The Neonatal Window of Opportunity: Setting Rupted, Drives Far-Ranging Clinical Consequences

e Journal of Immunology VOLUME 198, NUMBER 2 • JANUARY 15, 2017 • WWW.JIMMUNOL.ORG The Macro Inﬂuence of the Microbiome

am pleased to draw your attention to this, the first ever topical issue of Brief Reviews in The Journal of Immunology. I Through seven Brief Reviews, this issue of The JI highlights the pervasive impact of the gut microbiome on immune functions, both in the gut and at distal sites, and features a cover chosen from a themed competition for this issue. These compiled Brief Reviews underscore our recent appreci- ation that the far-reaching influence of commensal gut microbes on immune function reflects the coevolution of this symbiotic microbial ecosystem and its mammalian hosts. Authors Torow and

Hornef emphasize that the nature of this microbial community in Downloaded from human and mouse early neonates impacts immune homeostasis throughout life (1). Grigg and Sonnenberg explore the reciprocal inﬂuence of the identity and localization of the microbiome components on the balance between inﬂammation and immune homeostasis both within and outside the gut, and point to the

therapeutic potential for diseases of chronic inflammation pro- http://www.jimmunol.org/ vided by manipulation of the microbiota-host interactions (2). Elinav and coauthors describe the microbially derived and microbially modified small metabolites whose bioactivity mediates much of the influence on sterile host tissues described in the ac- companying reviews, in part by modulating nutrient content, digestion, inflammation, and epigenetic control of transcription (3). Recognizing the sweeping impact of lifestyle changes in the Western world, Plunkett and Nagler highlight the influence of the microbiota and environmental exposure to microbial products on by guest on January 26, 2017 the acquisition of food allergies (4), whereas the focus of the Brief Review by Paun, Yau, and Danska is the effect of the resident gut microbiota on the rapid rise in the incidence of type I diabetes Ag-specific CD8 T cells (red) and CXC3R11 APCs (green) in the small among genetically stable populations (5). Colpitts and Kasper intestine (visualized with a DNA stain in blue) 9 d after oral infection document the influence of the gut microbiome on autoimmunity with Ag-expressing Yersinia pseudotuberculosis. Image provided by in the CNS mediated by alterations in the balance of effector and Dr. T. Bergsbaken, University of Washington. suppressor cells along the bidirectional gut–brain axis (6). Landay and coauthors highlight the clinical consequences of microbiome in Washington, DC, May 12–16, 2017, for the focus of next dysbiosis through the documented cross-talk between the gut year’s topical issue of Brief Reviews. microbiome and the immune system during HIV infection, with an eye to some of the recent technical advances that have illu- Pamela J. Fink, Ph.D. minated this field (7). Taken together, these Brief Reviews provide Editor-in-Chief an impressive illustration of the ongoing dialogue between the gut microbiota and the immune system which, when balanced at the References ideal set point, maintains immune homeostasis, and when dis- 1. Torow, N., and M. W. Hornef. 2017. The neonatal window of opportunity: setting rupted, drives far-ranging clinical consequences. the stage for life-long host-microbial interaction and immune homeostasis. We are excited that this collection of topical Brief Reviews on J. Immunol. 557–563. 2. Grigg, J. B., and G. F. Sonnenberg. 2017. Host-microbiota interactions shape local “The Macro Influence of the Microbiome” inaugurates the 101st and systemic inflammatory diseases. J. Immunol. 564–571. year of continuous publication of The Journal of Immunology, 3. Blacher, E., M. Levy, E. Tatirovsky, and E. Elinav. 2017. Microbiome-modulated metabolites at the interface of host immunity. J. Immunol. 572–580. and underscores the breadth of the journal’s scope and the 4. Plunkett, C. H., and C. R. Nagler. 2017. The influence of the microbiome on eclectic research interests of its readership. We hope to make this allergic sensitization to food. J. Immunol. 581–589. topical issue an annual event. Look for an announcement at the 5. Paun, A., C. Yau, and J. S. Danska. 2017. The influence of the microbiome on type 1 diabetes. J. Immunol. 590–595. annual meeting of The American Association of Immunologists 6. Colpitts, S. L., and L. H. Kasper. 2017. Influence of the gut microbiome on autoimmunity in the central nervous system. J. Immunol. 596–604. 7. Liu, J., B. Williams, D. Frank, S. M. Dillon, C. C. Wilson, and A. L. Landay. 2017. Inside out: HIV, the gut microbiome, and the mucosal immune system. Copyright Ó2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00 J. Immunol. 605–614. www.jimmunol.org/cgi/doi/10.4049/jimmunol.1690022 Th eJournal of Brief Reviews Immunology

The Neonatal Window of Opportunity: Setting the Stage for Life-Long Host-Microbial Interaction and Immune Homeostasis Natalia Torow and Mathias W. Hornef The existence of a neonatal window was first high- Approximately three decades ago, researchers noticed dra- lighted by epidemiological studies that revealed the par- matic changes in the interaction of the human host with ticular importance of this early time in life for the pathogenic microorganisms in industrialized countries. The susceptibility to immune-mediated diseases in humans. number of infections decreased very significantly during the Recently, the first animal studies emerged that present second half of the twentieth century as a consequence of examples of early-life exposure–triggered persisting im- improved medical healthcare, including effective vaccine mune events, allowing a detailed analysis of the factors strategies and antibiotics, as well as better hygiene and living Downloaded from that define this particular time period. The enteric standards (1, 2). The concomitant increase in immune- microbiota and the innate and adaptive immune sys- mediated diseases, such as Crohn’s disease and asthma, as well tem represent prime candidates that impact on the path- as diabetes and multiple sclerosis, led to the hygiene hy- ogenesis of immune-mediated diseases and are known pothesis that proposed a causative relationship between the decrease in infectious diseases and the increasing burden of

to reach a lasting homeostatic equilibrium following a http://www.jimmunol.org/ dynamic priming period after birth. In this review, we immune-mediated and allergic diseases. Consistent with this idea, a steady increase in IBD and asthma was subsequently outline the postnatal establishment of the microbiota observed in other geographical areas with previous very low and maturation of the innate and adaptive immune sys- incidence following the implementation of effective health- tem and discuss examples of early-life exposure–triggered care systems and infection-control measures (3, 4). Even more immune-mediated diseases that start to shed light on strikingly, nematode infection, a highly endemic type of in- the critical importance of the early postnatal period fection in geographic areas with low IBD incidence, resulted for life-long immune homeostasis. The Journal of in a signiﬁcant clinical improvement in IBD patients (5).

Immunology, 2017, 198: 557–563. Later, epidemiological studies extended this view to include by guest on January 26, 2017 exposure to environmental microbial constituents, as well as mmune-mediated diseases, such as allergies and inflam- commensal bacteria (6, 7). This was first noted when farm matory bowel disease (IBD), are highly prevalent in western children were compared with their urban counterparts I countries and are associated with significant morbidity. (8). Raw milk consumption and exposure to the livestock- Despite decades of intensive research on the associated func- produced feces within the stable environment with a high tional and structural alterations, the disease etiology and, thus, microbial load and potent immunomodulatory activity were identified as critical factors (9). High endotoxin concentra- the decisive molecular mechanisms underlying disease initiation tions in animal feces and the farm environment were identi- have not been resolved. For example, the search for unknown fied as functional triggers of regulatory mechanisms and pathogenic microorganisms revealed some interesting candidates immune homeostasis, yet other less-well detectable microbial but no uniform causative agent. Also, genome-wide association stimuli or even viable bacteria might contribute to this effect studies identified a large number of susceptibility loci, but many (10, 11). Indeed, the reduction in the prevalence of certain individuals that carry these mutations never experience clinical infectious diseases was paralleled by major alterations in the symptoms. Therefore, the identified factors might enhance dis- enteric microbiota composition (12, 13). Loss of individual ease susceptibility and/or promote the progression and severity bacterial members of a healthy microbiota and a general re- of clinical symptoms rather than play a decisive role during the duction in bacterial diversity were noted and may contribute initiation of the disease. However, only the identification of the to enhanced disease susceptibility (13). Consistently, individual first step in disease pathogenesis will allow us to develop effective members of the microbiota were assigned a specific preventive strategies for future disease prevention. or disease-promoting function in immune modulation. For

Institute of Medical Microbiology, RWTH Aachen University Hospital, 52074 Address correspondence and reprint requests to Prof. Mathias W. Hornef, RWTH Aachen, Germany Aachen University Hospital, Pauwelsstrasse 30, 52074 Aachen, Germany. E-mail address: [email protected] ORCID: 0000-0003-3013-8724 (N.T.). Abbreviations used in this article: DC, dendritic cell; IBD, inﬂammatory bowel disease; Received for publication July 19, 2016. Accepted for publication October 17, 2016. iNKT, invariant NKT; pTreg, peripheral-induced Treg; Treg, regulatory T cell; tTreg, This work was supported by the Deutsche Forschungsgemeinschaft (Ho2236/8-1, Priority thymus-derived regulatory T cell. Program 1656, and Priority Program 1580 [to M.W.H.] and TO 1052/1-1 [to N.T.]). Copyright Ó 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601253 558 BRIEF REVIEWS: MICROBIAL EXPOSURE IN EARLY LIFE example, the segmented filamentous bacterium Faecalibacterium perturbations (30). Subsequently, population dynamics of the prausnitzii and the mucin-degrading bacterium Akkermansia commensal bacterial communities are shaped by newly in- muciniphilia were shown to promote mucosal immune cell troduced bacterial species, the local conditions in the neonatal maturation, immune homeostasis, or a beneficial host meta- intestine, and niche and nutrient availability (31). With the bolism in mice and men (14–16). Yet, fecal transplantation, cessation of breastfeeding and the introduction of solid food, despite impressive clinical benefits in certain patient cohorts, the microbiota composition changes dramatically, and the such as patients with recurrent Clostridium difficile infec- diversity rises, mirroring the increasing complexity of nutrition (17), has failed to provide a uniform and lasting clinical ents. At this stage, predominance of Clostridiales and response. Bacteroidetes that is typical of the adult colonic microbiota in Soon, epidemiological studies identified the particular in- mice is noted in most individuals (32). The diversity of the fluence of age. For example, farm exposure during fetal and adult microbiota is reached in humans after ∼1–3 y and early postnatal life exerted the strongest protective effect (8, correlates with an increase in the overall compositional sta- 18). Also, antibiotic use during pregnancy and early child- bility (27, 33). This compositional stability progressively hood was reported to represent a risk factor for allergic disease generates what is called colonization resistance. Newly intro- development (19). These findings introduced yet another duced bacteria barely gain access to niches in the intestinal model in which exogenous factors present during early de- lumen and nutrient supply and fail to become a permanent velopment could also directly or indirectly influence immune member of this ecosystem (34–36). This strongly enhances homeostasis and disease susceptibility during later life. In- the resistance to many enteric pathogens in adult individuals. deed, the immune system at birth differs significantly from Thus, the lower compositional stability and colonization re- Downloaded from that of adult individuals, and the postnatal period contributes sistance in neonates and young infants may explain the par- significantly to immune development and maturation. Also, ticular susceptibility of the neonate to certain foodborne microbial colonization starts at birth and eventually generates infectious agents (35, 37). It may also allow longer persistence a life-long, relatively stable ecosystem. This early priming was of probiotic bacteria and provide an explanation for the later described as the neonatal window to highlight the ex- stronger effects observed after administration of probiotics http://www.jimmunol.org/ clusive and nonredundant function of the early postnatal and in the human pediatric population (38). Although environ- infant period for life-long immune homeostasis (Fig. 1). A mental factors, such as mode of delivery, nutrition, or anti- number of well-controlled animal studies has begun to un- biotic treatment, significantly alter the human neonatal ravel the mechanisms that underlie the observations from microbiota composition, perturbations generally resolve epidemiological studies (Table I) (20–23). In this article, we within 2 y (29). The first known perturbing factor in life, discuss human and mouse studies that reveal the critical and cesarean section, was associated with an increased risk for nonredundant role of the neonatal window for immune ho- immune-mediated diseases, such as asthma, allergy, and meostasis and its implication for future research. celiac disease, in childhood (39–42). Also, the use of antibiotics was associated with atopic diseases, such as asthma by guest on January 26, 2017 Early microbial exposure and generation of the mature microbiota and atopic dermatitis, in human infants (43, 44). It is under It is generally accepted that the healthy fetus is devoid of debate whether microbiota alterations, such as those detected colonizing viable microorganisms. However, reports on the in cesarean section–delivered children, exert functional conse- PCR- or culture-based detection of bacteria in presumably quences. Only the generation of a disease state after transfer sterile mouse and human fetal or placental tissue samples of the altered microbiota is able to prove this direct func- stimulated discussion on the possible existence of a fetal or tional link. placental microbiome (reviewed in Ref. 24). The common idea is that only with rupture of membranes do viable com- Establishment of the mucosal innate immune defense mensal bacteria start to colonize the newborn’s body surfaces. The mucosal innate immune system undergoes major alter- It is thought that the fetus first encounters bacteria derived ations after birth that might help to define the neonatal from the maternal vaginal microbiota during passage through window. These alterations occur as part of a developmental the birth canal. Thus, the vaginal microbiota represents the program or in response to environmental stimuli. They are first inoculum to colonize the newborn. Of notable difference, particularly dominant in mice that exhibit a very short ges- cesarean section–delivered human neonates first encounter tational period and are born with an immature intestinal bacteria of the skin of the mother or health care worker; thus, epithelium. For example, the newborn small intestinal epi- in this case, the skin microbiota represents the primary source thelium exhibits enhanced permeability to soluble Ags and is (25). High interindividual variation, but low diversity and devoid of small intestinal crypts that only develop at weaning density, characterize this early colonization phase that is and harbor intestinal stem cells and antimicrobial peptide- dominated by bacteria specialized in milk fermentation, such producing Paneth cells (45, 46). The lack of Paneth cell– as Lactobacillus, Streptococcus, and Bifidobacterium, in humans derived antimicrobial peptides may be compensated for by (26, 27). Given the nutrient-rich environment provided by the cathelin-related antimicrobial peptide that is produced breast milk in the neonatal gut lumen, microbial density by murine enterocytes in the neonatal small intestine (47). reaches plateau levels only a few days later (28). In both Also, expression of mucins, building blocks of the intestinal mouse and man, the microbiota at this stage is highly sensitive mucus layer, is low in the neonate and rises at weaning (35). to exogenous perturbations (e.g., formula feeding or antibiotic Finally, epithelial innate immune recognition varies in an administration) that delay the development of a mature di- age-dependent manner. Epithelial expression levels of the pat- verse microbial community (29). Microbiota alterations, in tern recognition receptors Tlr3, Tlr4, and Tlr9 are expressed in turn, render the bacterial ecosystem less resilient to further an age-dependent fashion in mice (48, 49). Also, postnatal The Journal of Immunology 559 Downloaded from

FIGURE 1. The neonatal window of opportunity. Environmental factors (e.g., rural versus urban), nutrients (e.g., breast milk versus formula versus solid food), http://www.jimmunol.org/ and infections (e.g., antibiotic treatment) impact the dynamic development of the postnatal microbiome and the maturation of the innate and adaptive immune system during the neonatal phase. The mechanisms of these interactions remain mostly unresolved. The functional consequences skew the delicate balance toward an increased susceptibility to immune-mediated diseases, such as asthma, IBD, and allergy, far beyond the neonatal phase. exposure to bacterial endotoxin induces a transient state of neonates that explains the increased susceptibility to many innate immune tolerance of the intestinal epithelium that infectious diseases (55). However, we have come to under- helps to prevent an exaggerated Tlr-mediated immune stim- stand that the neonate is well able to mount adult-like im- ulation during early colonization in mice (50, 51). In contrast, mune responses under certain conditions, such as a reduced by guest on January 26, 2017 the structure of the human intestinal epithelial layer is fully pathogen inoculum or increased adjuvant dose (56–58). Also, developed at birth. Functional analyses in the healthy intes- specific features of the neonate mucosal immune system have tinal epithelium are limited by ethical concerns. Nevertheless, been unraveled that suggest a specialized (rather than imma- lower expression of TLR3 and enhanced expression of the ture) nature of the neonate immune system. NF-kB RelA inhibitor IkB-a were demonstrated in the hu- In the murine host, naive TCRab and B lymphocytes are man neonatal epithelium (49, 52). Also, postnatal innate released approximately at birth from the primary lymphoid immune tolerance by induction of the regulatory secretory organs and home to the periphery, including mucosal body leukocyte protease inhibitor was observed in the human upper sites (59). In contrast to the adult host, homing of lymphocytes gastrointestinal tract epithelium (53). Interestingly, an altered to the intestine is independent of microbial cues (60). At the concentration of immunostimulatory endotoxin within the mucosa, lymphocytes encounter myriads of mostly innocuous gut lumen as a consequence of an altered microbiota com- Ags derived from the environment, the rapidly evolving position was reported recently (13). Thus, an impaired bal- microbiota, and diet. Several mechanisms were identified ance between microbial stimulation and tolerance to innate that might influence the interaction of the neonatal immune immune stimuli might promote the prevalence of autoim- system with the postnatal environment. In the murine in- mune diseases in humans (11, 13). A recently reported novel testine, the general cell composition and immune tissue ar- approach analyzing enterocyte transcription in fecal material chitecturearemarkedlydifferentfromtheadultstate(60). of term and preterm human neonates by RNA-sequencing TCRab+ and B cells are localized exclusively to the Peyer’s identified differential expression, primarily in genes involved patches and exhibit a naive phenotype until weaning under in lipid metabolism and immunity (54). This approach also steady-state conditions (60). This is in stark contrast to the might help to examine age-dependent changes in cell differ- situation in the adult tissue, wherein microbiota and food Ags entiation and correlate mucosal immune signaling with bac- engage the adaptive immune system in a continuous interplay, terial colonization in the future. resulting in the presence of Ag-experienced lymphocytes in Peyer’s patches, lamina propria, and the intraepithelial space Maturation of the adaptive mucosal immune system during the at steady-state (60–63). In humans, the thymus produces postnatal period functional lymphocytes at the beginning of the second tri- Similarly, the adaptive immune system in the neonate host mester, and lymphocytes are found in all compartments of differs significantly from the adult situation in mice and men. the intestine at birth (64, 65). However, lymphocyte mat- This was long believed to represent a state of immaturity in uration is also delayed in human infants and, thus, may 560 BRIEF REVIEWS: MICROBIAL EXPOSURE IN EARLY LIFE

underlie similar mechanisms (66, 67). Ethical concerns limit the access to intestinal tissue samples from healthy

Refs. human neonates and have restricted in-depth functional (75) analyses. Homing of lymphocytes to the periphery is accompanied by a concomitant release of thymus-derived regulatory T cells (tTregs) at the fetal and neonatal stage in mice and men, respectively (68, 69). Human fetal lymph nodes contain Tregs at higher densities compared with their adult coun- AHR (74) Allergy (22)

Psoriasis (23) terparts and are essential to maintain self-tolerance (69). In Food allergy (85) AHR, colitis (20, 21) the murine host, tTregs are required to mediate life-long Disease susceptibility immunity to self-antigens and contribute to the general

Certain types of autoimmunity (68, 94) state of tolerance toward the enteric microbiota (68, 70). They suppress the homeostatic activation of naive T cells within the Peyer’s patches during the early postnatal period that is characterized by the establishment of the microbiota in the murine host (60). With weaning approximately 3 wk after birth, a second population of Tregs, Downloaded from peripheral-induced Tregs (pTregs), is observed that provides

-mediated diseases an additional regulatory control of the host-microbial inter- suppression (lung and derived sphingolipids; and DCs play in the intestine (71). Food Ags represent the major

skin commensals inducer for pTregs in the small intestine, whereas microbiota- Cxcl16 neonatal PD1+ Tregs Mechanism in the adult Lack of oral tolerance Allergy (46) expansion of iNKT cells and adaptive lymphocytes B. fragilis derived Ags prime pTregs in the colon (72). Whether pTregs in Peyer’s patches; hyper IgE

IL-22 production; barrier defect are already present among the gut Treg population in the http://www.jimmunol.org/ gut); nrae ILC3Increased and myeloid cells numbers Increased barrier permeability (82) human neonate is unclear (73). Also, in the murine lung, Helios+ tTregs dominate during the early phase of the host- environmental interaction that is prone to the development of an airway hyperresponsiveness. These cells expand greatly during the first week of life dependent on microbial exposure. Only after the second week of life do programmed death g ligand-1–dependent pTregs develop that efficiently control the susceptibility of the airways to environmental stimuli and inappropriate responses (74). Also, in the skin, significant by guest on January 26, 2017 ? T cell-dependent class switch of B cells expansion and activation of (presumably) tTregs are observed around the second week of life, promoting tolerance to commensal bacteria that colonize the dermal tissue (75). In- function; lack of IFN ILC3 and myeloid cells Mechanism in the neonate terestingly, the increase in Tregs in the murine skin coincides with the morphogenesis of hair follicles, which provide a reservoir of commensal bacteria (76).

trancriptional changes in intestinal tissue, increased An additional mechanism of tolerance induction acts transfer of bacterial constituents from dam to neonate; during fetal and postnatal development in mice and men: the tolerization against noninherited alloantigens and other Ags present in the maternal organism. During fetal development, this takes place by transfer of maternal cells via the placenta (77, 78). Interestingly, maternal-derived Ags can also be transferred postnatally by breastfeeding and lead to tolerance in transplantation, as well as allergy models (79, 80). IgA-producing plasma cells are only detected after weaning

monocolonization vs. SPF ? Lack of in mice (60–63). However, maternal Igs transmitted trans- placentally and by breast milk facilitate high concentrations Exposure modiﬁcations

complex Schaedler flora in serum and at mucosal body sites. Of note, the mother also B. fragilis represents the most important source of the microbiota by transferring the newborn’s first inoculum at birth and providing a constant supply of viable microorganisms in breast milk (25, 81). Although the specificity of the mucosal Abs is ill defined, it may allow the simultaneous transmission of commensal bacteria with their corresponding Igs. This could restrict bacterial penetration of the neonatal mucosa and Age compared may explain why maternal Igs contribute significantly to Birth vs. adultPostnatal vs. adultPrenatal Abx vs. Reversible SPF gestational colonization SPF vs. only GF Transfer of bacterial constituents across the placenta; Generation of tTreg in the neonatal thymus Enhanced ? suppressory capacity of Enhanced IL-22 production by innate 7d vs. adult SPF+Bacteroides vs. SPF Expansion and activation of Treg Neonatal Treg-mediated tolerance to 3d, 15d, 60d GF vs. SPF PD1-PD1L interaction between pTreg 2w Abx vs. SPF ? Suppression of lymphocyte mediated Birth vs. adult Abx/GF/ Birth vs. adultBirth vs. adult1w, 2w, and 3w GF/simple(r) Schaedler flora vs. more GF vs. monocolonized SPF only Vitamin A deficiency; impaired ? DC Lack of oral tolerance Allergy (88, 89)

