Review

From Disease fo

0022-2836/© 2019 Publish

Good Gone Bad: One Toxin Awayr Bacteroides fragilis

Ezequiel Valguarnera and Juliane Bubeck Wardenburg

Department of Pediatrics, Washington University School of Medicine, 660 S. Euclid Ave. Box 8208, St. Louis, MO 63110

Correspondence to Juliane Bubeck Wardenburg: jbubeck@wustl.eduhttps://doi.org/10.1016/j.jmb.2019.12.003Edited by John Ralston Brannon

Abstract

The human gut is colonized by hundreds of trillions of microorganisms whose acquisition begins during earlyinfancy. Species from the Bacteroides genus are ubiquitous commensals, comprising about thirty percent ofthe human gut microbiota. Bacteroides fragilis is one of the least abundant Bacteroides species, yet is themost common anaerobe isolated from extraintestinal infections in humans. A subset of B. fragilis strains carrya genetic element that encodes a metalloprotease enterotoxin named Bacteroides fragilis toxin, or BFT.Toxin-bearing strains, or Enterotoxigenic B. fragilis (ETBF) cause acute and chronic intestinal disease inchildren and adults. Despite this association with disease, around twenty percent of the human populationappear to be asymptomatic carriers of ETBF. BFT damages the colonic epithelial barrier by inducing cleavageof the zonula adherens protein E-cadherin and initiating a cell signaling response characterized byinflammation and c-Myc-dependent pro-oncogenic hyperproliferation. As a consequence, mice harboringgenetic mutations that predispose to colonic inflammation or tumor formation are uniquely susceptible to toxin-mediated injury. The recent observation of ETBF-bearing biofilms in colon biopsies from humans with coloncancer susceptibility loci strongly suggests that ETBF is a driver of colorectal cancer. This article will addressETBF biology from a host-pathobiont perspective, including clinical data, analysis of molecular mechanisms ofdisease, and the complex ecological context of the human gut.

© 2019 Published by Elsevier Ltd.

Human beings are persistently colonized bycommensal microorganisms. The gut microbiota, inparticular, is composed of tens of trillions of micro-organisms and is normally acquired from birth toaround 3 years of age [1]. Bacteroides fragilis is aubiquitous member of the human gut microbiota. Itbelongs to the phylum Bacteroidetes, which togetherwith the phylum Firmicutes constitutes about 80% ofthe total human gut microbiota [2,3]. B. fragilis is acommensal organism that can become an opportu-nistic pathogen in certain individuals. AlthoughB. fragilis comprises around 0.1e0.5% of total gutbacteria, it is the most frequently isolated anaerobefrom peritoneal and abdominal abscesses, as wellas from samples of bloodstream infections [4]. Thisspecies became notorious in the 1970s becausereports have shown its relevance in peritoneal,abdominal, intestinal, and blood infections, as wellas its propensity to develop antibiotic resistance[5e11]. Most B. fragilis strains will remain as a

ed by Elsevier Ltd.

commensal during the lifetime of an individual, withcolonization occurring during early childhood[12e14]. Nevertheless, a subset of strains canproduce a proteolytic enterotoxin, named B.fragilistoxin (BFT), or fragilysin, that causes secretorydiarrhea and colonic epithelial damage [15]. BFT isamong the most studied virulence factors ofB. fragilis, and current evidence suggests that thistoxin may be a driver for chronic colitis and colorectalcancer [16e18]. Toxin-producing strains, or enter-otoxigenic B. fragilis (ETBF), were discovered in the1980s and were found to be associated with diarrheain lambs, calves, pigs, foals, and humans and alsofound to be present in sewage waters [19e24]. Sincethose initial reports, multiple studies have enhancedour understanding of ETBF as a relevant pathogen inhumans and BFT as the key virulence factor indisease. Humans are variably colonized with non-toxigenic B. fragilis (NTBF) or ETBF strains, andmost frequently, a single B. fragilis strain type is

Journal of Molecular Biology (2020) 432, 765e785

766 Good gone bad

found in fecal samples from individual healthydonors [14]. Presently, knowledge of genetic deter-minants of human colonization by B. fragilis is in itsinfancy. The host and microbial factors that governtemporal and spatial plasticity of the B. fragilis nicheare not understood; similarly, determinants respon-sible for progression from asymptomatic carriage ofB. fragilis toward disease remain undefined. In thiswork, we will review current knowledge on toxigenicB. fragilis (ETBF) and discuss the dual nature of thisorganism as both a commensal and pathogen in thecontext of a complex ecological scenario.

Colonization of Humans by ETBF

ETBF has been identified in stool samples fromhealthy individuals and patients with diarrhea.Interestingly, the prevalence of ETBF is abouttwice as high in patients (20e30%) than in healthyindividuals (10e20%) [25e27]. Reports from distincturban and underdeveloped geographic locationsshow higher percentages of ETBF colonization inchildren older than 1 year with diarrhea than in age-matched controls [25,28e32]. Interestingly, in chil-dren younger than 1 year, prevalence is the lowest inthe general population and is not associated withdiarrhea, suggesting that the developmental trajec-tory is important for diarrheal association to occur[28,32]. An association of ETBF with chronicintestinal disease has been established for morethan 20 years, the first reported in patients withinflammatory bowel disease (IBD) [33]. ETBF is alsopositively associated with ulcerative colitis andcolonic neoplasia [34e41]. The association ofETBF in patients with colorectal cancer (CRC)includes sporadic and familial cases, indicating thatprogression of this disease, regardless of the onset,has physiologic commonalities. The disparitybetween high asymptomatic carriage of ETBF inhuman populations and the low number of cases ofETBF-associated IBD and sporadic CRC suggeststhat pathogenicity is not stochastic and depends onunknown host susceptibility determinants. It iscurrently not known if ETBF strains typically produceenterotoxin in asymptomatic carriers [42]. Studiesmeasuring amounts of secreted BFT in both healthyand diseased individuals will be required to establisha functional association between carriage of ETBFand specific manifestations of disease. Furthermore,the fact that NTBF accounts for the vast majority ofB. fragiliseassociated infections increases the com-plexity of studies that will be required to define bothepidemiologic and mechanistic contributions ofETBF to human disease.As non-toxigenic (NTBF readily colonizes the

human colon and is suggested to benefit thedevelopment of the host T-cell response throughstudies conducted in murine model systems [43,44],

the following question arisesdwhat is the benefit oftoxin production by B. fragilis in an organism thatexhibits a commensal existence? Several plausibleanswers to this question can be put forth: (1)intoxication facilitates colonic niche acquisition andsurvival. A report from our group showed that aparticular strain of ETBF (43859) can colonize aniche previously occupied by NTBF (TM4000) in atoxin-dependent manner [45]. The presence of thetoxin was not, however, a universal determinant ofintraluminal niche acquisition or competition, whichalso relies in part on the B. fragilis type VI secretionsystem (T6SS) and other genetic determinants[46e54]. A precedent for toxin-mediated modulationof the bacterial niche to facilitate survival hasrecently been demonstrated for Vibriocholerae [55].(2) Intoxication facilitates transmission of ETBF. Aspreviously mentioned, in children aged between 1and 5 years, the ETBF count is increased in patientswith diarrhea. Toxin production and the associateddiarrhea may represent a key strategy for transmis-sion of the pathogen from human to human, byincreasing fecal-oral contamination. Human intest-inal pathogens such as V.cholerae and Salmonellaenterica use diarrhea as a means to transmitinfection between human hosts, so this could alsobe the case for ETBF [56,57]. (3) Intoxication permitsan extraintestinal lifestyle for B. fragilis. As a leadingcause of anaerobic infection, B. fragilis exhibits theability to invade multiple tissue sites within the body[58]. Studies that have investigated the presence ofETBF in extraintestinal infections have not reached aclear consensus; however, they suggest that ETBFmay be more represented in bloodstream isolatesand samples of vaginal infections than in otherisolation sites [59e65]. BFT has been shown to betoxic in vitro to kidney and lung epithelial cells as wellas to the endothelium, suggesting a potential role inextraintestinal infections [66,67]. Although othergenetic determinants of colonization, including cap-sular polysaccharide, have been implicated inextraintestinal infections and abscess formation[68], such roles have not been clearly defined forBFT. A considerable amount of epidemiologicinvestigation focused within the human populationand paired mechanistic analyses in model systemswill be needed to provide clarity on the potentialevolutionary advantage of maintenance of BFT andits associated pathogenicity island within theB. fragilis genome.

Discovery of BFT and Toxin Activation

The first report of enterotoxicity by B. fragilis waspresented in 1984, describing diarrhea in lambs [19].At that time, there was a clear distinction betweenenterotoxigenic and nonenterotoxigenic strains, butthe etiology of the diarrhea was unknown. In 1992,

Fig. 1. Known molecular mechanisms that governsecretion of BFT. The RprX/RprY two-component systemis activated under certain as-yet-unidentified conditions,with concomitant phosphorylation of RprY (RprY-P) andbinding to the bft promoter, repressing gene expression. Ina noneRprY-P situation, bft is expressed and translatedas a signal peptide-containing protein into the periplasmicspace. Pre-BFT is most likely a lipoprotein that getstransported to the outer membrane. Surface-exposed pre-BFT can be cleaved by Fragipain or host gut proteases,releasing the active metalloprotease C-terminal domaininto the extracellular milieu. BFT, Bacteroides fragilis toxin;Fpn, fragipain.

767Good gone bad

Weikel et al. [69] were the first to demonstrate that acomponent in culture supernatants from enterotoxi-genic strains was causing the secretory phenotypepreviously observed when injecting lamb ileal loops(LILs) with bacteria. The authors developed an invitro assay of BFT cell damage using coloncarcinoma HT29 cell lines, allowing for simultaneousscreening of multiple enterotoxigenic strains [69].The toxin was initially purified from culture super-natants as a 20-KDa protein and characterized asenterotoxic and cytotoxic in LIL and HT29 models,respectively. Partial cloning and expression of thetoxin revealed that it is a heat-labile metalloprotease[70,71]. Diagnostics for detection of the toxin werealso developed, allowing for confirmation that theenterotoxin is present in ETBF strains isolated fromhumans with ETBF-associated diarrhea [72,73]. Thecomplete sequence of the bft gene was cloned, andcharacterization of the amino acid sequence andbiochemical analysis suggested that BFT is pro-duced as a protoxin that is processed releasing theactive C-terminal domain into the extracellular milieu[74,75]. Further studies showed that cleavage of theprotoxin is not dependent on the zinc-binding motif ofthe metalloprotease domain and that the whole C-terminal domain is essential for toxin activity [76,77].Detailed structural analysis determined that the N-terminal prodomain is likely involved in the secretionof the protoxin through the cell envelope and in theinhibition of toxin activity within the bacterial cell [78].Interestingly, the N-terminal domain contains alipoprotein signal peptide representing a uniquefold, whereas the C-terminal metalloproteasedomain is a xenolog of eukaryotic A Disintegrinand Metalloprotease (ADAM) proteases, suggestinghorizontal acquisition of bft by B. fragilis [78]. Asshown in Fig. 1, cleavage of the protoxin can occur invivo by host proteases from the intestinal lumen.Interestingly, in bloodstream, BFT can only beactivated by a single endogenous bacterial cysteineprotease named fragipain (Fpn) [67,79]. An ETBFfpn mutant strain was unable to cause lethality inmice when bacteria were administered intrave-nously, showing the importance of Fpn in the contextof ETBF-mediated sepsis [67]. The fpn gene ispresent in most NTBF and ETBF strains, suggestingan additional role of Fpn beyond BFT activation.

Genomic Context and TranscriptionalRegulation

Studies on the genetic locus that encodesBFThavebeen crucial to understand its origin and distributionwithin strains. The bft gene is part of the ~6-kbpB. fragilis pathogenicity island (BfPAI). The BfPAI isabsent in NTBF strains, and it also encodes anothermetalloprotease (mpII), whose role in pathogenicity isunclear [80,81]. Three bft variants have been identi-

fied, bft-1, bft-2, and bft-3 [62,82,83]. Most ETBFstrains carry a single bft variant; each variantdemonstrates different degrees of potency in vitroand in germ-free mice. Among these, BFT-2 exhibitsthe greatest potential to elicit tissue damage [84e86].Population-wide carriage has been studied, and BFT-1 was found to be the most widely distributed toxinvariant within ETBF human isolates, whereas BFT-3seems to be geographically restricted to southeastAsia [31,62,87,88].The BfPAI contains 12-bp repeats on both ends,

and it is always inserted in the same genomiclocalization. A 17-bp GC-rich sequence is also foundat the putative insertion site in NTBF strains [80]. TheBfPAI is present within the conjugative transposonCTn86, supporting the hypothesis of horizontalacquisition of the BfPAI by ETBF strains [89e92].Genome sequencing and comparison indicates highgenetic diversity between ETBF strains [92,93].Some ETBF strains are more closely related toNTBF strains than to other ETBF strains, furthersupporting that the BfPAI has been acquired throughmultiple independent horizontal transfer events [92].Thus, the only known genetic determinant thatallows for differentiation between ETBF and NTBFstrains is the presence or absence of the BfPAI,respectively.The BfPAI contains a promoter sequence

upstream of the bft gene that controls transcription

768 Good gone bad

[94]. A key mechanism for control of toxin productionis regulation of expression by the two-componentsystem (TCS) RprX/RprY. The TCS is a commonstrategy to couple environmental stimuli with generegulation in bacteria, in which a sensor histidinekinase (RprX) detects specific stimuli and phosphor-ylates a response regulator (RprY) that binds topromoter regions in the bacterial genome inducing orrepressing gene expression [95]. PhosphorylatedRprY binds to the promoter upstream of the bft geneand represses expression (Fig. 1). Overexpressionof RprY abrogates toxin expression, and deletion ofrprY causes increased nonregulated expression ofbft [96]. Hence, ETBF strains could be sensingenvironmental cues to determine the optimal situa-tion for toxin expression. BFT can be repressed byfermentable sugars and induced by heat andoxidative stress [45]. Although RprX/Y seems thekey to bft regulation in vivo, stimuli for this TCSinside the host remain unknown.

Fig. 2. BFT-host cell interactions.BFT binds to colonicepithelial cells (CECs) through an unknown receptor andtriggers cleavage of E-cadherin. BFT, Bacteroides fragilistoxin; ETBF, enterotoxigenic Bacteroides fragilis.

Toxin-Host Interactions

Since the discovery of BFT, the molecularmechanisms that govern damage of host cellshave been described. When the HT29 cell modelwas established, it was observed that BFT causedcell rounding and shedding, presumed to result fromdamage to the intercellular junction [66,71]. Morpho-logical changes in BFT-treated HT29 cells werefound to be a product of F- and G-actin rearrange-ment [97,98]. BFT binds to an unidentified cellreceptor in a protease-dependent manner andinduces cleavage of the extracellular domain of thezonula adherens protein E-cadherin; this cleavageevent occurs only in the context of intact cells (Fig. 2)[99e101]. Given the similarity of BFT to ADAM10, itis possible that BFT contributes directly to E-cadherin cleavage [78]. As there is no evidence fordirect processing of cellular E-cadherin by BFT yet,an alternative hypothesis is that activity of BFTtoward its receptor or another host protein couldtrigger a signaling pathway responsible for the lossof E-cadherin (Fig. 2) [100]. A study using humancolonic biopsies incubated with BFT on the luminalor serosal side of the epithelium showed moredamage on the latter, suggesting BFT receptorpolarization toward the basal cell region [102].BFT-dependent cleavage of E-cadherin causesloss of cell-cell contacts and cell rounding, whichrequires the intramembrane protease presenilin-1/gamma secretase complex [103]. The intracellulardomain of E-cadherin is normally bound to a- and b-catenins [104]. When b-catenin is dissociated fromE-cadherin, it can function as a transcription factor ina Wnt-dependent manner, inducing cell proliferationthrough activation of the c-Myc pathway [104](Fig. 3). BFT-mediated cleavage of E-cadherin

promotes migration of beta-catenin to the nucleus[105]. In addition, E-cadherin cleavage by BFTtriggers induction of mitogen-activated proteinkinases (MAPKs) and the NF-kappa B pathway,thus increasing secretion of interleukin (IL)-8, achemokine that attracts polymorphonuclear cells[106e111]. NF-kappa B activation controls fluidsecretion of intestinal cells through induction ofCOX2 and an increase in prostaglandin E2 levels[112]. COX2 and heme oxygenase-1 induction byBFT is related to a delay of apoptosis in intestinalepithelial cells [113,114]. BFT can also inducemechanisms of host defense such as beta-defensin2 and expression of the siderophore-binding anti-microbial protein lipocalin-2 [115,116] and alsoincreases autophagy in human umbilical veinendothelial cells through the MAPK, AP-1 [117].Signaling pathways affected by the toxin causedifferential gene expression and epigenetic changesin HT29 cells [118]. Variations in the host celltranscription profile and epigenetic marks are lostwhen toxin stimulation is withdrawn, suggesting thatcontinuous toxin secretion by ETBF could be acomponent of disease progression [118].The relevance of mechanistic findings in cell

culture systems has been elaborated in animalmodels of ETBF-associated disease. Rabizadehet al [119] and Rhee et al [120] showed for the firsttime in specific pathogenefree (SPF) C57BL/6 micethat ETBF causes acute and persistent colitisin mice, driven by cleavage of E-cadherin by BFTin vivo. Activation of STAT3 in vivo can be observed

Fig. 3. Pro-oncogenic signaling by BFT. CECsinduce cMyc through b-catenin and STAT3 activating acell proliferation program. cMyc also induces spermineoxidase (SMOX), in a BFT-dependent manner, generatinga source of ROS (Reactive Oxygen Species) that cancontribute to DNA damage. BFT, Bacteroides fragilis toxin;CEC, colonic epithelial cell.

