOD2 mutations, and the precise biologic role of NOD2 inormal intestinal physiology and in the pathogenesis ofrohn’s disease remains elusive.
Expression of some TLRs, including TLR4, is differentiallyltered in Crohn’s disease and ulcerative colitis.107,108 It has beenroposed that increased TLR4 expression in patients with IBDonfers lipopolysaccharide hyperresponsiveness.107 Alterna-ively, it might reflect a loss of response. The TLR4 gene isocalized on chromosome 9q32-33, a region harboring arohn’s disease susceptibility gene,109 and in selective popula-
ions, a TLR4 polymorphism has been associated with IBD.49,110
t is plausible that variant alleles in the TLR4 gene could induceunctional dysregulation in response to lipopolysaccharide.owever, the functional phenotypic consequences of TLR4olymorphisms in IBD remain unresolved. A preliminary reportrom a single German cohort study has suggested that Crohn’sisease might also be associated with TLR9 promoter polymor-hisms.111 Recent reports demonstrating a hyperreactivity toagellins in sera from Crohn’s disease patients90,112 suggest thathese TLR or other PRR polymorphisms could lead to impairedacterial clearance and thus increased load of bacterial antigens,
ncluding flagellin, in the lumen. This would be consistent withvidence for defective handling of bacteria and increased bacterial
umbers in the mucosa of patients with Crohn’s disease.90,112 C
herefore, it is probable that additional IBD susceptibility geneshat affect the interpretation of the microbial microenvironmentnd/or regulate the host response to it remain to be identified.
Modifying the Ecosystem asTherapeutic StrategyTraditional drug therapies for IBD address only one
spect of the pathogenesis by targeting the mucosal inflamma-ory response, but this strategy remains suboptimal in terms ofisease management and efficacy. Optimal disease managementight require the microbial contribution to be addressed. Ma-
ipulating the microbiota and exploiting host-microbial signal-ng pathways could provide benefit for the treatment of bothcute and chronic intestinal diseases. This underpins the ratio-ale for therapeutic manipulation of the microbiota with pro-iotic bacteria or other pharmabiotics.
Lactobacilli and bifidobacteria have traditionally been the mostommon probiotic candidates, but multi-strain cocktails (eg,SL#3), nonpathogenic E coli, and nonbacterial organisms such asaccharomyces boulardii and nematode parasites have been used forrobiotic effect.113,114 Pharmabiotic is an umbrella term to encom-ass any form of therapeutic exploitation of the gut microbiota. Inddition to live probiotic bacteria, pharmabiotics might includeacterial constituents such as DNA, probiotic-derived biologicallyctive metabolites, food ingredients that modulate the composi-ion of the microbiota (prebiotics), probiotic/prebiotic combina-ions (synbiotics), or genetically modified commensal bacteria.ecently, in the first human trial with genetically engineered ther-peutic bacteria, modified Lactococcus lactis was used to deliver IL-10ocally to the gut in 10 Crohn’s disease patients. The treatment wasafe, disease activity was reduced, and the modified bacteria wereiologically contained.115 Therefore, bacterial-based topical deliv-ry of biologically active proteins represents a highly promisingnd safe therapeutic strategy for combating mucosal diseases.
Increasing evidence supports a therapeutic role for probiotictrategies for treating enteric infections, post-antibiotic syn-romes, necrotizing enterocolitis, and irritable bowel syn-rome.116 –119 In murine models of colitis, probiotics have alsoemonstrated prophylactic effects that are associated with aeduction in proinflammatory cytokines and an induction ofegulatory cytokines.84,120 However, the role of probiotics inuman IBD is more complex. The best evidence for probioticfficacy in IBD is in the prevention of pouchitis,121 but resultsn clinical practice appear to be inconsistent. This might relateo variability in patient populations or the quality or choice ofrobiotic preparation. In ulcerative colitis, E coli Nissle 1917,actobacillus rhamnosus GG, and VSL#3 have shown efficacy sim-
lar to the drug mesalazine in maintaining remission.93,122,123
robiotics have induced remission of acute ulcerative colitislso.93,124 However, in a randomized, double-blind, placebo-ontrolled trial involving 157 patients, neither Bifidobacteriumnfantis 35624 nor Lactobacillus salivarius subspecies salivarius
CC118 demonstrated a significant benefit over placebo in theaintenance of steroid-induced remission of ulcerative coli-
is.125 These strains have attenuated disease severity in animalodels of IBD.85,126 It might be that the differences in efficacy
etween animal and human IBD might reflect the timing ofdministration, differences in disease severity, or effective pro-iotic dose/body weight.