Abx, antibiotic; AHR, airway hyperresponsiveness; GF, germ-free; ILC3, type 3control innate lymphoid cell; PD1, programmed cell death protein 1; SPF, speciﬁc pathogen–free. T and B cell activation in the neonatal intestine (60,

Table I. Experimental studies that demonstrate the influence of microbial exposure during the fetal/neonatal period on the susceptibility to immune 61). Maternal microbiota-induced Igs might thereby help to The Journal of Immunology 561 cope with the presence of proinflammatory microbial con- not been investigated (86, 87). Similarly, antibiotic admin- stituents and prevent inflammatory responses (82). Particu- istration during the postnatal period rendered mice more larly the transfer of microbiota-specific IgG2b and IgG3 susceptible in a model of imiquimod-induced psoriasis (23); together with IgA may contribute to the inhibition of mucosal however, IL-22 plays an adverse role in this model. A dis- T cell responses during the postnatal period in mice (83). turbed microbiota with enhanced IL-22 and IL-23 levels was identified to contribute to the more severe disease outcome. The impact of early-life exposure on later-life health and disease This may be explained by a distorted induction of immune susceptibility tolerance to skin commensal bacteria, presumably by thymus- The important role of microbiota-derived factors, particularly derived Tregs, by antibiotic treatment during the postnatal in immune development and maturation, is best illustrated in period (75). The effect would be expected to be age specific. animals raised in germ-free environments (84). Germ-free Indeed, antibiotic treatment of adult animals reduced the mice exhibit a number of significant differences in their im- expression levels of IL-17 and disease severity after imiquimod mune system, including the cellular composition and matu- administration (23). Whether similar age-specific mechanisms ration status at mucosal sites. Interestingly, the phenotype of are active at other body surfaces remains to be investigated. germ-free mice is only partially rescued by bacterial coloni- In addition to microbial factors, dietary factors contribute zation after weaning, demonstrating the critical role of early- significantly to the mucosal immune response in the neonate. life exposure for life-long immune homeostasis (Table I). Neonatal mice are more susceptible to allergic sensitization, and their ability to establish oral tolerance increases during the

Among these distinct features is the excessive homing of in- Downloaded from variant NK T (iNKT) cells to the lung and colon (but not first 3 wk of life (46, 74). The ability to induce oral tolerance small intestinal) tissue in germ-free animals. iNKT cells ex- (to food and microbiota-derived Ag) critically depends on the pand in gut and lung tissue of germ-free mice in response to presence of microbial stimuli during the postnatal phase (88, Cxcl16. The presence of the microbiota prior to weaning (but 89). A recent study also identified endogenously low levels of not thereafter) suppresses Cxcl16 expression by epigenetic vitamin A in the breast milk as a causative mechanism for impaired oral tolerance acquisition in neonates. Low dietary regulation, explaining the high iNKT cell count in germ-free http://www.jimmunol.org/ animals (20). An alternative Cxcl16-independent mechanism retinol was associated with enhanced para- and transepithelial to suppress the expansion of iNKT cells was identified after permeability, as well as reduced expression of RALDH by + monocolonization of neonates with the commensal Bacteroides CD103 dendritic cells (DCs) and decreased IFN-g induction + fragilis. B. fragilis glycosphingolipids that represent cognate in CD4 T cells (46). Vitamin A supplementation, in turn, Ags of iNKT cells are also able to suppress the expansion of enhanced epithelial barrier function, RALDH-expressing + iNKT cells in the gut (21). Importantly, enhanced iNKT cell CD103 DCs, and T cell proliferation and facilitated oral numbers confer increased susceptibility to mucosal inflam- tolerance induction shortly after birth. Several studies are un- mation in lung and colon tissue, thus linking neonatal mi- derway to examine the effect of early vitamin A supplementa- crobial exposure with the susceptibility to inflammatory tion in human children; a report from Ghana did not provide by guest on January 26, 2017 diseases in later life (20, 21). Another example is the elevated evidence for improved survival (90). systemic IgE levels found in germ-free animals. In the absence Most studies compared germ-free mice colonized at birth of microbiota, IgE class switch takes place in a T cell– and versus those colonized as adults. However, to understand the IL-4–dependent manner within Peyer’s patches. This phe- processes that take place within the neonatal window, it is notype can be reversed, and an atopic (anaphylactic) reaction critical to define its precise time frame. In fact, even fetal ex- can be prevented only if the mice are colonized with a rela- posure appears to exert a major influence on immune matura- tively diverse microbiota at birth (22). tion (82). Using a reversible colonization model, it was recently Administration of antibiotics transiently reduces the mi- shown that even in utero microbial constituents, such as aryl crobial load and alters the bacterial composition. Although still hydrocarbon ligands, can be detected and induce major tran- poorly understood, the developing microbiota in human ne- scriptional changes in the fetal gut and enhance the cellularity onates and infants may be particularly sensitive to antibiotic- of the innate immune system (82). The underlying functional induced disturbance, with potential lasting consequences (22, mechanisms through which microbial cues are conveyed to the 29). Antibiotic exposure during pregnancy or the first year of neonatal immune system remain largely undefined. Reversible life in humans slightly, but significantly, enhanced the risk for colonization at different time points during the postnatal phase childhood asthma and atopic dermatitis (43). Similarly, an- could provide useful insights into the actual critical age that tibiotic administration to mice prior to weaning (at 2 wk defines the neonatal window. of age) led to significant microbiota alterations and enhanced the sensitization to food allergens (85). Administration of Conclusions spore-forming clostridial species was sufficient to reverse the Environmental stimuli and microbial exposure during the early phenotype and protect from sensitization. In addition to en- postnatal period exert a major influence on life-long immune hanced production of IgA and Tregs, clostridial species homeostasis and disease susceptibility (Fig. 1). The first mi- stimulated innate lymphoid cell– and adaptive lymphocyte- crobial encounter appears to set the stage for the subsequent derived IL-22. In turn, IL-22 tightened the epithelial barrier long-term host-microbial relationship. Over the course of and reduced the penetration of sensitizing luminal Ags to evolution, this relationship has been strongly interwoven with inductive lymphoid organs, possibly by the induction of an- infectious diseases that represented the most important cause tibacterial C-type lectins (85). Although the critical role of of death. Only with the late nineteenth century did improved IL-22 in epithelial regeneration and antibacterial barrier sanitary conditions, such as drinking water supply, wastewater function is well described, its effect on Ag translocation has treatment, housing conditions, and, later, effective vaccine 562 BRIEF REVIEWS: MICROBIAL EXPOSURE IN EARLY LIFE strategies and antibiotic drugs, provide successful preventive Study Group. 2016. Variation in microbiome LPS immunogenicity contributes to autoimmunity in humans. [Published erratum appears in 2016 Cell 165: 1551.] Cell measures against infectious diseases. As a consequence, life 165: 842–853. expectancy increased dramatically, and infectious diseases now 14. Talham, G. L., H. Q. Jiang, N. A. Bos, and J. J. Cebra. 1999. Segmented filamentous bacteria are potent stimuli of a physiologically normal state of the murine rank far below cardiovascular diseases and cancer in developed gut mucosal immune system. Infect. Immun. 67: 1992–2000. countries. With the increasing incidence of immune-mediated 15. Everard, A., C. Belzer, L. Geurts, J. P. Ouwerkerk, C. Druart, L. B. Bindels, diseases, we may now experience the downside of our success Y. Guiot, M. Derrien, G. G. Muccioli, N. M. Delzenne, et al. 2013. Cross-talk between Akkermansia muciniphila and intestinal epithelium controls diet-induced in preventing the infectious disease burden (13). Therefore, obesity. Proc. Natl. Acad. Sci. USA 110: 9066–9071. we need to address the question of how to compensate for the 16. Sokol, H., B. Pigneur, L. Watterlot, O. Lakhdari, L. G. Bermu´dez-Humarań, J. J. Gratadoux, S. Blugeon, C. Bridonneau, J. P. Furet, G. Corthier, et al. 2008. lack of microbial exposure during early development without Faecalibacterium prausnitzii is an anti-inflammatory commensal bacterium identi- the risk for potential life-threatening infections. Clearly, we fied by gut microbiota analysis of Crohn disease patients. Proc. Natl. Acad. Sci. USA do not want to sacrifice our achievements in infection control 105: 16731–16736. 17. Fischer, M., D. Kao, C. Kelly, A. Kuchipudi, S. M. Jafri, M. Blumenkehl, D. Rex, and prevention (13). Nevertheless, we may have overshot the M. Mellow, N. Kaur, H. Sokol, et al. 2016. Fecal microbiota transplantation is safe mark with our attempts to reduce the risk of pathogen ex- and efficacious for recurrent or refractory Clostridium difficile infection in patients with inflammatory bowel disease. Inflamm. Bowel Dis. 22: 2402–2409. posure. Therefore, we should seek a more moderate attitude 18. Riedler, J., C. Braun-Fahrla¨nder, W. Eder, M. Schreuer, M. Waser, S. Maisch, D. that facilitates exposure of our children to benign nonpatho- Carr, R. Schierl, D. Nowak, and E. von Mutius; ALEX Study Team. 2001. Ex- posure to farming in early life and development of asthma and allergy: a cross- genic microorganisms (91). The first attempts, such as exposure sectional survey. Lancet 358: 1129–1133. to maternal vaginal fluid after caesarean section or the ad- 19. Risnes, K. R., K. Belanger, W. Murk, and M. B. Bracken. 2011. Antibiotic exposure ministration of probiotic bacteria to preterm neonates, are by 6 months and asthma and allergy at 6 years: findings in a cohort of 1,401 US children. [Published errata appear in 2011 Am. J. Epidemiol. 173: 846; 2011 Am. Downloaded from being made (25, 92, 93). An improved understanding of the J. Epidemiol. 173: 1475.] Am. J. Epidemiol. 173: 310–318. underlying molecular mechanisms and the identification of 20. Olszak, T., D. An, S. Zeissig, M. P. Vera, J. Richter, A. Franke, J. N. Glickman, R. Siebert, R. M. Baron, D. L. Kasper, and R. S. Blumberg. 2012. Microbial ex- strategies to reestablish a healthy host-microbial relationship posure during early life has persistent effects on natural killer T cell function. Science need further attention. Although recent studies provided first 336: 489–493. 21.An,D.,S.F.Oh,T.Olszak,J.F.Neves,F.Y.Avci,D.Erturk-Hasdemir,X.Lu, insights into mechanisms that lead to disease in the adult after S. Zeissig, R. S. Blumberg, and D. L. Kasper. 2014. Sphingolipids from a symbiotic inappropriate exposure during the postnatal phase, we still microbe regulate homeostasis of host intestinal natural killer T cells. Cell 156: 123–133. http://www.jimmunol.org/ lack understanding of what actually happens during the neo- 22. Cahenzli, J., Y. Ko¨ller, M. Wyss, M. B. Geuking, and K. D. McCoy. 2013. In- testinal microbial diversity during early-life colonization shapes long-term IgE levels. natal window itself. We need to address how a distorted Cell Host Microbe 14: 559–570. microbiota composition may persist and how microbial cues 23. Zanvit, P., J. E. Konkel, X. Jiao, S. Kasagi, D. Zhang, R. Wu, C. Chia, N. J. Ajami, D. P. Smith, J. F. Petrosino, et al. 2015. Antibiotics in neonatal life increase murine are transferred to the developing immune system in the ne- susceptibility to experimental psoriasis. Nat. Commun. 6: 8424. onate to confer disease resistance or susceptibility (94). 24. Tamburini, S., N. Shen, H. C. Wu, and J. C. Clemente. 2016. The microbiome in early life: implications for health outcomes. Nat. Med. 22: 713–722. 25. Dominguez-Bello, M. G., E. K. Costello, M. Contreras, M. Magris, G. Hidalgo, N. Fierer, and R. Knight. 2010. Delivery mode shapes the acquisition and structure Disclosures of the initial microbiota across multiple body habitats in newborns. Proc. Natl. Acad. The authors have no financial conflicts of interest. Sci. USA 107: 11971–11975. by guest on January 26, 2017 26. Del Chierico, F., P. Vernocchi, A. Petrucca, P. Paci, S. Fuentes, G. Pratico`, G. Capuani, A. Masotti, S. Reddel, A. Russo, et al. 2015. Phylogenetic and met- References abolic tracking of gut microbiota during perinatal development. PLoS One 10: 1. Bach, J. F. 2002. The effect of infections on susceptibility to autoimmune and e0137347. allergic diseases. N. Engl. J. Med. 347: 911–920. 27. Yatsunenko, T., F. E. Rey, M. J. Manary, I. Trehan, M. G. Dominguez-Bello, 2. Strachan, D. P. 1989. Hay fever, hygiene, and household size. BMJ 299: 1259– M. Contreras, M. Magris, G. Hidalgo, R. N. Baldassano, A. P. Anokhin, et al. 2012. 1260. Human gut microbiome viewed across age and geography. Nature 486: 222–227. 3. Ananthakrishnan, A. N. 2015. Epidemiology and risk factors for IBD. Nat. Rev. 28. Palmer, C., E. M. Bik, D. B. DiGiulio, D. A. Relman, and P. O. Brown. 2007. Gastroenterol. Hepatol. 12: 205–217. Development of the human infant intestinal microbiota. PLoS Biol. 5: e177. 4. Asher, I., and N. Pearce. 2014. Global burden of asthma among children. Int. 29. Bokulich, N. A., J. Chung, T. Battaglia, N. Henderson, M. Jay, H. Li, A. D Lieber, J. Tuberc. Lung Dis. 18: 1269–1278. F. Wu, G. I. Perez-Perez, Y. Chen, et al. 2016. Antibiotics, birth mode, and diet 5. Weinstock, J. V. 2012. Autoimmunity: the worm returns. Nature 491: 183–185. shape microbiome maturation during early life. Sci. Transl. Med. 8: 343ra82. 6. Braun-Fahrla¨nder, C., J. Riedler, U. Herz, W. Eder, M. Waser, L. Grize, S. Maisch, 30. Nobel,Y.R.,L.M.Cox,F.F.Kirigin,N.A.Bokulich,S.Yamanishi,I.Teitler,J.Chung, D. Carr, F. Gerlach, A. Bufe, et al; Allergy and Endotoxin Study Team. 2002. J. Sohn, C. M. Barber, D. S. Goldfarb, et al. 2015. Metabolic and metagenomic out- Environmental exposure to endotoxin and its relation to asthma in school-age comes from early-life pulsed antibiotic treatment. Nat. Commun. 6: 7486. children. N. Engl. J. Med. 347: 869–877. 31. Yassour, M., T. Vatanen, H. Siljander, A. M. Ha¨ma¨laïnen, T. Ha¨rko¨nen, 7. Conrad, M. L., R. Ferstl, R. Teich, S. Brand, N. Blumer,€ A. O. Yildirim, S. J. Ryha¨nen, E. A. Franzosa, H. Vlamakis, C. Huttenhower, D. Gevers, et al; C. C. Patrascan, A. Hanuszkiewicz, S. Akira, H. Wagner, et al. 2009. Maternal TLR DIABIMMUNE Study Group. 2016. Natural history of the infant gut microbiome signaling is required for prenatal asthma protection by the nonpathogenic microbe and impact of antibiotic treatment on bacterial strain diversity and stability. Sci. Acinetobacter lwoffii F78. J. Exp. Med. 206: 2869–2877. Transl. Med. 8: 343ra81. 8. Eriksson, J., L. Ekerljung, J. Lo¨tvall, T. Pullerits, G. Wennergren, E. Ro¨nmark, 32. Pantoja-Feliciano, I. G., J. C. Clemente, E. K. Costello, M. E. Perez, M. J. Blaser, K. Toreń, and B. Lundba¨ck. 2010. Growing up on a farm leads to lifelong pro- R. Knight, and M. G. Dominguez-Bello. 2013. Biphasic assembly of the murine tection against allergic rhinitis. Allergy 65: 1397–1403. intestinal microbiota during early development. ISME J. 7: 1112–1115. 9. Peters, M., M. Kauth, J. Schwarze, C. Ko¨rner-Rettberg, J. Riedler, D. Nowak, 33. de Meij, T. G., A. E. Budding, E. F. de Groot, F. M. Jansen, C. M. Frank C. Braun-Fahrla¨nder, E. von Mutius, A. Bufe, and O. Holst. 2006. Inhalation of Kneepkens, M. A. Benninga, J. Penders, A. A. van Bodegraven, and P. H. Savelkoul. stable dust extract prevents allergen induced airway inflammation and hyper- 2016. Composition and stability of intestinal microbiota of healthy children within responsiveness. Thorax 61: 134–139. a Dutch population. FASEB J. 30: 1512–1522. 10. Waser, M., R. Schierl, E. von Mutius, S. Maisch, D. Carr, J. Riedler, W. Eder, 34. Barthel, M., S. Hapfelmeier, L. Quintanilla-Martıńez, M. Kremer, M. Rohde, M. Schreuer, D. Nowak, and C. Braun-Fahrlander, ALEX Study Team. 2004. M. Hogardt, K. Pfeffer, H. Russmann,€ and W. D. Hardt. 2003. Pretreatment of mice Determinants of endotoxin levels in living environments of farmers’ children and with streptomycin provides a Salmonella enterica serovar typhimurium colitis model their peers from rural areas. Clin. Exp. Allergy 34: 389–397. that allows analysis of both pathogen and host. Infect. Immun. 71: 2839–2858. 11. Schuijs, M. J., M. A. Willart, K. Vergote, D. Gras, K. Deswarte, M. J. Ege, 35. Zhang, K., A. Dupont, N. Torow, F. Gohde, S. Leschner, S. Lienenklaus, S. Weiss, F. B. Madeira, R. Beyaert, G. van Loo, F. Bracher, et al. 2015. Farm dust and M. M. Brinkmann, M. Kuhnel,€ M. Hensel, et al. 2014. Age-dependent enterocyte endotoxin protect against allergy through A20 induction in lung epithelial cells. invasion and microcolony formation by Salmonella. [Published erratum appears in Science 349: 1106–1110. 2014 PLos Pathog. 10: e1004476.] PLoS Pathog. 10: e1004385. 12. Tyakht, A. V., E. S. Kostryukova, A. S. Popenko, M. S. Belenikin, A. V. Pavlenko, 36. Buffie, C. G., and E. G. Pamer. 2013. Microbiota-mediated colonization resistance A. K. Larin, I. Y. Karpova, O. V. Selezneva, T. A. Semashko, E. A. Ospanova, et al. against intestinal pathogens. Nat. Rev. Immunol. 13: 790–801. 2013. Human gut microbiota community structures in urban and rural populations 37. Dupont, A., F. Sommer, K. Zhang, U. Repnik, M. Basic, A. Bleich, M. Kuhnel,€ in Russia. Nat. Commun. 4: 2469. F. Ba¨ckhed, Y. Litvak, M. Fulde, et al. 2016. Age-dependent susceptibility to en- 13. Vatanen, T., A. D. Kostic, E. d’Hennezel, H. Siljander, E. A. Franzosa, M. Yassour, teropathogenic Escherichia coli (EPEC) infection in mice. PLoS Pathog. 12: R. Kolde, H. Vlamakis, T. D. Arthur, A. M. Ha¨ma¨laïnen, et al; DIABIMMUNE e1005616. The Journal of Immunology 563