769Good gone bad

at 24 h after infection in mucosal immune cells andcolonic epithelial cells [121], inducing c-myc expres-sion and concomitant cell proliferation (Fig. 3)[122,123]. Colonic histology of acute colitis (2 daysafter infection) in SPF mice shows rupture of cell-celladhesions and epithelial exfoliation, with the pre-sence of immune cell infiltrates [121]. Conversely,chronic colitis, observed between 7 days and up to16 months after infection, shows progressive hyper-plasia of the colonic crypts, consequent with BFT-dependent induction of a cell hyperproliferationprogram [120,121].

Bacterial Factors in Colonization andDisease

Species from the Bacteroides genus are acquiredearly in life and are commonly found in the lowergastrointestinal tract [1]. There are many potentialdeterminants of success for a given strain to becomeestablished in its niche, including host diet, devel-opment, antibiotic usage, and interactions with othermembers of the microbiota [12,124,125]. Althoughexternal factors, such as the aforementioned ones,are pivotal for colonization, B. fragilis carry geneticdeterminants of colonization that contribute to nicheoccupancy by interacting with the host and microbialcompetitors. In this section, we will emphasize two

main aspects of colonization: (1) the relevance of theB. fragilis capsule and other factors in host-patho-biont relationships and (2) mechanisms that mediateinterbacterial competition in the gut, in particularT6SSs and secreted antimicrobial toxins.

Capsular Polysaccharide

The B. fragilis capsule was described in the 1970sin the studies by Kasper [126], Kasper et al [127,129]and Lindberg et al [128] as a distinctive feature of thisorganism relative to other species from the genusBacteroides. These observations, coupled with thefact that B. fragilis is the most commonly isolatedBacteroides species from anaerobic infections inhumans despite its low relative abundance in themicrobiome, led to the hypothesis that the B. fragiliscapsule is linked to disease pathogenesis. Indeed,studies have indicated that capsule reduces phago-cytosis by immune cells, thereby increasing bacterialfitness outside the colonic lumen [130]. Consistentwith this finding, capsular polysaccharide is the maincontributor to extraintestinal abscess formation byB. fragilis [128,130e132].B. fragilis harbors 8 distinct genomic loci, each of

which encodes the required enzymes for synthesisof a specific capsular polysaccharide variant.Expression of these loci is regulated throughpromoter inversion to “on” or “off” configurations[133,134]. Capsular polysaccharide levels areincreased when B. fragilis is animal passaged,indicating its key role in survival within the host[130]. Monocolonization experiments in germ-freemice show that any capsular polysaccharide issufficient for niche establishment [135]; however,this finding does not indicate that all capsule variantsare immunologically equivalent within the host or thata single capsule variant is sufficient for colonizationin a complex ecosystem [135,136]. B. fragilis cap-sular polysaccharides are composed of repeatingunits of zwitterionic glycans [132]. Polysaccharide A(PSA), unlike most glycan antigens, can be pre-sented within the context of MHCII antigen-present-ing cells, inducing formation of regulatory T cells thatcontribute to immune tolerance toward B. fragilis[137e139]. PSA induces IL-10 production throughthe Toll-like receptor (TLR) 2 pathway and repressesproduction of IL-17 in germ-free mice, promoting alow inflammation environment [44]. The lack of PSAhowever does not decrease IL-10 in SPF micecolonized with NTBF [140,141]. For ETBF, recentresearch using snap-frozen samples from humancolonic tissue biopsies showed a negative associa-tion between PSA expression and carriage of bftgene [140], suggesting the potential to exacerbateinflammation. Together, these studies suggest thereare likely strain- and context-dependent regulatoryelements that modulate the host response toB. fragilis polysaccharides.

770 Good gone bad

Polysaccharide Utilization Loci

A genetic screening by Lee et al [142] identified the“commensal colonization factor” (ccf) operon that isconserved among many Bacteroides species andconfers stable niche acquisition in mice. Thestructure of the ccf operon resembles typicalpolysaccharide utilization loci (PULs) from Bacter-oides [142]. PULs are specialized loci dedicated tothe breakdown and assimilation of complex glycans;these constitute close to 20% of the genomic contentof some Bacteroides species [143]. Different PULscan process distinct polysaccharides, allowing Bac-teroides species to process both dietary and hostglycans [143e147]. Versatility in nutrient utilizationincreases the chances of survival during host dietchanges; thus, PUL diversity within Bacteroidesspecies is the key to niche acquisition and stableniche occupancy. Similar to most PULs, the ccfoperon consists of a transcriptional regulator sigma/antisigma factor pair (ccfA and ccfB) that controlsgene expression. Downstream of ccfA/B is theTonB-dependent receptor that transports breakdownproducts into the cell periplasm (ccfC) and the SusD-like accessory lipoprotein (ccfD) that is required forcontrol of nutrient transport by ccfC. The last gene inthe locus is a putative chitobiase (ccfE) that cleavesa complex substrate polymer into oligomers for cellacquisition. Functional characterization of ccfshowed that ccfA, ccfC, and ccfD are required forstable colonization, precluding secondary coloniza-tion events by Bacteroides from the same species.On the contrary, ccfE is not required for ccf function,most likely owing to enzymatic redundancy fromother B. fragilis gene products [142]. A recent studyshowed that the ccf locus controls capsule variants,repressing PSA and inducing polysaccharide C(PSC) expression. ccf function allows IgA-depen-dent niche establishment in response to PSC,enabling B. fragilis cells to localize closer to the gutepithelium [142,148]. The specific environmentalcues that induce ccf expression in vivo remain tobe defined.

B. fragilis Hemolysins

Many gram-negative and gram-positive bacteriasecrete enzymes that lyse red blood cells, known ashemolysins [149e151]. B. fragilis is not anexception as many strains carry hemolysin orthologgenes [152]. Hemolysins A and B (HlyA and HlyB,respectively) have been shown to be enzymaticallyactive in vitro against red blood cells [152]. B. fragilismutant strains lacking genes hlyA/B show reducedfitness in vitro and in vivo, indicating that hemolysinsmay be involved in colonization [153]. At present,there is no clear evidence showing a specific role ofhemolysins in the pathogenesis of disease causedby B. fragilis.

Neuraminidase

Bacterial glycosidases are ubiquitous and areused by many organisms to degrade complexpolysaccharides for nutritional and ecological pur-poses [154]. B. fragilis and other Bacteroides sp.Can cleave sialic acid from host glycoproteins bysecreting neuraminidase [155,156]. Sialic acid iscommonly present on host glycoproteins [157]. Themost studied neuraminidase from B. fragilis isencoded by the nanH gene. Deletion of nanHrenders a mutant strain that is outgrown by thewild-type strain in vitro and in vivo [158]. Sialic acidrelease from host glycoproteins and utilization islikely to function as an extra measure of nutrientversatility, crucial to a gut commensal. Neuramini-dase seems to also increase binding of B. fragilis tomammalian epithelial cells, by releasing sialic acidand uncovering other glycan moieties [159,160]. Thepresence of the nanH gene has also been used as ameans of additional taxonomic information forclassification of B. fragilis isolates from infectionsites and stool samples [161].

Proteases

B. fragilis has been shown to produce otherproteases than BFT and fragipain. A subset ofcysteine proteases of the C10 family (BFP) wasfound in genomes of B. fragilis strains [162]. Genesbfp1e4 are present within mobile genetic elements,which is indicative of genetic acquisition by B. fragilisthrough horizontal transmission. bfp gene expres-sion, particularly bfp4, is induced in vitro withincreased oxygen concentrations, suggesting aputative role of BFP in adaptation to environmentalchanges [163]. Bfp genes have been found both inbacterial isolates from infected sites and in stoolsamples from healthy donors [162]; however, iso-genic bfp deletion strains have not been evaluated toassess the role of BFP in pathogenicity. A fibrinogen-degrading protease has also been identified inB. fragilis, although its role in virulence is unknown[164,165].

Adhesive Molecules

Bacterial cells adhere to substrates via adhesins,proteins that bind specifically to receptors from hostcells [166,167]. Many adhesins are expressed as apart of fimbriae or pili [168]. In commensal gutbacteria such as Bacteroides, binding to intestinalmucus and epithelial cells favors stable gut nichecolonization. In humans, piliated strains are morecommonly associated with abscess and healthystool samples, whereas nonpiliated strains areenriched in isolates from blood infections [169].Adhesive-piliated Bacteroides are more easily pha-gocytosed by neutrophils; thus, loss of cell

771Good gone bad

adhesiveness could promote extraintestinal disse-mination and immune evasion [170]. Specific bindingto red blood cells, or hemagglutination, has beenobserved more frequently in blood isolates than inabscess or healthy stool isolates [171e173]. Sialicacid has been shown to be required for hemagglu-tinin phenotypes, and it is proposed as the receptorfor lectin-like adhesins from B. fragilis [159]. Thegenetic basis for cell adhesiveness in B. fragilis stillremains unknown.B. fragilis also presents extracellular matrix

(ECM)ebinding proteins, suggesting a role in extra-intestinal survival. Binding to fibronectin, the mostabundant protein in the ECM, is mediated by aprotein similar to a TonB-dependent receptor,BF1991 [174,175]. Surprisingly, a mutant strainlacking bf1991 is more adhesive to fibronectin thatthe wild-type strain, indicative of redundant fibro-nectin-binding proteins. Bf1991 mutants are moresusceptible in vitro to phagocytosis by macrophages[175]. Binding of B. fragilis to laminin-1 and collagen-1 has also been reported [176e178]. Similar to otherhuman pathogens, B. fragilis can interact withplasma proteins related to coagulation [179]. BF-FBP is a 54-KDa protein that binds fibrinogen, themajor component of fibrin abscess formation [165].Other proteins can bind plasminogen and high-molecular-mass kininogen, yet the role in B. fragilismanipulation of host coagulation is not clear[180,181]. The diversity of host targets thatB. fragilis can potentially bind to allow us tohypothesize that different binding patterns might beinduced in vivo as a response to environmental andhost cues.

Response to Oxidative Stress

Gut bacteria are exposed to oxygen outside of thehost and within the intestinal cavity owing to theoxygen concentration gradient across the lumen[182]. Bacteroides clinical isolates are more aero-tolerant than nonclinical isolates, suggesting a rolefor such adaptation in pathogenesis [183]. Oxidativestress response (OSR) mechanisms might beespecially protective in the context of extraintestinalinfections, where tissues are more oxygenated thanthe gut lumen [184,185]. The anaerobic/aerobic fluxcan modify gene expression profiles, potentiallyaltering pathogenicity of the organism and itsassociation with abdominal and peritoneal infections[186,187]. Although B. fragilis can grow in nanomolaroxygen concentrations [188], exposure to higherlevels can arrest its growth [189]; hence, OSR isrequired to avoid oxidative cellular damage [184].OSR is achieved by a vast repertoire of proteins withdifferent functions, including superoxide dismutase[190], catalase [191], peroxidases [185,192], ironstorage proteins [193e196], and thioredoxins [197].The transcription factor OxyR is responsible for the

induction of many genes in the OSR pathway[185,186,198]. Another transcription factor bmoRhas been reported as a component of OSR, inducinggenes required for the maintenance of the intracel-lular redox state [199,200].

Response to Other Environmental Stimuli

B. fragilis has shown to be responsive to thepresence of bile salts, increasing cell adhesion andcoaggregation in vitro [201,202]. There is noevidence of bile salt tolerance in Bacteroides as apathogenic trait, rather than part of an environmentalstress response. Transcriptional regulators such asthe mar system mediate resistance to antimicrobialsand other environmental stressors [199,203].

Lipopolysaccharide

In gram-negative bacteria, the outer layer of theouter membrane is composed of phospholipids andlipooligosaccharides or lipopolysaccharides (LOSsor LPSs, respectively) [204]. LPS molecules arecomposed of a lipid moiety knows as Lipid A, a coreglycan, and an outer glycan [205,206]. Lipid A, orendotoxin, triggers an inflammatory response,dependent on recognition and signaling by TLR4 ofimmune host cells [206e208]. The endotoxin ofBacteroidesinduces a much lower degree of LPS/TLR4-mediated inflammation than Lipid A fromspecies of Enterobacteria owing to differences inits chemical structure [208e210]. Monoclonal anti-bodies against B. fragilis LPSs have been shown torescue from peritoneal infections and bacteremia innonimmune mice [211].

Mechanisms of InterbacterialCompetition

T6SS in Bacteroides

Bacteria that inhabit the gut are present in largenumbers, and competition for a stable place in suchan environment is commonly established [212,213].Secretion of toxic proteins is a common mechanismof competition [214e216], perhaps best exemplifiedthrough multiple studies that have recently indicatedthe importance of the T6SS in Bacteroides nicheestablishment [46e48] and more broadly in inter-bacterial competition within the intestine [217e220].The T6SS resembles an inverted phage, withsequence and structural analysis indicative oforthology between these systems [221]. The T6SSfunctions by secreting effectors (toxins) directly fromthe cytosol of the bacterial cell into another cell in acontact-dependent manner [222]. Secretion of T6SSeffectors by a bacterial cell is accompanied by

772 Good gone bad

synthesis of specific immunity proteins that conferresistance to attack by sister cells. Each effectorprotein contains a cognate immunity protein, typi-cally encoded by neighboring genes. Although it hasbeen shown that the T6SS can be used by certainbacteria to inject toxins into eukaryotic cells, mostspecies use the T6SS as a mechanism of inter-bacterial killing [223].In the species of the order Bacteroidales, the

genes encoding the T6SS can be present in threedifferent genetic arrangements known as genomicarchitectures (GAs), GA1, GA2, and GA3. WhileGA1 and GA2 are found in many species of the orderBacteroidales, GA3 is restricted to B. fragilis [49]. Anevolutionary explanation to B. fragilis GA3 restrictionis that GA1 and GA2 can be mobilized betweenbacterial cells through genetic elements, but GA3does not contain such features of transmission [49].T6SS genes from Bacteroides sp. are present in upto 75% of the population as per sequence analysisfrom the Human Microbiome Project. This highprevalence of T6SS-bearing Bacteroides in humanssuggests that this locus plays an important role in thecompetition for colonic niche establishment. Mostadult human hosts are colonized by a single strain ofB. fragilis; however, during infancy, strain domi-nance can shift, evidenced by differential relativeabundance of E-I pairs over time [14]. This findinggenerates a compelling argument that the T6SS is acritical feature for initial niche acquisition during earlychildhood. The same study shows an associationbetween T6SS-bearing B. fragilis and an increase inother species of Bacteroides in the same ecosys-tem, suggesting that GA3 effector-immunity pairs areused in vivo for competition mostly betweenB. fragilis strains [14]. This could be due to otherBacteroides species occupying different geographicniches within the colon, limiting the negative effect ofthe T6SS from B. fragilis [47]. In addition, otherBacteroides species could be less sensitive in vivo toB. fragilis T6SS-dependent killing than B. fragilisstrains carrying different GA3 E-I pairs [47]. A veryrecent report has shed light on the mechanistic basisof coexistence between B. fragilis and other speciesfrom the same genus. Many Bacteroides specieshave acquired additional immunity proteins againstGA3 from B. fragilis, gaining the capacity to defendfrom GA3 attacks and potent ial ly exploitB. fragiliseoccupied niches [224]. The role of theT6SS in bacterial competition makes it a potentiallyuseful tool to manipulate carriage of pathogens byprobiotic interventions. We have shown that coin-oculation and competition between wild-type ETBF(strain 43858) and NTBF (strain NCTC 9343) inantibiotic-treated mice reduces the burden andtoxicity of ETBF in a T6SS-dependent manner.When NTBF lacks an active T6SS or the effectorprotein from the operon, ETBF is able to colonizesuccessfully and drive colonic disease [50]. This

property of colonization resistance is stronglydependent on the strain and temporal order ofcolonization of both the NTBF and ETBF strains tobe tested [50].