The evidence for probiotics in Crohn’s disease is even weaker.
ontrolled studies have not established efficacy for L rhamnosus
GCaccpdsb
pbtmb
pa
FpotpmiapIi
280 O’HARA AND SHANAHAN CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 3
G or Lactobacillus johnsonii LA1 as maintenance therapies forrohn’s disease.127,128 Crohn’s disease is a complex condition, vari-ble in its location as well as its manifestation. Variability in theomposition and diversity of the microbiota along and over theross-sectional axis of the gastrointestinal tract suggests that de-ending on the topographic distribution of lesions in Crohn’sisease, a single probiotic might not be equally suited to differentubsets of patients. Colonic location of disease seems to respond
igure 2. Possible mechanisms contributing to the association of mutarotein, NOD2, which is expressed in the cytosol of cells of the macrophf bacterial peptidoglycan leads to the activation of the transcription fahrough TLR2. (A) In normal mucosa, TLR2-driven activation of NF-�B iroduction of the proinflammatory Th1 cytokines IL-12 and IFN-�. Conediated inflammation occurs through IL-12 as a result of NOD2 to neg
n antigen-presenting cells expressing mutated NOD2, MDP elevates Npoptosis, which together might mediate the pathogenesis of Croharticularly by Paneth’s cells at the base of the small intestinal crypts, res
mpaired NOD-2–dependent �-defensin production might lead to increammunologic response to the microbiota.
etter to antibiotics and might, as a result, be more susceptible to v
harmabiotic therapy. It might be that multi-strain probiotic com-inations are required, and strain selection, dosing, and coloniza-ion issues need more intensive investigation. Moreover, there
ight be a role for probiotics in the enhancement of epithelialarrier function in the very early stages of Crohn’s disease.
Several unresolved issues impede the clinical evaluation ofrobiotics.67,129 These include determination of optimal dosend vehicle of delivery, development of reliable predictors of in
in the CARD15 gene with Crohn’s disease. CARD15 encodes the NODonocyte lineage, DCs, and enterocytes. Stimulation of NOD2 by MDP
NF-�B, and peptidoglycan can signal also to antigen-presenting cellsatively regulated by non-mutated NOD2, thereby preventing excessively, in NOD2-deficient cells, IL-10 secretion is decreased, and Th1-cell–ly regulate TLR2 signaling. (B) Alternatively, it has been postulated thatactivation and results in excessive IL-1� production and uncontrolled
isease. (C) Deficient expression of NOD2 by intestinal enterocytes,n decreased secretion of antimicrobial peptides known as �-defensins.acterial colonization and a type of bacterial overgrowth that triggers an
tionsage/mctor
s negverseativeF-�Bn’s dults ised b
ivo performance, regulation and verification of product stabil-
ipcFaodmTs
poNcidscpciocisc
jppwscceccbttmchrtacicspb
March 2007 THERAPEUTIC POTENTIAL OF THE GUT MICROBIOTA 281
ty, determination of which combinations of probiotics or otherharmabiotics are synergistic or antagonistic, and strain-strainomparisons of probiotic performance in different indications.urthermore, the microbial, immunologic, and functional char-cteristics of individual probiotic strains and their mechanismsf action in different clinical settings require clarification. In-ividual variability in composition of the enteric microbiotaight also be a determining factor for optimal strain selection.o protect consumers, there is also a pressing need for more
tringent regulation of unsubstantiated health claims.Bacterial-mediated host immune responses involve several
athways and signal transduction systems, and several modesf action by which commensal and probiotic bacteria dampenF-�B signaling have already been elucidated. More detailed
larification of probiotic mechanisms will be facilitated by thedentification of the molecular basis of prokaryotic-eukaryoticialogue. Furthermore, the increasing availability of commen-al/probiotic genomes should facilitate the identification ofommensal effector molecules or other components withharmabiotic potential. The possibility of using these mole-ules to specifically target distinct points of intracellular signal-ng cascades might alleviate inflammation in a target area andvercome the global immunosuppressive effects associated withurrent therapies. In patients with severe IBD, the use of genet-cally engineered or “designer” probiotics might offer a newtrategy for more targeted delivery of anti-inflammatory mole-ules to the inflamed mucosa.
Conclusions and Future PerspectivesThe rationale of pharmabiotic therapies appears to be
ustified, and gastroenterologists need to be aware of an ex-anding scope for the microbiota in clinical practice. The ap-eal of probiotic strategies in IBD is their safety in comparisonith immunosuppressive drugs and corticosteroids. However,
afety concerns surrounding other pharmabiotic strategies in-luding the modification of food-grade bacteria to achieve spe-ific functional activity in vivo need to be resolved. Clinicalvidence of efficacy also requires substantiation, and scientifi-ally accredited medical evidence should underlie a clinician’shoice of pharmabiotic strategy. Moreover, mining the meta-olic activity of the indigenous microbiota might yield novelherapeutic agents. Nevertheless, therapeutic manipulation ofhe indigenous microbiota with any form of pharmabiotic re-
ains suboptimal as a result of incomplete understanding ofommensal bacteria, their anti-inflammatory properties, andost-microbial interactions. The importance of PRRs in theegulation of intestinal innate immunity and homeostatic in-eractions with the commensal microbiota has only begun to beppreciated, but it is clear that engagement with host immuneells is central to pharmabiotic action. Further studies of phys-ologic interactions within the complex network of host cell–ommensal–PRR–ligand signaling in gut health and diseasehould lead to the optimal exploitation of pharmabiotic ap-roaches to alleviate mucosal inflammation in IBD and possi-ly other intestinal diseases.