38. Kristensen, N. B., T. Bryrup, K. H. Allin, T. Nielsen, T. H. Hansen, and O. Pedersen. 68. Yang, S., N. Fujikado, D. Kolodin, C. Benoist, and D. Mathis. 2015. Immune 2016. Alterations in fecal microbiota composition by probiotic supplementation in tolerance. regulatory T cells generated early in life play a distinct role in maintaining healthy adults: a systematic review of randomized controlled trials. Genome Med. 8: 52. self-tolerance. Science 348: 589–594. 39. Thavagnanam, S., J. Fleming, A. Bromley, M. D. Shields, and C. R. Cardwell. 69. Michae¨lsson, J., J. E. Mold, J. M. McCune, and D. F. Nixon. 2006. Regulation of 2008. A meta-analysis of the association between caesarean section and childhood T cell responses in the developing human fetus. J. Immunol. 176: 5741–5748. asthma. Clin. Exp. Allergy 38: 629–633. 70. Cebula, A., M. Seweryn, G. A. Rempala, S. S. Pabla, R. A. McIndoe, T. L. Denning, 40. Pistiner, M., D. R. Gold, H. Abdulkerim, E. Hoffman, and J. C. Celedoń. 2008. L. Bry, P. Kraj, P. Kisielow, and L. Ignatowicz. 2013. Thymus-derived regulatory Birth by cesarean section, allergic rhinitis, and allergic sensitization among children T cells contribute to tolerance to commensal microbiota. Nature 497: 258–262. with a parental history of atopy. J. Allergy Clin. Immunol. 122: 274–279. 71. Weiss, J. M., A. M. Bilate, M. Gobert, Y. Ding, M. A. Curotto de Lafaille, C. N. 41. Sevelsted, A., J. Stokholm, K. Bønnelykke, and H. Bisgaard. 2015. Cesarean section Parkhurst, H. Xiong, J. Dolpady, A. B. Frey, M. G. Ruocco, et al. 2012. Neuropilin and chronic immune disorders. Pediatrics 135: e92–e98. 1 is expressed on thymus-derived natural regulatory T cells, but not mucosa- 42. Decker, E., G. Engelmann, A. Findeisen, P. Gerner, M. Laass, D. Ney, C. Posovszky, generated induced Foxp3+ T reg cells. J. Exp. Med. 209: 1723–1742, S1. L. Hoy, and M. W. Hornef. 2010. Cesarean delivery is associated with celiac disease 72. Kim, K. S., S. W. Hong, D. Han, J. Yi, J. Jung, B. G. Yang, J. Y. Lee, M. Lee, and but not inflammatory bowel disease in children. Pediatrics 125: e1433–e1440. C. D. Surh. 2016. Dietary antigens limit mucosal immunity by inducing regulatory 43.Murk,W.,K.R.Risnes,andM.B.Bracken.2011.Prenatalorearly-lifeexposureto T cells in the small intestine. Science 351: 858–863. antibiotics and risk of childhood asthma: a systematic review. Pediatrics 127: 1125–1138. 73. Weitkamp, J. H., E. Rudzinski, T. Koyama, H. Correa, P. Matta, B. Alberty, and 44. Flohr, C., D. Pascoe, and H. C. Williams. 2005. Atopic dermatitis and the ‘hygiene D. B. Polk. 2009. Ontogeny of FOXP3(+) regulatory T cells in the postnatal human hypothesis’: too clean to be true? Br. J. Dermatol. 152: 202–216. small intestinal and large intestinal lamina propria. Pediatr. Dev. Pathol. 12: 443–449. € 45. Muncan, V., J. Heijmans, S. D. Krasinski, N. V. Buller, M. E. Wildenberg, S. Meisner, 74. Gollwitzer, E. S., S. Saglani, A. Trompette, K. Yadava, R. Sherburn, K. D. McCoy, M. Radonjic, K. A. Stapleton, W. H. Lamers, I. Biemond, et al. 2011. Blimp1 reg- L. P. Nicod, C. M. Lloyd, and B. J. Marsland. 2014. Lung microbiota promotes ulates the transition of neonatal to adult intestinal epithelium. Nat. Commun. 2: 452. tolerance to allergens in neonates via PD-L1. Nat. Med. 20: 642–647. 46. Turfkruyer, M., A. Rekima, P. Macchiaverni, L. Le Bourhis, V. Muncan, G. R. van den 75. Scharschmidt, T. C., K. S. Vasquez, H. A. Truong, S. V. Gearty, M. L. Pauli, Brink, M. K. Tulic, and V. Verhasselt. 2016. Oral tolerance is inefficient in neonatal A. Nosbaum, I. K. Gratz, M. Otto, J. J. Moon, J. Liese, et al. 2015. A wave of mice due to a physiological vitamin A deficiency. Mucosal Immunol. 9: 479–491. regulatory T cells into neonatal skin mediates tolerance to commensal microbes. € 47. Meńard, S., V. Fo¨rster, M. Lotz, D. Gutle,C.U.Duerr,R.L.Gallo,B.Henriques- Immunity 43: 1011–1021. € Normark, K. Putsep, M. Andersson, E. O. Glocker, and M. W. Hornef. 2008. Devel- 76. Lange-Asschenfeldt, B., D. Marenbach, C. Lang, A. Patzelt, M. Ulrich, Downloaded from opmental switch of intestinal antimicrobial peptide expression. J. Exp. Med. 205: 183–193. A. Maltusch, D. Terhorst, E. Stockfleth, W. Sterry, and J. Lademann. 2011. Dis- 48. Gribar, S. C., C. P. Sodhi, W. M. Richardson, R. J. Anand, G. K. Gittes, tribution of bacteria in the epidermal layers and hair follicles of the human skin. M. F. Branca, A. Jakub, X. H. Shi, S. Shah, J. A. Ozolek, and D. J. Hackam. 2009. Skin Pharmacol. Physiol. 24: 305–311. Reciprocal expression and signaling of TLR4 and TLR9 in the pathogenesis and 77. Billingham, R. E., L. Brent, and P. B. Medawar. 1956. The antigenic stimulus in treatment of necrotizing enterocolitis. J. Immunol. 182: 636–646. transplantation immunity. Nature 178: 514–519. 49. Pott, J., S. Stockinger, N. Torow, A. Smoczek, C. Lindner, G. McInerney, 78. Mold, J. E., J. Michae¨lsson, T. D. Burt, M. O. Muench, K. P. Beckerman, F. Ba¨ckhed, U. Baumann, O. Pabst, A. Bleich, and M. W. Hornef. 2012. Age- M. P. Busch, T. H. Lee, D. F. Nixon, and J. M. McCune. 2008. Maternal allo- dependent TLR3 expression of the intestinal epithelium contributes to rotavirus antigens promote the development of tolerogenic fetal regulatory T cells in utero. http://www.jimmunol.org/ susceptibility. PLoS Pathog. 8: e1002670. Science 322: 1562–1565. € ´ 50. Lotz, M., D. Gutle, S. Walther, S. Menard, C. Bogdan, and M. W. Hornef. 2006. 79. Aoyama, K., M. Koyama, K. Matsuoka, D. Hashimoto, T. Ichinohe, M. Harada, Postnatal acquisition of endotoxin tolerance in intestinal epithelial cells. J. Exp. Med. K. Akashi, M. Tanimoto, and T. Teshima. 2009. Improved outcome of allogeneic 203: 973–984. bone marrow transplantation due to breastfeeding-induced tolerance to maternal 51. Chassin, C., M. Kocur, J. Pott, C. U. Duerr, D. Gutle,€ M. Lotz, and M. W. Hornef. antigens. Blood 113: 1829–1833. 2010. miR-146a mediates protective innate immune tolerance in the neonate intestine. 80. Verhasselt, V., V. Milcent, J. Cazareth, A. Kanda, S. Fleury, D. Dombrowicz, Cell Host Microbe 8: 358–368. N. Glaichenhaus, and V. Julia. 2008. Breast milk-mediated transfer of an antigen 52.Claud,E.C.,L.Lu,P.M.Anton,T.Savidge,W.A.Walker,andB.J.Cherayil.2004. induces tolerance and protection from allergic asthma. Nat. Med. 14: 170–175. Developmentally regulated IkappaB expression in intestinal epithelium and suscepti- 81. Heikkila¨, M. P., and P. E. Saris. 2003. Inhibition of Staphylococcus aureus by the bility to flagellin-induced inflammation. Proc.Natl.Acad.Sci.USA101: 7404–7408. commensal bacteria of human milk. J. Appl. Microbiol. 95: 471–478. 53. Menckeberg, C. L., J. Hol, Y. Simons-Oosterhuis, H. R. Raatgeep, L. F. de Ruiter, 82. Gomez de Aguero,€ M., S. C. Ganal-Vonarburg, T. Fuhrer, S. Rupp, Y. Uchimura, D. J. Lindenbergh-Kortleve, A. M. Korteland-van Male, S. El Aidy, P. P. van

H.Li,A.Steinert,M.Heikenwalder,S.Hapfelmeier,U.Sauer,etal.2016.Thematernal by guest on January 26, 2017 Lierop, M. Kleerebezem, et al. 2015. Human buccal epithelium acquires microbial microbiota drives early postnatal innate immune development. Science 351: 1296–1302. hyporesponsiveness at birth, a role for secretory leukocyte protease inhibitor. Gut 83. Koch, M. A., G. L. Reiner, K. A. Lugo, L. S. Kreuk, A. G. Stanbery, E. Ansaldo, 64: 884–893. T. D. Seher, W. B. Ludington, and G. M. Barton. 2016. Maternal IgG and IgA 54. Knight, J. M., L. A. Davidson, D. Herman, C. R. Martin, J. S. Goldsby, I. V. Ivanov, S. M. Donovan, and R. S. Chapkin. 2014. Non-invasive analysis of intestinal devel- antibodies dampen mucosal T helper cell responses in early life. Cell 165: 827–841. opment in preterm and term infants using RNA-sequencing. Sci. Rep. 4: 5453. 84. Hooper, L. V., D. R. Littman, and A. J. Macpherson. 2012. Interactions between 55. Lawn, J. E., S. Cousens, and J. Zupan, Lancet Neonatal Survival Steering Team. the microbiota and the immune system. Science 336: 1268–1273. 2005. 4 million neonatal deaths: when? where? why? Lancet 365: 891–900. 85. Stefka,A.T.,T.Feehley,P.Tripathi,J.Qiu,K.McCoy,S.K.Mazmanian,M.Y.Tjota, 56. Sarzotti, M., D. S. Robbins, and P. M. Hoffman. 1996. Induction of protective G. Y. Seo, S. Cao, B. R. Theriault, et al. 2014. Commensal bacteria protect against food CTL responses in newborn mice by a murine retrovirus. Science 271: 1726–1728. allergen sensitization. Proc. Natl. Acad. Sci. USA 111: 13145–13150. 57. Ridge, J. P., E. J. Fuchs, and P. Matzinger. 1996. Neonatal tolerance revisited: 86. Lindemans, C. A., M. Calafiore, A. M. Mertelsmann, M. H. O’Connor, turning on newborn T cells with dendritic cells. Science 271: 1723–1726. J. A. Dudakov, R. R. Jenq, E. Velardi, L. F. Young, O. M. Smith, G. Lawrence, 58. Forsthuber, T., H. C. Yip, and P. V. Lehmann. 1996. Induction of TH1 and TH2 et al. 2015. Interleukin-22 promotes intestinal-stem-cell-mediated epithelial re- immunity in neonatal mice. Science 271: 1728–1730. generation. Nature 528: 560–564. 59. Garcia, A. M., S. A. Fadel, S. Cao, and M. Sarzotti. 2000. T cell immunity in 87. Giacomin, P. R., R. H. Moy, M. Noti, L. C. Osborne, M. C. Siracusa, T. Alenghat, neonates. Immunol. Res. 22: 177–190. B. Liu, K. A. McCorkell, A. E. Troy, G. D. Rak, et al. 2015. Epithelial-intrinsic 60. Torow,N.,K.Yu,K.Hassani,J.Freitag,O.Schulz,M.Basic,A.Brennecke,T.Sparwasser, IKKa expression regulates group 3 innate lymphoid cell responses and antibacterial N. Wagner, A. Bleich, et al. 2015. Active suppression of intestinal CD4(+)TCRab(+) immunity. J. Exp. Med. 212: 1513–1528. T-lymphocyte maturation during the postnatal period. Nat. Commun. 6: 7725. 88. Karlsson, M. R., H. Kahu, L. A. Hanson, E. Telemo, and U. I. Dahlgren. 1999. 61. Harris, N. L., I. Spoerri, J. F. Schopfer, C. Nembrini, P. Merky, J. Massacand, Neonatal colonization of rats induces immunological tolerance to bacterial antigens. J. F. Urban, Jr., A. Lamarre, K. Burki, B. Odermatt, et al. 2006. Mechanisms of Eur. J. Immunol. 29: 109–118. neonatal mucosal antibody protection. J. Immunol. 177: 6256–6262. 89. Sudo, N., S. Sawamura, K. Tanaka, Y. Aiba, C. Kubo, and Y. Koga. 1997. The 62. Ferguson, A., and D. M. Parrott. 1972. The effect of antigen deprivation on requirement of intestinal bacterial flora for the development of an IgE production thymus-dependent and thymus-independent lymphocytes in the small intestine of system fully susceptible to oral tolerance induction. J. Immunol. 159: 1739–1745. the mouse. Clin. Exp. Immunol. 12: 477–488. 90.Edmond,K.M.,S.Newton,C.Shannon,M.O’Leary,L.Hurt,G.Thomas, 63. Rogier, E. W., A. L. Frantz, M. E. Bruno, L. Wedlund, D. A. Cohen, S. Amenga-Etego, C. Tawiah-Agyemang, L. Gram, C. N. Hurt, et al. 2015. Effect of A. J. Stromberg, and C. S. Kaetzel. 2014. Secretory antibodies in breast milk early neonatal vitamin A supplementation on mortality during infancy in Ghana promote long-term intestinal homeostasis by regulating the gut microbiota and host (Neovita): a randomised, double-blind, placebo-controlled trial. Lancet 385: 1315–1323. gene expression. Proc. Natl. Acad. Sci. USA 111: 3074–3079. 91. Lynch, S. J., M. R. Sears, and R. J. Hancox. 2016. Thumb-sucking, nail-biting, and 64. Blom, B., P. C. Res, and H. Spits. 1998. T cell precursors in man and mice. Crit. atopic sensitization, asthma, and hay fever. Pediatrics 138: e20160443. Available at: Rev. Immunol. 18: 371–388. http://pediatrics.aappublications.org/content/138/2/e20160443. 65. Spencer, J., T. T. MacDonald, T. Finn, and P. G. Isaacson. 1986. The development 92. Dominguez-Bello, M. G., K. M. De Jesus-Laboy, N. Shen, L. M. Cox, A. Amir, of gut associated lymphoid tissue in the terminal ileum of fetal human intestine. A. Gonzalez, N. A. Bokulich, S. J. Song, M. Hoashi, J. I. Rivera-Vinas, et al. 2016. Clin. Exp. Immunol. 64: 536–543. Partial restoration of the microbiota of cesarean-born infants via vaginal microbial 66.Heninger,A.K.,P.Monti,C.Wilhelm,P.Schwaiger,D.Kuehn,A.G.Ziegler,and transfer. Nat. Med. 22: 250–253. E. Bonifacio. 2013. Activation of islet autoreactive naı¨ve T cells in infants is influenced 93. Denkel, L. A., F. Schwab, L. Garten, C. Geffers, P. Gastmeier, and B. Piening. by homeostatic mechanisms and antigen-presenting capacity. Diabetes 62: 2059–2066. 2016. Protective effect of dual-strain probiotics in preterm infants: a multi-center 67. Mold, J. E., S. Venkatasubrahmanyam, T. D. Burt, J. Michae¨lsson, J. M. Rivera, time series analysis. PLoS One 11: e0158136. S. A. Galkina, K. Weinberg, C. A. Stoddart, and J. M. McCune. 2010. Fetal and 94. Lahl, K., C. Loddenkemper, C. Drouin, J. Freyer, J. Arnason, G. Eberl, adult hematopoietic stem cells give rise to distinct T cell lineages in humans. Science A. Hamann, H. Wagner, J. Huehn, and T. Sparwasser. 2007. Selective depletion of 330: 1695–1699. Foxp3+ regulatory T cells induces a scurfy-like disease. J. Exp. Med. 204: 57–63. Th eJournal of Brief Reviews Immunology

Host-Microbiota Interactions Shape Local and Systemic Inflammatory Diseases John B. Grigg and Gregory F. Sonnenberg Recent advances in understanding how the mammalian with the microbiota to maintain a state of homeostasis that is immune system and intestinal microbiota functionally mutually beneficial. For example, the requirement for micro- interact have yielded novel insights for human health biota in the proper development of the immune system was first and disease. Modern technologies to quantitatively mea- demonstrated in animals reared in microorganism-free envi- sure specific members and functional characteristics ronments, known as germ-free. Germ-free animals display a of the microbiota in the gastrointestinal tract, along variety of intestinal immune defects, including impaired de- with fundamental and emerging concepts in the field velopment of GALTs, lower amounts of secreted Ig, and reduced intraepithelial CD8+ T cells (5). Additionally, evidence of immunology, have revealed numerous ways in which Downloaded from host-microbiota interactions proceed beneficially, neu- has supported the notion of the gut microbiota having a strong influence over the development of the immune system outside trally, or detrimentally for mammalian hosts. It is clear + that the gut microbiota has a strong influence on the shape of the intestine (6). In germ-free mice, splenic CD4 Th cells are skewed toward the Th2 cell subset and promote enhanced and quality of the immune system; correspondingly, the allergic responses and type 2 immunity (6). Germ-free mice immune system guides the composition and localization of also have decreased total numbers of peripheral CD4+ Tcells, http://www.jimmunol.org/ the microbiota. In the following review, we examine the including Th17 cells (7) and regulatory T cell (Treg) com- evidence that these interactions encompass homeostasis partments (8, 9). Conversely, the intestinal immune system also and inflammation in the intestine and, in certain cases, actively shapes the composition and compartmentalization of extraintestinal tissues. Lastly, we discuss translational ther- the microbiota through various mechanisms (10–13). Overall, apies stemming from research on host-microbiota interac- these observations demonstrate that the colonizing microbiota tions that could be used for the treatment of chronic and host immune system have a complex, dynamic, and re- inflammatory diseases. The Journal of Immunology, ciprocal dialogue. 2017, 198: 564–571. Members of the microbiota are recognized by the innate immune system through their conserved pathogen-associated mo- by guest on January 26, 2017 he human body hosts a remarkable variety (1) and lecular patterns, referred to in this article as microbe-associated quantity (2) of microorganisms collectively referred molecular patterns (MAMPs) (14), to encompass such ligands in T to as the microbiota. The microbiota encompasses Ar- normally nonpathogenic organisms of the microbiota. MAMPs chaea, Bacteria, Eukarya, and viruses that form a complex eco- are recognized by germline-encoded pattern recognition recep- system thought to have coevolved with mammalian hosts over tors distributed spatiotemporally across various cell types and time. Commensal bacteria are the most well-defined member of tissues. Despite this ability to directly respond to microbiota- the microbiota. Among the various body surfaces where com- derived signals, several features of the immune system act in mensal bacteria reside, the gastrointestinal (GI) tract contains the cooperation with the intestinal barrier to protect the body from highest densities, which are estimated to range between 1011 and opportunistic pathogens and to limit the immune system from 1014 cells per gram of luminal content (3). This enormous overreacting to beneficial microbiota in the gut (Fig. 1A). Such cellular and genetic component of the human body is now well features include the following: a thick mucus lining the lumen recognized to provide indispensible functions in digestion, nu- of the gut epithelial cells that physically excludes most micro- trition status, and protection against invasive pathogens (4). organisms (15), secreted IgA that recognizes and binds microbe- The mammalian immune system is also significantly specific epitopes and facilitates their removal (16), and secreted enriched in the GI tract and engages in a complex dialogue antimicrobial peptides that directly neutralize microorganisms

Gastroenterology and Hepatology Division, Joan and Sanford I. Weill Department of Address correspondence and reprint requests to Dr. Gregory F. Sonnenberg, Weill Medicine, Weill Cornell Medicine, New York, NY 10021; Department of Microbiology Cornell Medical College, New York, NY 10021. E-mail address: gfsonnenberg@med. and Immunology, Weill Cornell Medicine, New York, NY 10065; and The Jill Roberts cornell.edu Institute for Research in Inflammatory Bowel Disease, Weill Cornell Medicine, New Abbreviations used in this article: AhR, aryl hydrocarbon receptor; CD, Crohn’s disease; York, NY 10021 EAE, experimental autoimmune encephalitis; FMT, fecal microbiota transplant therapy; Received for publication September 19, 2016. Accepted for publication October 31, GI, gastrointestinal; IBD, inflammatory bowel disease; IEC, intestinal epithelial cell; 2016. ILC, innate lymphoid cell; ILC3, group 3 ILC; MAMP, microbe-associated molecular pattern; PSA, polysaccharide A; RA, rheumatoid arthritis; SCFA, short-chain fatty acid; This work was supported by National Institutes of Health Grants DP5OD012116, SFB, segmented filamentous bacteria; Treg, regulatory T cell; WT, wild-type. R01AI123368, R21DK110262, and U01AI095608, the National Institute of Allergy and Infectious Diseases Mucosal Immunology Studies Team, the Crohn’s and Colitis Ó Foundation of America, the Searle Scholars Program, and an American Asthma Foun- Copyright 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00 dation Scholar Award.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601621 The Journal of Immunology 565

(17, 18). In addition to their pathogen-protective effects, these phenocopy aspects of human disease when experimentally features help to maintain sequestration of the microbiota, thus induced in animal models (31). These studies support the reducing the likelihood of the mammalian immune system concept that diverse impairments of hematopoietic and non- mounting an overreactive response to commensal bacteria. hematopoietic cell signaling pathways underlie abnormal host- However, when the epithelial barrier is compromised by chem- microbiota interactions. ical, pathogenic, or inflammatory insults, the immune system Environmental and lifestyle risk factors also play a role in must deal with the resulting influx of commensal and opportu- disease development (23), as evidenced by modest disease nistic microorganisms. In most contexts, the immune system concordance between monozygotic twins who develop ulcer- responds appropriately to protect the host from invasive mi- ative colitis and CD in adulthood (32). For example, diet is crobes while maintaining long-term tolerance to the microbiota. implicated in causative and preventative roles in IBD through Not surprisingly, sustained breakdown of the intestinal barrier is various mechanisms (33). Of note, short- and long-term die- linked to several chronic inflammatory diseases, although the tary patterns can modify the composition of the gut microbiota mechanisms are still being determined (19, 20) (Fig. 1B). (34–36). Diet’s link to IBD, among other environmental risk In this review, we assess how functional interactions between factors that alter the gut microbiota, such as antibiotic use, has the mammalian immune system and the microbiota in the gut motivated investigation of associations between microbial can drive inflammatory diseases locally and systemically. dysbiosis and intestinal inflammation. Dysbiosis is defined as a Comparatively, we also examine settings in which host-microbiota microbial imbalance resulting in a shift (i.e., loss or outgrowth interactions can prevent or constrain autoimmune disease. Lastly, of a species) and overall reduction in microbial diversity. Using Downloaded from we briefly discuss the evidence advocating for the therapeutic 16S sequencing of gut fecal content, it was observed that IBD modulation of host immune factors, as well as the manipulation patients have reduced colonic microbial diversity and a detect- of microbiota or microbiota-derived biomolecules for treating able shift in bacterial enterotypes compared with healthy indi- chronic inflammatory disease. viduals (37). For example, in a cohort of CD patients, Frank et al. (38) detected a relative decrease in Firmicutes and Bac- Aberrant host-microbiota interactions underlie intestinal teroides in the intestinal microbiota, as well as a relative increase http://www.jimmunol.org/ inflammation in the proinflammatory bacteria Enterobacteriaceae, relative to Inflammatory bowel disease (IBD) is a family of chronic in- controls(Fig.1B).Additionally,fecal metabolite analysis revealed flammatory disorders of the GI tract. In the clinic, IBD is a decrease in butyrate-producing bacterium in CD patients (39, frequently diagnosed as Crohn’s disease (CD), affecting any 40). However, no single bacterial strain or combinations of strains part of the GI tract, or ulcerative colitis, in which pathology is have been shown to directly cause or prevent IBD in humans. restricted primarily to the colon (20). As with most complex Several animal studies have demonstrated that dysbiosis can diseases, IBD is thought to occur from a combination of ge- drive inflammatory pathogenesis in the intestine. These studies netic (21, 22), environmental, and lifestyle-associated risk factors involve the demonstration that a proinflammatory consortium (23) that culminate in dysregulated host innate and adaptive of microbiota, generally resulting from immune impairment, by guest on January 26, 2017 immune responses to the intestinal microbiota. Despite the can transfer disease phenotype to healthy wild-type (WT) re- complexity of IBD’s etiology, the host-microbiota interactions cipient animals. For instance, one of the first animal studies to that drive disease pathogenesis are becoming better understood implicate an IBD-causative consortium of bacteria involved the through studies in human IBD patients and animal models of horizontal transfer of microbiota from a spontaneous model of intestinal damage and inflammation. colitis (mice deficient in both T-bet and Rag2; referred to as Genetic analyses have identified loss-of-function mutations TRUC mice) into a healthy recipient mouse which resulted in and polymorphisms in key immune tolerance–related genes transfer of colitis (41). Garrett et al. (42) later identified more and immune-response elements that can lead to early-onset specifically that Klebsiella pneumoniae and Proteus mirabilis, IBD or increase disease susceptibility in adulthood (21, 22). which grow out in TRUC mice, act in conjunction with the Many primary immunodeficiencies first manifest in the GI presence of normal gut flora to drive the colitogenic effect upon tract (24). For example, individuals with loss-of-function mu- transfer. In another model, NLRP6 inflammasome–deficient tations in IL-10/IL-10R signaling present with very early–onset mice displayed spontaneous colitis that was transferrable to IBD as a result of their inability to regulate inflammatory im- WT neonates or adults via cross-fostering or cohousing (43). mune responses to commensal bacteria in the GI tract (25–27). Prevotellaceae was implicated as the primary driver of the in- Indeed, several other primary immunodeficiencies, such as com- flammatory effect in this study (43). In another study, Couturier- bined T and B cell deficiencies, are linked to early-onset GI Maillard et al. found that mice with a deficiency in NOD2 disorders (24), and many more are continually being identified had a dysbiotic consortium of microbiota that could transfer through whole exome sequencing (28). Genome-wide association colitis to healthy recipient mice (44). These studies under- studies have also identified a number of polymorphisms that are score that genetic disruption of immune pathways in the in- associated with an increased susceptibility to developing IBD in testine is sufficient to initiate colitogenic microbial dysbiosis. early life or adulthood (21, 22). The susceptibility loci include These microbial consortiums can then transfer disease even in genes and gene pathways involved in intestinal barrier function, the context of a functional immune system. innate immune recognition, adaptive immunity, and cellular Specific microbial species can be characterized by their ca- homeostasis (21). For example, genetic alterations in NOD2, pacity to provoke inflammatory responses in the intestine. an intracellular pattern recognition receptor for bacterial pep- Segmented filamentous bacteria (SFB) preferentially induce tidoglycans, confer increased susceptibility to developing CD in the differentiation of proinflammatory CD4+ Th17 cells in the adulthood (29, 30). To extend the genetic evidence, many of lamina propria of the ileum through the TLR5 innate path- the identified genetic alterations found in IBD patients can way, serum amyloid A, and direct epithelial cell adhesion 566 BRIEF REVIEWS: HOST-MICROBIOTA INTERACTIONS REGULATE INFLAMMATION Downloaded from http://www.jimmunol.org/ by guest on January 26, 2017

FIGURE 1. Host–microbiota interactions underlie homeostasis and inflammation in the intestine and extraintestinal tissues. (A) At homeostasis, gut bacteria are compartmentalized within the lumen through exclusion by the mucus, neutralization by antimicrobial peptides produced by IECs, and release of secretory IgA (sIgA) from intestinal-resident B cells. In response to various cues, ILC3s and Th17 cells in the intestine produce IL-22, which acts on IECs to promote compartmentalization of the microbiota. Tregs produce IL-10 and are induced by microbially derived SCFAs and PSA or by the bacterial species B. fragilis and Clostridium. Intestinal activation of Tregs can protect against neuroinflammation in the CNS during EAE. (B) During chronic intestinal inflammation, loss of intestinal barrier function results in bacterial translocation across the epithelium, release of commensally derived MAMPs, proinflammatory cytokine and chemokine activation, and Th17 and B cell responses. Specific bacteria exacerbate intestinal inflammation, including Prevotellaceae, Enterobacteriaceae, and the Th17-inducing SFB. Loss of tolerance to self-Ags can occur because of lowered thresholds for autoactivation (bystander effect) that can mediate autoimmunity in extraintestinal tissues. Bystander effects initiate and exacerbate Th17-mediated inflammation in mouse models of EAE and RA. In RA, Th17 and follicular helper T cell responses aid in autoantibody production in secondary lymph nodes. Licensing of cross-reactive T cell responses that recognize microbially derived peptides and react to self-peptides can also initiate autoimmunity in extraintestinal tissues, exemplified in a mouse model of experimental autoimmune uveitis (EAU).