Antimicrobial Proteins in Bacteroides

In addition to contact-dependent killing, Bacter-oides can secrete soluble toxins. In particular,Bacteroides genomes carry genes that encodetoxins that belong to a group named Bacteroi-dales-secreted antimicrobial proteins (BSAPs)[51e53]. BSAPs are orthologs of membrane attackcomplex/perforin (MACPF) domainecontaining pro-teins and have been shown to be important forstrain competition between Bacteroides species invitro and in vivo [51e53]. Bacteroides species canalso secrete a ubiquitin-like protein that mediatesinterbacterial killing [54]. The wide variety ofcontact-dependent and contact-independentmechanisms and effector molecules for bacterialcompetition reinforces the concept that a singlegenetic determinant will not suffice to outcompetemany diverse organisms from a given niche. Multi-ple genetic factors likely govern niche occupancyby B. fragilis; hence, genomic diversity betweenstrains reduces the possibility of a single strain to beused as a general probiotic to displace ETBF fromthe gut ecosystem. It seems clear at this point thatcuration of gene function is still one of the majorbottlenecks in understanding the role of the15e20% strain-specific portion of B. fragilis gen-omes in colonization.

Host Factors in Colonization andDisease

When B. fragilis initially encounters the colonduring early infancy, it is immersed in an environ-ment where constant and deep changes areoccurring to host and microbial cells [225,226].Maturation of the colonic epithelial lining and mucuslayer are the key to avoid excessive exposure of thehost to microbes and prevent inflammation[227,228]. The main structural component of themucus is the glycoprotein MUC2, secreted bygoblet cells [229]. MUC2 and other mucins form adense mucus layer in direct contact with epithelialcells and a loose mucus layer around 50 mm apartfrom the epithelium [230]. The loose mucus layerallows penetration by many microorganisms,whereas the dense layer restricts access to theepithelium to a small group of bacteria [230,231].Thickness and penetrability of the mucus is highlyinfluenced by microbiota composition, highlightingthe complex relationship between the variablesaffecting gut health [232,233]. Challenge with ETBFin Muc2 mucus-deficient mice leads to high

Fig. 4. Effects of BFT on the immune response.Activation of b-catenin/Wnt, STAT3, and NF-kb pathwaysis required for cell damage and inflammation. Macro-phages within the lamina propria secrete proinflammatorycytokines that promote a Th17 cellular phenotype. The NF-kB pathway requires IL-17 signaling in CECs. CECssecrete CXC chemokines such as IL-8, promoting recruit-ment of myeloid cells to the site of damage. BFT,Bacteroides fragilis toxin; CEC, colonic epithelial cell; IL,interleukin; MAPK, mitogen-activated protein kinase.

773Good gone bad

lethality, indicating the key role of the mucus layer inpreventing ETBF-associated disease [96]. Con-stant and dynamic communication between hostcells and microbiota during intestinal maturationcontributes to achievement of a symbiotic steadystate that can be disrupted by perturbation of eitherthe host or the microbiome constituents. In thissection, we will focus on the known host determi-nants for progression of disease caused by ETBF.

Association of ETBF With Chronic Disease inHumans

The association between colitis and CRC withcarriage of ETBF has led to the hypothesis that thisstrain, because of its tissue-damaging toxin, is acausal agent of CRC instead of a bystander such asStreptococcus gallolyticus [234]. Most cases ofCRC (~90%) are sporadic whereas a minorpercentage are familial in nature. In both sporadicand familial CRC, mutations in the adenomatouspolyposis coli (APC) gene are present in 70e80% ofcases [235,236]. The lack of functional APC is oneof the main drivers of polyp disease and early-onsetCRC [237]. APC forms a protein complex thattargets free cytosolic b-catenin to degradationthrough ubiquitinylation [237]. When E-cadherin iscleaved in a BFT-dependent manner, more solubleb-catenin is available for Wnt/b-catenin signaling,inducing a proliferative response through c-myc(Fig. 3) [105]. Thus, mutations that abrogatefunctional APC, together with the effect of BFT onE-cadherin and the free cytosolic pool of b-catenin,create the conditions for cellular hyperproliferationand crypt hyperplasia. The use of multiple intestinalneoplasia (Min) mouse models carrying mutationsin APC has promoted a clearer causal relationshipbetween ETBF and CRC [18,238,239]. These micespontaneously develop tumors in the small intestineafter 2e3 months in the presence of their endogen-ous microbiota. However, after oral gavage withETBF, colonic disease appears ~4 weeks afterinoculation [239]. BFT-driven damage of coloniccells in APC-mutant mice triggers a hyperprolifera-tive response (Fig. 3) and inflammatory cascade(Fig. 4) in a Stat3- and IL-17edependent manner[16,18,239]. Secretion of IL-8 and other CXCchemokines recruits immature polymorphonuclearcells, leading to exacerbation of inflammation andcell damage (Fig. 4) [16,18]. These experimentsindicate that ETBF can remodel the colonic epithe-lium drastically in a predisposed model organismtoward a state of disease. A very recent studylooked at colon biopsies from human patients withfamilial adenomatous polyposis and found evi-dence of causal association between ETBF anddevelopment of CRC [17]. In these individuals,ETBF was present in biofilms together with pro-

tease-producing (colibactin) pks and E. coli, restrict-ing the presence of other microorganisms (Fig. 5)[17]. Coinoculation of both bacteria in an azoxy-methane-treated mice model of colon cancerincreased severity of the disease, dependent onETBF-driven secretion of IL-17 [17]. Most recently,mucosal bacterial communities from both healthyindividuals and patients with colon cancer wereshown to induce tumors in three Apc mousemodels, implying that such communities would becontributors to disease in predisposed individuals[240]. Major perturbations of gut epithelial home-ostasis seem to be accompanied of more severeETBF-dependent phenotypes.The intrinsic predisposition of each individual to

colonic inflammation and dysbiosis is likely togovern whether ETBF displays a phenotype con-sistent with asymptomatic colonization or severecolonic disease. As shown in Fig. 5, host suscept-ibility to ETBF is determined by the state of themammalian host cells per se and the symbioticmicroorganisms that together form a complex anddynamic ecosystem. During periods of gut home-ostasis, either toxin production could be abolishedor become nondamaging in a host that presents a

Fig. 5. Host and microbial factors that modulate degree of commensalism or pathogenicity of ETBF. ETBFcarriage can result in a risk of colonic disease owing to secretion of BFT and cell damage, but host predisposition is the keyto disease by ETBF. A healthy steady-state microbial community that is not permissive to colonization by ETBF shouldcontain NTBF bacteria that compete for the colonic niche with ETBF. NTBF requires the capsule to colonize and establishescompetition through the T6SS and BSAPs. In addition, colonization by bacterial commensals during early infancy drivessecretion of anti-inflammatory IL-10, contributing to immune tolerance and microbial symbiosis. On the other hand, damagecaused by BFT and pathogenicity can be worsened by stable colonization by ETBF through bactericidal mechanisms suchas T6SS and BSAPs and specific host susceptibility factors including APC mutations, defects in the epithelial mucus layer,and colonization by other deleterious microbes such as protease-positive (pksþ) E. coli. BFT, Bacteroides fragilis toxin;ETBF, enterotoxigenic Bacteroides fragilis; T6SS, type VI secretion system; BSAP, Bacteroidales-secreted antimicrobialprotein; APC, adenomatous polyposis coli; NTBF, non-toxigenic B. fragilis; IL, interleukin.

774 Good gone bad

healthy mucus layer and a symbiotic microbiome.Perturbations of the gut environment could triggerpathogenicity of ETBF, resulting in inflammationand dysbiosis. Changes in bacterial compositionand host cell metabolism could be fueling theecosystem with metabolites that can be hijackedby ETBF, in a similar way to what has been reportedfor S.enterica and V.cholerae [55,241]. A case forthat is polyamine metabolism, affected in micecolonized with ETBF in a BFT-dependent manner.Spermine oxidase catalyzes formation of ReactiveOxygen Species (ROS), and its expression isincreased in HT29 cells exposed to purified BFT,and in C57BL/6 mice, causing DNA damage andinflammation (Fig. 3) [242]. Probiotic interventionscould be a very valuable tool to delay progression ofdisease caused by ETBF in predisposed indivi-duals. It has recently been shown that in Min mice,coinoculation of mice with NTBF (strain NCTC9343) and ETBF (strain 86-5443-2-2, or 86), orinoculation of NTBF after ETBF colonization, doesnot rescue the progression of chronic colitis andCRC [141]. Therefore, previous niche occupancy byETBF highly reduces the success of a secondacquisition event by a probiotic NTBF strain[50,141]. In this study, dominance of ETBF overNTBF is partly explained by the secretion of aMACPF non-T6SS toxin, highlighting how poten-tially complex probiotic interventions would be inclinical settings, where genetic diversity is high, and

most niche acquisition determinants remain fairlyunexplored [141].

ETBF as Part of a Complex EcologicalScenario

B. fragilis is one of the least abundant species ofBacteroides spp. in humans, comprising around0.5% of total gut bacteria, yet is the most commonlyisolated Bacteroides species from samples of infec-tions. As previously mentioned, human metagen-ome data in children show that this period of life ispivotal for initial niche acquisition by B. fragilis and ischaracterized by fierce strain competition throughthe T6SS [14]. From an evolutionary perspective, it istempting to speculate that environmental pressurefor niche establishment could have positivelyselected the presence of additional factors such asBFT that allows ETBF to trespass the colonicepithelial barrier and gain access to other niches. Alink between the site of infection by B. fragilis and itsoriginal niche localization in the gut remains to befound.The generation of efficacious probiotic interven-

tions requires a detailed molecular understanding ofthe repertoire of genetic determinants of nicheacquisition within many strains of B. fragilis. Geno-mic acquisition of bft has occurred through multipleindependent events of horizontal transmission [92].

775Good gone bad

Thus, diverse genetic backgrounds with differentcapacities for niche establishment and strain com-petition can produce BFT and potentially causedisease. A desirable probiotic strain against ETBFshould attempt to outcompete various strains byusing “designed” combinations of genetic determi-nants of colonization; given the complexity of hostcolonization, these may require a personalizedmedicine perspective to effectively treat those atrisk of ETBF-associated disease. To this end,knowledge within the field should progress towardunderstanding genetic determinants of colonizationand how these can be strategically manipulated tosubvert a given strain of ETBF. Moving toward suchan approach will require genomic analysis of multiplehuman isolates and a comprehensive understandingof the determinants of niche competition.One of the major challenges to the field is to

understand the initial niche acquisition interactionsof B. fragilis with the human host and to define howthese interactions modulate the immune systemtoward a state of tolerance that promotes nichestability and symbiosis. While these events mostlikely occur during infancy, further studies will benecessary to understand the kinetics of nicheacquisition in the human population. In mousemodels, the time frame for early immunologicalrecognition and generation of tolerance to a Bacter-oides antigen spans between 10 and 20 days afterbirth [243]. Nonexposure to gut bacterial antigensduring that time frame or prolonged exposurebeyond weaning is detrimental to tolerance by theimmune system, even to gut bacteria that becomedominant later in life [243]. Longitudinal studiesinvolving large cohorts of infants will be necessary todefine ETBF prevalence and to be able to narrowdown the temporal aspect of initial host-pathobiontinteractions [225]. The current lack of such studies isone of the major caveat in the comprehension of howearly childhood acquisition of ETBF impacts diseaseprogression in adults. Studies to date in mousemodels would suggest that ETBF-driven disease is achronic condition in predisposed individuals, result-ing from loss of homeostasis of the colonic cells andmicrobiota over long periods of time [244]. Hence,understanding early events in stable niche occu-pancy by ETBF is probably pivotal for understandingthe risk of progression of disease. We are stillunaware of how BFT expression and secretion isregulated in vivo in humans. It is unknown whetherlow BFT production or toxin resistance by the host isresponsible for such a high proportion of asympto-matic carriers of ETBF. The process of long-termand stable colonization of humans by B. fragilisremains as uncharted territory.The field is now poised with an opportunity to

develop and use methods with higher spatial

resolution than enumeration of colony-formingunits from stool samples to understand the rele-vance of bacterial niche occupancy to human healthand disease. Recent research has clearly shownthat fecal matter does not necessarily reflect theabundance and degree of colonization of specificgut mucosal tissues by bacteria [245,246], high-lighting the need for detailed investigation of themicrobial niche with a focus on anatomic resolution.The use of new tools that enable analysis of in vivoregulation of gene expression and visualization ofBacteroides cells directly from intestinal tissuesections will be crucial to dissect the process ofhost colonization in a temporal, spatial, and geneticmanner [247,248]. A novel observation in humansshows that B. fragilis continues adapting withinhealthy individuals via de novo mutations. Long-itudinal sampling from a human cohort has shownthe appearance of novel strain variants that coexistin the same host, increasing our grasp on micro-biome biodiversity within a single individual [249]. Inthis review, we have focused on genetic determi-nants from the host-pathobiont side of the B. fragilis/ETBF story. There are other important factors forcolonization, such as host diet, antibiotics, andbacteriophages, that are crucial in the developmentand stability of the lower Gastrointestinal (GI)ecosystem. A challenge for the field will thereforebe the use of emerging complex data analysisplatforms to integrate host and microbial factors in atemporal manner and to better comprehend howpathobionts such as ETBF can acquire a niche inthe gut, can become a stable member of themicrobiome, and can cause disease progressionin predisposed hosts.

Acknowledgments

This work was supported by the Department ofPediatrics at Washington University and by the grantR01 AI138565-A1 (to J.B.W.).

Received 18 April 2019;Received in revised form 27 November 2019;

Accepted 5 December 2019Available online 17 December 2019

Keywords:Bacteroides fragilis;

BFT;Fragilysin;

Metalloprotease;Gut microbiota

776 Good gone bad

References

[1] T. Yatsunenko, F.E. Rey, M.J. Manary, I. Trehan,M.G. Dominguez-Bello, M. Contreras, M. Magris,G. Hidalgo, R.N. Baldassano, A.P. Anokhin, A.C. Heath,B. Warner, J. Reeder, J. Kuczynski, J.G. Caporaso,C.A. Lozupone, C. Lauber, J.C. Clemente, D. Knights,R. Knight, J.I. Gordon, Human gut microbiome viewedacross age and geography, Nature 486 (2012) 222, https://doi.org/10.1038/nature11053.

[2] C.A. Lozupone, J.I. Stombaugh, J.I. Gordon, J.K. Jansson,R. Knight, Diversity, stability and resilience of the human gutmicrobiota, Nature 489 (2012) 220e230, https://doi.org/10.1038/nature11550.

[3] J.A. Gilbert, M.J. Blaser, J.G. Caporaso, J.K. Jansson, S. VLynch, R. Knight, Current understanding of the humanmicrobiome, Nat. Med. 24 (2018) 392e400, https://doi.org/10.1038/nm.4517.

[4] W.E. Moore, L. V Holdeman, Human fecal flora: the normalflora of 20 Japanese-Hawaiians, Appl. Microbiol. 27 (1974)961e979.

[5] D.A. Leigh, Clinical importance of infections due toBacteroides fragilis and role of antibiotic therapy, Br. Med.J. 3 (1974) 225e228.

[6] R.W. Cooke, Bacteroides fragilis septicaemia and meningi-tis in early infancy, Arch. Dis. Child. 50 (1975) 241e243.

[7] D.A. Leigh, Wound infections due to Bacteroides fragilisfollowing intestinal surgery, Br. J. Surg. 62 (1975) 375e378.

[8] H. Olsen, Bacteroides bacteraemia. A clinical and bacter-iological analysis of 51 patients, Scand. J. Infect. Dis. 8(1976) 107e111.