References
1. Elson CO. Genes, microbes, and T cells: new therapeutic targetsin Crohn’s disease. N Engl J Med 2002;346:614–616.
2. Duchmann R, Kaiser I, Hermann E, et al. Tolerance exists to-
wards resident intestinal flora but is broken in activeinflammatory bowel disease (IBD). Clin Exp Immunol 1995;102:448–455.
3. Rutgeerts P, Hiele M, Geboes K, et al. Controlled trial of metro-nidazole treatment for prevention of Crohn’s recurrence afterileal resection. Gastroenterology 1995;108:1617–1621.
4. Shanahan F. Nutrient tasting and signaling mechanisms in thegut V: mechanisms of immunologic sensation of intestinal con-tents. Am J Physiol Gastrointest Liver Physiol 2000;278:G191–G196.
5. Haller D, Russo MP, Sartor RB, et al. IKK beta and phosphati-dylinositol 3-kinase/Akt participate in non-pathogenic Gram-neg-ative enteric bacteria-induced RelA phosphorylation and NF-kappa B activation in both primary and intestinal epithelial celllines. J Biol Chem 2002;277:38168–38178.
6. Kagnoff MF, Eckmann L. Epithelial cells as sensors for microbialinfection. J Clin Invest 1997;100:6–10.
7. Neish AS, Gewirtz AT, Zeng H, et al. Prokaryotic regulation ofepithelial responses by inhibition of IkappaB-alpha ubiquitina-tion. Science 2000;289:1560–1563.
8. Kelly D, Campbell JI, King TP, et al. Commensal anaerobic gutbacteria attenuate inflammation by regulating nuclear-cytoplas-mic shuttling of PPAR-gamma and RelA. Nat Immunol 2004;5:104–112.
9. Otte JM, Podolsky DK. Functional modulation of enterocytes bygram-positive and gram-negative microorganisms. Am J PhysiolGastrointest Liver Physiol 2004;286:G613–G626.
10. Ma D, Forsythe P, Bienenstock J. Live Lactobacillus reuteri isessential for the inhibitory effect on tumor necrosis factor alpha-induced interleukin-8 expression. Infect Immun 2004;72:5308–5314.
11. O’Hara AM, O’Regan P, Fanning A, et al. Functional modulationof human intestinal epithelial cell responses by Bifidobacteriuminfantis and Lactobacillus salivarius. Immunology 2006;118:202–215.
12. Savage DC. Microbial ecology of the gastrointestinal tract. AnnuRev Microbiol 1977;31:107–133.
13. Eckburg PB, Bik EM, Bernstein CN, et al. Diversity of the humanintestinal microbial flora. Science 2005;308:1635–1638.
14. Moore WE, Moore LH. Intestinal floras of populations that havea high risk of colon cancer. Appl Environ Microbiol 1995;61:3202–3207.
15. Hopkins MJ, Sharp R, Macfarlane GT. Age and disease relatedchanges in intestinal bacterial populations assessed by cellculture, 16S rRNA abundance, and community cellular fatty acidprofiles. Gut 2001;48:198–205.
16. Zoetendal EG, Akkermans ADL, Akkermans-van Vilet WM, et al.The host genotype affects the bacterial community in the humangastrointestinl tract. Microb Ecol Health Dis 2001;13:129–134.
17. Berg RD. The indigenous gastrointestinal microflora. TrendsMicrobiol 1996;4:430–435.
18. Stephen AM, Cummings JH. The microbial contribution to hu-man faecal mass. J Med Microbiol 1980;13:45–56.
19. Backhed F, Ley RE, Sonnenburg JL, et al. Host-bacterial mutu-alism in the human intestine. Science 2005;307:1915–1920.
20. Whitman WB, Coleman DC, Wiebe WJ. Prokaryotes: the unseenmajority. Proc Natl Acad Sci U S A 1998;95:6578–6583.
21. Mountzouris KC, McCartney AL, Gibson GR. Intestinal microfloraof human infants and current trends for its nutritional modula-tion. Br J Nutr 2002;87:405–420.
22. Gronlund MM, Lehtonen OP, Eerola E, et al. Fecal microflora inhealthy infants born by different methods of delivery: permanentchanges in intestinal flora after cesarean delivery. J PediatrGastroenterol Nutr 1999;28:19–25.
23. Hooper LV, Gordon JI. Commensal host-bacterial relationships
in the gut. Science 2001;292:1115–1118.