(45, 46) (Fig. 1B). Recently, it was found that a variety of bacte- (Fig. 1B). It is important to note that ILCs and Th17 cells also rial species exhibiting epithelial cell adhesion (such as Citrobacter mediate protection from pathogens and proper containment of rodentium, enterohemorrhagic Escherichia coli, and Candida commensal bacteria, thus promoting homeostasis in the gut albicans) could similarly induce Th17 cells in the intestine (49), which is examined later in this review. Furthermore, SFB (47). Epithelial adhesion was an indispensible bacterial char- were well characterized to significantly induce Th17 cell re- acteristic for this effect because adhesion-defective bacte- sponses but also were demonstrated to promote the develop- rial mutants failed to induce Th17 cell responses (47). Finally, ment of intestinal Tregs (8). Thus, a comprehensive analysis of Helicobacter hepaticus also was demonstrated to induce proin- microbiota-induced responses should carefully be considered, flammatory innate lymphoid cell (ILC) and Th17 cell re- and single species cannot always be defined solely as pro- or anti- sponses through the induction of cytokines IL-1b and IL-23 inflammatory. Moreover, dysbiotic blooms of bacterial species in in the colon (48). Thus, the induction of proinflammatory the gut, such as Enterobacteriaceae, can be secondary to intestinal immune cells and their subsequent effector functions under- inflammation from various insults (50), which may owe to the lie some of the pathogenic effects of these bacterial members unique ability of these bacteria to feed off the by-products of host The Journal of Immunology 567 inflammation (51). Such observations obscure cause-and-effect in germinal center formation and the production of auto- relationships between microbial dysbiosis and inflammation. antibodies that mediate disease (62). More recently, SFB col- In summary, IBD arises from a framework of genetic pre- onization in K/BxN mice was found to induce the activation of dispositions, environmental risk factors, and dysbiotic micro- follicular helper T cells in the Peyer’s patches, which sub- biota that underpins its chronic nature. Continued interrogation sequently egress to the spleen and aid in the production of of the complexity of host-microbiota interactions that promote autoantibodies (63). A unique variation of the bystander model or constrain IBD in the various contexts of human disease also was demonstrated by Campisi et al. (64): C. rodentium should yield more rationally informed preventative or therapeutic infection causes self-antigen release from apoptotic host cells, approaches. which are then processed and presented by APCs alongside bacterial peptides, resulting in the licensing of autoreactive Involvement of the gut microbiota in initiating or exacerbating Th17 cells. This reinforces the concept that severe inflam- extraintestinal inflammatory diseases mation in the intestine, arising from infection in this case, can In parallel with research on how host-microbiota interactions provide the initiating conditions for the development of local drive inflammatory diseases in the intestine, mounting evi- and systemic autoimmunity. dence suggests that these interactions also impact systemic Molecular mimicry, or an adaptive response that recognizes inflammatory disease (52–55) (Fig. 1B). Indeed, IBD patients and responds to foreign-derived non-self and self-antigens, also frequently have extraintestinal disease manifestations involving has been proposed as a general mechanism for autoimmunity (65, 66). Understandably, cross-reactivity of adaptive responses the joints, skin, and eyes (56). Multi-disease cohort genome- Downloaded from wide association studies have revealed significant overlap in has direct benefit to the host when it results in broader pro- genetic susceptibility loci for IBD with a variety of inflamma- tection against phylogenically related pathogens; in contrast, it tory diseases involving extraintestinal tissues (21, 57). Addition- can adversely result in an inappropriate response to self-antigens. ally, similar to ulcerative colitis, some autoimmune diseases, such Using mouse models, several groups have provided evidence that as rheumatoid arthritis (RA), show less disease concordance be- microbiota-derived Ags may provide the antigenic basis for the initiation of systemic autoimmune disease (55). The seminal tween monozygotic twins than do other autoimmune diseases http://www.jimmunol.org/ (58).Thissuggestsastrongrolefor environmental factors in observation that molecular mimicry to common microbial pep- disease development. Alterations in the gut microbiota are asso- tides can induce autoimmunity came from a series of experi- ciated with several chronic inflammatory diseases outside of the ments demonstrating that structurally related microbial peptides intestine (5). Although causal relationships have not been dem- could activate MBP-specific T cells (in the Ob TCR-DR2b onstrated, these observations provoke the theory that disrupted mouse model), which then induce neurodegeneration in mice host-microbiota interactions in the gut, arising from genetic or (67). More recently, Horai et al. (68) showed that gut micro- environmental perturbations, may underlie or impact the course biota can provide the antigenic material for cross-reactivity to a of systemic autoimmune diseases. self-antigen. Using a spontaneous model of autoimmune uveitis Several research efforts in animal models have illuminated (TCR transgenic for the retinal protein IRBP), the investigators by guest on January 26, 2017 + how the gut microbiota can have a causative or protective role found that IRBP-specific CD4 T cells are first activated in the in autoimmune disease outside of the intestine. In the following gut, migrate to the eye, and drive pathogenic autoimmune sections we categorize such evidence into two, nonmutually uveitis (68). Interestingly, Kadowaki et al. (69) demonstrated exclusive groupings: bystander effects (Ag nonspecific) and that myelin protein (MOG)-specific CD4+ intraepithelial molecular mimicry (Ag specific). Although these have already lymphocytes can be activated and proliferate in response to gut been proposed and demonstrated as ways in which infectious Ags. In this context, CD4+ T cells differentiate into a regulatory pathogens, such as viruses, may lead to chronic autoimmune Th17 cell phenotype that expresses CTLA4 and TGFBR1. disease in humans (59), less research has elucidated mechanisms Upon transfer into WT mice, MOG-specific CD4+ intraepithelial in which resident or transient microbiota can have similar roles lymphocytes infiltrate the CNS and upregulate LAG3 expression, in systemic inflammatory diseases. reducing neuroinflammation (69). These experiments demon- In certain contexts, gut microbiota can exert an adjuvant strate that molecular mimicry can activate proinflammatory and effect in the priming of autoreactive adaptive immune re- immunoregulatory pathways that influence autoimmune disease sponses. Animal models support the concept that microbiota in extraintestinal sites. provide a necessary bystander role in initiating autoimmune Despite the correlative evidence that commensal bacteria can diseases in extraintestinal sites (Fig. 1B). For instance, mice directly initiate autoimmunity in extraintestinal tissues, no treated with antibiotics (60) or reared in germ-free conditions single bacterium or consortium of bacteria has been specifically (61) show reduced induction of experimental autoimmune identified in humans. Additional research and experimental encephalitis (EAE). Recolonization of germ-free mice with tools are warranted to clarify the mechanisms of bystander SFB alone induces Th17 cells in the gut and enhances neu- effects and molecular mimicry in extraintestinal autoimmune rodegeneration in the CNS upon active EAE induction (61). disease settings. Recently published articles demonstrated SFB can have a similar impact, although through a different systematic approaches to identify commensal bacteria that mechanism, to induce autoimmunity in a mouse model of incite colitis (70) or diet-dependent enteropathy (71). These spontaneous autoimmune arthritis (K/BxN) (62). Germ-free approaches consist of sequencing IgA-targeted bacterial taxa K/BxN mice are resistant to the development of arthritis, pri- from the fecal microbiota and then validating the bacterium’s marily as a result of reduced systemic germinal center formation immunological impact in gnotobiotic mice, which may prove and subsequent loss of autoantibody production. Recolonization useful for the identification of specific intestinal commensal bac- with SFB restores autoimmune arthritis through the activation of teria involved in autoimmune diseases outside of the gut. Fur- Th17 cells in the intestine, which then traffic to the spleen to aid thermore, stratification of IgA sequencing into T cell–independent 568 BRIEF REVIEWS: HOST-MICROBIOTA INTERACTIONS REGULATE INFLAMMATION and T cell–dependent IgA-production mechanisms (72) may protective and regulatory immune responses (82, 83). Of note, help to delineate bacterial members that have the strongest ca- group 3 ILCs (ILC3s) can modify the composition and ana- pacity to invoke autoreactive T cell responses. tomical localization of the microbiota. ILC3s respond to a variety of inflammatory cytokines (i.e., IL-1b, IL-23, IL-6) and Protective host-microbiota interactions can prevent or diminish local microbiota-derived metabolites (i.e., aryl hydrocarbon receptor and systemic inflammatory responses [AhR] ligands) (83). Following activation, ILC3s produce In addition to the detrimental outcomes of host-microbiota multiple effector molecules, including the cytokine IL-22, interactions, attention has focused on interactions that lead which acts directly on IECs to produce antimicrobial pep- to beneficial physiological states associated with mammalian tides, increase mucus production from goblet cells, and increase health. Recently identified mechanisms including microbiota- fucosylation of the mucus (84–86). During steady-state, the derived metabolites, regulatory immune cell types, systemic Ig, physiological outcome of this response is to maintain proper and intrinsic immune functions that coordinate to enforce localization and composition of commensal bacteria (87, 88). peripheral tolerance have been elucidated from experimental In response to breakdown of the intestinal barrier from various models of inflammatory diseases. insults, such as infection with C. rodentium, this pathway re- The presence of microbiota in the gut of conventionalized inforces compartmentalization of the pathogen to prevent its mice was shown to have an essential role in generating the systemic dissemination from the intestine (89, 90). CD4+ Foxp3+ Treg compartment (73), a key cellular medi- Recently, it was demonstrated that microbiota-derived indole metabolites of tryptophan in the diet can target AhR in ator of regulatory immune responses. Several research groups Downloaded from have identified specific species of commensal bacteria, or Th17 cells and ILC3s to promote production of IL-22 (91). consortia of commensal bacteria, that regulate the number, Mice deficient for the adaptor protein CARD9 develop quality, and TCR repertoire of intestinal Tregs. As reviewed spontaneous colitis and display a loss of bacterial species able recently (74), it is clear that many members of the commensal to convert tryptophan into ligands for AhR. Supplementation of three Lactobacillus strains capable of metabolizing trypto- microbiota have a Treg-inducing capacity; well-documented 2 2 phan into AhR ligands protected CARD9 / mice against in- examples include Bacteroides fragilis (75) and Clostridium (9) http://www.jimmunol.org/ (Fig. 1A). Recently, Faith et al. (76) devised a systematic flammation in the colon (91). These results expand the evidence approach to elucidate combinations of human-associated mi- that particular metabolites in the gut microenvironment are key crobial species that can promote intestinal Treg responses in to the proper regulation of gut homeostasis. Notably, ILC3s can gnotobiotic mice. This method may permit the identification also directly limit microbiota-specific T cell responses to main- of new immunoregulatory phenotypes that are associated with tain intestinal homeostasis through Ag presentation on MHC particular members of the human microbiota. class II (92, 93). Future investigation of ILC3s and other regu- In parallel with the identification of groups or specific latory pathways that influence host-microbiota interactions in members of the microbiota, more reductionist approaches have the GI tract could provoke the development of novel treatment identified microbiota-derived molecules and metabolites that options for inflammatory disease. by guest on January 26, 2017 stimulate Treg responses. The best-characterized molecules Systemic Ig responses to commensal microbiota also were by which the microbiota promote Treg differentiation are found to be essential for the maintenance of beneficial host- bacterial-derived polysaccharide A (PSA) and short-chain microbiota interactions. For example, maternally acquired fatty acids (SCFAs) (Fig. 1A). PSA was the first documented IgA and IgG in neonatal mice leads to dampened T cell– microbiota-derived molecule that directs the development of a dependent immune responses against commensal bacteria balanced T cell compartment in mice (6). Later, it was found (94). Additionally, systemic IgG responses to Gram-negative that colonization of mice with PSA-sufficient strains of the bacterial commensals, acquired early and over the course of commensal B. fragilis or purified PSA protect against experi- life, were shown to provide cross-protection against Gram- mentally induced colitis and that the protective effect required negative pathogens, such as E. coli and Salmonella, in mice the presence of a functional IL-10–producing CD4+ Tcell (95). These observations reinforce the concept that microbial compartment (6). Furthermore, prophylactic or therapeutic composition and the timing of host-commensal interactions administration of PSA could protect mice from the induction provide the foundation for balanced immunity in the intes- of EAE, which was dependent on an IL-10–producing Treg tine. Lastly, the microbiota was observed to promote its own population (77). This suggests that the effects of PSA-mediated compartmentalization within the intestinal lumen, as well as Treg induction can have a systemic influence over peripheral provide protection to the host from pathogens (96, 97). Re- tolerance in tissues outside of the intestine. SCFAs are another cent identification of the gut–vascular barrier system in the group of immunoregulatory molecules that are primarily de- small intestine (98), which restricts dissemination of gut bac- rived from the microbiota-mediated digestion of dietary fiber teria, should provide a new therapeutic framework for con- and promote the differentiation of peripheral Tregs (78–80). straining disease manifestations that are due to bacterial Interestingly, SCFAs, such as acetate, can be detected in the translocation across the intestinal epithelium. blood circulation (81), suggesting that microbiota-derived SCFAs can have far-ranging effects outside of the intestine. Manipulation of intestinal microbiota and host-microbiota Along with the well-documented role of Tregs, other re- interactions to treat inflammatory diseases cently identified cell types and cytokine–cytokine receptor Established immunosuppressive medicines, such as glucocor- pathways that connect hematopoietic and nonhematopoietic ticoids, still represent effective front-line therapies to treat cells mediate tolerance at the intestinal barrier. For example, inflammatory disease (99), but they have clear disadvantages ILCs maintain dialogue with intestinal epithelial cells (IECs), as a long-term treatment option. Not surprisingly, the medical intestinal dendritic cells, and other cell types to coordinate and biotechnology sectors have taken an interest in translating The Journal of Immunology 569 knowledge of host-microbiota interactions into better standards many inflammatory diseases. Furthermore, rationally designed of care and medicines to treat autoimmune diseases. For ex- therapies that modulate or re-establish beneficial interactions are a ample, therapies aiming to adoptively transfer Tregs or promote promising approach for the treatment of intestinal and extra- their in vivo induction in patients are being explored as thera- intestinal inflammatory diseases. Emergent technologies aim to peutic approaches for IBD (100). Additionally, mAbs have been focus research efforts on identifying the scope and relevance of developed to block cytokine pathways implicated in chronic these interactions more systematically and unambiguously. inflammation. For instance, blockade of the Th1 and Th17 cell pathways by targeting the shared anti-p40 subunit of IL-12/IL- Acknowledgments 23 with the mAb ustekinumab is rapidly advancing through We thank members of the Sonnenberg Laboratory for discussions and critical clinical trials as a novel treatment for moderate-to-severe CD reading of the manuscript. (101, 102). ILCs have received increasing attention as novel targets for the treatment of inflammatory diseases given some of Disclosures their analogous signaling pathways with T cells (103). Clinical The authors have no financial conflicts of interest. investigation on how specific subsets of ILCs respond to approved mAb therapies that target cytokine/cytokine receptor pathways, or novel small molecules targeting transcriptional References regulators, will extend and refine the paradigm of ILC involve- 1. Human Microbiome Project Consortium. 2012. Structure, function and diversity of the healthy human microbiome. Nature 486: 207–214. ment in provoking and resolving inflammatory disease in the 2. Sender, R., S. Fuchs, and R. Milo. 2016. Are we really vastly outnumbered? intestine and in extraintestinal tissues (104–106). revisiting the ratio of bacterial to host cells in humans. Cell 164: 337–340. Downloaded from 3. Hill, D. A., and D. Artis. 2010. Intestinal bacteria and the regulation of immune In parallel with therapeutic strategies of directly modulating cell homeostasis. Annu. Rev. Immunol. 28: 623–667. host immune factors, recent approaches have used mechanistic 4. Round, J. L., and S. K. Mazmanian. 2009. The gut microbiota shapes intestinal immune responses during health and disease. Nat. Rev. Immunol. 9: 313–323. understanding of intestinal microbiota, or microbiota-derived 5. Kamada, N., S. U. Seo, G. Y. Chen, and G. Nuń˜ez. 2013. Role of the gut microbiota products, for the development of novel treatments (107–109). in immunity and inflammatory disease. Nat. Rev. Immunol. 13: 321–335. Among others, proposed and emerging therapies include 6. Mazmanian, S. K., C. H. Liu, A. O. Tzianabos, and D. L. Kasper. 2005. An immunomodulatory molecule of symbiotic bacteria directs maturation of the host http://www.jimmunol.org/ modification of diet (81, 110), more targeted antibiotics to immune system. Cell 122: 107–118. preserve microbiota integrity (111), supplementation of immu- 7. Ivanov, I. I., Rde. L. Frutos, N. Manel, K. Yoshinaga, D. B. Rifkin, R. B. Sartor, B. B. Finlay, and D. R. Littman. 2008. Specific microbiota direct the differenti- noregulatory metabolites, administration of live biotherapeutic ation of IL-17-producing T-helper cells in the mucosa of the small intestine. Cell products (112), and fecal microbiota transplant therapy (FMT) Host Microbe 4: 337–349. 8. Gaboriau-Routhiau, V., S. Rakotobe, E. Lećuyer, I. Mulder, A. Lan, C. Bridonneau, (113–116). Although FMT studies have reported promising V. Rochet, A. Pisi, M. De Paepe, G. Brandi, et al. 2009. The key role of segmented results in preclinical and clinical settings for infectious diseases, filamentous bacteria in the coordinated maturation of gut helper T cell responses. Immunity 31: 677–689. such as Clostridium difficile, preliminary clinical trials of FMT 9. Atarashi, K., T. Tanoue, T. Shima, A. Imaoka, T. Kuwahara, Y. Momose, for IBD revealed limited efficacy (114, 117). No microbiota- G. Cheng, S. Yamasaki, T. Saito, Y. Ohba, et al. 2011. Induction of colonic based or microbiota-derived medical product has been approved regulatory T cells by indigenous Clostridium species. Science 331: 337–341. by guest on January 26, 2017 10. Brown, E. M., M. Sadarangani, and B. B. Finlay. 2013. The role of the immune by governmental regulatory agencies for the prevention or treat- system in governing host-microbe interactions in the intestine. Nat. Immunol. 14: ment of inflammatory disease. 660–667. 11. Fung, T. C., D. Artis, and G. F. Sonnenberg. 2014. Anatomical localization of Among other challenges faced (107), a prerequisite for commensal bacteria in immune cell homeostasis and disease. Immunol. Rev. 260: translational development includes distinguishing the tem- 35–49. 12. Zhang, H., J. B. Sparks, S. V. Karyala, R. Settlage, and X. M. Luo. 2015. Host poral and kinetic influences of the microbiota on the host adaptive immunity alters gut microbiota. ISME J. 9: 770–781. immune system (118). Animal studies emphasize that the 13. Dolle´, L., H. Q. Tran, L. Etienne-Mesmin, and B. Chassaing. 2016. Policing of functions of several immune cell types arise during a critical gut microbiota by the adaptive immune system. BMC Med. 14: 27. + 14. Ausubel, F. M. 2005. Are innate immune signaling pathways in plants and animals period in infancy (118). For example, a subset of RORgt conserved? Nat. Immunol. 6: 973–979. Tregs arises in the colon early after birth and prior to weaning 15. Hansson, G. C. 2012. Role of mucus layers in gut infection and inflammation. Curr. Opin. Microbiol. 15: 57–62. (119–121). Additionally, germ-free mice display increased 16. Mantis, N. J., N. Rol, and B. Corthe´sy. 2011. Secretory IgA’s complex roles in frequencies of invariant NKT cells in the colon, which pre- immunity and mucosal homeostasis in the gut. Mucosal Immunol. 4: 603–611. 17. Bevins, C. L., and N. H. Salzman. 2011. Paneth cells, antimicrobial peptides and disposes the mice to environmentally triggered colitis (122). maintenance of intestinal homeostasis. Nat. Rev. Microbiol. 9: 356–368. Hyperresponsive invariant NKT cell responses in germ-free 18. Mukherjee, S., and L. V. Hooper. 2015. Antimicrobial defense of the intestine. Immunity 42: 28–39. mice are reversible through administration of a normal micro- 19. Maynard, C. L., C. O. Elson, R. D. Hatton, and C. T. Weaver. 2012. Reciprocal biota, or B. fragilis–derived Ags, but only during the first 2 wk of interactions of the intestinal microbiota and immune system. Nature 489: 231– life (122). Nevertheless, many beneficial influences of the 241. 20. Maloy, K. J., and F. Powrie. 2011. Intestinal homeostasis and its breakdown in microbiota and microbiota-derived biomolecules on immune inflammatory bowel disease. Nature 474: 298–306. cell subsets are age independent (118) and conceivably repre- 21. Khor, B., A. Gardet, and R. J. Xavier. 2011. Genetics and pathogenesis of inflammatory bowel disease. Nature 474: 307–317. sent the most appropriate treatment options for reversing im- 22. Liu, T. C., and T. S. Stappenbeck. 2016. Genetics and pathogenesis of inflam- mune defects present in adulthood. matory bowel disease. Annu. Rev. Pathol. 11: 127–148. 23. Ananthakrishnan, A. N. 2015. Epidemiology and risk factors for IBD. Nat. Rev. Gastroenterol. Hepatol. 12: 205–217. Conclusions 24. Agarwal, S., and L. Mayer. 2013. Diagnosis and treatment of gastrointestinal disorders in patients with primary immunodeficiency. Clin. Gastroenterol. Hepatol. Interrogation of host-microbiota interactions in the intestine 11: 1050–1063. has revealed unexpected and novel insights into human health 25. Glocker, E. O., D. Kotlarz, K. Boztug, E. M. Gertz, A. A. Scha¨ffer, F. Noyan, M. Perro, J. Diestelhorst, A. Allroth, D. Murugan, et al. 2009. Inflammatory and disease. Human genetic, epidemiologic, and microbial bowel disease and mutations affecting the interleukin-10 receptor. N. Engl. J. Med. analyses, paired with complementary animal disease models, 361: 2033–2045. 26. Shah, N., J. Kammermeier, M. Elawad, and E. O. Glocker. 2012. Interleukin-10 support the concept that disrupted host-microbiota interac- and interleukin-10-receptor defects in inflammatory bowel disease. Curr. Allergy tions with the immune system underlie the chronic nature of Asthma Rep. 12: 373–379. 570 BRIEF REVIEWS: HOST-MICROBIOTA INTERACTIONS REGULATE INFLAMMATION