[9] B. Olsson, C.E. Nord, T. Wadstrom, Formation of beta-lactamase in Bacteroides fragilis: cell-bound and extracel-lular activity, Antimicrob. Agents Chemother. 9 (1976)727e735.

[10] D.J. Blazevic, Antibiotic susceptibility of the subspecies ofBacteroides fragilis, Antimicrob. Agents Chemother. 9(1976) 481e484.

[11] J.S. Salaki, R. Black, F.P. Tally, J.W. Kislak, Bacteroidesfragilis resistant to the administration of clindamycin, Am. J.Med. 60 (1976) 426e428.

[12] T. Vatanen, D.R. Plichta, J. Somani, P.C. Munch,T.D. Arthur, A.B. Hall, S. Rudolf, E.J. Oakeley, X. Ke,R.A. Young, H.J. Haiser, R. Kolde, M. Yassour,K. Luopajarvi, H. Siljander, S.M. Virtanen, J. Ilonen,R. Uibo, V. Tillmann, S. Mokurov, N. Dorshakova,J.A. Porter, A.C. McHardy, H. Lahdesmaki, H. Vlamakis,C. Huttenhower, M. Knip, R.J. Xavier, Genomic variationand strain-specific functional adaptation in the human gutmicrobiome during early life, Nat. Microbiol. 4 (2019)470e479, https://doi.org/10.1038/s41564-018-0321-5.

[13] J. Lloyd-Price, A. Mahurkar, G. Rahnavard, J. Crabtree,J. Orvis, A.B. Hall, A. Brady, H.H. Creasy, C. McCracken,M.G. Giglio, D. McDonald, E.A. Franzosa, R. Knight,O. White, C. Huttenhower, Strains, functions and dynamicsin the expanded human microbiome project, Nature 550(2017) 61, https://doi.org/10.1038/nature23889.

[14] A.J. Verster, B.D. Ross, M.C. Radey, Y. Bao,A.L. Goodman, J.D. Mougous, E. Borenstein, The land-scape of type VI secretion across human gut microbiomesreveals its role in community composition, Cell HostMicrobe 22 (2017) 411e419, https://doi.org/10.1016/j.chom.2017.08.010, e4.

[15] C.L. Sears, S. Islam, A. Saha, M. Arjumand, N.H. Alam,A.S.G. Faruque, M.A. Salam, J. Shin, D. Hecht,A. Weintraub, R.B. Sack, F. Qadri, Association of enter-otoxigenic Bacteroides fragilis infection with inflammatorydiarrhea, Clin. Infect. Dis. 47 (2008) 797e803, https://doi.org/10.1086/591130.

[16] E. Thiele Orberg, H. Fan, A.J. Tam, C.M. Dejea,C.E. Destefano Shields, S. Wu, L. Chung, B.B. Finard,X. Wu, P. Fathi, S. Ganguly, J. Fu, D.M. Pardoll,C.L. Sears, F. Housseau, The myeloid immune signatureof enterotoxigenic Bacteroides fragilis-induced murinecolon tumorigenesis, Mucosal Immunol. 10 (2017)421e433, https://doi.org/10.1038/mi.2016.53.

[17] C.M. Dejea, P. Fathi, J.M. Craig, A. Boleij, R. Taddese,A .L . Ge is , X . Wu, C.E . DeSte fano Sh ie lds ,E.M. Hechenbleikner, D.L. Huso, R.A. Anders,F.M. Giardiello, E.C. Wick, H. Wang, S. Wu, D.M. Pardoll,F. Housseau, C.L. Sears, Patients with familial adenoma-tous polyposis harbor colonic biofilms containing tumori-genic bacteria, Science 359 (2018) 592e597, https://doi.org/10.1126/science.aah3648.

[18] L. Chung, E. Thiele Orberg, A.L. Geis, J.L. Chan, K. Fu,C.E. DeStefano Shields, C.M. Dejea, P. Fathi, J. Chen,B.B. Finard, A.J. Tam, F. McAllister, H. Fan, X. Wu,S. Ganguly, A. Lebid, P. Metz, S.W. Van Meerbeke,D.L. Huso, E.C. Wick, D.M. Pardoll, F. Wan, S. Wu,C.L. Sears, F. Housseau, Bacteroides fragilis toxin co-ordinates a pro-carcinogenic inflammatory cascade viatargeting of colonic epithelial cells, Cell Host Microbe 23(2018) 203e214, https:/ /doi .org/10.1016/j .chom.2018.01.007, e5.

[19] L.L. Myers, B.D. Firehammer, D.S. Shoop, M.M. Border,Bacteroides fragilis: a possible cause of acute diarrhealdisease innewborn lambs, Infect. Immun.44 (1984)241e244.

[20] L.L. Myers, D.S. Shoop, B.D. Firehammer, M.M. Border,Association of enterotoxigenic Bacteroides fragilis withdiarrheal disease in calves, J. Infect. Dis. 152 (1985)1344e1347.

[21] L.L. Myers, D.S. Shoop, Association of enterotoxigenicBacteroides fragilis with diarrheal disease in young pigs,Am. J. Vet. Res. 48 (1987) 774e775, https://doi.org/10.1016/S0167-2991(08)64389-X.

[22] L.L. Myers, D.S. Shoop, L.L. Stackhouse, F.S. Newman,R.J. Flaherty, G.W. Letson, R.B. Sack, Isolation ofenterotoxigenic Bacteroides fragilis from humans withdiarrhea, J. Clin. Microbiol. 25 (1987) 2330e2333, https://doi.org/10.1177/0146167293195003.

[23] L.L. Myers, D.S. Shoop, T.D. Byars, Diarrhea associatedwith enterotoxigenic Bacteroides fragilis in foals, Am. J. Vet.Res. 48 (1987) 1565e1567.

[24] D.S. Shoop, L.L. Myers, J.B. LeFever, Enumeration ofenterotoxigenic Bacteroides fragilis in municipal sewage,Appl. Environ. Microbiol. 56 (1990) 2243e2244.

[25] R.B. Sack, L.L. Myers, J. Almeido-Hill, D.S. Shoop,W.C. Bradbury, R. Reid, M. Santosham, EnterotoxigenicBacteroides fragilis: epidemiologic studies of its role as ahuman diarrhoeal pathogen, J. Diarrhoeal Dis. Res. 10(1992) 4e9, https://doi.org/10.1126/science.1184944.

[26] R.B. Sack, M.J. Albert, K. Alam, P.K. Neogi, M.S. Akbar,Isolation of enterotoxigenic Bacteroides fragilis from Ban-gladeshi children with diarrhea: a controlled study, J. Clin.Microbiol. 32 (1994) 960e963.

[27] G. Zhang, B. Svenungsson, A. Karnell, A. Weintraub,Prevalence of enterotoxigenic Bacteroides fragilis in adult

777Good gone bad

patients with diarrhea and healthy controls, Clin. Infect. Dis.29 (1999) 590e594.

[28] V.H. San Joaquin, J.C. Griffis, C. Lee, C.L. Sears,Association of bacteroides fragilis with childhood diarrhea,Scand. J. Infect. Dis. 27 (1995) 211e215, https://doi.org/10.3109/00365549509019011.

[29] S.K. Niyogi, P. Dutta, U. Mitra, D.K. Pal, Association ofenterotoxigenic Bacteroides fragilis with childhood diar-rhoea, Indian J. Med. Res. 105 (1997) 167e169.

[30] P. Pathela, K.Z. Hasan, E. Roy, K. Alam, F. Huq,A.K. Siddique, R.B. Sack, Enterotoxigenic Bacteroidesfragilis-associated diarrhea in children 0-2 years of age inrural Bangladesh, J. Infect. Dis. 191 (2005) 1245e1252,https://doi.org/10.1086/428947.

[31] T. Vu Nguyen, P. Le Van, C. Le Huy, A. Weintraub,Diarrhea caused by enterotoxigenic Bacteroides fragilis inchildren less than 5 years of age in Hanoi, Vietnam.,Anaerobe. 11 (2005) 109e114, https://doi.org/10.1016/j.anaerobe.2004.10.004.

[32] B. Durmaz, M. Dalgalar, R. Durmaz, Prevalence ofenterotoxigenic Bacteroides fragilis in patients with diar-rhea: a controlled study, Anaerobe 11 (2005) 318e321,https://doi.org/10.1016/j.anaerobe.2005.06.001.

[33] T.P. Prindiville, R.A. Sheikh, S.H. Cohen, Y.J. Tang,M.C. Cantrell, J.J. Silva, Bacteroides fragilis enterotoxingene sequences in patients with inflammatory boweldisease, Emerg. Infect. Dis. 6 (2000) 171e174, https://doi.org/10.3201/eid0602.000210.

[34] N.U. Toprak, A. Yagci, B.M. Gulluoglu, M.L. Akin,P. Demirkalem, T. Celenk, G. Soyletir, A possible role ofBacteroides fragilis enterotoxin in the aetiology of color-ectal cancer, Clin. Microbiol. Infect. 12 (2006) 782e786,https://doi.org/10.1111/j.1469-0691.2006.01494.x.

[35] A. Boleij, E.M. Hechenbleikner, A.C. Goodwin, R. Badani,E.M. Stein, M.G. Lazarev, B. Ellis, K.C. Carroll, E. Albesiano,E.C. Wick, E.A. Platz, D.M. Pardoll, C.L. Sears, TheBacteroides fragilis toxin gene is prevalent in the colonmucosa of colorectal cancer patients, Clin. Infect. Dis. 60(2015) 208e215, https://doi.org/10.1093/cid/ciu787.

[36] K.S. Vil joen, A. Dakshinamurthy, P. Goldberg,J.M. Blackburn, Quantitative profiling of colorectal can-cer-associated bacteria reveals associations betweenfusobacterium spp., enterotoxigenic Bacteroides fragilis(ETBF) and clinicopathological features of colorectalcancer, PLoS One 10 (2015), e0119462, https://doi.org/10.1371/journal.pone.0119462.

[37] Y. Zhou, H. He, H. Xu, Y. Li, Z. Li, Y. Du, J. He, Y. Zhou,H. Wang, Y. Nie, Association of oncogenic bacteria withcolorectal cancer in South China, Oncotarget 7 (2016)80794e80802, https://doi.org/10.18632/oncotarget.13094.

[38] J.I. Keenan, A. Aitchison, R. V Purcell, R. Greenlees,J.F. Pearson, F.A. Frizelle, Screening for enterotoxigenicBacteroides fragilis in stool samples, Anaerobe 40 (2016)50e53, https://doi.org/10.1016/j.anaerobe.2016.05.004.

[39] S. Zamani, S. Hesam Shariati, M.R. Zali, H. AsadzadehAghdaei, A. Sarabi Asiabar, S. Bokaie, B. Nomanpour,L.A. Sechi, M.M. Feizabadi, Detection of enterotoxigenicBacteroides fragilis in patients with ulcerative colitis, GutPathog. 9 (2017) 53, https://doi.org/10.1186/s13099-017-0202-0.

[40] R. V Purcell, J. Pearson, A. Aitchison, L. Dixon,F.A. Frizelle, J.I. Keenan, Colonization with enterotoxigenicBacteroides fragilis is associated with early-stage color-

ectal neoplasia, PLoS One 12 (2017), e0171602, https://doi.org/10.1371/journal.pone.0171602.

[41] F. Haghi, E. Goli, B. Mirzaei, H. Zeighami, The associationbetween fecal enterotoxigenic B. fragilis with colorectalcancer, BMC Canc. 19 (2019) 879, https://doi.org/10.1186/s12885-019-6115-1.

[42] A. Pantosti, M.G. Menozzi, A. Frate, L. Sanfilippo,F. D'Ambrosio, M. Malpeli, Detection of enterotoxigenicBacteroides fragilis and its toxin in stool samples fromadults and children in Italy, Clin. Infect. Dis. 24 (1997)12e16, https://doi.org/10.1093/clinids/24.1.12.

[43] J.L. Round, S.K. Mazmanian, Inducible Foxp3þ regulatoryT-cell development by a commensal bacterium of theintestinal microbiota, Proc. Natl. Acad. Sci. U. S. A 107(2010) 12204e12209, h t tps : / /do i .o rg /10 .1073/pnas.0909122107.

[44] J.L. Round, S.M. Lee, J. Li, G. Tran, B. Jabri, T.A. Chatila,S.K. Mazmanian, The Toll-like receptor 2 pathway estab-lishes colonization by a commensal of the human micro-biota, Science 332 (2011) 974e977, https://doi.org/10.1126/science.1206095.

[45] B.W. Casterline, A.L. Hecht, V.M. Choi, J. Bubeck Warden-burg, The Bacteroides fragilis pathogenicity island linksvirulence and strain competition, Gut Microb. 8 (2017)374e383, https://doi.org/10.1080/19490976.2017.1290758.

[46] A.B. Russell, A.G. Wexler, B.N. Harding, J.C. Whitney,A.J. Bohn, Y.A. Goo, B.Q. Tran, N.A. Barry, H. Zheng,S.B. Peterson, S. Chou, T. Gonen, D.R. Goodlett,A.L. Goodman, J.D. Mougous, A type VI secretion-relatedpathway in Bacteroidetes mediates interbacterial antagon-ism, Cell Host Microbe 16 (2014) 227e236, https://doi.org/10.1016/j.chom.2014.07.007.

[47] A.G. Wexler, Y. Bao, J.C. Whitney, L.-M. Bobay,J.B. Xavier, W.B. Schofield, N.A. Barry, A.B. Russell,B.Q. Tran, Y.A. Goo, D.R. Goodlett, H. Ochman,J.D. Mougous, A.L. Goodman, Human symbionts injectand neutralize antibacterial toxins to persist in the gut,Proc. Natl. Acad. Sci. U. S. A 113 (2016) 3639e3644,https://doi.org/10.1073/pnas.1525637113.

[48] M. Chatzidaki-Livanis, N. Geva-Zatorsky, L.E. Comstock,Bacteroides fragilis type VI secretion systems use noveleffector and immunity proteins to antagonize human gutBacteroidales species, Proc. Natl. Acad. Sci. U. S. A 113( 2016 ) 3 627e3632 , h t t p s : / / d o i . o r g / 1 0 . 1 073 /pnas.1522510113.

[49] M.J. Coyne, K.G. Roelofs, L.E. Comstock, Type VIsecretion systems of human gut Bacteroidales segregateinto three genetic architectures, two of which are containedon mobile genetic elements, BMC Genomics 17 (2016) 58,https://doi.org/10.1186/s12864-016-2377-z.

[50] A.L. Hecht, B.W. Casterline, Z.M. Earley, Y.A. Goo,D.R. Goodlett, J. Bubeck Wardenburg, Strain competitionrestricts colonization of an enteric pathogen and preventscolitis, EMBO Rep. 17 (2016) 1281e1291, https://doi.org/10.15252/embr.201642282.

[51] M. Chatzidaki-Livanis, M.J. Coyne, L.E. Comstock, Anantimicrobial protein of the gut symbiont Bacteroides fragiliswith a MACPF domain of host immune proteins, Mol.Microbiol. 94 (2014) 1361e1374, https://doi.org/10.1111/mmi.12839.

[52] K.G. Roelofs, M.J. Coyne, R.R. Gentyala, M. Chatzidaki-Livanis, L.E. Comstock, Bacteroidales secreted antimicro-bial proteins target surface molecules necessary for gut

778 Good gone bad

colonization and mediate competition in vivo, mBio 7(2016), https://doi.org/10.1128/mBio.01055-16.

[53] A.M. Shumaker, V.L. McEneany, M.J. Coyne, P.A. Silver,L.E. Comstock, Identification of a fifth antibacterial toxinproduced by a single Bacteroides fragilis strain, J. Bacteriol.(2019), https://doi.org/10.1128/JB.00577-18.

[54] M. Chatzidaki-Livanis, M.J. Coyne, K.G. Roelofs,R.R. Gentyala, J.M. Caldwell, L.E. Comstock, Gut sym-biont Bacteroides fragilis secretes a eukaryotic-like ubiqui-tin protein that mediates intraspecies antagonism, mBio 8(2017), https://doi.org/10.1128/mBio.01902-17.

[55] F. Rivera-Chavez, J.J. Mekalanos, Cholera toxin promotespathogen acquisition of host-derived nutrients, Nature 572(2019) 244e248, https://doi.org/10.1038/s41586-019-1453-3.

[56] M. Diard, W.-D. Hardt, Basic processes in salmonella-hostinteractions: within-host evolution and the transmission ofthe virulent genotype, Microbiol. Spectr. 5 (2017), https://doi.org/10.1128/microbiolspec.MTBP-0012-2016.