282 O’HARA AND SHANAHAN CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 3
24. Xu J, Gordon JI. Inaugural article: honor thy symbionts. Proc NatlAcad Sci U S A 2003;100:10452–10459.
25. Rook GA, Brunet LR. Microbes, immunoregulation, and the gut.Gut 2005;54:317–320.
26. Umesaki Y, Okada Y, Matsumoto S, et al. Segmented filamen-tous bacteria are indigenous intestinal bacteria that activateintraepithelial lymphocytes and induce MHC class II moleculesand fucosyl asialo GM1 glycolipids on the small intestinal epi-thelial cells in the ex-germ-free mouse. Microbiol Immunol1995;39:555–562.
27. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, et al. Recogni-tion of commensal microflora by toll-like receptors is required forintestinal homeostasis. Cell 2004;118:229–241.
28. Mazmanian SK, Liu CH, Tzianabos AO, et al. An immunomodu-latory molecule of symbiotic bacteria directs maturation of thehost immune system. Cell 2005;122:107–118.
29. Shanahan F. The host-microbe interface within the gut. BestPract Res Clin Gastroenterol 2002;16:915–931.
30. Falk PG, Hooper LV, Midtvedt T, et al. Creating and maintainingthe gastrointestinal ecosystem: what we know and need toknow from gnotobiology. Microbiol Mol Biol Rev 1998;62:1157–1170.
31. Madsen K, Cornish A, Soper P, et al. Probiotic bacteria enhancemurine and human intestinal epithelial barrier function. Gastro-enterology 2001;121:580–591.
32. Hooper LV, Wong MH, Thelin A, et al. Molecular analysis ofcommensal host-microbial relationships in the intestine. Sci-ence 2001;291:881–884.
33. Cario E, Gerken G, Podolsky DK. Toll-like receptor 2 enhancesZO-1-associated intestinal epithelial barrier integrity via proteinkinase C. Gastroenterology 2004;127:224–238.
34. Bocci V. The neglected organ: bacterial flora has a crucial im-munostimulatory role. Perspect Biol Med 1992;35:251–260.
35. Gordon JI, Hooper LV, McNevin MS, et al. Epithelial cell growthand differentiation. III. Promoting diversity in the intestine: con-versations between the microflora, epithelium, and diffuseGALT. Am J Physiol 1997;273:G565–G570.
36. Conly JM, Stein K, Worobetz L, et al. The contribution of vitaminK2 (menaquinones) produced by the intestinal microflora tohuman nutritional requirements for vitamin K. Am J Gastroen-terol 1994;89:915–923.
37. Hill MJ. Intestinal flora and endogenous vitamin synthesis. Eur JCancer Prev 1997;6(Suppl 1):S43–S45.
38. Roberfroid MB, Bornet F, Bouley C, et al. Colonic microflora:nutrition and health: summary and conclusions of an Interna-tional Life Sciences Institute (ILSI) [Europe] workshop held inBarcelona, Spain. Nutr Rev 1995;53:127–130.
39. Backhed F, Ding H, Wang T, et al. The gut microbiota as anenvironmental factor that regulates fat storage. Proc Natl AcadSci U S A 2004;101:15718–15723.
40. Berg RD. Bacterial translocation from the gastrointestinal tract.Adv Exp Med Biol 1999;473:11–30.
41. Hope ME, Hold GL, Kain R, et al. Sporadic colorectal cancer:role of the commensal microbiota. FEMS Microbiol Lett 2005;244:1–7.
42. Takahashi A, Wada A, Ogushi K, et al. Production of beta-defensin-2 by human colonic epithelial cells induced by Salmo-nella enteritidis flagella filament structural protein. FEBS Lett2001;508:484–488.
43. Suzuki K, Meek B, Doi Y, et al. Aberrant expansion of seg-mented filamentous bacteria in IgA-deficient gut. Proc Natl AcadSci U S A 2004;101:1981–1986.
44. Stadnyk AW. Intestinal epithelial cells as a source of inflamma-tory cytokines and chemokines. Can J Gastroenterol 2002;16:241–246.
45. Shanahan F. Physiological basis for novel drug therapies used
to treat the inflammatory bowel diseases I. Pathophysiological
basis and prospects for probiotic therapy in inflammatory boweldisease. Am J Physiol Gastrointest Liver Physiol 2005;288:G417–G421.
46. Rescigno M, Urbano M, Valzasina B, et al. Dendritic cells ex-press tight junction proteins and penetrate gut epithelial mono-layers to sample bacteria. Nat Immunol 2001;2:361–367.
47. Macpherson AJ, Uhr T. Induction of protective IgA by intestinaldendritic cells carrying commensal bacteria. Science 2004;303:1662–1665.
48. Rimoldi M, Chieppa M, Salucci V, et al. Intestinal immunehomeostasis is regulated by the crosstalk between epithelialcells and dendritic cells. Nat Immunol 2005;6:507–514.
49. Cario E. Bacterial interactions with cells of the intestinal mu-cosa: toll-like receptors and NOD2. Gut 2005;54:1182–1193.