27. Engelhardt, K. R., and B. Grimbacher. 2014. IL-10 in humans: lessons from the 53. Ferreira, C. M., A. T. Vieira, M. A. Vinolo, F. A. Oliveira, R. Curi, and gut, IL-10/IL-10 receptor deficiencies, and IL-10 polymorphisms. Curr. Top. Fdos. S. Martins. 2014. The central role of the gut microbiota in chronic in- Microbiol. Immunol. 380: 1–18. flammatory diseases. J. Immunol. Res. 2014: 689492. 28. Kelsen, J. R., R. N. Baldassano, D. Artis, and G. F. Sonnenberg. 2015. Main- 54. Ho, J. T., G. C. Chan, and J. C. Li. 2015. Systemic effects of gut microbiota and taining intestinal health: the genetics and immunology of very early onset in- its relationship with disease and modulation. BMC Immunol. 16: 21. flammatory bowel disease. Cell Mol. Gastroenterol. Hepatol. 1: 462–476. 55. Za´rate-Blade´s, C. R., R. Horai, and R. R. Caspi. 2016. Regulation of autoim- 29.Hugot,J.P.,M.Chamaillard,H.Zouali,S.Lesage,J.P.Ce´zard, J. Belaiche, S. Almer, munity by the microbiome. DNA Cell Biol. 35: 455–458. C. Tysk, C. A. O’Morain, M. Gassull, et al. 2001. Association of NOD2 leucine-rich 56. Vavricka, S. R., A. Schoepfer, M. Scharl, P. L. Lakatos, A. Navarini, and repeat variants with susceptibility to Crohn’s disease. Nature 411: 599–603. G. Rogler. 2015. Extraintestinal manifestations of inflammatory bowel disease. 30. Ogura, Y., D. K. Bonen, N. Inohara, D. L. Nicolae, F. F. Chen, R. Ramos, Inflamm. Bowel Dis. 21: 1982–1992. H. Britton, T. Moran, R. Karaliuskas, R. H. Duerr, et al. 2001. A frameshift 57. Richard-Miceli, C., and L. A. Criswell. 2012. Emerging patterns of genetic overlap mutation in NOD2 associated with susceptibility to Crohn’s disease. Nature 411: across autoimmune disorders. Genome Med. 4: 6. 603–606. 58. Bogdanos, D. P., D. S. Smyk, E. I. Rigopoulou, M. G. Mytilinaiou, M. A. Heneghan, 31. Mizoguchi, A., T. Takeuchi, H. Himuro, T. Okada, and E. Mizoguchi. 2016. C. Selmi, and M. E. Gershwin. 2012. Twin studies in autoimmune disease: genetics, Genetically engineered mouse models for studying inflammatory bowel disease. gender and environment. J. Autoimmun. 38: J156–J169. J. Pathol. 238: 205–219. 59. Getts, D. R., E. M. Chastain, R. L. Terry, and S. D. Miller. 2013. Virus infection, 32. Spehlmann, M. E., A. Z. Begun, J. Burghardt, P. Lepage, A. Raedler, and antiviral immunity, and autoimmunity. Immunol. Rev. 255: 197–209. S. Schreiber. 2008. Epidemiology of inflammatory bowel disease in a German twin 60. Ochoa-Repa´raz, J., D. W. Mielcarz, L. E. Ditrio, A. R. Burroughs, D. M. Foureau, cohort: results of a nationwide study. Inflamm. Bowel Dis. 14: 968–976. S. Haque-Begum, and L. H. Kasper. 2009. Role of gut commensal microflora in 33. Lee, D., L. Albenberg, C. Compher, R. Baldassano, D. Piccoli, J. D. Lewis, and the development of experimental autoimmune encephalomyelitis. J. Immunol. 183: G. D. Wu. 2015. Diet in the pathogenesis and treatment of inflammatory bowel 6041–6050. diseases. Gastroenterology 148: 1087–1106. 61. Lee, Y. K., J. S. Menezes, Y. Umesaki, and S. K. Mazmanian. 2011. Proin- 34. De Filippo, C., D. Cavalieri, M. Di Paola, M. Ramazzotti, J. B. Poullet, flammatory T-cell responses to gut microbiota promote experimental autoimmune S. Massart, S. Collini, G. Pieraccini, and P. Lionetti. 2010. Impact of diet in encephalomyelitis. Proc. Natl. Acad. Sci. USA 108(Suppl. 1): 4615–4622. shaping gut microbiota revealed by a comparative study in children from Europe 62. Wu, H. J., I. I. Ivanov, J. Darce, K. Hattori, T. Shima, Y. Umesaki, D. R. Littman,

and rural Africa. Proc. Natl. Acad. Sci. USA 107: 14691–14696. C. Benoist, and D. Mathis. 2010. Gut-residing segmented filamentous bacteria Downloaded from 35. Wu, G. D., J. Chen, C. Hoffmann, K. Bittinger, Y. Y. Chen, S. A. Keilbaugh, drive autoimmune arthritis via T helper 17 cells. Immunity 32: 815–827. M. Bewtra, D. Knights, W. A. Walters, R. Knight, et al. 2011. Linking long-term 63. Teng, F., C. N. Klinger, K. M. Felix, C. P. Bradley, E. Wu, N. L. Tran, dietary patterns with gut microbial enterotypes. Science 334: 105–108. Y. Umesaki, and H. J. Wu. 2016. Gut microbiota drive autoimmune arthritis by 36. David, L. A., C. F. Maurice, R. N. Carmody, D. B. Gootenberg, J. E. Button, promoting differentiation and migration of Peyer’s patch T follicular helper cells. B. E. Wolfe, A. V. Ling, A. S. Devlin, Y. Varma, M. A. Fischbach, et al. 2014. Diet Immunity 44: 875–888. rapidly and reproducibly alters the human gut microbiome. Nature 505: 559–563. 64. Campisi, L., G. Barbet, Y. Ding, E. Esplugues, R. A. Flavell, and J. M. Blander. 37. Manichanh, C., L. Rigottier-Gois, E. Bonnaud, K. Gloux, E. Pelletier, L. Frangeul, 2016. Apoptosis in response to microbial infection induces autoreactive TH17 R. Nalin, C. Jarrin, P. Chardon, P. Marteau, et al. 2006. Reduced diversity of cells. Nat. Immunol. 17: 1084–1092. faecal microbiota in Crohn’s disease revealed by a metagenomic approach. 65. Oldstone, M. B. 1987. Molecular mimicry and autoimmune disease. Cell 50: 819– http://www.jimmunol.org/ Gut 55: 205–211. 820. 38. Frank, D. N., A. L. St Amand, R. A. Feldman, E. C. Boedeker, N. Harpaz, and 66. Yurkovetskiy, L. A., J. M. Pickard, and A. V. Chervonsky. 2015. Microbiota and N. R. Pace. 2007. Molecular-phylogenetic characterization of microbial commu- autoimmunity: exploring new avenues. Cell Host Microbe 17: 548–552. nity imbalances in human inflammatory bowel diseases. Proc. Natl. Acad. Sci. USA 67. Harkiolaki, M., S. L. Holmes, P. Svendsen, J. W. Gregersen, L. T. Jensen, 104: 13780–13785. R. McMahon, M. A. Friese, G. van Boxel, R. Etzensperger, J. S. Tzartos, et al. 39. Fujimoto, T., H. Imaeda, K. Takahashi, E. Kasumi, S. Bamba, Y. Fujiyama, and 2009. T cell-mediated autoimmune disease due to low-affinity crossreactivity to A. Andoh. 2013. Decreased abundance of Faecalibacterium prausnitzii in the gut common microbial peptides. Immunity 30: 348–357. microbiota of Crohn’s disease. J. Gastroenterol. Hepatol. 28: 613–619. 68. Horai, R., C. R. Za´rate-Blade´s, P. Dillenburg-Pilla, J. Chen, J. L. Kielczewski, 40. Wang, W., L. Chen, R. Zhou, X. Wang, L. Song, S. Huang, G. Wang, and B. Xia. P. B. Silver, Y. Jittayasothorn, C. C. Chan, H. Yamane, K. Honda, and 2014. Increased proportions of Bifidobacterium and the Lactobacillus group and R. R. Caspi. 2015. Microbiota-dependent activation of an autoreactive T cell re- loss of butyrate-producing bacteria in inflammatory bowel disease. J. Clin. ceptor provokes autoimmunity in an immunologically privileged site. Immunity by guest on January 26, 2017 Microbiol. 52: 398–406. 43: 343–353. 41. Garrett, W. S., G. M. Lord, S. Punit, G. Lugo-Villarino, S. K. Mazmanian, S. Ito, 69. Kadowaki, A., S. Miyake, R. Saga, A. Chiba, H. Mochizuki, and T. Yamamura. J. N. Glickman, and L. H. Glimcher. 2007. Communicable ulcerative colitis in- 2016. Gut environment-induced intraepithelial autoreactive CD4(+) T cells sup- duced by T-bet deficiency in the innate immune system. Cell 131: 33–45. press central nervous system autoimmunity via LAG-3. Nat. Commun. 7: 11639. 42. Garrett,W.S.,C.A.Gallini,T.Yatsunenko,M.Michaud,A.DuBois,M.L.Delaney, 70. Palm, N. W., M. R. de Zoete, T. W. Cullen, N. A. Barry, J. Stefanowski, L. Hao, S. Punit, M. Karlsson, L. Bry, J. N. Glickman, et al. 2010. Enterobacteriaceae act in P. H. Degnan, J. Hu, I. Peter, W. Zhang, et al. 2014. Immunoglobulin A coating concert with the gut microbiota to induce spontaneous and maternally trans- identifies colitogenic bacteria in inflammatory bowel disease. Cell 158: 1000–1010. mitted colitis. Cell Host Microbe 8: 292–300. 71. Kau, A. L., J. D. Planer, J. Liu, S. Rao, T. Yatsunenko, I. Trehan, M. J. Manary, 43. Elinav, E., T. Strowig, A. L. Kau, J. Henao-Mejia, C. A. Thaiss, C. J. Booth, T. C. Liu, T. S. Stappenbeck, K. M. Maleta, et al. 2015. Functional character- D. R. Peaper, J. Bertin, S. C. Eisenbarth, J. I. Gordon, and R. A. Flavell. 2011. ization of IgA-targeted bacterial taxa from undernourished Malawian children that NLRP6 inflammasome regulates colonic microbial ecology and risk for colitis. Cell produce diet-dependent enteropathy. Sci. Transl. Med. 7: 276ra24. 145: 745–757. 72. Bunker, J. J., T. M. Flynn, J. C. Koval, D. G. Shaw, M. Meisel, B. D. McDonald, 44. Couturier-Maillard, A., T. Secher, A. Rehman, S. Normand, A. De Arcangelis, I. E. Ishizuka, A. L. Dent, P. C. Wilson, B. Jabri, et al. 2015. Innate and adaptive R. Haesler, L. Huot, T. Grandjean, A. Bressenot, A. Delanoye-Crespin, et al. humoral responses coat distinct commensal bacteria with immunoglobulin A. 2013. NOD2-mediated dysbiosis predisposes mice to transmissible colitis and Immunity 43: 541–553. colorectal cancer. J. Clin. Invest. 123: 700–711. 73. Strauch, U. G., F. Obermeier, N. Grunwald, S. Gurster,€ N. Dunger, M. Schultz, 45. Ivanov, I. I., K. Atarashi, N. Manel, E. L. Brodie, T. Shima, U. Karaoz, D. Wei, D. P. Griese, M. Ma¨hler, J. Scho¨lmerich, and H. C. Rath. 2005. Influence of K. C. Goldfarb, C. A. Santee, S. V. Lynch, et al. 2009. Induction of intestinal intestinal bacteria on induction of regulatory T cells: lessons from a transfer model Th17 cells by segmented filamentous bacteria. Cell 139: 485–498. of colitis. Gut 54: 1546–1552. 46. Sano, T., W. Huang, J. A. Hall, Y. Yang, A. Chen, S. J. Gavzy, J. Y. Lee, 74. Tanoue, T., K. Atarashi, and K. Honda. 2016. Development and maintenance of J. W. Ziel, E. R. Miraldi, A. I. Domingos, et al. 2015. An IL-23R/IL-22 circuit intestinal regulatory T cells. Nat. Rev. Immunol. 16: 295–309. regulates epithelial serum amyloid A to promote local effector Th17 responses. 75. Round, J. L., S. M. Lee, J. Li, G. Tran, B. Jabri, T. A. Chatila, and [Published erratum appears in 2016 Cell 164(1–2): 324.] Cell 163: 381–393. S. K. Mazmanian. 2011. The Toll-like receptor 2 pathway establishes colonization 47. Atarashi, K., T. Tanoue, M. Ando, N. Kamada, Y. Nagano, S. Narushima, by a commensal of the human microbiota. Science 332: 974–977. W. Suda, A. Imaoka, H. Setoyama, T. Nagamori, et al. 2015. Th17 cell induction 76. Faith, J. J., P. P. Ahern, V. K. Ridaura, J. Cheng, and J. I. Gordon. 2014. by adhesion of microbes to intestinal epithelial cells. Cell 163: 367–380. Identifying gut microbe-host phenotype relationships using combinatorial com- 48. Coccia, M., O. J. Harrison, C. Schiering, M. J. Asquith, B. Becher, F. Powrie, and munities in gnotobiotic mice. Sci. Transl. Med. 6: 220ra11. K. J. Maloy. 2012. IL-1b mediates chronic intestinal inflammation by promoting 77. Ochoa-Repa´raz, J., D. W. Mielcarz, Y. Wang, S. Begum-Haque, S. Dasgupta, the accumulation of IL-17A secreting innate lymphoid cells and CD4(+) Th17 D. L. Kasper, and L. H. Kasper. 2010. A polysaccharide from the human com- cells. J. Exp. Med. 209: 1595–1609. mensal Bacteroides fragilis protects against CNS demyelinating disease. Mucosal 49. Sonnenberg, G. F., and D. Artis. 2012. Innate lymphoid cell interactions with Immunol. 3: 487–495. microbiota: implications for intestinal health and disease. Immunity 37: 601–610. 78. Smith, P. M., M. R. Howitt, N. Panikov, M. Michaud, C. A. Gallini, M. Bohlooly-Y, 50. Lupp, C., M. L. Robertson, M. E. Wickham, I. Sekirov, O. L. Champion, J. N. Glickman, and W. S. Garrett. 2013. The microbial metabolites, short-chain fatty E. C. Gaynor, and B. B. Finlay. 2007. Host-mediated inflammation disrupts the acids, regulate colonic Treg cell homeostasis. Science 341: 569–573. intestinal microbiota and promotes the overgrowth of Enterobacteriaceae. Cell Host 79. Arpaia, N., C. Campbell, X. Fan, S. Dikiy, J. van der Veeken, P. deRoos, H. Liu, Microbe 2: 119–129. J. R. Cross, K. Pfeffer, P. J. Coffer, and A. Y. Rudensky. 2013. Metabolites 51. Winter, S. E., M. G. Winter, M. N. Xavier, P. Thiennimitr, V. Poon, produced by commensal bacteria promote peripheral regulatory T-cell generation. A. M. Keestra, R. C. Laughlin, G. Gomez, J. Wu, S. D. Lawhon, et al. 2013. Host- Nature 504: 451–455. derived nitrate boosts growth of E. coli in the inflamed gut. Science 339: 708–711. 80. Furusawa, Y., Y. Obata, S. Fukuda, T. A. Endo, G. Nakato, D. Takahashi, 52. Wu, H. J., and E. Wu. 2012. The role of gut microbiota in immune homeostasis Y. Nakanishi, C. Uetake, K. Kato, T. Kato, et al. 2013. Commensal microbe-derived and autoimmunity. Gut Microbes 3: 4–14. butyrate induces the differentiation of colonic regulatory T cells. Nature 504: 446–450. The Journal of Immunology 571

81. Richards, J. L., Y. A. Yap, K. H. McLeod, C. R. Mackay, and E. Marinõ. 2016. 101. O’Toole, A., and A. C. Moss. 2015. Optimizing biologic agents in ulcerative Dietary metabolites and the gut microbiota: an alternative approach to control colitis and Crohn’s disease. Curr. Gastroenterol. Rep. 17: 32. inflammatory and autoimmune diseases. Clin. Transl. Immunology 5: e82. 102. Simon, E. G., S. Ghosh, M. Iacucci, and G. W. Moran. 2016. Ustekinumab for 82. Sonnenberg, G. F., and D. Artis. 2015. Innate lymphoid cells in the initiation, the treatment of Crohn’s disease: can it find its niche? Therap. Adv. Gastroenterol. regulation and resolution of inflammation. Nat. Med. 21: 698–708. 9: 26–36. 83. Klose, C. S., and D. Artis. 2016. Innate lymphoid cells as regulators of immunity, 103. Goldberg, R., N. Prescott, G. M. Lord, T. T. MacDonald, and N. Powell. 2015. inflammation and tissue homeostasis. Nat. Immunol. 17: 765–774. The unusual suspects–innate lymphoid cells as novel therapeutic targets in IBD. 84. Pickard, J. M., C. F. Maurice, M. A. Kinnebrew, M. C. Abt, D. Schenten, Nat. Rev. Gastroenterol. Hepatol. 12: 271–283. T. V. Golovkina, S. R. Bogatyrev, R. F. Ismagilov, E. G. Pamer, P. J. Turnbaugh, 104. Perry, J. S., S. Han, Q. Xu, M. L. Herman, L. B. Kennedy, G. Csako, and and A. V. Chervonsky. 2014. Rapid fucosylation of intestinal epithelium sustains B. Bielekova. 2012. Inhibition of LTi cell development by CD25 blockade is host-commensal symbiosis in sickness. Nature 514: 638–641. associated with decreased intrathecal inflammation in multiple sclerosis. Sci. 85. Pham, T. A., S. Clare, D. Goulding, J. M. Arasteh, M. D. Stares, H. P. Browne, Transl. Med. 4: 145ra106. J. A. Keane, A. J. Page, N. Kumasaka, L. Kane, et al; Sanger Mouse Genetics 105. Villanova, F., B. Flutter, I. Tosi, K. Grys, H. Sreeneebus, G. K. Perera, Project. 2014. Epithelial IL-22RA1-mediated fucosylation promotes intestinal A. Chapman, C. H. Smith, P. Di Meglio, and F. O. Nestle. 2014. Character- colonization resistance to an opportunistic pathogen. Cell Host Microbe 16: 504– ization of innate lymphoid cells in human skin and blood demonstrates increase of 516. NKp44+ ILC3 in psoriasis. J. Invest. Dermatol. 134: 984–991. 86. Goto, Y., T. Obata, J. Kunisawa, S. Sato, I. I. Ivanov, A. Lamichhane, 106. Withers, D. R., M. R. Hepworth, X. Wang, E. C. Mackley, E. E. Halford, N. Takeyama, M. Kamioka, M. Sakamoto, T. Matsuki, et al. 2014. Innate E. E. Dutton, C. L. Marriott, V. Brucklacher-Waldert, M. Veldhoen, J. Kelsen, lymphoid cells regulate intestinal epithelial cell glycosylation. Science 345: et al. 2016. Transient inhibition of ROR-gt therapeutically limits intestinal in- 1254009. flammation by reducing TH17 cells and preserving group 3 innate lymphoid cells. 87. Sonnenberg, G. F., L. A. Monticelli, T. Alenghat, T. C. Fung, N. A. Hutnick, Nat. Med. 22: 319–323. J. Kunisawa, N. Shibata, S. Grunberg, R. Sinha, A. M. Zahm, et al. 2012. Innate 107. Olle, B. 2013. Medicines from microbiota. Nat. Biotechnol. 31: 309–315. lymphoid cells promote anatomical containment of lymphoid-resident commensal 108. Marchesi, J. R., D. H. Adams, F. Fava, G. D. Hermes, G. M. Hirschfield, bacteria. Science 336: 1321–1325. G. Hold, M. N. Quraishi, J. Kinross, H. Smidt, K. M. Tuohy, et al. 2016. The gut 88. Fung, T. C., N. J. Bessman, M. R. Hepworth, N. Kumar, N. Shibata, D. Kobuley, microbiota and host health: a new clinical frontier. Gut 65: 330–339.