[57] J.G. Conner, J.K. Teschler, C.J. Jones, F.H. Yildiz, Stayingalive: Vibrio cholerae's cycle of environmental survival,transmission, and dissemination, Microbiol. Spectr. 4(2016), https://doi.org/10.1128/microbiolspec.VMBF-0015-2015.

[58] J. Papaparaskevas, A. Katsandri, A. Pantazatou, I. Stefanou,A. Avlamis, N.J. Legakis, A. Tsakris, Epidemiological char-acteristics of infections causedbyBacteroides,Prevotella andFusobacterium species: a prospective observational study,Anaerobe 17 (2011) 113e117, https://doi.org/10.1016/J.ANAEROBE.2011.05.013.

[59] A. Pantosti, M. Cerquetti, R. Colangeli, F. D'Ambrosio,Detection of intestinal and extra-intestinal strains ofenterotoxigenic Bacteroides fragilis by the HT-29 cytotoxi-city assay, J. Med. Microbiol. 41 (1994) 191e196, https://doi.org/10.1099/00222615-41-3-191.

[60] N. Kato, A. Karuniawati, R. Jotwani, H. Kato, K. Watanabe,K. Ueno, Isolation of enterotoxigenic Bacteroides fragilisfrom extraintestinal sites by cell culture assay, Clin. Infect.Dis. 20 (Suppl 2) (1995) S141.

[61] L.M. Mundy, C.L. Sears, Detection of toxin production byBacteroides fragilis: assay development and screening ofextraintestinal clinical isolates, Clin. Infect. Dis. 23 (1996)269e276.

[62] G.T. Chung, A.A. Franco, S. Wu, G.E. Rhie, R. Cheng,H.B. Oh, C.L. Sears, Identification of a third metallopro-tease toxin gene in extraintestinal isolates of Bacteroidesfragilis, Infect. Immun. 67 (1999) 4945e4949.

[63] M.C. Claros, Z.C. Claros, D.W. Hecht, D.M. Citron,E.J.C. Goldstein, J.J. Silva, Y. Tang-Feldman,A.C. Rodloff, Characterization of the Bacteroides fragilispathogenicity island in human blood culture isolates,Anaerobe 12 (2006) 17e22, https://doi.org/10.1016/j.anae-robe.2005.06.005.

[64] I. Foulon, D. Pierard, G. Muyldermans, K. Vandoorslaer,O. Soetens, P. Rosseel, S. Lauwers, Prevalence offragilysin gene in Bacteroides fragilis isolates from bloodand other extraintestinal samples, J. Clin. Microbiol. 41(2003) 4428e4430.

[65] N. Polanco, L. Manzi, O. Carmona, [Possible role ofenterotoxigenic Bacteroides fragilis in the etiology ofinfectious vaginitis], Investig. Clin. 53 (2012) 28e37.

[66] R.J.J. Obiso, A.O. Azghani, T.D. Wilkins, The Bacteroidesfragilis toxin fragilysin disrupts the paracellular barrier ofepithelial cells, Infect. Immun. 65 (1997) 1431e1439.

[67] V.M. Choi, J. Herrou, A.L. Hecht, W.P. Teoh, J.R. Turner,S. Crosson, J. Bubeck Wardenburg, Activation of Bacter-oides fragilis toxin by a novel bacterial protease contributesto anaerobic sepsis in mice, Nat. Med. 22 (2016) 563e567,https://doi.org/10.1038/nm.4077.

[68] A.B. Onderdonk, D.L. Kasper, R.L. Cisneros, J.G. Bartlett,The capsular polysaccharide of Bacteroides fragilis as avirulence factor: comparison of the pathogenic potential ofencapsulated and unencapsulated strains, J. Infect. Dis.136 (1977) 82e89, https://doi.org/10.1093/infdis/136.1.82.

[69] C.S. Weikel, F.D. Grieco, J. Reuben, L.L. Myers, R.B. Sack,Human colonic epithelial cells, HT29/C1, treated with crudeBacteroides fragilis enterotoxin dramatically alter theirmorphology, Infect. Immun. 60 (1991) 321e327.

[70] R.L. Van Tassell, D.M. Lyerly, T.D. Wilkins, Purification andcharacterization of an enterotoxin from Bacteroides fragilis,Infect. Immun. 60 (1992) 1343e1350.

[71] J.S. Moncrief, R. Obiso, L.A. Barroso, J.J. Kling,R.L. Wright, R.L. Van Tassell, D.M. Lyerly, T.D. Wilkins,The enterotoxin of Bacteroides fragilis is a metalloprotease,Infect. Immun. 63 (1995) 175e181.

[72] R.L. Van Tassell, D.M. Lyerly, T.D. Wilkins, Characteriza-tion of enterotoxigenic Bacteroides fragilis by a toxin-specific enzyme-linked immunosorbent assay, Clin. Diagn.Lab. Immunol. 1 (1994) 578e584.

[73] S. Mootien, P.M. Kaplan, Monoclonal antibodies specific forBacteroides fragilis enterotoxins BFT1 and BFT2 and theiruse in immunoassays, PLoS One 12 (2017), e0173128,https://doi.org/10.1371/journal.pone.0173128.

[74] J.J. Kling, R.L. Wright, J.S. Moncrief, T.D. Wilkins, Cloningand characterization of the gene for the metalloproteaseenterotoxin of Bacteroides fragilis, FEMS Microbiol. Lett.146 (1997) 279e284, https://doi.org/10.1111/j.1574-6968.1997.tb10205.x.

[75] A.A. Franco, L.M. Mundy, M. Trucksis, S. Wu, J.B. Kaper,C.L. Sears, Cloning and characterization of the Bacteroidesfragilis metalloprotease toxin gene, Infect. Immun. 65(1997) 1007e1013.

[76] A.A. Franco, S.L. Buckwold, J.W. Shin, M. Ascon, C.L. Sears,Mutation of the zinc-binding metalloprotease motif affectsBacteroides fragilis toxinactivity but doesnot affect propeptideprocessing, Infect. Immun. 73 (2005) 5273e5277, https://doi.org/10.1128/IAI.73.8.5273-5277.2005.

[77] C.L. Sears, S.L. Buckwold, J.W. Shin, A.A. Franco, The C-terminal region of Bacteroides fragilis toxin is essential to itsbiological activity, Infect. Immun. 74 (2006) 5595e5601,https://doi.org/10.1128/IAI.00135-06.

[78] T. Goulas, J.L. Arolas, F.X. Gomis-Rüth, Structure, functionand latency regulation of a bacterial enterotoxin potentiallyderived from a mammalian adamalysin/ADAM xenolog,Proc. Natl. Acad. Sci. 108 (2011) 1856, https://doi.org/10.1073/pnas.1012173108. LP e 1861.

[79] J. Herrou, V.M. Choi, J. Bubeck Wardenburg, S. Crosson,Activation mechanism of the Bacteroides fragilis cysteinepeptidase, fragipain, Biochemistry 55 (2016) 4077e4084,https://doi.org/10.1021/acs.biochem.6b00546.

[80] J.S. Moncrief, A.J. Duncan, R.L. Wright, L.A. Barroso,T.D. Wilkins, Molecular characterization of the fragilysinpathogenicity islet of enterotoxigenic Bacteroides fragilis,Infect. Immun. 66 (1998) 1735e1739.

[81] A.A. Franco, R.K. Cheng, G.T. Chung, S. Wu, H.B. Oh,C.L. Sears, Molecular evolution of the pathogenicity islandof enterotoxigenic Bacteroides fragilis strains, J. Bacteriol.181 (1999) 6623e6633, 0021-9193/99/$04.00?0.

779Good gone bad

[82] N. Kato, C.X. Liu, H. Kato, K. Watanabe, Y. Tanaka,T. Yamamoto, K. Suzuki, K. Ueno, A new subtype of themetalloprotease toxin gene and the incidence of the threebft subtypes among Bacteroides fragilis isolates in Japan,FEMS Microbiol. Lett. 182 (2000) 171e176, https://doi.org/10.1111/j.1574-6968.2000.tb08892.x.

[83] S. Wu, L.A. Dreyfus, A.O. Tzianabos, C. Hayashi,C.L. Sears, Diversity of the metalloprotease toxin producedby enterotoxigenic Bacteroides fragilis, Infect. Immun. 70(2002) 2463e2471.

[84] A.S. Scotto d'Abusco, M. Del Grosso, S. Censini,A. Covacci, A. Pantosti, The alleles of the bft gene aredistributed differently among enterotoxigenic Bacteroidesfragilis strains from human sources and can be present indouble copies, J. Clin. Microbiol. 38 (2000) 607e612.

[85] C.L. Sears, The toxins of Bacteroides fragilis, Toxicon 39(2001) 1737e1746.

[86] V. Nakano, D.A. Gomes, R.M.E. Arantes, J.R. Nicoli,M.J. Avila-Campos, Evaluation of the pathogenicity of theBacteroides fragilis toxin gene subtypes in gnotobioticmice, Curr. Microbiol. 53 (2006) 113e117, https://doi.org/10.1007/s00284-005-0321-6.

[87] N. Ulger Toprak, D. Rajendram, A. Yagci, S. Gharbia,H.N. Shah, B.M. Gulluoglu, L.M. Akin, P. Demirkalem,T. Celenk, G. Soyletir, The distribution of the bft alleles amongenterotoxigenic Bacteroides fragilis strains from stool speci-mens and extraintestinal sites, Anaerobe 12 (2006) 71e74,https://doi.org/10.1016/j.anaerobe.2005.11.001.

[88] K.P. Sarvari, J. Soki, M. Ivan, C. Miszti, K. Latkoczy,S.Z. Melegh, E. Urban, Detection of enterotoxin andprotease genes among Hungarian clinical Bacteroidesfragilis isolates, Anaerobe 48 (2017) 98e102, https://doi.org/10.1016/j.anaerobe.2017.07.005.

[89] A.A. Franco, The Bacteroides fragilis pathogenicity island iscontained in a putative novel conjugative transposon,J. Bacteriol. 186 (2004) 6077e6092, https://doi.org/10.1128/JB.186.18.6077-6092.2004.

[90] S.L. Buckwold, N.B. Shoemaker, C.L. Sears, A.A. Franco,Identification and characterization of conjugative transpo-sons CTn86 and CTn9343 in Bacteroides fragilis strains,Appl. Environ. Microbiol. 73 (2007) 53e63, https://doi.org/10.1128/AEM.01669-06.

[91] J. Soki, Z. Eitel, G. Terhes, E. Nagy, E. Urban, Occurrenceand analysis of rare cfiA-bft doubly positive Bacteroidesfragilis strains, Anaerobe 23 (2013) 70e73, https://doi.org/10.1016/j.anaerobe.2013.06.008.

[92] J. V Pierce, H.D. Bernstein, Genomic diversity of enterotoxi-genic strains of Bacteroides fragilis, PLoS One 11 (2016),e0158171, https://doi.org/10.1371/journal.pone.0158171.

[93] P. Obuch-Woszczatynski, R.G.F. Wintermans, A. VanBelkum, H. Endtz, H. Pituch, D. Kreft, F. Meisel-Mikolajc-zyk, M. Luczak, Enterotoxigenic Bacteroides fragilis (ETBF)strains isolated in The Netherlands and Poland aregenetically diverse, Pol. J. Microbiol. 53 (2004) 35e39.

[94] A.A. Franco, R.K. Cheng, A. Goodman, C.L. Sears,Modulation of bft expression by the Bacteroides fragilispathogenicity island and its flanking region, Mol. Microbiol.45 (2002) 1067e1077.

[95] E.A. Groisman, Feedback control of two-component reg-ulatory systems, Annu. Rev. Microbiol. 70 (2016) 103e124,https://doi.org/10.1146/annurev-micro-102215-095331.

[96] A.L. Hecht, B.W. Casterline, V.M. Choi, J. Bubeck Warden-burg, A two-component system regulates Bacteroidesfragilis toxin to maintain intestinal homeostasis and prevent

lethal disease, Cell Host Microbe 22 (2017) 443e448,https://doi.org/10.1016/j.chom.2017.08.007, e5.

[97] S.S. Koshy, M.H. Montrose, C.L. Sears, Human intestinalepithelial cells swell and demonstrate actin rearrangementin response to the metalloprotease toxin of Bacteroidesfragilis, Infect. Immun. 64 (1996) 5022e5028.

[98] R.F. Saidi, K. Jaeger, M.H. Montrose, S. Wu, C.L. Sears,Bacteroides fragilis toxin rearranges the actin cytoskeletonof HT29/C1 cells without direct proteolysis of actin ordecrease in F-actin content, Cell Motil. Cytoskelet. 37(1997) 159e165, https://doi.org/10.1002/(SICI)1097-0169,1997)37:2<159::AID-CM8>3.0.CO;2-3.

[99] S. Wu, K.C. Lim, J. Huang, R.F. Saidi, C.L. Sears,Bacteroides fragilis enterotoxin cleaves the zonula adhe-rens protein, E-cadherin. Proc. Natl. Acad. Sci. U. S. A 95(1998) 14979e14984.

[100] S. Wu, J. Shin, G. Zhang, M. Cohen, A. Franco, C.L. Sears,The Bacteroides fragilis toxin binds to a specific intestinalepithelial cell receptor, Infect. Immun. 74 (2006)5382e5390, https://doi.org/10.1128/IAI.00060-06.

[101] D. Kharlampieva, V. Manuvera, O. Podgorny, E. Grafskaia,S. Kovalchuk, O. Pobeguts, I. Altukhov, V. Govorun,V. Lazarev, Recombinant fragilysin isoforms cause E-cadherin cleavage of intact cells and do not cleave isolatedE-cadherin, Microb. Pathog. 83e84 (2015) 47e56, https://doi.org/10.1016/j.micpath.2015.05.003.

[102] M. Riegler, M. Lotz, C. Sears, C. Pothoulakis, I. Castagliuolo,C.C.Wang,R.Sedivy, T.Sogukoglu,E.Cosentini,G.Bischof,W. Feil, B. Teleky, G. Hamilton, J.T. LaMont, E. Wenzl,Bacteroides fragilis toxin 2 damages human colonic mucosain vitro, Gut 44 (1999) 504, https://doi.org/10.1136/gut.44.4.504. LP e 510.

[103] S. Wu, K.-J. Rhee, M. Zhang, A. Franco, C.L. Sears,Bacteroides fragilis toxin stimulates intestinal epithelial cellshedding and gamma-secretase-dependent E-cadherincleavage, J. Cell Sci. 120 (2007) 1944e1952, https://doi.org/10.1242/jcs.03455.

[104] S.H.M. Wong, C.M. Fang, L.-H. Chuah, C.O. Leong,S.C. Ngai, E-cadherin: its dysregulation in carcinogenesisand clinical implications, Crit. Rev. Oncol. Hematol. 121(2018) 11e22, h t tps : / /do i .org /10 .1016/ j .c r i t re -vonc.2017.11.010.

[105] S. Wu, P.J. Morin, D. Maouyo, C.L. Sears, Bacteroidesfragilis enterotoxin induces c-Myc expression and cellularproliferation, Gastroenterology 124 (2003) 392e400,https://doi.org/10.1053/gast.2003.50047.

[106] L. Sanfilippo, C.K. Li, R. Seth, T.J. Balwin, M.G. Menozzi,Y.R. Mahida, Bacteroides fragilis enterotoxin induces theexpression of IL-8 and transforming growth factor-beta(TGF-beta) by human colonic epithelial cells, Clin. Exp.Immunol. 119 (2000) 456e463.

[107] J.M. Kim, Y.K. Oh, Y.J. Kim, H.B. Oh, Y.J. Cho, Polarizedsecretion of CXC chemokines by human intestinal epithelialcells in response to Bacteroides fragilis enterotoxin: NF-kappa B plays a major role in the regulation of IL-8expression, Clin. Exp. Immunol. 123 (2001) 421e427.

[108] J.M. Kim, S.J. Cho, Y.-K. Oh, H.-Y. Jung, Y.-J. Kim, N. Kim,Nuclear factor-kappa B activation pathway in intestinalepithelial cells is a major regulator of chemokine geneexpression and neutrophil migration induced by Bacter-oides fragilis enterotoxin, Clin. Exp. Immunol. 130 (2002)59e66.