50. Takeda K, Akira S. Toll-like receptors in innate immunity. IntImmunol 2005;17:1–14.
51. Abreu MT, Fukata M, Arditi M. TLR signaling in the gut in healthand disease. J Immunol 2005;174:4453–4460.
52. Takeda K, Akira S. TLR signaling pathways. Semin Immunol2004;16:3–9.
53. Inohara N, Chamaillard M, McDonald C, et al. NOD-LRR pro-teins: role in host-microbial interactions and inflammatory dis-ease. Annu Rev Biochem 2005;74:355–383.
54. Pasare C, Medzhitov R. Toll pathway-dependent blockade ofCD4�CD25� T cell-mediated suppression by dendritic cells.Science 2003;299:1033–1036.
55. Fukata M, Michelsen KS, Eri R, et al. Toll-like receptor-4 isrequired for intestinal response to epithelial injury and limitingbacterial translocation in a murine model of acute colitis. Am JPhysiol Gastrointest Liver Physiol 2005;288:G1055–G1065.
56. Cario E, Podolsky DK. Intestinal epithelial TOLLerance versusinTOLLerance of commensals. Mol Immunol 2005;42:887–893.
57. Coombes JL, Robinson NJ, Maloy KJ, et al. Regulatory T cellsand intestinal homeostasis. Immunol Rev 2005;204:184–194.
58. Hart AL, Lammers K, Brigidi P, et al. Modulation of humandendritic cell phenotype and function by probiotic bacteria. Gut2004;53:1602–1609.
59. Hart AL, Kamm MA, Knight SC, et al. Quantitative and functionalcharacteristics of intestinal-homing memory T cells: analysis ofCrohn’s disease patients and healthy controls. Clin Exp Immu-nol 2004;135:137–145.
60. Groux H, O’Garra A, Bigler M, et al. A CD4� T-cell subset inhibitsantigen-specific T-cell responses and prevents colitis. Nature1997;389:737–742.
61. Duchmann R, May E, Heike M, et al. T cell specificity and crossreactivity towards enterobacteria, bacteroides, bifidobacterium,and antigens from resident intestinal flora in humans. Gut1999;44:812–818.
62. Petrof EO, Kojima K, Ropeleski MJ, et al. Probiotics inhibitnuclear factor-kappaB and induce heat shock proteins in colonicepithelial cells through proteasome inhibition. Gastroenterology2004;127:1474–1487.
63. Dubuquoy L, Jansson EA, Deeb S, et al. Impaired expression ofperoxisome proliferator-activated receptor gamma in ulcerativecolitis. Gastroenterology 2003;124:1265–1276.
64. Cobrin GM, Abreu MT. Defects in mucosal immunity leading toCrohn’s disease. Immunol Rev 2005;206:277–295.
65. Targan SR, Karp LC. Defects in mucosal immunity leading toulcerative colitis. Immunol Rev 2005;206:296–305.
66. Tysk C, Lindberg E, Jarnerot G, et al. Ulcerative colitis andCrohn’s disease in an unselected population of monozygoticand dizygotic twins: a study of heritability and the influence ofsmoking. Gut 1988;29:990–996.
67. Shanahan F. Probiotics in inflammatory bowel disease: thera-peutic rationale and role. Adv Drug Deliv Rev 2004;56:
809–818.
1
1
1
1
1
1
1
1
1
March 2007 THERAPEUTIC POTENTIAL OF THE GUT MICROBIOTA 283
68. Naser SA, Ghobrial G, Romero C, et al. Culture of Mycobacte-rium avium subspecies paratuberculosis from the blood of pa-tients with Crohn’s disease. Lancet 2004;364:1039–1044.
69. Darfeuille-Michaud A, Boudeau J, Bulois P, et al. High preva-lence of adherent-invasive Escherichia coli associated with ilealmucosa in Crohn’s disease. Gastroenterology 2004;127:412–421.
70. Suenaga K, Yokoyama Y, Nishimori I, et al. Serum antibodies toMycobacterium paratuberculosis in patients with Crohn’s dis-ease. Dig Dis Sci 1999;44:1202–1207.
71. Olsen I, Wiker HG, Johnson E, et al. Elevated antibody re-sponses in patients with Crohn’s disease against a 14-kDasecreted protein purified from Mycobacterium avium subsp.paratuberculosis. Scand J Immunol 2001;53:198–203.
72. Polymeros D, Bogdanos DP, Day R, et al. Does cross-reactivitybetween mycobacterium avium paratuberculosis and human intes-tinal antigens characterize Crohn’s disease? Gastroenterology2006;131:85–96.
73. Shanahan F, O’Mahony J. The mycobacteria story in Crohn’sdisease. Am J Gastroenterol 2005;100:1537–1538.
74. Thomas GA, Swift GL, Green JT, et al. Controlled trial of antitu-berculous chemotherapy in Crohn’s disease: a five year followup study. Gut 1998;42:497–500.
75. Martin HM, Campbell BJ, Hart CA, et al. Enhanced Escherichiacoli adherence and invasion in Crohn’s disease and colon can-cer. Gastroenterology 2004;127:80–93.