K. Wang, C. G. Ziegler, J. Goc, T. Shima, et al. 2016. Lymphoid-tissue-resident 109. Scott, K. P., J. M. Antoine, T. Midtvedt, and S. van Hemert. 2015. Manipulating Downloaded from commensal bacteria promote members of the IL-10 cytokine family to establish the gut microbiota to maintain health and treat disease. Microb. Ecol. Health Dis. mutualism. Immunity 44: 634–646. 26: 25877. 89. Zheng, Y., P. A. Valdez, D. M. Danilenko, Y. Hu, S. M. Sa, Q. Gong, 110. Olendzki, B. C., T. D. Silverstein, G. M. Persuitte, Y. Ma, K. R. Baldwin, and A. R. Abbas, Z. Modrusan, N. Ghilardi, F. J. de Sauvage, and W. Ouyang. 2008. D. Cave. 2014. An anti-inﬂammatory diet as treatment for inﬂammatory bowel Interleukin-22 mediates early host defense against attaching and effacing bacterial disease: a case series report. Nutr. J. 13: 5. pathogens. Nat. Med. 14: 282–289. 111. Yao, J., R. A. Carter, G. Vuagniaux, M. Barbier, J. W. Rosch, and C. O. Rock. 90. Sonnenberg, G. F., L. A. Monticelli, M. M. Elloso, L. A. Fouser, and D. Artis. 2016. A pathogen-selective antibiotic minimizes disturbance to the microbiome. 2011. CD4(+) lymphoid tissue-inducer cells promote innate immunity in the gut. Antimicrob. Agents Chemother. 60: 4264–4273.

Immunity 34: 122–134. 112. Ross, J. J., P. E. Boucher, S. P. Bhattacharyya, D. J. Kopecko, E. M. Sutkowski, http://www.jimmunol.org/ 91. Lamas, B., M. L. Richard, V. Leducq, H. P. Pham, M. L. Michel, G. Da Costa, P. J. Rohan, D. K. Chandler, and J. Vaillancourt. 2008. Considerations in the devel- C. Bridonneau, S. Jegou, T. W. Hoffmann, J. M. Natividad, et al. 2016. CARD9 opment of live biotherapeutic products for clinical use. Curr.IssuesMol.Biol.10: 13–16. impacts colitis by altering gut microbiota metabolism of tryptophan into aryl 113. Borody, T. J., and A. Khoruts. 2011. Fecal microbiota transplantation and hydrocarbon receptor ligands. Nat. Med. 22: 598–605. emerging applications. Nat. Rev. Gastroenterol. Hepatol. 9: 88–96. 92. Hepworth, M. R., L. A. Monticelli, T. C. Fung, C. G. Ziegler, S. Grunberg, 114. Moayyedi, P., M. G. Surette, P. T. Kim, J. Libertucci, M. Wolfe, C. Onischi, R. Sinha, A. R. Mantegazza, H. L. Ma, A. Crawford, J. M. Angelosanto, et al. D. Armstrong, J. K. Marshall, Z. Kassam, W. Reinisch, and C. H. Lee. 2015. Fecal 2013. Innate lymphoid cells regulate CD4+ T-cell responses to intestinal com- microbiota transplantation induces remission in patients with active ulcerative mensal bacteria. Nature 498: 113–117. colitis in a randomized controlled trial. Gastroenterology 149: 102–109.e6. 93. Hepworth, M. R., T. C. Fung, S. H. Masur, J. R. Kelsen, F. M. McConnell, 115. Anderson, J. L., R. J. Edney, and K. Whelan. 2012. Systematic review: faecal J. Dubrot, D. R. Withers, S. Hugues, M. A. Farrar, W. Reith, et al. 2015. Immune microbiota transplantation in the management of inflammatory bowel disease. tolerance. Group 3 innate lymphoid cells mediate intestinal selection of com- Aliment. Pharmacol. Ther. 36: 503–516. mensal bacteria-specific CD4+ T cells. Science 348: 1031–1035. 116. Suskind, D. L., M. J. Brittnacher, G. Wahbeh, M. L. Shaffer, H. S. Hayden, by guest on January 26, 2017 94. Koch, M. A., G. L. Reiner, K. A. Lugo, L. S. Kreuk, A. G. Stanbery, E. Ansaldo, X. Qin, N. Singh, C. J. Damman, K. R. Hager, H. Nielson, and S. I. Miller. 2015. T. D. Seher, W. B. Ludington, and G. M. Barton. 2016. Maternal IgG and IgA Fecal microbial transplant effect on clinical outcomes and fecal microbiome in antibodies dampen mucosal T helper cell responses in early life. Cell 165: 827– active Crohn’s disease. Inflamm. Bowel Dis. 21: 556–563. 841. 117. Rossen, N. G., S. Fuentes, M. J. van der Spek, J. G. Tijssen, J. H. Hartman, 95.Zeng,M.Y.,D.Cisalpino,S.Varadarajan,J.Hellman,H.S.Warren, A. Duflou, M. Lowenberg, G. R. van den Brink, E. M. Mathus-Vliegen, W. M. de M. Cascalho, N. Inohara, and G. Nuń˜ez. 2016. Gut microbiota-induced immu- Vos, et al. 2015. Findings from a randomized controlled trial of fecal transplan- noglobulin G controls systemic infection by symbiotic bacteria and pathogens. tation for patients with ulcerative colitis. Gastroenterology 149: 110–118.e4. Immunity 44: 647–658. 118. Gensollen, T., S. S. Iyer, D. L. Kasper, and R. S. Blumberg. 2016. How coloni- 96. Diehl, G. E., R. S. Longman, J. X. Zhang, B. Breart, C. Galan, A. Cuesta, zation by microbiota in early life shapes the immune system. Science 352: 539–544. S. R. Schwab, and D. R. Littman. 2013. Microbiota restricts trafficking of bacteria 119. Kim, K. S., S. W. Hong, D. Han, J. Yi, J. Jung, B. G. Yang, J. Y. Lee, M. Lee, and to mesenteric lymph nodes by CX(3)CR1(hi) cells. Nature 494: 116–120. C. D. Surh. 2016. Dietary antigens limit mucosal immunity by inducing regula- 97. Knoop, K. A., K. G. McDonald, D. H. Kulkarni, and R. D. Newberry. 2016. tory T cells in the small intestine. Science 351: 858–863. Antibiotics promote inflammation through the translocation of native commensal 120. Ohnmacht, C., J. H. Park, S. Cording, J. B. Wing, K. Atarashi, Y. Obata, V. Gaboriau- colonic bacteria. Gut 65: 1100–1109. Routhiau, R. Marques, S. Dulauroy, M. Fedoseeva, et al. 2015. The microbiota reg- 98. Spadoni, I., E. Zagato, A. Bertocchi, R. Paolinelli, E. Hot, A. Di Sabatino, ulates type 2 immunity through RORgt+ T cells. Science 349: 989–993. F. Caprioli, L. Bottiglieri, A. Oldani, G. Viale, et al. 2015. A gut-vascular barrier 121. Sefik, E., N. Geva-Zatorsky, S. Oh, L. Konnikova, D. Zemmour, A. M. McGuire, controls the systemic dissemination of bacteria. Science 350: 830–834. D. Burzyn, A. Ortiz-Lopez, M. Lobera, J. Yang, et al. 2015. Individual intestinal 99. Coutinho, A. E., and K. E. Chapman. 2011. The anti-inflammatory and immu- symbionts induce a distinct population of RORg+ regulatory T cells. Science 349: nosuppressive effects of glucocorticoids, recent developments and mechanistic 993–997. insights. Mol. Cell. Endocrinol. 335: 2–13. 122. Olszak, T., D. An, S. Zeissig, M. P. Vera, J. Richter, A. Franke, J. N. Glickman, 100. Geem, D., A. Harusato, K. Flannigan, and T. L. Denning. 2015. Harnessing R. Siebert, R. M. Baron, D. L. Kasper, and R. S. Blumberg. 2012. Microbial regulatory T cells for the treatment of inflammatory bowel disease. Inflamm. Bowel exposure during early life has persistent effects on natural killer T cell function. Dis. 21: 1409–1418. Science 336: 489–493. Th eJournal of Brief Reviews Immunology

Microbiome-Modulated Metabolites at the Interface of Host Immunity Eran Blacher,1 Maayan Levy,1 Evgeny Tatirovsky,1 and Eran Elinav The mammalian gastrointestinal tract and associated development and function. Conversely, nutritional compo- mucosal immune system harbor a large repertoire of sition tremendously impacts the gut microbial composition metabolites of prokaryotic and eukaryotic origin that and function (1). Other metabolites are endogenously pro- play important roles in eukaryotic development and duced or modified through diverse metabolic processes, by physiology. These often bioactive small molecules orig- the host or its microbiota (2). Collectively, it is estimated inate from nutrition- and environmental-related sour- that more than 50% of fecal and urinary metabolites are ces, or are endogenously produced and modulated by derived from or modified by the gut microbiome (3). Some microbiome-associated metabolites are bioactive and the host and its microbiota. A complex network of in- Downloaded from teractions exists between the intestinal mucosal immune affect the host cellular processes including differentiation, system and the microbiota. This intimate cross-talk may migration, proliferation, and apoptosis, thereby featuring be driven by metabolite secretion and signaling, and fea- pleotropic physiological or pathophysiological effects on the eukaryotic host. A number of metabolites impact mucosal and tures profound influences on host immunity and phys- systemic immune maturation and function, in steady state and iology, including the endocrine, metabolic, and nervous during disease (4–8). The host has evolved multiple metab- http://www.jimmunol.org/ system function in health and disease. Alterations in olite sensing platforms, and downstream immune signaling microbiome-associated metabolite levels and activity pathways that confer reactivity to microbiome-modulated are implicated in the pathogenesis of a growing number metabolites (9, 10). These sensing platforms are expressed of illnesses. In this review we discuss the origin and in different combinations in mucosal cellular subsets, such as influence of microbiome-modulated metabolites, with intestinal epithelial cells, macrophages, dendritic cells (DCs), an emphasis on immune cell development and func- T cells, and innate lymphoid cells (ILCs), where they play tion. We further highlight the emerging data potentially critical roles in host-microbiome mutualistic cross-talk. In implicating metabolite misbalance with host-microbiome– addition to local metabolite effects on gastrointestinal mucosal associated disease. The Journal of Immunology, 2017, immune function, many systemically absorbed metabolites may by guest on January 26, 2017 198: 572–580. reach remote organs and modulate the immune responses in sterile host regions such as the CNS (11, 12). As such, metab- he gut microbiome is a microbial ecosystem that has olites may provide a missing link between the gut microbiome diverse effects on physiological host functions, par- compositional and functional configuration to its remote effects T ticularly immune development and activity. The on host physiology and disease risk in seemingly unrelated sterile molecular basis of host-microbiome interactions is only just organs. The type, composition, and concentration of metabolites beginning to be unraveled, and is mediated by both cell to cell coupled with the host sensor molecules repertoire orchestrate the interactions and the production, modification, and sensing of net physiological response at a given physiological context. a large variety of bioactive small molecules, termed metabo- In this review we provide an overview of gut microbiome- lites. Many gastrointestinal metabolites originate from dietary modulated metabolites, their physiological effects on major or environmental sources. They encounter the gut micro- immune functions, and recent observations possibly linking biome as part of the digestion and intestinal transit process, metabolite misbalances with risk of immune-mediated and and link host nutrition with physiology, including immune immune-associated disease.

Department of Immunology, Weizmann Institute of Science, Rehovot 76100, Israel Career Development Chair and is a senior fellow at the Canadian Institute for Advanced Research. 1E.B., M.L., and E.T. contributed equally to this work. Address correspondence and reprint requests to Prof. Eran Elinav, Department of Im- ORCIDs: 0000-0003-3207-0153 (E.B.); 0000-0001-6476-9183 (E.T.). munology, Weizmann Institute of Science, Herzl Street 100, Rehovot 76100, Israel. Received for publication July 18, 2016. Accepted for publication October 13, 2016. E-mail address: [email protected] E.E. was supported by Y. and R. Ungar, the Abisch Frenkel Foundation for the Promotion Abbreviations used in this article: AHR, aryl hydrocarbon receptor; ASD, autistic spectrum of Life Sciences, the Gurwin Family Fund for Scientific Research, the Leona M. and Harry disorder; BBB, blood–brain barrier; DC, dendritic cell; FXR, farnesoid X receptor; B. Helmsley Charitable Trust, the Crown Endowment Fund for Immunological Research, HDAC, histone deacetylase; IBD, inflammatory bowel disease; IEC, intestinal epithelial the estate of J. Gitlitz, the estate of L. Hershkovich, the Benoziyo Endowment Fund for the cell; ILC, innate lymphoid cell; NA, nicotinic acid; NAFLD, nonalcoholic fatty liver Advancement of Science, the Adelis Foundation, J.L. and V. Schwartz, A. and G. Markovitz, disease; NASH, nonalcoholic steatohepatitis; SCFA, short chain fatty acid; SFB, segmented A. and C. Adelson, the French National Center for Scientific Research, D.L. Schwarz, the filamentous bacteria; TMAO, trimethylamine-N-oxide; Treg, regulatory T cell; Trp, tryp- V.R. Schwartz Research Fellow Chair, L. Steinberg, J.N. Halpern, A. Edelheit, grants tophan; WT, wild type. funded by the European Research Council, a Marie Curie Career Integration grant, the German–Israeli Foundation for Scientific Research and Development, the Israel Science Foun- Copyright Ó 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00 dation, the Minerva Foundation, the Rising Tide Foundation, the Helmholtz Association, and the European Foundation for the Study of Diabetes. E.E. is the incumbent Rina Gudinski

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601247 The Journal of Immunology 573

Microbiome modulation of the gastrointestinal metabolite produce SCFAs that locally reach a millimolar concentration (19). configuration Additional fermentation byproducts like succinate and lactate are One of the essential functions provided by the gut microbiota also generated during SCFAs production, and are used by the includes the modulation of gastrointestinal metabolites, in- microbiota for maintenance and survival (20). SCFAs, in turn, cluding their synthesis, digestion, fermentation, and secondary mediate a number of important functions for the eukaryotic metabolism. Numerous studies have investigated the forces host, including usage as an energy source by intestinal epithelial that drive the development of the adult microbial population cells, and a variety of anti-inflammatory properties in T cells, in the intestine from birth (13, 14), yet the dynamics of the regulatory T cells (Tregs), neutrophils, and macrophages, where microbiome impact on the host metabolome remain largely they affect migration, cytolytic activity, cytokine production, unknown. The centrality of microbial activity in shaping the and epigenetic regulation of gene expression (discussed below). gut metabolome is demonstrated by profound metabolite Another example of bioactive de novo microbial-synthesized alterations found in germ-free as compared with gnotobiotic metabolites are vitamins, small bioactive nutrients that are and colonized animals, including differences in diverse me- extracted from the diet and may be modulated by the tabolite biochemical groups noted in feces, urine, and the microbiome or de novo produced by intestinal commensals. systemic circulation (15–18). Multiple commensal microbial One such important example involves vitamin K, a central taxonomies, including Enterobacteriaceae, Enterococus spp., cofactor regulating the mammalian coagulation cascade (21) and Lactobacillus spp., have been shown to influence gastro- and immunity (22). The vitamin K group consists of vitamin intestinal metabolite concentrations (17), by processes span- K1 (phylloquinone), derived from food, and vitamin K2 Downloaded from ning from utilization of amino acids as a nitrogen source, to (menaquinone), which can be also produced from vitamin K1 byproduct generation of secondary metabolites such as short by most of the gut microbiota species, including Enterobacter chain fatty acids (SCFAs). In the following section we will sp., Eubacterium lentum, Veillonella sp. and Bacteroides sp (23). highlight key examples of these microbiome effects on dietary Members of the vitamin K group are absorbed in the small and endogenously generated gut metabolites, and how these intestine in a process requiring bile salts (24). The importance microbiome effects contribute to the downstream effector of microbial production of vitamin K to its overall intestinal http://www.jimmunol.org/ functions of these metabolites on host immunity (Figs. 1, 2). pool is supported by the fact that broad-spectrum antibiotic De novo microbiome production of metabolites. The gut microbiota treatment induces significantly reduced vitamin K levels (25). produces a number of metabolites, some of which feature distinct Furthermore, a primary vitamin K deficiency animal model bioactivefunctionsontheeukaryotichost.Aprimeexampleof was not successfully generated using a vitamin K–deficient such de-novo synthesized metabolites include SCFAs, which are diet alone, unless germ-free mice were employed in the process, the products of microbial fermentation of nondigestible nutritional further supporting the important role of the gut microbiome in fibers, and include acetate (C2), propionate (C3), and butyrate its processing (26). However, the transformation of vitamin K1

(C4). Gut commensal bacteria, such as those belonging to to vitamin K2 may not be solely dependent on the microbiota, by guest on January 26, 2017 the genera Butyrivibrio, Clostridium,andEubacterium,can as it occurs in gnotobiotic rats (27).

FIGURE 1. Microbiota-associated metabolites shape mucosal immunity. Metabolites participate in a complex host-microbiome network of com- munications orchestrating immune responses. The most well-studied metabolites SCFA and NA are suggested to affect multiple facets of the immune response, such as macrophage and DC function and cytokine secretion, Treg differentiation, intestinal goblet cell mucin secretion, inﬂammasome- mediated IL-18 activation, and neutrophil chemotaxis via the NF-kBpathway. 574 BRIEF REVIEWS: METABOLITES, MICROBIOME, AND IMMUNITY

Other vitamins, such as vitamin B1 (thiamine) (28) and folic One recently discovered role of microbiome-modulated acid (vitamin B9) (29), may also be synthesized by some metabolites on the intestinal immune response involves sig- members of the microbiome, thereby contributing to the naling through the Nlrp6 inflammasome. The bile acid related overall vitamin pull. Vitamin B12 can be produced by bac- metabolite taurine was suggested to shape the host-microbiota teria, e.g., Propionibacterium freudenreichii, Salmonella enter- interface through activation of Nlrp6 inflammasome signal- ica, Listeria innocua,andLactobacillus reuteri, and may also be ing, resulting in intestinal epithelial cell IL-18 secretion and degraded by multiple members of the microbiome (30). Its downstream modulation of anti-microbial peptide transcrip- binding to the eukaryotic intrinsic factor as part of the di- tion (43), thereby impacting the microbiome composition gestion process is believed to protect it from microbial deg- and risk of auto-inflammation. The microbiome-modulated radation in the distal small intestine (30). metabolites histamine and spermine, in turn, inhibit the Microbiome modulation of nutritionally derived metabolites. Mul- NLRP6 inflammasome signaling, suggesting a mechanism by tiple bioactive gastrointestinal small molecules originate from which the combination of bioactive metabolites at given dietary sources, undergo microbial modifications and have contexts drive the host-microbiome interface thorough regu- recently been shown to feature important immune functions. lation of immune sensing and downstream immune modu- One example is the immunomodulatory amino acid trypto- lation. Collectively, postbiotic metabolite intervention may phan (Trp), abundantly found in foods such as milk, eggs, red target a variety of host-related pathways, including those ex- meat, and vegetables (e.g., broccoli). Trp is known to undergo emplified above, representing a promising potential new ther- metabolism by Lactobacilli, giving rise to indole-3-aldehyde, apeutic modality of microbiome-related metabolic and Downloaded from which can bind to the aryl hydrocarbon receptor (AHR), inflammatory disorders. followed by its transport through the epithelial cell layer by a transporter containing angiotensin I converting enzyme 2 Mechanisms of microbiota-modulated metabolite immune regulation (9, 31). Deficiency in murine angiotensin I converting enzyme Microbiome-modulated metabolites may impact the host 2, which controls the levels of neutral amino acids in the intestine, immune response through several mechanisms that are only just beginning to be unraveled. In addition to metabolite-

including Trp, results in higher susceptibility for intestinal http://www.jimmunol.org/ inflammation and an altered gut microbial composition induced signaling in immune cell subsets that may trigger a (32). Inflammation in this context is transmissible through cascade of inflammatory changes, recent studies have identified transfer of the microbiota to germ-free mice, whereas dietary direct metabolite-mediated effects on immune cell meta- Trp replenishment rescues microbial dysbiosis (32). bolism, often resulting in substantial, previously unappreci- Microbiome modulation of host-derived metabolites. The micro- ated, functional outcomes. biome may also modulate metabolites that are produced by the Immune cell metabolic reprogramming. One of the recently eukaryotic host. An important example of such endogenously characterized roles of commensal microbiota is to provide generated gastrointestinal metabolites are bile acids, which are energy to intestinal epithelial cells (IECs), through fermen- produced from cholesterol by the liver, further conjugated to tation of dietary fibers by degradation of undigested complex by guest on January 26, 2017 glycine (in humans) or taurine (in mice) and transported to the carbohydrates into SCFAs. In the absence of the microbiota gallbladder, common bile duct, and the proximal small in- and associated SCFAs, germ-free mice feature an altered energy testine (33, 34). They are considered essential for solubilizing metabolism characterized by preferential fermentation of dietary fat and cholesterol, thereby accelerating digestion and glucose into lactate (44), thereby leading to enhanced IEC absorption. autophagy as a consequence of nutrient starvation (45, 46). Secondary microbial-mediated metabolism of primary to Monoinoculation of germ-free mice with the butyrate-producing secondary bile acids increases their diversity (35), as exem- Butyrivibrio fibrisolvens rescued IECs from autophagy and plified by germ-free mice having lower levels and reduced mitochondrial respiration insufficiency (45). Similar to epithe- diversity of secondary bile acids as compared with colonized lial cells, energy metabolism in leukocytes is affected by microbiota- mice (36). In the intestinal epithelium, many different bile regulated metabolites. T cell responses are dependent on acids, including microbiome-modulated deoxycholic acid and nutrient catabolism, whereas intestinal T cells in germ-free mice lithocholic acid (37), can bind the G-protein coupled receptor present an immature phenotype characterized by impaired TGR5 as well as the nuclear receptor farnesoid X receptor cytolytic activity (47, 48). (FXR). FXR was suggested to regulate epithelial cell integrity The molecular mechanisms by which SCFAs, such as buty- and bacterial composition because in the absence of FXR, rate, serve as energy sources and consequent functional mod- mice present a disrupted epithelial barrier function and dys- ulators of intestinal epithelial cells and immune cells are biosis (38). In addition, secondary bile acid signaling may beginning to be unraveled (45). Butyrate acts as a local substrate impact the host immune response by modulating proin- for the production of energy through the tricarboxylic acid flammatory genes through NF-kB signaling (39, 40). As such, cycle, ATP generation, and b-oxidation as well as through the the addition of an FXR ligand to LPS-treated macrophages suppression of autophagy in intestinal epithelial cells (45). downregulated the NF-kB–controlled genes IL-1b, TNF-a, 2 2 COX-1, COX-2, and iNOS. Moreover, Fxr / mice pre- Transcriptional and epigenetic modulation of immune-related genes sented a significant exacerbation of colitis symptoms, which Another key mechanism by which gut microbiome–modulated could be mitigated by treating wild-type (WT) mice with an metabolites influence the immune response involves regula- FXR analog (40). Furthermore, bile acid signaling was re- tion of immune cell transcriptional programming through cently suggested to contribute to the effects of the microbiota impacting their epigenetic landscape. One such example re- on the metabolic syndrome, which may partially stem from its lates to the transcription of mucin-related genes by butyrate, immune modulatory effects (41, 42). thereby contributing to goblet cell differentiation and mucus The Journal of Immunology 575