[109] S. Wu, J. Powell, N. Mathioudakis, S. Kane, E. Fernandez,C.L. Sears, Bacteroides fragilis enterotoxin induces intest-

780 Good gone bad

inal epithelial cell secretion of interleukin-8 through mito-gen-activated protein kinases and a tyrosine kinase-regulated nuclear factor-kappaB pathway, Infect. Immun.72 (2004) 5832e5839, ht tps: / /doi .org/10.1128/IAI.72.10.5832-5839.2004.

[110] J.M. Kim, H.Y. Jung, J.Y. Lee, J. Youn, C.-H. Lee, K.-H. Kim, Mitogen-activated protein kinase and activatorprotein-1 dependent signals are essential for Bacteroidesfragilis enterotoxin-induced enteritis, Eur. J. Immunol. 35(2005) 2648e2657, https://doi.org/10.1002/eji.200526321.

[111] S. Hwang, S.-Y. Gwon, M.S. Kim, S. Lee, K.-J. Rhee,Bacteroides fragilis toxin induces IL-8 secretion in HT29/C1cells through disruption of E-cadherin junctions, ImmuneNetw 13 (2013) 213e217, https://doi.org/10.4110/in.2013.13.5.213.

[112] J.M. Kim, J.Y. Lee, Y.M. Yoon, Y.-K. Oh, J.S. Kang, Y.-J. Kim, K.-H. Kim, Bacteroides fragilis enterotoxin inducescyclooxygenase-2 and fluid secretion in intestinal epithelialcells through NF-kappaB activation, Eur. J. Immunol. 36(2006) 2446e2456, https://doi.org/10.1002/eji.200535808.

[113] J.M. Kim, J.Y. Lee, Y.-J. Kim, Inhibition of apoptosis inBacteroides fragilis enterotoxin-stimulated intestinal epithe-lial cells through the induction of c-IAP-2, Eur. J. Immunol.38 (2008) 2190e2199, ht tps: / /doi .org/10.1002/eji.200838191.

[114] S.H. Ko, D.J. Rho, J.I. Jeon, Y.-J. Kim, H.A. Woo, Y.K. Lee,J.M. Kim, Bacteroides fragilis enterotoxin upregulatesheme oxygenase-1 in intestinal epithelial cells via amitogen-activated protein kinase- and NF-kappaB-Depen-dent pathway, leading to modulation of apoptosis, Infect.Immun. 84 (2016) 2541e2554, https://doi.org/10.1128/IAI.00191-16.

[115] Y.M. Yoon, J.Y. Lee, D. Yoo, Y.-S. Sim, Y.-J. Kim, Y.-K. Oh, J.S. Kang, S. Kim, J.S. Kim, J.M. Kim, Bacteroidesfragilis enterotoxin induces human beta-defensin-2 expres-sion in intestinal epithelial cells via a mitogen-activatedprotein kinase/I kappaB kinase/NF-kappaB-dependentpathway, Infect. Immun. 78 (2010) 2024e2033, https://doi.org/10.1128/IAI.00118-10.

[116] D.Y. Yoo, S.H. Ko, J. Jung, Y.-J. Kim, J.S. Kim, J.M. Kim,Bacteroides fragilis enterotoxin upregulates lipocalin-2expression in intestinal epithelial cells, Lab. Investig. 93(2013) 384e396, https://doi.org/10.1038/labinvest.2013.1.

[117] S.H. Ko, J.I. Jeon, H.S. Myung, Y.-J. Kim, J.M. Kim,Bacteroides fragilis enterotoxin induces formation ofautophagosomes in endothelial cells but interferes withfusion with lysosomes for complete autophagic flux througha mitogen-activated protein kinase-, AP-1-, and C/EBPhomologous protein-dependent pat, Infect. Immun. 85(2017), https://doi.org/10.1128/IAI.00420-17.

[118] J. Allen, S. Hao, C.L. Sears, W. Timp, Epigenetic changesinduced by Bacteroides fragilis toxin (BFT), Infect. Immun.(2019), https://doi.org/10.1128/IAI.00447-18.

[119] S. Rabizadeh, K.-J. Rhee, S. Wu, D. Huso, C.M. Gan,J.E. Golub, X. Wu, M. Zhang, C.L. Sears, Enterotoxigenicbacteroides fragilis: a potential instigator of colitis, Inflamm.Bowel Dis. 13 (2007) 1475e1483, https://doi.org/10.1002/ibd.20265.

[120] K.-J. Rhee, S. Wu, X. Wu, D.L. Huso, B. Karim,A.A. Franco, S. Rabizadeh, J.E. Golub, L.E. Mathews,J. Shin, R.B. Sartor, D. Golenbock, A.R. Hamad, C.M. Gan,F. Housseau, C.L. Sears, Induction of persistent colitis by ahuman commensal, enterotoxigenic Bacteroides fragilis, in

wild-type C57BL/6 mice, Infect. Immun. 77 (2009)1708e1718, https://doi.org/10.1128/IAI.00814-08.

[121] E.C. Wick, S. Rabizadeh, E. Albesiano, X. Wu, S. Wu,J. Chan, K.-J. Rhee, G. Ortega, D.L. Huso, D. Pardoll,F. Housseau, C.L. Sears, Stat3 activation in murine colitisinduced by enterotoxigenic Bacteroides fragilis, Inflamm.Bowel Dis. 20 (2014) 821e834, https://doi.org/10.1097/MIB.0000000000000019.

[122] T. Bowman, M.A. Broome, D. Sinibaldi, W. Wharton,W.J. Pledger, J.M. Sedivy, R. Irby, T. Yeatman,S.A. Courtneidge, R. Jove, Stat3-mediated Myc expressionis required for Src transformation and PDGF-inducedmitogenesis, Proc. Natl. Acad. Sci. 98 (2001) 7319,https://doi.org/10.1073/pnas.131568898. LP e 7324.

[123] G.S. Yochum, C.M. Sherrick, M. MacPartlin, R.H. Goodman,A b-catenin/TCF-coordinated chromatin loop at<em>MYC</em> integrates 50 and 30 Wnt responsive enhancers, Proc.Natl. Acad. Sci. 107 (2010) 145, https://doi.org/10.1073/pnas.0912294107. LP e 150.

[124] M. Yassour, T. Vatanen, H. Siljander, A.-M. Hamalainen,T. Harkonen, S.J. Ryhanen, E.A. Franzosa, H. Vlamakis,C. Huttenhower, D. Gevers, E.S. Lander, M. Knip,R.J. Xavier, Natural history of the infant gut microbiomeand impact of antibiotic treatment on bacterial straindiversity and stability, Sci. Transl. Med. 8 (2016) 343ra81,https://doi.org/10.1126/scitranslmed.aad0917.

[125] D. Rothschild, O. Weissbrod, E. Barkan, A. Kurilshikov,T. Korem, D. Zeevi, P.I. Costea, A. Godneva, I.N. Kalka,N. Bar, S. Shilo, D. Lador, A.V. Vila, N. Zmora, M. Pevsner-Fischer, D. Israeli, N. Kosower, G. Malka, B.C. Wolf,T. Avnit-Sagi, M. Lotan-Pompan, A. Weinberger,Z. Halpern, S. Carmi, J. Fu, C. Wijmenga, A. Zhernakova,E. Elinav, E. Segal, Environment dominates over hostgenetics in shaping human gut microbiota, Nature 555(2018) 210, https://doi.org/10.1038/nature25973.

[126] D.L. Kasper, The polysaccharide capsule of Bacteroidesfragilis subspecies fragilis: immunochemical and morpho-logic definition, J. Infect. Dis. 133 (1976) 79e87.

[127] D.L. Kasper, M.E. Hayes, B.G. Reinap, F.O. Craft,A.B. Onderdonk, B.F. Polk, Isolation and identification ofencapsulated strains of Bacteroides fragilis, J. Infect. Dis.136 (1977) 75e81.

[128] A.A. Lindberg, P. Berthold, C.E. Nord, A. Weintraub,Encapsulated strains of Bacteroides fragilis in clinicalspecimens, Med. Microbiol. Immunol. 167 (1979) 29e36.

[129] D.L. Kasper, A. Weintraub, A.A. Lindberg, J. Lonngren,Capsular polysaccharides and lipopolysaccharides fromtwo Bacteroides fragilis reference strains: chemical andimmunochemical characterization, J. Bacteriol. 153 (1983)991e997.

[130] G.L. Simon, M.S. Klemper, D.L. Kasper, S.L. Gorbach,Alterations in opsonophagocytic killing by neutrophilsof Bacteroides fragilis associated with animal and labora-tory passage: effect of capsular polysaccharide, J. Infect.Dis. 145 (1982) 72e77, https://doi.org/10.1093/infdis/145.1.72.

[131] A.O. Tzianabos, A.B. Onderdonk, B. Rosner, R.L. Cisneros,D.L. Kasper, Structural features of polysaccharides thatinduce intra-abdominal abscesses, Science 262 (1993)416e419.

[132] Y. Wang, W.M. Kalka-Moll, M.H. Roehrl, D.L. Kasper,Structural basis of the abscess-modulating polysaccharideA2 from Bacteroides fragilis, Proc. Natl. Acad. Sci. U. S. A

781Good gone bad

97 (2000) 13478e13483, https://doi.org/10.1073/pnas.97.25.13478.

[133] C.M. Krinos, M.J. Coyne, K.G. Weinacht, A.O. Tzianabos,D.L. Kasper, L.E. Comstock, Extensive surface diversity ofa commensal microorganism by multiple DNA inversions,Nature 414 (2001) 555e558, https://doi.org/10.1038/35107092.

[134] M.J. Coyne, A.O. Tzianabos, B.C. Mallory, V.J. Carey,D.L. Kasper, L.E. Comstock, Polysaccharide biosynthesislocus required for virulence of Bacteroides fragilis, Infect.Immun. 69 (2001) 4342e4350, https://doi.org/10.1128/IAI.69.7.4342-4350.2001.

[135] M.J. Coyne, M. Chatzidaki-Livanis, L.C. Paoletti,L.E. Comstock, Role of glycan synthesis in colonizationof the mammalian gut by the bacterial symbiont Bacter-oides fragilis, Proc. Natl. Acad. Sci. U. S. A 105 (2008)13099e13104, https://doi.org/10.1073/pnas.0804220105.

[136] C.H. Liu, S.M. Lee, J.M. Vanlare, D.L. Kasper,S.K. Mazmanian, Regulation of surface architecture bysymbiotic bacteria mediates host colonization, Proc. Natl.Acad. Sci. U. S. A 105 (2008) 3951e3956, https://doi.org/10.1073/pnas.0709266105.

[137] B.A. Cobb, Q. Wang, A.O. Tzianabos, D.L. Kasper,Polysaccharide processing and presentation by the MHCIIpathway, Cell 117 (2004) 677e687, https://doi.org/10.1016/J.CELL.2004.05.001.

[138] B.A. Cobb, D.L. Kasper, Characteristics of carbohydrateantigen binding to the presentation protein HLA-DR,Glycobiology 18 (2008) 707e718, https://doi.org/10.1093/glycob/cwn050.

[139] Yong-Hoon Choi, Michael H. Roehrl, Dennis L. Kasper,Julia Y. Wang, A Unique Structural Pattern Shared by T-Cell-Activating and Abscess-Regulating Zwitterionic Poly-saccharidesy, 2002, https://doi.org/10.1021/BI020491V.

[140] L.E. Blandford, E.L. Johnston, J.D. Sanderson, W.G. Wade,A.J. Lax, Promoter orientation of the immunomodulatoryBacteroides fragilis capsular polysaccharide A (PSA) is off inindividuals with inflammatory bowel disease (IBD), GutM ic rob . ( 2019 ) 1e9 , h t t ps : / / do i . o rg /10 .1080 /19490976.2018.1560755.

[141] J.L. Chan, S. Wu, A.L. Geis, G. V Chan, T.A.M. Gomes,S.E. Beck, X. Wu, H. Fan, A.J. Tam, L. Chung, H. Ding,H. Wang, D.M. Pardoll, F. Housseau, C.L. Sears, Non-toxigenic Bacteroides fragilis (NTBF) administration reducesbacteria-driven chronic colitis and tumor development inde-pendent of polysaccharide A, Mucosal Immunol. 12 (2019)164e177, https://doi.org/10.1038/s41385-018-0085-5.

[142] S.M. Lee, G.P. Donaldson, Z. Mikulski, S. Boyajian, K. Ley,S.K. Mazmanian, Bacterial colonization factors controlspecificity and stability of the gut microbiota, Nature 501(2013) 426e429, https://doi.org/10.1038/nature12447.

[143] E.C. Martens, E.C. Lowe, H. Chiang, N.A. Pudlo, M. Wu,N.P. McNulty, D.W. Abbott, B. Henrissat, H.J. Gilbert,D.N. Bolam, J.I. Gordon, Recognition and degradation ofplant cell wall polysaccharides by two human gut sym-bionts, PLoS Biol. 9 (2011), e1001221, https://doi.org/10.1371/journal.pbio.1001221.

[144] J.L. Sonnenburg, J. Xu, D.D. Leip, C.-H. Chen,B.P. Westover, J. Weatherford, J.D. Buhler, J.I. Gordon,Glycan foraging in vivo by an intestine-adapted bacterialsymbiont, Science 307 (2005) 1955e1959, https://doi.org/10.1126/science.1109051.

[145] E.C. Martens, H.C. Chiang, J.I. Gordon, Mucosal glycanforaging enhances fitness and transmission of a sacchar-

olytic human gut bacterial symbiont, Cell Host Microbe 4( 2 0 0 8 ) 4 4 7e4 5 7 , h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 /j.chom.2008.09.007.

[146] A. Marcobal, M. Barboza, E.D. Sonnenburg, N. Pudlo,E.C. Martens, P. Desai, C.B. Lebrilla, B.C. Weimer,D.A. Mills, J.B. German, J.L. Sonnenburg, Bacteroides inthe infant gut consume milk oligosaccharides via mucus-utilization pathways, Cell Host Microbe 10 (2011) 507e514,https://doi.org/10.1016/j.chom.2011.10.007.

[147] Y. Cao, E.R. Rocha, C.J. Smith, Efficient utilization ofcomplex N-linked glycans is a selective advantage for<em>Bacteroides fragilis</em> in extraintestinal infec-tions, Proc. Natl. Acad. Sci. 111 (2014) 12901, https://doi.org/10.1073/pnas.1407344111. LP e 12906.

[148] G.P. Donaldson, M.S. Ladinsky, K.B. Yu, J.G. Sanders,B.B. Yoo, W.-C. Chou, M.E. Conner, A.M. Earl, R. Knight,P.J. Bjorkman, S.K. Mazmanian, Gut microbiota utilizeimmunoglobulin A for mucosal colonization, Science 360(2018) 795e800, https://doi.org/10.1126/science.aaq0926.

[149] M.R. Popoff, Clostridial pore-forming toxins: powerfulvirulence factors, Anaerobe 30 (2014) 220e238, https://doi.org/10.1016/j.anaerobe.2014.05.014.

[150] B.J. Berube, J. Bubeck Wardenburg, Staphylococcusaureus alpha-toxin: nearly a century of intrigue, Toxins 5(2013) 1140e1166.

[151] M. Bielaszewska, T. Aldick, A. Bauwens, H. Karch,Hemolysin of enterohemorrhagic Escherichia coli: struc-ture, transport, biological activity and putative role invirulence, Int. J. Med. Microbiol. 304 (2014) 521e529,https://doi.org/10.1016/j.ijmm.2014.05.005.

[152] K.P. Robertson, C.J. Smith, A.M. Gough, E.R. Rocha,Characterization of Bacteroides fragilis hemolysins andregulation and synergistic interactions of HlyA and HlyB,Infect. Immun. 74 (2006) 2304e2316, https://doi.org/10.1128/IAI.74.4.2304-2316.2006.

[153] L.A. Lobo, A.L. Jenkins, C. Jeffrey Smith, E.R. Rocha,Expression of Bacteroides fragilis hemolysins in vivo and roleofHlyBA inan intra-abdominal infectionmodel,Microbiology2(2013) 326e337, https://doi.org/10.1002/mbo3.76.

[154] A.S. Luis, E.C. Martens, Interrogating gut bacterial gen-omes for discovery of novel carbohydrate degradingenzymes, Curr. Opin. Chem. Biol. 47 (2018) 126e133,https://doi.org/10.1016/j.cbpa.2018.09.012.

[155] A.G. FRASER, R. BROWN, Neuraminidase production bybacteroidaceae, J. Med. Microbiol. 14 (1981) 63e76,https://doi.org/10.1099/00222615-14-1-63.