76. Bringer MA, Glasser AL, Tung CH, et al. The Crohn’s disease-associated adherent-invasive Escherichia coli strain LF82 repli-cates in mature phagolysosomes within J774 macrophages.Cell Microbiol 2006;8:471–484.
77. Ryan P, Bennett MW, Aarons S, et al. PCR detection of Myco-bacterium paratuberculosis in Crohn’s disease granulomas iso-lated by laser capture microdissection. Gut 2002;51:665–670.
78. Ryan P, Kelly RG, Lee G, et al. Bacterial DNA within granulomasof patients with Crohn’s disease: detection by laser capturemicrodissection and PCR. Am J Gastroenterol 2004;99:1539–1543.
79. Blumberg RS, Saubermann LJ, Strober W. Animal models ofmucosal inflammation and their relation to human inflammatorybowel disease. Curr Opin Immunol 1999;11:648–656.
80. Kosiewicz MM, Nast CC, Krishnan A, et al. Th1-type responsesmediate spontaneous ileitis in a novel murine model of Crohn’sdisease. J Clin Invest 2001;107:695–702.
81. Onderdonk AB, Franklin ML, Cisneros RL. Production of experi-mental ulcerative colitis in gnotobiotic guinea pigs with simpli-fied microflora. Infect Immun 1981;32:225–231.
82. Strober W, Fuss IJ, Blumberg RS. The immunology of mucosalmodels of inflammation. Annu Rev Immunol 2002;20:495–549.
83. Cong Y, Brandwein SL, McCabe RP, et al. CD4� T cells reactiveto enteric bacterial antigens in spontaneously colitic C3H/He-JBir mice: increased T helper cell type 1 response and ability totransfer disease. J Exp Med 1998;187:855–864.
84. McCarthy J, O’Mahony L, O’Callaghan L, et al. Double blind,placebo controlled trial of two probiotic strains in interleukin 10knockout mice and mechanistic link with cytokine balance. Gut2003;52:975–980.
85. Sheil B, McCarthy J, O’Mahony L, et al. Is the mucosal route ofadministration essential for probiotic function? subcutaneousadministration is associated with attenuation of murine colitisand arthritis. Gut 2004;53:694–700.
86. Madsen KL, Doyle JS, Jewell LD, et al. Lactobacillus speciesprevents colitis in interleukin 10 gene-deficient mice. Gastroen-terology 1999;116:1107–1114.
87. Schultsz C, Van Den Berg FM, Ten Kate FW, et al. The intestinalmucus layer from patients with inflammatory bowel diseaseharbors high numbers of bacteria compared with controls. Gas-
troenterology 1999;117:1089–1097. 1
88. Swidsinski A, Ladhoff A, Pernthaler A, et al. Mucosal flora ininflammatory bowel disease. Gastroenterology 2002;122:44–54.
89. Macpherson A, Khoo UY, Forgacs I, et al. Mucosal antibodies ininflammatory bowel disease are directed against intestinal bac-teria. Gut 1996;38:365–375.
90. Lodes MJ, Cong Y, Elson CO, et al. Bacterial flagellin is adominant antigen in Crohn disease. J Clin Invest 2004;113:1296–1306.
91. Gionchetti P, Rizzello F, Venturi A, et al. Oral bacteriotherapy asmaintenance treatment in patients with chronic pouchitis: adouble-blind, placebo-controlled trial. Gastroenterology 2000;119:305–309.
92. Lammers KM, Vergopoulos A, Babel N, et al. Probiotic therapy inthe prevention of pouchitis onset: decreased interleukin-1beta,interleukin-8, and interferon-gamma gene expression. InflammBowel Dis 2005;11:447–454.
93. Bibiloni R, Fedorak RN, Tannock GW, et al. VSL#3 Probiotic-mixture induces remission in patients with active ulcerativecolitis. Am J Gastroenterol 2005;100:1539–1546.
95. Macdonald TT, Monteleone G. Immunity, inflammation, and al-lergy in the gut. Science 2005;307:1920–1925.
96. Bairead E, Harmon DL, Curtis AM, et al. Association of NOD2with Crohn’s disease in a homogenous Irish population. Eur JHum Genet 2003;11:237–244.
97. Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation inNOD2 associated with susceptibility to Crohn’s disease. Nature2001;411:603–606.
98. Inoue N, Tamura K, Kinouchi Y, et al. Lack of common NOD2variants in Japanese patients with Crohn’s disease. Gastroen-terology 2002;123:86–91.
99. Pauleau AL, Murray PJ. Role of nod2 in the response of macro-phages to toll-like receptor agonists. Mol Cell Biol 2003;23:7531–7539.
00. Kobayashi KS, Chamaillard M, Ogura Y, et al. Nod2-dependentregulation of innate and adaptive immunity in the intestinaltract. Science 2005;307:731–734.
01. Li J, Moran T, Swanson E, et al. Regulation of IL-8 and IL-1betaexpression in Crohn’s disease associated NOD2/CARD15 mu-tations. Hum Mol Genet 2004;13:1715–1725.