FIGURE 2. Bile acid effects on the host-microbiota interface and immune functions. Bile acids are produced by the liver, secreted into the proximal small intestine, and modulated by the microbiome, thereby impacting multiple immune processes. Bile acids bind FXR on intestinal epithelial cells, through which they regulate barrier integrity and bacterial community structure. In macrophages, bile acid signaling downregulates NF-kB–controlled proinflammatory responses. The bile acid metabolite taurine shapes the host-microbiota interface through activation of Nlrp6 inflammasome signaling, resulting in intestinal epithelial cell IL-18 secretion and downstream modulation of antimicrobial peptide transcription. Histamine and spermine negatively modulate the Nlrp6 inflammasome. Downloaded from production (49–51), which are essential in host innate im- Some microbiome-mediated transcriptional and epigenetic mune defense against invading pathogens. Furthermore, his- modifications of the host may involve metabolite signaling, yet tone deacetylase (HDAC) inhibitory activity of butyrate such metabolites and their putative roles in these modifications induces gene expression changes in DCs including the sup- have not been identified to date. A notable example of genome- pression of IL-12, IL-6, and Relb, which contributes to Treg wide epigenetic effects mediated by the gut microbiome and differentiation (52). Butyrate also plays a role in modulating impacting the mucosal immune response involves ILCs (64). http://www.jimmunol.org/ the immune response in macrophages, by inhibiting tran- The transcriptional landscape of ILC subsets has been re- scription of proinflammatory mediators in colon lamina propria cently shown to involve a complex and subset-specific epigenetic macrophages such as Il6, Nos2, and Il12 (53), a mechanism modification program. Importantly, microbiota depletion by contributing to the maintenance of tolerance toward com- broad-spectrum antibiotic treatment or in germ-free mice results mensal bacteria. in massive restructuring of the global epigenetic landscape of In addition to its possible effects on cell metabolism, bu- small intestinal ILCs, involving several thousand histone modi- tyrate was shown to influence peripheral Tregs through in- fications at enhancer and promoter sites. Also, the impact of the creased histone H3 acetylation of the Foxp3 promoter region intestinal microbiota on altered gene expression in ILCs is dif- by guest on January 26, 2017 and at the intragenic enhancer elements conserved noncoding ferent between distinct, newly recognized, transcriptional subsets. DNA sequence 1, associated with gene expression (54, 55). As This suggests that within the small intestinal ILC population, such, butyrate was demonstrated to indirectly suppress co- different clusters of cells feature distinct microbiota responsive- lonic inflammation through GPR109A receptor-mediated ness patterns, possibly through defined sets of epigenetic modi- regulation of Treg differentiation (56). fications (64). Whether some of these microbiome-mediated The epigenetic suppressive activity of SCFAs may be epigenetic alterations in ILCs are a result of metabolite functions vertically transferrable to offspring, as high-fiber feeding of merit future studies. Another notable transcriptional outcome of pregnant female mice suppressed the expression of genes microbiome colonization that may involve a yet unidentified related to airway response in the fetal lung (57). Similarly, metabolite signaling is the microbiome-mediated modulation of SCFAs were also suggested to influence leukocyte function the TH17 immune response. Following colonization with seg- and neutrophil chemotaxis (58), through inhibition of HDAC mented filamentous bacteria (SFB), which are known to adhere activity leading to NF-kB inactivation and suppression of to intestinal mucosa in a host-species manner, TH17 levels of proinflammatory cytokines and nitric-oxide (58, 59). A sim- germ-free mice are restored (65), leading to induction of tran- ilar suppression of HDACs resulting in an anti-inflammatory scription of TH17 associated genes including Il17A, Il21, Ccr6, phenotype was observed in DCs and intestinal macrophages and Nos2, as well as serum amyloid A and anti-microbial pep- (53, 60, 61). tides (65). SFB-induced TH17 cells were shown to recognize In addition to being the receptor for butyrate, GPR109A is SFB Ags, through yet unrecognized mechanisms, whereas DC also the receptor for microbial-derived nicotinic acid (NA, expression of MHCII is necessary for TH17 cell induction (66). or niacin). The microbial-derived butyrate and NA engage Metabolite immune modulation through yet unknown mechanisms. GPR109A on epithelial cells, macrophages, and DCs and As metabolite-mediated functions on the host are only just trigger the secretion of IL-18 from epithelial cells, and IL-10 beginning to be discovered, some effects are not yet fully from DCs and macrophages, which further leads to sup- defined mechanistically. For example, some mechanisms of pression of inflammation as IL-10 favors the development of SCFAs modulatory effects on host immune functions have Tregs from naive CD4+ cells (56). The SCFA acetate can also been deciphered, and are described above, but others remain bind GPR43 on IECs, which leads to NLRP3 inflammasome cryptic and can be driven by intracellular metabolic pathways, assembly and the downstream production of IL-18 (62). As epigenetic modification, transcription regulation, or by other such, acetate was shown to promote epithelial cell integrity, yet undefined mechanism(s). Different SCFAs can be differ- which contributes to protection against infection (63). entially sensed through the receptors GPR41, GPR43, and 576 BRIEF REVIEWS: METABOLITES, MICROBIOME, AND IMMUNITY

GPR109A (10). Butyrate is the ligand for GPR109A, expressed pathologies primarily affecting the digestive system [such as by IECs and T cells, where its signaling induces substantial effects inflammatory bowel diseases (IBD) and colorectal cancer], on Treg abundance and activity, yet the mechanism of this effect metabolic diseases (diabetes mellitus and nonalcoholic fatty is still not completely understood (56). Likewise, colonic mac- liver disease), and even cardiovascular and neurologic disorders rophages and DCs from mice deficient in GPR109A are (68–73). As such, it is reasonable to assume that at least some defective in their ability to induce differentiation of naive of the effects of the microbiome may be mediated by altered T cells into IL-10–expressing Tregs, resulting in a general re- metabolite profiles associated with these microbial disease duction in colonic Tregs and increased severity of dextrane signatures. Such altered metabolite levels and combina- sodium sulfate–induced colitis, through mechanisms that tions may affect immune function and metabolism as well as are still elusive (56). Germ-free and colonized mice treated multiple other seemingly unrelated physiological functions. with SCFAs such as acetate feature reduced inflammation in a Although the complex network of interactions between the dextrane sodium sulfate–induced colitis model as well as in a microbiome, its associated metabolite profile and the im- T cell transfer colitis model in a GPR43-dependent manner. mune system in the context of disease is just beginning to be Mechanistically, SCFA administration increases the numbers unraveled, several studies, exemplified below, have identified of Treg cells in germ-free mice as well as the secretion of its some of the pathways underlying these interactions (Fig. 3). key suppressive effector cytokine IL-10, possibly through Inflammatory bowel disease. The incidence of IBDs, a group of effects on immune cell metabolism (54, 67). Furthermore, a 2 2 auto-inflammatory diseases involving different regions of the lack of GPR43 expression in Gpr43 / mice severely alters gastrointestinal tract, has been steadily increasing globally (70). Downloaded from SCFA-mediated expansion of colonic Tregs (54). Thus, SCFAs One of the characteristic associated hallmarks of IBD is intestinal and their GPRs represent one pathway through which the dysbiosis — a deviation from the healthy composition and commensal microbiota regulates the inflammatory response, function of the microbiome. Dysbiosis may contribute to the yet the intracellular and immune-mediated mechanisms of these impaired cross-talk between immune cells and the microbiome metabolite effects merit further studies. Similar to their local in IBD, and may involve aberrant signaling by different immu- effect in the intestine, circulating SCFAs dampen the severity of nomodulatory metabolites. For example, in an animal model http://www.jimmunol.org/ allergic airway inflammation with reduced levels of IL-4, IL-5, of ulcerative colitis, an interplay between the commensal IL-13, and IL-17a in the lung, possibly through increased bone microbiota and AHR-expressing group 3 ILCs is mediated marrow generation of macrophages and DCs; this and other by microbiota-produced metabolites (31, 74). The absence potential mechanisms merit further exploration (60). of AHR or its ligands induces changes in microbial composition, leading to exacerbated colitis development in mice Microbiome-modulated metabolites and the risk of disease (75). A subset of commensal bacteria utilizes Trp as an energy Defined compositional and functional microbiome alterations source and produces the metabolite indole-3-aldeyde, which

have been associated with the pathogenesis of common further activates AHR in ILCs, leading to secretion of IL-22 by guest on January 26, 2017 multifactorial diseases. These diseases include, among others, impacting both mucosal healing and the anti-microbial

FIGURE 3. Microbiome-modulated metabolites and disease. Altered levels of microbiome-modulated metabolites have been associated with immune-mediated and immune-associated disease risk. Metabolite-mediated effects on disease processes may be localized to the gastrointestinal tract, such as in IBD or food allergy, or systemically influence remote tissues like the brain, heart and the liver. The Journal of Immunology 577 peptides repertoire including lipocalin-2, S100A8, and S100A9 Similar to the above animal model, NAFLD patients were in mice (31, 74). Another possible metabolite-mediated mechanism found to have higher levels of TNF-a,IFN-g, and IL-6 and of auto-inflammation involves SCFA-mediated inhibition of disrupted microvilli morphology associated with increased gut HDACs, independently of their receptors GPR41 and GPR43, permeability (83). These results suggest that dysbiosis may not leading to T cell activation of the mTOR-S6K pathway (76). only affect the local intestinal inflammatory response, but may Acetate supplementation in drinking water was able to mitigate also systemically influence sterile tissues through metabolites anti-CD3–induced colitis severity in the ileum of WT but not driving pathological conditions, inflammatory and metabolic 2 2 of Il10 / mice, indicating that this protective effect requires alike. Other manifestations of the metabolic syndrome, such IL-10 (76). Conversely, oral butyrate treatment enhanced as obesity and glucose intolerance, were recently suggested to IL-23 levels in DCs and exacerbated colitis symptoms in an IBD involve an altered inflammatory response in tissues such as animal model (77). As the effects of SCFAs are complex and adipose tissue, the pancreas, and the liver (84). An intriguing, sometimes diverse, indirect, and combinatorial, future studies yet unexplored possibility is that some of these inflammatory will have to delineate their human relevance and therapeutic changes may be mediated by dysbiosis and its associated altered potential in different combinations and clinical contexts. metabolite configuration. Exploring metabolite roles in these Another example of the role of the microbiome-immune- contexts may enable to identify new therapeutic and diagnostic metabolite axis in driving intestinal auto-inflammation is pro- modalities to these common idiopathic multifactorial disorders. vided by the NLRP6 inflammasome model in mice. As described Cardiovascular disease. The catastrophic consequences of ath- above, a properly functional NLRP6 inflammasome signaling, erosclerosis, myocardial infarction, and cerebrovascular ac- Downloaded from orchestrated by the microbiome-modulated metabolites taurine, cident still account for the majority of deaths worldwide (85). histamine, and spermine (43) in mouse IEC, contributes to a Atherosclerotic plaques in arterial walls are caused by an imbal- healthy microbiome population. Conversely, NLRP6-deficient anced lipid homeostasis leading to accumulation of cholesterol– mice feature dysbiosis and associated intestinal auto-inflammation, containing low-density lipoprotein particles, accompanied by which is transmittable via fecal transplantation into normal chronic inflammation (86). In the past decade changes in WT mice, thereby inducing exacerbated colitis symptoms, which microbiome composition and function have been associated http://www.jimmunol.org/ are ameliorated by restoration of normal microbiota (43). Mul- with cardiovascular disease pathogenesis. One notable example tiple other unknown metabolites in mice and in human IBD of such an association involves trimethylamine-N-oxide (TMAO) patients may impact immune activation and function, thereby metabolism. TMAO is a metabolite that in humans is generated affecting the pathogenesis of different IBD subsets, and merit by microbial metabolism of dietary phosphatidylcholine or further studies. L-carnitine, which are abundant in meat and high-fat diets. Nonalcoholic fatty liver disease. Nonalcoholic fatty liver disease Mice fed L-carnitine featured increased levels of TMAO (NAFLD) is the most common hepatic disease in the developed leading to enhanced atherosclerosis. Germ-free mice or world, and is tightly associated with other features of metabolic

antibiotic-treated mice were protected from the disease, thereby by guest on January 26, 2017 syndrome including obesity, hyperlipidemia, and adult-onset linking the gut microbiota, nutrition, and risk for cardio- diabetes mellitus. Whereas most affected individuals remain vascular disease (87). Similar findings were observed in hu- asymptomatic, a significant minority of NAFLD cases develop mans, in which a time-dependent increase in plasma TMAO a progressive inflammatory liver disease termed nonalcoholic (and other phosphatidylcholine metabolites) was documented steatohepatitis (NASH), ultimately leading to liver dysfunction, following ingestion of isotopic-labeled food. These increased cirrhosis, and life-risking complications (78). The microbiome metabolite levels were inhibited by antibiotic treatment, and was suggested in both mouse models and humans to regulate reappeared after antibiotic withdrawal (88). Increased fasting some manifestations of NAFLDandNASHviaitssystemic TMAO plasma levels were significantly associated with major effects on the immune system (79, 80). In one such animal adverse cardiovascular events such as overall mortality, myo- model, inflammasome-deficient mice feature dysbiosis me- cardial infarction or stroke. These results demonstrate the diated by an altered metabolite profile (81). When induced potential central role of this microbiome-derived metabolite with NAFLD, these mice developed a context-specific form of dysbiosis characterized by a massive expansion of in the risk factor for cardiovascular disease (88, 89). Porphyromonadaceae accompanied with enhanced TLR4 and Food allergy. Allergies to nutritional components constitute a TLR9 ligands influxing into the portal circulation and leading common immune-mediated condition affecting the pediatric and to massive TNF-a secretion, hepatic inflammation, and the adult population alike. Mucosal DCs contribute to allergy to food progression to NASH (82). Interestingly, this microbiome- Ags by regulation of Treg differentiation. A high-fiber diet together mediated transition into NASH was abolished by wide-spectrum with vitamin A was recently shown to drive the microbiota toward + antibiotic treatment and transferrable by cohabitation of a configuration that is supportive for tolerogenic CD103 DCs, inflammasome-deficient mice with WT mice. As dysbiosis which provides tolerance toward food Ags in a mouse model, in this model has been recently linked to microbiome- thereby protecting from food allergy. This protective effect modulated metabolite misbalance, it hints toward the pos- depends on epithelial GPR43 and immune GPR109A, because 2/2 sibility of an indirect link existing between gastrointestinal the protective effect was not observed in Gpr43 or in 2/2 metabolite alterations, dysbiosis, and influx of microbial Gpr109a mice. In this model, changes in the microbiota products into the liver leading to inflammatory conseq- following a high-fiber diet were also associated with enhanced uences (43). Likewise, in NAFLD human patients, dysbiosis IgA production and T follicular helper response (90), further wasobservedwithanexpansionofStreptococcus, Anaerobacter, implicating the microbiome in this food allergy protective effect. Lactobacillus and Escherichia genera as compared with healthy Neurodegenerative disease. Although neurological disorders are subjects, accompanied by reduction of Alistipes and Prevotella. increasingly suggested to feature important immune components, 578 BRIEF REVIEWS: METABOLITES, MICROBIOME, AND IMMUNITY research on neurological immune modulation remains in its bioactivity of metabolites in the gastrointestinal tract and how infancy. Some of the observed immunomodulatory effects on the it may impact local and systemic inflammatory processes. In- CNS are mediated by resident immune cell subsets in the CNS, tegration of the accumulating knowledge on microbial com- which impact disease progression, and particularly the microglia, munity structure in different disease scenarios with the the resident myeloid cells of the CNS. Interestingly, microglia corresponding changes in the metabolome and its bioactivity were recently suggested to be affected by metabolites such as may enable addressing fundamental questions regarding the SCFAs in a mouse model. SCFAs secreted by the gut microbiota molecular mechanisms by which the microbiome impacts modulated microglial activation and maturation gene-expression physiology, pathophysiology and even its own community profile in the mouse steady state (91). The NA receptor function (97). Such studies will likely involve an integration of GPR109A was recently found to be expressed by microglia advanced next-generation sequencing related genomics, high- (92) and feature increased expression colocalizing with microglial throughput metabolomics, and gnotobiotic experimentation markers in the substantia nigra of Parkinson’s disease patients, that is critical in demonstrating causality and differentiating possibly participating in disease modulation (92). Likewise, between primary driver microbial impact on disease patho- b-hydroxybutyrate treatment given to a Parkinson’s disease genesis, and secondary passenger microbial changes. Such in- rat model improved motor skills and protected nigro-striatal tegrative studies may potentially yield novel microbiota-based neurons by reducing neuroinflammation (92). Addition of diagnostics and therapies of common disease. As such, colo- b-hydroxybutyrate to mesencephalic neuron-glia cultures nization of germ-free animals with microbiota from humans reduced the harmful effects of LPS-induced microglial (notwithstanding its inherent technical limitations) has recently activation through GPR109A inhibition of NF-kB pathway, emerged as a promising way to simplify studying these complex Downloaded from resulting in a reduction of proinflammatory enzymes (COX-2, disorders (98). Humanized germ-free animals may allow better iNOS) and cytokines (IL-6, TNF-a,IL-1b) (93). Butyrate and long-term elucidation of the roles of specific human-resident other SCFAs influx from the gastrointestinal tract through the bacterial species and their associated metabolites in contribut- portal circulation or are absorbed directly into the systemic ing to disease development and progression. circulation through the distal colonic blood supply, and may Furthermore, a detailed and comprehensive microbiome http://www.jimmunol.org/ access the CNS through the blood–brain barrier (BBB). Mice characterization using a combination of 16S rDNA analysis, fed a fermented-fiber diet presented with a reduced endotoxin- shotgun metagenomic sequencing of the microbiome meta- related sickness behavior (94), and interestingly featured genome, and metatranscriptome, coupled with metabolomics increased CNS IL-4 mRNA levels probably mediated by and characterization of the host transcriptome, may be inte- enhancement of histone acetylation increasing IL-1RA levels, grated into predictive computational modeling that may fa- which inhibits the production of IL-1b proinflammatory cilitate early diagnosis, patient stratification, and personalized cytokine (94). Furthermore, the tight-junction forming proteins treatment approaches (99). Last, heightened understanding of occludin and claudin-5 were reduced in the brains of germ- the effects of various metabolites on homeostasis and disease free mice, leading to an increased permeability of their may pave the way for their future application as a means of by guest on January 26, 2017 BBB (95). Treatment with butyrate-producing Clostridium postbiotic intervention that may allow the administration of tyrobutyricum bacteria was able to elevate occludin and metabolite combinations as a preventive or therapeutic mea- claudin-5 levels in germ-free mice brains and to restore sure in microbiome-associated disease, thereby avoiding the their BBB permeability to the level of specific pathogen-free need for dealing with the significant variability in individual mice (95). human microbiome configurations. Interestingly, other neuropsychiatric disorders such as autistic spectrum disorder (ASD) were also suggested to involve Acknowledgments modulation by microbiota-dependent metabolites (72). ASD- We apologize to those authors whose relevant work could not be included ow- prone mice feature dysbiosis, increased gut permeability, and ing to space constraints. We thank the members of the Elinav laboratory for altered serum metabolites (96). Treatment with Bacteroides discussions. fragilis restored gut permeability pathology and ASD neurologic symptoms, probably by regulation of 4-ethylphenylsulfate, a Disclosures metabolite contributing to some ASD-like symptoms. These The authors have no financial conflicts of interest. results suggest that the gut microbiome may influence some CNS neurological functions, including neuro-inflammation, References via systemic metabolite regulation (96). Whether some of 1. Veldhoen, M., and V. Brucklacher-Waldert. 2012. Dietary influences on intestinal these effects may involve altered activation of the CNS im- immunity. Nat. Rev. Immunol. 12: 696–708. 2. Lee, W.-J., and K. Hase. 2014. Gut microbiota-generated metabolites in animal mune response, and whether similar effects may be observed health and disease. Nat. Chem. Biol. 10: 416–424. in humans remains elusive. 3. Zheng, X., G. Xie, A. Zhao, L. Zhao, C. Yao, N. H. L. Chiu, Z. Zhou, Y. Bao, W. Jia, J. K. Nicholson, and W. Jia. 2011. The footprints of gut microbial- mammalian co-metabolism. J. Proteome Res. 10: 5512–5522. Conclusions 4. Gargaro, M., M. Pirro, R. Romani, T. Zelante, and F. Fallarino. 2016. AhR- In this review we highlight some examples by which dietary, host dependent pathways in immune regulation. Am. J. Transplant. 16: 2270–2276. 5. Gaudet, R. G., A. Sintsova, C. M. Buckwalter, N. Leung, A. Cochrane, J. Li, derived– and microbiota-modulated metabolites may affect A. D. Cox, J. Moffat, and S. D. Gray-Owen. 2015. Cytosolic detection of the numerous aspects of the host immune response. In many of bacterial metabolite HBP activates TIFA-dependent innate immunity. Science 348: these observations, the mechanisms through which commensal 1251–1255. bacteria regulate host immunity remain unclear and merit 6. Hall, J. A., J. L. Cannons, J. R. Grainger, L. M. Dos Santos, T. W. Hand, S. Naik, E. A. Wohlfert, D. B. Chou, G. Oldenhove, M. Robinson, et al. 2011. Essential role future investigation. Likewise, further investigation is required for retinoic acid in the promotion of CD4(+) T cell effector responses via retinoic to determine the repertoire, bio-geographical distribution, and acid receptor alpha. Immunity 34: 435–447. The Journal of Immunology 579