[156] N. Juge, L. Tailford, C.D. Owen, Sialidases from gutbacteria: a mini-review, Biochem. Soc. Trans. 44 (2016)166e175, https://doi.org/10.1042/BST20150226.

[157] B.L. Haines-Menges, W.B. Whitaker, J.B. Lubin, E.F. Boyd,Host sialic acids: a delicacy for the pathogen withdiscerning taste, Microbiol. Spectr. 3 (2015), https://doi.org/10.1128/microbiolspec.MBP-0005-2014.

[158] V.G. Godoy, M.M. Dallas, T.A. Russo, M.H. Malamy, A rolefor Bacteroides fragilis neuraminidase in bacterial growth intwo model systems, Infect. Immun. 61 (1993) 4415e4426.http://www.ncbi.nlm.nih.gov/pubmed/8406832. (Accessed31 July 2019).

[159] R.M.C.P. Domingues, S.M.B. Cavalcanti, A.F.B. Andrade,M.C.S. Ferreira, Sialic acid as receptor of Bacteroidesfragilis lectin-like adhesin, Zentralblatt Bakteriol. 277 (1992)340e344, https://doi.org/10.1016/S0934-8840(11)80912-6.

[160] C.A. Guzm�an, M. Plat�e, C. Pruzzo, M.J. Gauthier, Role ofneuraminidase-dependent adherence in Bacteroides fragi-

782 Good gone bad

lis attachment to human epithelial cells, FEMS Microbiol.Lett. 59 (1990) 187e192, https://doi.org/10.1111/j.1574-6968.1989.tb03107.x.

[161] R. Jotwani, N. Kato, H. Kato, K. Watanabe, K. Ueno,Detection of Bacteroides fragilis in clinical specimens bypolymerase chain reaction amplification of the neuramini-dase gene, Curr. Microbiol. 31 (1995) 215e219, https://doi.org/10.1007/BF00298376.

[162] R.F. Thornton, T.F. Kagawa, P.W.O'Toole, J.C. Cooney, Thedissemination of C10 cysteine protease genes in Bacteroidesfragilis bymobile genetic elements, BMCMicrobiol. 10 (2010)122, https://doi.org/10.1186/1471-2180-10-122.

[163] R.F. Thornton, E.C. Murphy, T.F. Kagawa, P.W. O'Toole,J.C. Cooney, The effect of environmental conditions onexpression of Bacteroides fragilis and Bacteroides thetaio-taomicron C10 protease genes, BMC Microbiol. 12 (2012)190, https://doi.org/10.1186/1471-2180-12-190.

[164] Y. Chen, T. Kinouchi, K. Kataoka, S. Akimoto, Y. Ohnishi,Purification and characterization of a fibrinogen-degradingprotease in Bacteroides fragilis strain YCH46, Microbiol.Immunol. 39 (1995) 967e977, https://doi.org/10.1111/j.1348-0421.1995.tb03300.x.

[165] S. Houston, G.W. Blakely, A. McDowell, L. Martin,S. Patrick, Binding and degradation of fibrinogen byBacteroides fragilis and characterization of a 54 kDafibrinogen-binding protein, Microbiology 156 (2010)2516e2526, https://doi.org/10.1099/mic.0.038588-0.

[166] P. Klemm, M.A. Schembri, Bacterial adhesins: function andstructure, Int. J. Med. Microbiol. 290 (2000) 27e35, https://doi.org/10.1016/S1438-4221(00)80102-2.

[167] K. Moonens, H. Remaut, Evolution and structural dynamicsof bacterial glycan binding adhesins, Curr. Opin. Struct.Biol. 44 (2017) 48e58, https://doi.org/10.1016/j.sbi.2016.12.003.

[168] M.K. Hospenthal, T.R.D. Costa, G. Waksman,A comprehensive guide to pilus biogenesis in Gram-negative bacteria, Nat. Rev. Microbiol. 15 (2017)365e379, https://doi.org/10.1038/nrmicro.2017.40.

[169] I. Brook, L.A. Myhal, C.H. Dorsey, Encapsulation and pilusformation of Bacteroides spp. in normal flora abscessesand blood, J. Infect. (1992), https://doi.org/10.1016/0163-4453(92)91479-U.

[170] C. Pruzzo, B. Dainelli, M. Ricchetti, Piliated Bacteroidesfragilis strains adhere to epithelial cells and are moresensitive to phagocytosis by human neutrophils thannonpiliated strains, Infect. Immun. 43 (1984) 189e194.http://www.ncbi.nlm.nih.gov/pubmed/6140223. (Accessed31 July 2019).

[171] W.A.C. Vel, F. Namavar, A.M.J.J. Verweij-Van Vught,A.N.B. Pubben, D.M. Maclaren, Haemagglutination by theBacteroides fragilis group, J. Med. Microbiol. 21 (1986)105e107, https://doi.org/10.1099/00222615-21-2-105.

[172] V. Nakano, M.J. Avila-Campos, Virulence markers andantimicrobial susceptibility of bacteria of the Bacteroidesfragilis group isolated from stool of children with diarrheain S~ao Paulo, Brazil, Mem. Inst. Oswaldo Cruz 99 (2004)3 0 7e3 1 2 , h t t p s : / / d o i . o r g / 1 0 . 1 5 9 0 / S 0 0 7 4 -02762004000300012.

[173] C. Pruzzo, C.A. Guzm�an, B. Dainelli, Incidence ofhemagglutination activity among pathogenic and non-pathogenic Bacteroides fragilis strains and role of capsuleand pili in HA and adherence, FEMS Microbiol. Lett. 59(1989) 113e118, https://doi.org/10.1016/0378-1097(89)90469-2.

[174] H. Pauer, E. de O. Ferreira, J. dos Santos-Filho,M .B . Por te la , R .B . Z inga l i , R .M.A . Soares ,R.M.C.P. Domingues, A TonB-dependent outer membraneprotein as a Bacteroides fragilis fibronectin-binding mole-cule, FEMS Immunol. Med. Microbiol. 55 (2009) 388e395,https://doi.org/10.1111/j.1574-695X.2009.00532.x.

[175] H. Pauer, S.N.V. Cavalcanti, F.L. Teixeira, J. Santos-Filho,R.C. Vommaro, A.C.S.C. Oliveira, E.O. Ferreira,R.R.M.C.P. Domingues, Inactivation of a fibronectin-bind-ing TonB-dependent protein increases adhesion propertiesof Bacteroides fragilis, J. Med. Microbiol. 62 (2013)1524e1530, https://doi.org/10.1099/jmm.0.054692-0.

[176] P. Eiring, B. Manncke, K. Gerbracht, H. Werner, Bacter-oides fragilis adheres to laminin significantly stronger thanbacteroides thetaiotaomicron and other species of thegenus, Zentralblatt Bakteriol. 282 (1995) 279e286, https://doi.org/10.1016/S0934-8840(11)80128-3.

[177] E. de O. Ferreira, L. Araújo Lobo, D. Barreiros Petr�opolis,K.E. dos, S. Avelar, M.C. Ferreira, F.C. e Silva Filho,R.M.C.P. Domingues, A Bacteroides fragilis surfaceglycoprotein mediates the interaction between the bacter-ium and the extracellular matrix component laminin-1, Res.Microbiol. 157 (2006) 960e966, https://doi.org/10.1016/J.RESMIC.2006.09.005.

[178] E. de O. Ferreira, F.L. Teixeira, F. Cordeiro, L. Araujo Lobo,E.R. Rocha, J.C. Smith, R.M.C.P. Domingues, The Bfp60surface adhesin is an extracellular matrix and plasminogenprotein interacting in Bacteroides fragilis, Int. J. Med.Microbiol. 303 (2013) 492e497, https://doi.org/10.1016/J.IJMM.2013.06.007.

[179] E. de O. Ferreira, J.B. de Carvalho, R.J.M. Peixoto,L.A. Lobo, R.B. Zingalli, C.J. Smith, E.R. Rocha,R.M.C.P. Domingues, The interaction of Bacteroidesfragilis with components of the human fibrinolytic system,FEMS Immunol. Med. Microbiol. 56 (2009) 48e55, https://doi.org/10.1111/j.1574-695X.2009.00546.x.

[180] E.C.Murphy,M.Morgelin, J.C.Cooney, I.-M.Frick, Interactionof Bacteroides fragilis and Bacteroides thetaiotaomicron withthe kallikrein-kinin system, Microbiology 157 (2011)2094e2105, https://doi.org/10.1099/mic.0.046862-0.

[181] R. Sijbrandi, M. Stork, J. Luirink, B.R. Otto, Pbp, a cell-surface exposed plasminogen binding protein of Bacter-oides fragilis, Microb. Infect. 10 (2008) 514e521, https://doi.org/10.1016/J.MICINF.2008.01.015.

[182] E.S. Friedman, K. Bittinger, T. V Esipova, L. Hou, L. Chau,J. Jiang, C. Mesaros, P.J. Lund, X. Liang, G.A. FitzGerald,M. Goulian, D. Lee, B.A. Garcia, I.A. Blair, S.A. Vinogradov,G.D. Wu, Microbes vs. chemistry in the origin of theanaerobic gut lumen, Proc. Natl. Acad. Sci. U. S. A 115( 2018 ) 4170e4175 , h t t p s : / / d o i . o r g / 10 . 1073 /pnas.1718635115.

[183] R.D. Rolfe, D.J. Hentges, J.T. Barrett, B.J. Campbell,Oxygen tolerance of human intestinal anaerobes, Am. J.Clin. Nutr. 30 (1977) 1762e1769, https://doi.org/10.1093/ajcn/30.11.1762.

[184] E. Rocha, C.D. Herren, D.J. Smalley, C. Smith, Thecomplex oxidative stress response of Bacteroides fragilis:the role of OxyR in control of gene expression, Anaerobe 9(2003) 165e173, https://doi.org/10.1016/S1075-9964(03)00118-5.

[185] C.D. Herren, E.R. Rocha, C.J. Smith, Genetic analysis ofan important oxidative stress locus in the anaerobeBacteroides fragilis, Gene 316 (2003) 167e175, https://doi.org/10.1016/S0378-1119(03)00759-5.

783Good gone bad

[186] C.J. Sund, E.R. Rocha, A.O. Tzinabos, W.G. Wells,J.M. Gee, M.A. Reott, D.P. O'Rourke, C.J. Smith, TheBacteroides fragilis transcriptome response to oxygen andH2O2: the role of OxyR and its effect on survival andvirulence, Mol. Microbiol. 67 (2007) 129e142, https://doi.org/10.1111/j.1365-2958.2007.06031.x.

[187] E. de Oliveira Ferreira, E. Yates, M. Jourlin, J. Liu,R.M.C. Pilotto Domingues, M. Goldner, Anaerobe/aerobeenvironmental flux determines protein expression profilesof Bacteroides fragilis, a redox pathogen, Anaerobe 17(2011) 4e14, h t tps : / /do i .o rg /10 .1016/J .ANAE-ROBE.2010.09.004.

[188] A.D. Baughn, M.H. Malamy, The strict anaerobe Bacter-oides fragilis grows in and benefits from nanomolarconcentrations of oxygen, Nature 427 (2004) 441e444,https://doi.org/10.1038/nature02285.

[189] B.M. Meehan, A.D. Baughn, R. Gallegos, M.H. Malamy,Inactivation of a single gene enables microaerobic growthof the obligate anaerobe Bacteroides fragilis, Proc. Natl.Acad. Sci. 109 (2012) 12153e12158, https://doi.org/10.1073/pnas.1203796109.

[190] F.P. Tally, B.R. Goldin, N. V Jacobus, S.L. Gorbach,Superoxide dismutase in anaerobic bacteria of clinicalsignificance, Infect. Immun. 16 (1977) 20e25. http://www.ncbi.nlm.nih.gov/pubmed/326669. (Accessed 31 July2019).

[191] E.R. Rocha, T. Selby, J.P. Coleman, C.J. Smith, Oxidativestress response in an anaerobe, Bacteroides fragilis: a rolefor catalase in protection against hydrogen peroxide,J. Bacteriol. 178 (1996) 6895e6903, https://doi.org/10.1128/jb.178.23.6895-6903.1996.

[192] E.R. Rocha, C.J. Smith, Role of the alkyl hydroperoxidereductase (ahpCF) gene in oxidative stress defense of theobligate Anaerobe bacteroides fragilis, J. Bacteriol. 181(1999) 5701e5710. http://www.ncbi.nlm.nih.gov/pubmed/10482511. (Accessed 31 July 2019).

[193] E.R. Rocha, S.C. Andrews, J.N. Keen, J.H. Brock, Isolationof a ferritin from Bacteroides fragilis, FEMS Microbiol. Lett.95 (1992) 207e212, https://doi.org/10.1111/j.1574-6968.1992.tb05367.x.

[194] E.R. Rocha, C.J. Smith, Transcriptional regulation of theBacteroides fragilis ferritin gene (ftnA) by redox stress,Microbiology 150 (2004) 2125e2134, https://doi.org/10.1099/mic.0.26948-0.

[195] G.H. Gauss, M.A. Reott, E.R. Rocha, M.J. Young,T. Douglas, C.J. Smith, C.M. Lawrence, Characterizationof the Bacteroides fragilis bfr gene product identifies abacterial DPS-like protein and suggests evolutionary linksin the ferritin superfamily, J. Bacteriol. 194 (2012) 15e27,https://doi.org/10.1128/JB.05260-11.

[196] M.I. Betteken, E.R. Rocha, C.J. Smith, Dps and DpsLmediate survival in vitro and in vivo during the prolongedoxidative stress response in Bacteroides fragilis,J. Bacteriol. 197 (2015) 3329e3338, https://doi.org/10.1128/JB.00342-15.

[197] M.A. Reott, A.C. Parker, E.R. Rocha, C.J. Smith, Thior-edoxins in redox maintenance and survival during oxidativestress of Bacteroides fragilis, J. Bacteriol. 191 (2009)3384e3391, https://doi.org/10.1128/JB.01665-08.

[198] E.R. Rocha, G. Owens, C.J. Smith, The redox-sensitivetranscriptional activator OxyR regulates the peroxideresponse regulon in the obligate anaerobe Bacteroidesfragilis, J. Bacteriol. 182 (2000) 5059e5069, https://doi.org/10.1128/jb.182.18.5059-5069.2000.

[199] F.L. Teixeira, D.N. Silva, H. Pauer, L.Q. Ferreira, O.Ferreira E. de, R.M. Domingues, L.A. Lobo, The role ofBmoR, a MarR Family Regulator, in the survival ofBacteroides fragilis during oxidative stress, Int. J. Med.Microbiol. 303 (2013) 443e448, https://doi.org/10.1016/J.IJMM.2013.05.007.

[200] F.L. Teixeira, H. Pauer, S.B. Costa, C.J. Smith,R.M.C.P. Domingues, E.R. Rocha, L.A. Lobo, Deletion ofBmoR affects the expression of genes related to thiol/disulfide balance in Bacteroides fragilis, Sci. Rep. 8 (2018)14405, https://doi.org/10.1038/s41598-018-32880-7.

[201] L. Pumbwe, C.A. Skilbeck, V. Nakano, M.J. Avila-Campos,R.M.F. Piazza, H.M. Wexler, Bile salts enhance bacterialco-aggregation, bacterial-intestinal epithelial cell adhesion,biofilm formation and antimicrobial resistance of Bacter-oides fragilis, Microb. Pathog. 43 (2007) 78e87, https://doi.org/10.1016/J.MICPATH.2007.04.002.

[202] R.F. Boente, H. Pauer, D.N.S. Silva, J.S. Filho, V. Sandim,L.C.M. Antunes, R.B.R. Ferreira, R.B. Zingali ,R.M.C.P. Domingues, L.A. Lobo, Differential proteomicanalysis of outer membrane enriched extracts of Bacter-oides fragilis grown under bile salts stress, Anaerobe 39(2016) 84e90, ht tps: / /doi .org/10.1016/J.ANAE-ROBE.2016.03.003.

[203] C.M.G. Silva, D.N. dos S Silva, S.B. da Costa, J.S. deS . A l m e i d a , R . F . B o e n t e , F . L . T e i x e i r a ,R.M.C.P. Domingues, L.A. Lobo, Inactivation of MarR genehomologs increases susceptibility to antimicrobials inBacteroides fragilis, Braz. J. Microbiol. 49 (2018)200e206, https://doi.org/10.1016/J.BJM.2017.05.005.

[204] A. Konovalova, D.E. Kahne, T.J. Silhavy, Outer membranebiogenesis, Annu. Rev. Microbiol. 71 (2017) 539e556,https://doi.org/10.1146/annurev-micro-090816-093754.