02. Hisamatsu T, Suzuki M, Reinecker HC, et al. CARD15/NOD2functions as an antibacterial factor in human intestinal epithe-lial cells. Gastroenterology 2003;124:993–1000.
03. Netea MG, Kullberg BJ, de Jong DJ, et al. NOD2 mediatesanti-inflammatory signals induced by TLR2 ligands: implicationsfor Crohn’s disease. Eur J Immunol 2004;34:2052–2059.
04. Watanabe T, Kitani A, Murray PJ, et al. NOD2 is a negativeregulator of Toll-like receptor 2-mediated T helper type 1 re-sponses. Nat Immunol 2004;5:800–808.
05. Maeda S, Hsu LC, Liu H, et al. Nod2 mutation in Crohn’sdisease potentiates NF-kappaB activity and IL-1beta process-ing. Science 2005;307:734–738.
06. Wehkamp J, Harder J, Weichenthal M, et al. NOD2 (CARD15)mutations in Crohn’s disease are associated with diminishedmucosal alpha-defensin expression. Gut 2004;53:1658–1664.
07. Cario E, Podolsky DK. Differential alteration in intestinal epithe-lial cell expression of toll-like receptor 3 (TLR3) and TLR4 ininflammatory bowel disease. Infect Immun 2000;68:7010–7017.
08. Cario E, Rosenberg IM, Brandwein SL, et al. Lipopolysaccharideactivates distinct signaling pathways in intestinal epithelial celllines expressing Toll-like receptors. J Immunol 2000;164:966–972.
09. Cho JH, Nicolae DL, Gold LH, et al. Identification of novel
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
ms4
pA0
284 O’HARA AND SHANAHAN CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 3
susceptibility loci for inflammatory bowel disease onchromosomes 1p, 3q, and 4q: evidence for epistasis between1p and IBD1. Proc Natl Acad Sci U S A 1998;95:7502–7507.
10. Franchimont D, Vermeire S, El Housni H, et al. Deficient host-bacteria interactions in inflammatory bowel disease? the toll-like receptor (TLR)-4 Asp299gly polymorphism is associatedwith Crohn’s disease and ulcerative colitis. Gut 2004;53:987–992.
11. Torok HP, Glas J, Tonenchi L, et al. Crohn’s disease is associ-ated with a toll-like receptor-9 polymorphism. Gastroenterology2004;127:365–366.
12. Sitaraman SV, Klapproth JM, Moore DA 3rd, et al. Elevatedflagellin-specific immunoglobulins in Crohn’s disease. Am JPhysiol Gastrointest Liver Physiol 2005;288:G403–G406.
13. Shanahan F. Probiotics and inflammatory bowel disease: is therea scientific rationale? Inflamm Bowel Dis 2000;6:107–115.
14. Shanahan F. Host-flora interactions in inflammatory bowel dis-ease. Inflamm Bowel Dis 2004;10(Suppl 1):S16–S24.
15. Braat H, Rottiers P, Hommes DW, et al. A phase I trial withtransgenic bacteria expressing interleukin-10 in Crohn’s dis-ease. Clin Gastroenterol Hepatol 2006;4:754–759.
16. Allen SJ, Okoko B, Martinez E, et al. Probiotics for treatinginfectious diarrhoea. Cochrane Database Syst Rev 2004:CD003048.
17. D’Souza AL, Rajkumar C, Cooke J, et al. Probiotics in preventionof antibiotic associated diarrhoea: meta-analysis. BMJ 2002;324:1361.
18. O’Mahony L, McCarthy J, Kelly P, et al. Lactobacillus and bi-fidobacterium in irritable bowel syndrome: symptom responsesand relationship to cytokine profiles. Gastroenterology 2005;128:541–551.
19. Lin HC, Su BH, Chen AC, et al. Oral probiotics reduce theincidence and severity of necrotizing enterocolitis in very lowbirth weight infants. Pediatrics 2005;115:1–4.
20. Schultz M, Veltkamp C, Dieleman LA, et al. Lactobacillus plan-tarum 299V in the treatment and prevention of spontaneouscolitis in interleukin-10-deficient mice. Inflamm Bowel Dis 2002;8:71–80.
21. Gionchetti P, Rizzello F, Helwig U, et al. Prophylaxis of pouchitisonset with probiotic therapy: a double-blind, placebo-controlledtrial. Gastroenterology 2003;124:1202–1209.
22. Kruis W, Fric P, Pokrotnieks J, et al. Maintaining remission ofulcerative colitis with the probiotic Escherichia coli Nissle 1917is as effective as with standard mesalazine. Gut 2004;53:1617–1623.
23. Zocco MA, dal Verme LZ, Cremonini F, et al. Efficacy of Lacto-bacillus GG in maintaining remission of ulcerative colitis. Ali-ment Pharmacol Ther 2006;23:1567–1574.
24. Rembacken BJ, Snelling AM, Hawkey PM, et al. Non-pathogenicEscherichia coli versus mesalazine for the treatment of ulcer-ative colitis: a randomised trial. Lancet 1999;354:635–639.
25. Shanahan F, Guarner F, Von Wright A, Vilpponen-Salmela T, etal. A one year, randomised, double-blind, placebo controlledtrial of a lactobacillus or a bifidobacterium probiotic for mainte-nance of steroid-induced remission of ulcerative colitis. Gastro-enterology 2006;130:A-44.
26. Sheil B, MacSharry J, O’Callaghan L, et al. Role of interleukin(IL-10) in probiotic-mediated immune modulation: an assess-ment in wild-type and IL-10 knock-out mice. Clin Exp Immunol2006;144:273–280.
27. Bousvaros A, Guandalini S, Baldassano RN, et al. A random-ized, double-blind trial of Lactobacillus GG versus placebo inaddition to standard maintenance therapy for children with
Crohn’s disease. Inflamm Bowel Dis 2005;11:833–839. m
28. Marteau P, Lemann M, Seksik P, et al. Ineffectiveness of Lac-tobacillus johnsonii LA1 for prophylaxis of postoperative recur-rence in Crohn’s disease: a randomised, double blind, placebocontrolled GETAID trial. Gut 2006;55:842–847.
29. Shanahan F. Probiotics: a perspective on problems and pitfalls.Scand J Gastroenterol Suppl 2003:34–36.
30. Abreu MT, Vora P, Faure E, et al. Decreased expression ofToll-like receptor-4 and MD-2 correlates with intestinal epithelialcell protection against dysregulated proinflammatory gene ex-pression in response to bacterial lipopolysaccharide. J Immunol2001;167:1609–1616.
31. Wang JH, Doyle M, Manning BJ, et al. Induction of bacteriallipoprotein tolerance is associated with suppression of toll-likereceptor 2 expression. J Biol Chem 2002;277:36068–36075.
32. Cario E, Brown D, McKee M, et al. Commensal-associatedmolecular patterns induce selective toll-like receptor-traffickingfrom apical membrane to cytoplasmic compartments in polar-ized intestinal epithelium. Am J Pathol 2002;160:165–173.
33. Otte JM, Cario E, Podolsky DK. Mechanisms of cross hypore-sponsiveness to Toll-like receptor bacterial ligands in intestinalepithelial cells. Gastroenterology 2004;126:1054–1070.
34. Gewirtz AT, Navas TA, Lyons S, et al. Cutting edge: bacterialflagellin activates basolaterally expressed TLR5 to induce epi-thelial proinflammatory gene expression. J Immunol 2001;167:1882–1885.
35. Melmed G, Thomas LS, Lee N, et al. Human intestinal epithelialcells are broadly unresponsive to Toll-like receptor 2-dependentbacterial ligands: implications for host-microbial interactions inthe gut. J Immunol 2003;170:1406–1415.
36. Garlanda C, Riva F, Polentarutti N, et al. Intestinal inflammationin mice deficient in Tir8, an inhibitory member of the IL-1 recep-tor family. Proc Natl Acad Sci U S A 2004;101:3522–3526.
37. Polentarutti N, Rol GP, Muzio M, et al. Unique pattern of expres-sion and inhibition of IL-1 signaling by the IL-1 receptor familymember TIR8/SIGIRR. Eur Cytokine Netw 2003;14:211–218.
38. Brint EK, Xu D, Liu H, et al. ST2 is an inhibitor of interleukin 1receptor and Toll-like receptor 4 signaling and maintains endo-toxin tolerance. Nat Immunol 2004;5:373–379.
40. Boone DL, Turer EE, Lee EG, et al. The ubiquitin-modifyingenzyme A20 is required for termination of Toll-like receptorresponses. Nat Immunol 2004;5:1052–1060.
41. Wang YY, Li L, Han KJ, et al. A20 is a potent inhibitor of TLR3-and Sendai virus-induced activation of NF-kappaB and ISRE andIFN-beta promoter. FEBS Lett 2004;576:86–90.
42. Netea MG, Ferwerda G, de Jong DJ, et al. Nucleotide-bindingoligomerization domain-2 modulates specific TLR pathways forthe induction of cytokine release. J Immunol 2005;174:6518–6523.
43. van Heel DA, Ghosh S, Hunt KA, et al. Synergy between TLR9and NOD2 innate immune responses is lost in genetic Crohn’sdisease. Gut 2005;54:1553–1557.
Address requests for reprints to: Professor Fergus Shanahan, Ali-entary Pharmabiotic Centre, Clinical Science Building, Cork Univer-
ity Hospital, Cork, Ireland. e-mail: [email protected]; fax: �353-21-345300.Supported in part by Science Foundation Ireland. F.S. is also sup-
orted by the Irish Health Research Board, the Higher Educationuthority of Ireland, and the European Union (PROGID QLK-2000-0563) and has been affiliated with a campus-based company (Ali-