[205] P. Sperandeo, A.M. Martorana, A. Polissi, Lipopolysac-charide biosynthesis and transport to the outer membraneof gram-negative bacteria, Subcell. Biochem. (2019) 9e37,https://doi.org/10.1007/978-3-030-18768-2_2.

[206] B.W. Simpson, M.S. Trent, Pushing the envelope: LPSmodifications and their consequences, Nat. Rev. Microbiol.17 (2019) 403e416, https://doi.org/10.1038/s41579-019-0201-x.

[207] Y.-C. Lu, W.-C. Yeh, P.S. Ohashi, LPS/TLR4 signaltransduction pathway, Cytokine 42 (2008) 145e151,https://doi.org/10.1016/j.cyto.2008.01.006.

[208] M.J. Wannemuehler, S.M. Michalek, E. Jiril lo,S.I. Williamson, M. Hirasawa, J.R. McGhee, LPS regula-tion of the immune response: Bacteroides endotoxininduces mitogenic, polyclonal, and antibody responses inclassical LPS responsive but not C3H/HeJ mice,J. Immunol. 133 (1984) 299e305. http://www.ncbi.nlm.nih.gov/pubmed/6202784. (Accessed 31 July 2019).

[209] D.L. Kasper, Chemical and biological characterization ofthe lipopolysaccharide of Bacteroides fragilis subspeciesfragilis, J. Infect. Dis. 134 (1976) 59e66, https://doi.org/10.1093/infdis/134.1.59.

[210] A.A. Lindberg, A. Weintraub, U. Z€ahringer, E. Th Rietschel,Structure-Activity Relationships in Lipopolysaccharidesof Bacteroides Fragilis, n.d. https://www.jstor.org/stable/pdf/4455470.pdf?refreqid¼excelsior%3A44d8641961de40aa8dd9fbbe16b7be53 (accessed July 31, 2019)..

[211] B. Brake, C. Larcher, T.F. Schulz, W. Prodinger,M.P. Dierich, Species specific monoclonal antibodiesto Bacteroides fragilis lipopolysaccharide protectmice from severe infection, Zentralblatt Bakteriol. 277

784 Good gone bad

(1992) 320e328, https://doi.org/10.1016/S0934-8840(11)80909-6.

[212] H. Seedorf, N.W. Griffin, V.K. Ridaura, A. Reyes, J. Cheng,F.E. Rey, M.I. Smith, G.M. Simon, R.H. Scheffrahn,D. Woebken, A.M. Spormann, W. Van Treuren,L.K. Ursell, M. Pirrung, A. Robbins-Pianka, B.L. Cantarel,V. Lombard, B. Henrissat, R. Knight, J.I. Gordon, Bacteriafrom diverse habitats colonize and compete in the mousegut, Cell 159 (2014) 253e266, https://doi.org/10.1016/j.cell.2014.09.008.

[213] M.A. Bauer, K. Kainz, D. Carmona-Gutierrez, F. Madeo,Microbial wars: competition in ecological niches and withinthe microbiome, Microb. Cell (Graz, Austria) 5 (2018)215e219, https://doi.org/10.15698/mic2018.05.628.

[214] M.E. Hibbing, C. Fuqua, M.R. Parsek, S.B. Peterson,Bacterial competition: surviving and thriving in the microbialjungle, Nat. Rev. Microbiol. 8 (2010) 15e25, https://doi.org/10.1038/nrmicro2259.

[215] A. Jamet, X. Nassif, New players in the toxin field:polymorphic toxin systems in bacteria, mBio 6 (2015),https://doi.org/10.1128/mBio.00285-15 e00285-15.

[216] E.C. Garcia, Contact-dependent interbacterial toxins deli-ver a message, Curr. Opin. Microbiol. 42 (2018) 40e46,https://doi.org/10.1016/j.mib.2017.09.011.

[217] T.G. Dong, B.T. Ho, D.R. Yoder-Himes, J.J. Mekalanos,Identification of T6SS-dependent effector and immunityproteins by Tn-seq in Vibrio cholerae, Proc. Natl. Acad. Sci.U. S. A 110 (2013) 2623e2628, https://doi.org/10.1073/pnas.1222783110.

[218] T.G. Sana, N. Flaugnatti, K.A. Lugo, L.H. Lam, A. Jacobson,V. Baylot, E. Durand, L. Journet, E. Cascales, D.M. Monack,Salmonella Typhimurium utilizes a T6SS-mediated antibac-terialweapon toestablish in thehost gut,Proc.Natl. Acad.Sci.U. S. A 113 (2016) E5044eE5051, https://doi.org/10.1073/pnas.1608858113.

[219] S.L. Logan, J. Thomas, J. Yan, R.P. Baker, D.S. Shields,J.B. Xavier, B.K. Hammer, R. Parthasarathy, The Vibriocholerae type VI secretion system can modulate hostintestinal mechanics to displace gut bacterial symbionts,Proc. Natl. Acad. Sci. U. S. A 115 (2018) E3779eE3787,https://doi.org/10.1073/pnas.1720133115.

[220] M.J. Coyne, L.E. Comstock, Type VI secretion systems andthe gut microbiota, Microbiol. Spectr. 7 (2019), https://doi.org/10.1128/microbiolspec.PSIB-0009-2018.

[221] Y. Cherrak, C. Rapisarda, R. Pellarin, G. Bouvier,B. Bardiaux, F. Allain, C. Malosse, M. Rey, J. Chamot-Rooke, E. Cascales, R. Fronzes, E. Durand, Biogenesisand structure of a type VI secretion baseplate, Nat.Microbiol. 3 (2018) 1404e1416, https://doi.org/10.1038/s41564-018-0260-1.

[222] C.J. Alteri, H.L.T. Mobley, The versatile type VI secretionsystem, Microbiol. Spectr. 4 (2016), https://doi.org/10.1128/microbiolspec.VMBF-0026-2015.

[223] A. Hachani, T.E. Wood, A. Filloux, Type VI secretion andanti-host effectors, Curr. Opin. Microbiol. 29 (2016) 81e93,https://doi.org/10.1016/j.mib.2015.11.006.

[224] B.D. Ross, A.J. Verster, M.C. Radey, D.T. Schmidtke,C.E. Pope, L.R. Hoffman, A.M. Hajjar, S.B. Peterson,E. Borenstein, J.D. Mougous, Human gut bacteria containacquired interbacterial defence systems, Nature 575 (2019)224e228, https://doi.org/10.1038/s41586-019-1708-z.

[225] C.J. Stewart, N.J. Ajami, J.L. O'Brien, D.S. Hutchinson,D.P. Smith, M.C. Wong, M.C. Ross, R.E. Lloyd,H. Doddapaneni, G.A. Metcalf, D. Muzny, R.A. Gibbs,

T. Vatanen, C. Huttenhower, R.J. Xavier, M. Rewers,W. Hagopian, J. Toppari, A.-G. Ziegler, J.-X. She,B. Akolkar, A. Lernmark, H. Hyoty, K. Vehik,J.P. Krischer, J.F. Petrosino, Temporal development ofthe gut microbiome in early childhood from the TEDDYstudy, Nature 562 (2018) 583e588, https://doi.org/10.1038/s41586-018-0617-x.

[226] K. Korpela, W.M. de Vos, Early life colonization of thehuman gut: microbes matter everywhere, Curr. Opin.Microbiol. 44 (2018) 70e78, https://doi.org/10.1016/j.mib.2018.06.003.

[227] H. Renz, P. Brandtzaeg, M. Hornef, The impact of perinatalimmune development on mucosal homeostasis andchronic inflammation, Nat. Rev. Immunol. 12 (2011)9e23, https://doi.org/10.1038/nri3112.

[228] S. Rokhsefat, A. Lin, E.M. Comelli, Mucin-microbiotainteraction during postnatal maturation of the intestinalecosystem: clinical implications, Dig. Dis. Sci. 61 (2016)1473e1486, https://doi.org/10.1007/s10620-016-4032-6.

[229] L. Arike, G.C. Hansson, The densely O-glycosylated MUC2mucin protects the intestine and provides food for thecommensal bacteria, J. Mol. Biol. 428 (2016) 3221e3229,https://doi.org/10.1016/j.jmb.2016.02.010.

[230] M.E. V Johansson, J.M.H. Larsson, G.C. Hansson, The twomucus layers of colon are organized by the MUC2 mucin,whereas the outer layer is a legislator of hostemicrobialinteractions, Proc. Natl. Acad. Sci. 108 (2011) 4659, https://doi.org/10.1073/pnas.1006451107. LP e 4665.

[231] T. Pelaseyed, J.H. Bergstrom, J.K. Gustafsson, A. Ermund,G.M.H. Birchenough, A. Schutte, S. van der Post,F. Svensson, A.M. Rodriguez-Pineiro, E.E.L. Nystrom,C. Wising, M.E. V Johansson, G.C. Hansson, The mucusand mucins of the goblet cells and enterocytes provide thefirst defense line of the gastrointestinal tract and interactwith the immune system, Immunol. Rev. 260 (2014) 8e20,https://doi.org/10.1111/imr.12182.

[232] A.M. Rodríguez-Pi~neiro, M.E. V Johansson, The colonicmucus protection depends on the microbiota, Gut Microb. 6( 2 0 1 5 ) 3 2 6e3 3 0 , h t t p s : / / d o i . o r g / 1 0 . 1 0 8 0 /19490976.2015.1086057.

[233] H.E. Jakobsson, A.M. Rodríguez-Pi~neiro, A. Schütte,A. Ermund, P. Boysen, M. Bemark, F. Sommer,F. B€ackhed, G.C. Hansson, M.E. V Johansson, Thecomposition of the gut microbiota shapes the colon mucusbarrier, EMBO Rep. 16 (2015) 164e177, https://doi.org/10.15252/embr.201439263.

[234] L. Aymeric, F. Donnadieu, C. Mulet, L. du Merle, G. Nigro,A. Saffarian, M. Berard, C. Poyart, S. Robine, B. Regnault,P. Trieu-Cuot, P.J. Sansonetti, S. Dramsi, Colorectalcancer specific conditions promote Streptococcus galloly-ticus gut colonization, Proc. Natl. Acad. Sci. U. S. A 115(2018) E283eE291, https://doi.org/10.1073/pnas.1715112115.

[235] K.W. Kinzler, B. Vogelstein, Lessons from hereditarycolorectal cancer, Cell 87 (1996) 159e170.

[236] A.J. Rowan, H. Lamlum, M. Ilyas, J. Wheeler, J. Straub,A. Papadopoulou, D. Bicknell, W.F. Bodmer, I.P. Tomlinson,APC mutations in sporadic colorectal tumors: a mutational“hotspot” and interdependence of the “two hits”, Proc. Natl.Acad. Sci. U. S. A 97 (2000) 3352e3357.

[237] L. Zhang, J.W. Shay, Multiple roles of APC and itstherapeutic implications in colorectal cancer, J. Natl.Cancer Inst. 109 (2017), https://doi.org/10.1093/jnci/djw332.

785Good gone bad

[238] A.R. Moser, E.M. Mattes, W.F. Dove, M.J. Lindstrom,J.D. Haag, M.N. Gould, ApcMin, a mutation in the murineApc gene, predisposes to mammary carcinomas and focalalveolar hyperplasias, Proc. Natl. Acad. Sci. U. S. A 90(1993) 8977e8981.

[239] S. Wu, K.-J. Rhee, E. Albesiano, S. Rabizadeh, X. Wu, H.-R. Yen, D.L. Huso, F.L. Brancati, E. Wick, F. McAllister,F. Housseau, D.M. Pardoll, C.L. Sears, A human coloniccommensal promotes colon tumorigenesis via activation ofT helper type 17 T cell responses, Nat. Med. 15 (2009)1016e1022, https://doi.org/10.1038/nm.2015.

[240] S. Tomkovich, C.M. Dejea, K. Winglee, J.L. Drewes,L. Chung, F. Housseau, J.L. Pope, J. Gauthier, X. Sun,M. Muhlbauer, X. Liu, P. Fathi, R.A. Anders, S. Besharati,E. Perez-Chanona, Y. Yang, H. Ding, X. Wu, S. Wu,J.R. White, R.Z. Gharaibeh, A.A. Fodor, H. Wang,D.M. Pardoll, C. Jobin, C.L. Sears, Human colon mucosalbiofilms from healthy or colon cancer hosts are carcino-genic, J. Clin. Investig. 130 (2019), https://doi.org/10.1172/JCI124196.

[241] Y. Litvak, M.X. Byndloss, A.J. Baumler, Colonocytemetabolism shapes the gut microbiota, Science 362(2018), https://doi.org/10.1126/science.aat9076.

[242] A.C. Goodwin, C.E. Destefano Shields, S. Wu, D.L. Huso,X. Wu, T.R. Murray-Stewart, A. Hacker-Prietz,S. Rabizadeh, P.M. Woster, C.L. Sears, R.A.J. Casero,Polyamine catabolism contributes to enterotoxigenic Bac-teroides fragilis-induced colon tumorigenesis, Proc. Natl.Acad. Sci. U. S. A 108 (2011) 15354e15359, https://doi.org/10.1073/pnas.1010203108.

[243] K.A. Knoop, J.K. Gustafsson, K.G. McDonald, D.H. Kulkarni,P.E. Coughlin, S. McCrate, D. Kim, C.-S. Hsieh, S.P. Hogan,C.O. Elson, P.I. Tarr, R.D. Newberry, Microbial antigenencounter during a preweaning interval is critical for toleranceto gut bacteria, Sci. Immunol. 2 (2017), https://doi.org/10.1126/sciimmunol.aao1314.

[244] C.E. DeStefano Shields, S.W. Van Meerbeke,F. Housseau, H. Wang, D.L. Huso, R.A.J. Casero,

H.M. O'Hagan, C.L. Sears, Reduction of murine colontumorigenesis driven by enterotoxigenic Bacteroidesfragilis using cefoxitin treatment, J. Infect. Dis. 214(2016) 122e129, https://doi.org/10.1093/infdis/jiw069.

[245] N. Zmora, G. Zilberman-Schapira, J. Suez, U. Mor, M. Dori-Bachash, S. Bashiardes, E. Kotler, M. Zur, D. Regev-Lehavi, R.B.-Z. Brik, S. Federici, Y. Cohen, R. Linevsky,D. Rothschild, A.E. Moor, S. Ben-Moshe, A. Harmelin,S. Itzkovitz, N. Maharshak, O. Shibolet, H. Shapiro,M. Pevsner-Fischer, I. Sharon, Z. Halpern, E. Segal,E. Elinav, Personalized gut mucosal colonization resis-tance to empiric probiotics is associated with unique hostand microbiome features, Cell 174 (2018) 1388e1405,https://doi.org/10.1016/j.cell.2018.08.041, e21.

[246] J. Suez, N. Zmora, G. Zilberman-Schapira, U. Mor, M. Dori-Bachash, S. Bashiardes, M. Zur, D. Regev-Lehavi, R. Ben-Zeev Brik, S. Federici, M. Horn, Y. Cohen, A.E. Moor,D. Zeevi, T. Korem, E. Kotler, A. Harmelin, S. Itzkovitz,N. Maharshak, O. Shibolet, M. Pevsner-Fischer,H. Shapiro, I. Sharon, Z. Halpern, E. Segal, E. Elinav,Post-antibiotic gut mucosal microbiome reconstitution isimpaired by probiotics and improved by autologous FMT,Cell 174 (2018) 1406e1423, https://doi.org/10.1016/j.cell.2018.08.047, e16.

[247] W.R. Whitaker, E.S. Shepherd, J.L. Sonnenburg, Tunableexpression tools enable single-cell strain distinction in thegut microbiome, Cell 169 (2017) 538e546, https://doi.org/10.1016/j.cell.2017.03.041, e12.

[248] B. Lim, M. Zimmermann, N.A. Barry, A.L. Goodman,Engineered regulatory systems modulate gene expressionof human commensals in the gut, Cell 169 (2017)547e558, https://doi.org/10.1016/J.CELL.2017.03.045,e15.

[249] S. Zhao, T.D. Lieberman, M. Poyet, K.M. Kauffman,S.M. Gibbons, M. Groussin, R.J. Xavier, E.J. Alm, Adaptiveevolution within gut microbiomes of healthy people, CellHost Microbe 25 (2019) 656e667s, https://doi.org/10.1016/J.CHOM.2019.03.007, e8.