7. Lamas, B., M. L. Richard, V. Leducq, H.-P. Pham, M.-L. Michel, G. Da Costa, 36. Swann, J. R., E. J. Want, F. M. Geier, K. Spagou, I. D. Wilson, J. E. Sidaway, C. Bridonneau, S. Jegou, T. W. Hoffmann, J. M. Natividad, et al. 2016. CARD9 J. K. Nicholson, and E. Holmes. 2011. Systemic gut microbial modulation of bile impacts colitis by altering gut microbiota metabolism of tryptophan into aryl hy- acid metabolism in host tissue compartments. Proc. Natl. Acad. Sci. USA 108(Suppl. drocarbon receptor ligands. Nat. Med. 22: 598–605. 1): 4523–4530. 8. Mader, D., M. J. Rabiet, F. Boulay, and A. Peschel. 2010. Formyl peptide receptor- 37. Wang, H., J. Chen, K. Hollister, L. C. Sowers, and B. M. Forman. 1999. En- mediated proinflammatory consequences of peptide deformylase inhibition in dogenous bile acids are ligands for the nuclear receptor FXR/BAR. Mol. Cell 3: 543– Staphylococcus aureus. Microbes Infect. 12: 415–419. 553. 9. Thorburn, A. N., L. Macia, and C. R. Mackay. 2014. Diet, metabolites, and 38. Inagaki, T., A. Moschetta, Y.-K. Lee, L. Peng, G. Zhao, M. Downes, R. T. Yu, “western-lifestyle” inflammatory diseases. Immunity 40: 833–842. J. M. Shelton, J. A. Richardson, J. J. Repa, et al. 2006. Regulation of antibacterial 10. Tan, J., C. McKenzie, M. Potamitis, A. N. Thorburn, C. R. Mackay, and L. Macia. defense in the small intestine by the nuclear bile acid receptor. Proc. Natl. Acad. Sci. 2014. The role of short-chain fatty acids in health and disease. Adv. Immunol. 121: USA 103: 3920–3925. 91–119. 39. Pols, T. W. H., M. Nomura, T. Harach, G. Lo Sasso, M. H. Oosterveer, 11. Rothhammer, V., I. D. Mascanfroni, L. Bunse, M. C. Takenaka, J. E. Kenison, C. Thomas, G. Rizzo, A. Gioiello, L. Adorini, R. Pellicciari, et al. 2011. TGR5 L. Mayo, C.-C. Chao, B. Patel, R. Yan, M. Blain, et al. 2016. Type I interferons and activation inhibits atherosclerosis by reducing macrophage inflammation and lipid microbial metabolites of tryptophan modulate astrocyte activity and central nervous loading. Cell Metab. 14: 747–757. system inflammation via the aryl hydrocarbon receptor. Nat. Med. 22: 586–597. 40. Vavassori, P., A. Mencarelli, B. Renga, E. Distrutti, and S. Fiorucci. 2009. The bile 12. Butovsky, O., A. E. Talpalar, K. Ben-Yaakov, and M. Schwartz. 2005. Activation of acid receptor FXR is a modulator of intestinal innate immunity. J. Immunol. 183: microglia by aggregated b-amyloid or lipopolysaccharide impairs MHC-II expres- 6251–6261. sion and renders them cytotoxic whereas IFN-g and IL-4 render them protective. 41. Ridaura, V. K., J. J. Faith, F. E. Rey, J. Cheng, A. E. Duncan, A. L. Kau, Mol. Cell. Neurosci. 29: 381–393. N. W. Griffin, V. Lombard, B. Henrissat, J. R. Bain, et al. 2013. Gut microbiota 13. Koren, O., J. K. Goodrich, T. C. Cullender, A. Spor, K. Laitinen, H. K. Ba¨ckhed, from twins discordant for obesity modulate metabolism in mice. Science 341: A. Gonzalez, J. J. Werner, L. T. Angenent, R. Knight, et al. 2012. Host remodeling 1241214. of the gut microbiome and metabolic changes during pregnancy. Cell 150: 470– 42. Yoshimoto, S., T. M. Loo, K. Atarashi, H. Kanda, S. Sato, S. Oyadomari, 480. Y. Iwakura, K. Oshima, H. Morita, M. Hattori, et al. 2013. Obesity-induced gut 14. Ley, R. E., D. A. Peterson, and J. I. Gordon. 2006. Ecological and evolutionary microbial metabolite promotes liver cancer through senescence secretome. Nature forces shaping microbial diversity in the human intestine. Cell 124: 837–848. 499: 97–101. Downloaded from 15. Martin, F. P. J., N. Sprenger, I. Montoliu, S. Rezzi, S. Kochhar, and 43. Levy, M., C. A. Thaiss, D. Zeevi, L. Dohnalova´, G. Zilberman-Schapira, J. K. Nicholson. 2010. Dietary modulation of gut functional ecology studied by J. A. Mahdi, E. David, A. Savidor, T. Korem, Y. Herzig, et al. 2015. Microbiota- fecal metabonomics. J. Proteome Res. 9: 5284–5295. modulated metabolites shape the intestinal microenvironment by regulating NLRP6 16. Marcobal, A., T. Yusufaly, S. Higginbottom, M. Snyder, J. L. Sonnenburg, and inflammasome signaling. Cell 163: 1428–1443. G. I. Mias. 2015. Metabolome progression during early gut microbial colonization 44. Donohoe, D. R., A. Wali, B. P. Brylawski, and S. J. Bultman. 2012. Microbial of gnotobiotic mice. Sci. Rep. 5: 11589. regulation of glucose metabolism and cell-cycle progression in mammalian colo- 17. Matsumoto, M., R. Kibe, T. Ooga, Y. Aiba, S. Kurihara, E. Sawaki, Y. Koga, and nocytes. PLoS One 7: e46589. Y. Benno. 2012. Impact of intestinal microbiota on intestinal luminal metabolome.

45. Donohoe, D. R., N. Garge, X. Zhang, W. Sun, T. M. O’Connell, M. K. Bunger, http://www.jimmunol.org/ Sci. Rep. 2: 233. and S. J. Bultman. 2011. The microbiome and butyrate regulate energy metabolism 18. Nicholls, A. W., R. J. Mortishire-Smith, and J. K. Nicholson. 2003. NMR and autophagy in the mammalian colon. Cell Metab. 13: 517–526. spectroscopic-based metabonomic studies of urinary metabolite variation in accli- 46. Sanderson, I. R. 2004. Short chain fatty acid regulation of signaling genes expressed matizing germ-free rats. Chem. Res. Toxicol. 16: 1395–1404. by the intestinal epithelium. J. Nutr. 134: 2450S–2454S. 19. Pryde, S. E., S. H. Duncan, G. L. Hold, C. S. Stewart, and H. J. Flint. 2002. The 47. Buck, M. D., D. O’Sullivan, and E. L. Pearce. 2015. T cell metabolism drives microbiology of butyrate formation in the human colon. FEMS Microbiol. Lett. 217: immunity. J. Exp. Med. 212: 1345–1360. 133–139. 48. Chung, H., S. J. Pamp, J. A. Hill, N. K. Surana, S. M. Edelman, E. B. Troy, 20. Topping, D. L., and P. M. Clifton. 2001. Short-chain fatty acids and human co- N. C. Reading, E. J. Villablanca, S. Wang, J. R. Mora, et al. 2012. Gut immune lonic function: roles of resistant starch and nonstarch polysaccharides. Physiol. Rev. maturation depends on colonization with a host-specific microbiota. Cell 149: 81: 1031–1064. 1578–1593. 21. Card, D. J., R. Gorska, J. Cutler, and D. J. Harrington. 2014. Vitamin K meta- 49. Gaudier, E., A. Jarry, H. M. Blottie`re, P. de Coppet, M. P. Buisine, J. P. Aubert, bolism: current knowledge and future research. Mol. Nutr. Food Res. 58: 1590– C. Laboisse, C. Cherbut, and C. Hoebler. 2004. Butyrate specifically modulates by guest on January 26, 2017 1600. MUC gene expression in intestinal epithelial goblet cells deprived of glucose. Am. J. 22. Xu, Z., Q. Qiu, J. Tian, J. S. Smith, G. M. Conenello, T. Morita, and A. P. Byrnes. Physiol. Gastrointest. Liver Physiol. 287: G1168–G1174. 2013. Coagulation factor X shields adenovirus type 5 from attack by natural anti- 50. Willemsen, L. E. M., M. A. Koetsier, S. J. van Deventer, and E. A. van Tol. 2003. bodies and complement. Nat. Med. 19: 452–457. Short chain fatty acids stimulate epithelial mucin 2 expression through differential 23. Biesalski, H. K. 2016. Nutrition meets the microbiome: micronutrients and the microbiota. Ann. N. Y. Acad. Sci. 1372: 53–64. effects on prostaglandin E(1) and E(2) production by intestinal myofibroblasts. Gut 24. Beulens, J. W. J., S. L. Booth, E. G. H. M. van den Heuvel, E. Stoecklin, A. Baka, 52: 1442–1447. 51. Wrzosek, L., S. Miquel, M.-L. Noordine, S. Bouet, M. Joncquel Chevalier-Curt, and C. Vermeer. 2013. The role of menaquinones (vitamin K2) in human health. Br. J. Nutr. 110: 1357–1368. V. Robert, C. Philippe, C. Bridonneau, C. Cherbuy, C. Robbe-Masselot, et al. 25. Conly, J., and K. Stein. 1994. Reduction of vitamin K2 concentrations in human 2013. Bacteroides thetaiotaomicron and Faecalibacterium prausnitzii influence the liver associated with the use of broad spectrum antimicrobials. Clin. Invest. Med. 17: production of mucus glycans and the development of goblet cells in the colonic 531–539. epithelium of a gnotobiotic model rodent. BMC Biol. 11: 61. 26. Komai, M., H. Shirakawa, and S. Kimura. 1988. Newly developed model for vi- 52. Arpaia, N., C. Campbell, X. Fan, S. Dikiy, J. van der Veeken, P. deRoos, H. Liu, tamin K deficiency in germfree mice. Int. J. Vitam. Nutr. Res. 58: 55–59. J. R. Cross, K. Pfeffer, P. J. Coffer, and A. Y. Rudensky. 2013. Metabolites pro- 27. Davidson, R. T., A. L. Foley, J. A. Engelke, and J. W. Suttie. 1998. Conversion of duced by commensal bacteria promote peripheral regulatory T-cell generation. dietary phylloquinone to tissue menaquinone-4 in rats is not dependent on gut Nature 504: 451–455. bacteria. J. Nutr. 128: 220–223. 53. Chang, P. V., L. Hao, S. Offermanns, and R. Medzhitov. 2014. The microbial 28. Champagne, C. P., T. A. Tompkins, N. D. Buckley, and J. M. Green-Johnson. metabolite butyrate regulates intestinal macrophage function via histone deacetylase 2010. Effect of fermentation by pure and mixed cultures of Streptococcus thermo- inhibition. Proc. Natl. Acad. Sci. USA 111: 2247–2252. philus and Lactobacillus helveticus on isoflavone and B-vitamin content of a fer- 54. Smith, P. M., M. R. Howitt, N. Panikov, M. Michaud, C. A. Gallini, M. Bohlooly- mented soy beverage. Food Microbiol. 27: 968–972. Y, J. N. Glickman, and W. S. Garrett. 2013. The microbial metabolites, short-chain 29. Pompei, A., L. Cordisco, A. Amaretti, S. Zanoni, D. Matteuzzi, and M. Rossi. fatty acids, regulate colonic treg cell homeostasis. Science 341: 569–573. 2007. Folate production by bifidobacteria as a potential probiotic property. Appl. 55. Furusawa, Y., Y. Obata, S. Fukuda, T. A. Endo, G. Nakato, D. Takahashi, Environ. Microbiol. 73: 179–185. Y. Nakanishi, C. Uetake, K. Kato, T. Kato, et al. 2013. Commensal microbe- 30. Degnan, P. H., M. E. Taga, and A. L. Goodman. 2014. Vitamin B12 as a mod- derived butyrate induces the differentiation of colonic regulatory T cells. Nature ulator of gut microbial ecology. Cell Metab. 20: 769–778. 504: 446–450. 31. Zelante, T., R. G. Iannitti, C. Cunha, A. De Luca, G. Giovannini, G. Pieraccini, 56. Singh, N., A. Gurav, S. Sivaprakasam, E. Brady, R. Padia, H. Shi, M. Thangaraju, R. Zecchi, C. D’Angelo, C. Massi-Benedetti, F. Fallarino, et al. 2013. Tryptophan P. D. Prasad, S. Manicassamy, D. H. Munn, et al. 2014. Activation of Gpr109a, catabolites from microbiota engage aryl hydrocarbon receptor and balance mucosal receptor for niacin and the commensal metabolite butyrate, suppresses colonic in- reactivity via interleukin-22. Immunity 39: 372–385. flammation and carcinogenesis. Immunity 40: 128–139. 32. Hashimoto, T., T. Perlot, A. Rehman, J. Trichereau, H. Ishiguro, M. Paolino, 57. Thorburn, A. N., C. I. McKenzie, S. Shen, D. Stanley, L. Macia, L. J. Mason, V. Sigl, T. Hanada, R. Hanada, S. Lipinski, et al. 2012. ACE2 links amino acid L. K. Roberts, C. H. Y. Wong, R. Shim, R. Robert, et al. 2015. Evidence that malnutrition to microbial ecology and intestinal inflammation. Nature 487: 477– asthma is a developmental origin disease influenced by maternal diet and bacterial 481. metabolites. Nat. Commun. 6: 7320. 33. de Aguiar Vallim, T. Q., E. J. Tarling, and P. A. Edwards. 2013. Pleiotropic roles of 58. Vinolo, M. A. R., H. G. Rodrigues, E. Hatanaka, F. T. Sato, S. C. Sampaio, and bile acids in metabolism. Cell Metab. 17: 657–669. R. Curi. 2011. Suppressive effect of short-chain fatty acids on production of 34. Wahlstro¨m, A., S. I. Sayin, H.-U. Marschall, and F. Ba¨ckhed. 2016. Intestinal proinflammatory mediators by neutrophils. J. Nutr. Biochem. 22: 849–855. crosstalk between bile acids and microbiota and its impact on host metabolism. Cell 59. Usami, M., K. Kishimoto, A. Ohata, M. Miyoshi, M. Aoyama, Y. Fueda, and Metab. 24: 41–50. J. Kotani. 2008. Butyrate and trichostatin A attenuate nuclear factor kappaB acti- 35. Ridlon, J. M., D.-J. Kang, and P. B. Hylemon. 2006. Bile salt biotransformations vation and tumor necrosis factor alpha secretion and increase prostaglandin E2 by human intestinal bacteria. J. Lipid Res. 47: 241–259. secretion in human peripheral blood mononuclear cells. Nutr. Res. 28: 321–328. 580 BRIEF REVIEWS: METABOLITES, MICROBIOME, AND IMMUNITY

60. Trompette, A., E. S. Gollwitzer, K. Yadava, A. K. Sichelstiel, N. Sprenger, 80. Wigg, A. J., I. C. Roberts-Thomson, R. B. Dymock, P. J. McCarthy, R. H. Grose, C. Ngom-Bru, C. Blanchard, T. Junt, L. P. Nicod, N. L. Harris, and and A. G. Cummins. 2001. The role of small intestinal bacterial overgrowth, in- B. J. Marsland. 2014. Gut microbiota metabolism of dietary fiber influences allergic testinal permeability, endotoxaemia, and tumour necrosis factor alpha in the airway disease and hematopoiesis. Nat. Med. 20: 159–166. pathogenesis of non-alcoholic steatohepatitis. Gut 48: 206–211. 61. Singh, N., M. Thangaraju, P. D. Prasad, P. M. Martin, N. A. Lambert, T. Boettger, 81. Elinav, E., T. Strowig, A. L. Kau, J. Henao-Mejia, C. A. Thaiss, C. J. Booth, S. Offermanns, and V. Ganapathy. 2010. Blockade of dendritic cell development by D. R. Peaper, J. Bertin, S. C. Eisenbarth, J. I. Gordon, and R. A. Flavell. 2011. bacterial fermentation products butyrate and propionate through a transporter NLRP6 inflammasome regulates colonic microbial ecology and risk for colitis. Cell (Slc5a8)-dependent inhibition of histone deacetylases. J. Biol. Chem. 285: 27601– 145: 745–757. 27608. 82. Henao-Mejia, J., E. Elinav, C. Jin, L. Hao, W. Z. Mehal, T. Strowig, C. A. Thaiss, 62. Macia, L., J. Tan, A. T. Vieira, K. Leach, D. Stanley, S. Luong, M. Maruya, C. Ian A. L. Kau, S. C. Eisenbarth, M. J. Jurczak, et al. 2012. Inflammasome-mediated McKenzie, A. Hijikata, C. Wong, et al. 2015. Metabolite-sensing receptors GPR43 dysbiosis regulates progression of NAFLD and obesity. Nature 482: 179–185. and GPR109A facilitate dietary fibre-induced gut homeostasis through regulation of 83. Jiang, W., N. Wu, X. Wang, Y. Chi, Y. Zhang, X. Qiu, Y. Hu, J. Li, and Y. Liu. the inflammasome. Nat. Commun. 6: 6734. 2015. Dysbiosis gut microbiota associated with inflammation and impaired mucosal 63. Fukuda, S., H. Toh, K. Hase, K. Oshima, Y. Nakanishi, K. Yoshimura, T. Tobe, immune function in intestine of humans with non-alcoholic fatty liver disease. Sci. J. M. Clarke, D. L. Topping, T. Suzuki, et al. 2011. Bifidobacteria can protect from Rep. 5: 8096. enteropathogenic infection through production of acetate. Nature 469: 543–547. 84. Thaiss, C. A., N. Zmora, M. Levy, and E. Elinav. 2016. The microbiome and 64. Gury-BenAri, M., C. A. Thaiss, N. Serafini, D. R. Winter, A. Giladi, D. Lara- innate immunity. Nature 535: 65–74. Astiaso, M. Levy, T. M. Salame, A. Weiner, E. David, et al. 2016. The spectrum 85. Lusis, A. J. 2000. Atherosclerosis. Nature 407: 233–241. and regulatory landscape of intestinal innate lymphoid cells are shaped by the 86. Moore, K. J., F. J. Sheedy, and E. A. Fisher. 2013. Macrophages in atherosclerosis: microbiome. Cell 166: 1231–1246.e13. 65. Ivanov, I. I., K. Atarashi, N. Manel, E. L. Brodie, T. Shima, U. Karaoz, D. Wei, a dynamic balance. Nat. Rev. Immunol. 13: 709–721. K. C. Goldfarb, C. A. Santee, S. V. Lynch, et al. 2009. Induction of intestinal Th17 87. Koeth, R. A., Z. Wang, B. S. Levison, J. A. Buffa, E. Org, B. T. Sheehy, E. B. Britt, cells by segmented filamentous bacteria. Cell 139: 485–498. X. Fu, Y. Wu, L. Li, et al. 2013. Intestinal microbiota metabolism of L-carnitine, a 66. Goto, Y., C. Panea, G. Nakato, A. Cebula, C. Lee, M. G. Diez, T. M. Laufer, nutrient in red meat, promotes atherosclerosis. Nat. Med. 19: 576–585. L. Ignatowicz, and I. I. Ivanov. 2014. Segmented filamentous bacteria antigens 88. Tang, W. H. W., Z. Wang, B. S. Levison, R. A. Koeth, E. B. Britt, X. Fu, Y. Wu, and S. L. Hazen. 2013. Intestinal microbial metabolism of phosphatidylcholine and

presented by intestinal dendritic cells drive mucosal Th17 cell differentiation. Im- Downloaded from munity 40: 594–607. cardiovascular risk. N. Engl. J. Med. 368: 1575–1584. 67. Maslowski, K. M., A. T. Vieira, A. Ng, J. Kranich, F. Sierro, D. Yu, H. C. Schilter, 89. Wang, Z., A. B. Roberts, J. A. Buffa, B. S. Levison, W. Zhu, E. Org, X. Gu, M. S. Rolph, F. Mackay, D. Artis, et al. 2009. Regulation of inflammatory re- Y. Huang, M. Zamanian-Daryoush, M. K. Culley, et al. 2015. Non-lethal inhibi- sponses by gut microbiota and chemoattractant receptor GPR43. Nature 461: tion of gut microbial trimethylamine production for the treatment of atheroscle- 1282–1286. rosis. Cell 163: 1585–1595. 68. Giongo, A., K. A. Gano, D. B. Crabb, N. Mukherjee, L. L. Novelo, G. Casella, 90. Tan, J., C. McKenzie, P. J. Vuillermin, G. Goverse, C. G. Vinuesa, R. E. Mebius, J. C. Drew, J. Ilonen, M. Knip, H. Hyo¨ty, et al. 2011. Toward defining the au- L. Macia, and C. R. Mackay. 2016. Dietary fiber and bacterial SCFA enhance oral toimmune microbiome for type 1 diabetes. ISME J. 5: 82–91. tolerance and protect against food allergy through diverse cellular pathways. Cell

69. Henao-Mejia, J., E. Elinav, C. A. Thaiss, P. Licona-Limon, and R. A. Flavell. 2013. Rep. 15: 2809–2824. http://www.jimmunol.org/ Role of the intestinal microbiome in liver disease. J. Autoimmun. 46: 66–73. 91. Erny, D., A. L. Hrabeˇ de Angelis, D. Jaitin, P. Wieghofer, O. Staszewski, E. David, 70. Kostic, A. D., R. J. Xavier, and D. Gevers. 2014. The microbiome in inflammatory H. Keren-Shaul, T. Mahlakoiv, K. Jakobshagen, T. Buch, et al. 2015. Host bowel disease: current status and the future ahead. Gastroenterology 146: 1489–1499. microbiota constantly control maturation and function of microglia in the CNS. 71. Sears, C. L., and W. S. Garrett. 2014. Microbes, microbiota, and colon cancer. Cell Nat. Neurosci. 18: 965–977. Host Microbe 15: 317–328. 92. Wakade, C., R. Chong, E. Bradley, B. Thomas, and J. Morgan. 2014. Upregulation 72. Parracho, H. M. R. T., M. O. Bingham, G. R. Gibson, and A. L. McCartney. 2005. of GPR109A in Parkinson’s disease. PLoS One 9: e109818. Differences between the gut microflora of children with autistic spectrum disorders 93.Fu,S.-P.,J.-F.Wang,W.-J.Xue,H.-M.Liu,B.R.Liu,Y.-L.Zeng,S.-N.Li, and that of healthy children. J. Med. Microbiol. 54: 987–991. B.-X. Huang, Q.-K. Lv, W. Wang, and J.-X. Liu. 2015. Anti-inflammatory effects 73. Tang, W. H. W., and S. L. Hazen. 2014. The contributory role of gut microbiota in of BHBA in both in vivo and in vitro Parkinson’s disease models are mediated by cardiovascular disease. J. Clin. Invest. 124: 4204–4211. GPR109A-dependent mechanisms. J. Neuroinflammation 12: 9. 74. Qiu, J., X. Guo, Z.-M. E. Chen, L. He, G. F. Sonnenberg, D. Artis, Y.-X. Fu, and 94. Sherry, C. L., S. S. Kim, R. N. Dilger, L. L. Bauer, M. L. Moon, R. I. Tapping, by guest on January 26, 2017 L. Zhou. 2013. Group 3 innate lymphoid cells inhibit T-cell-mediated intestinal G. C. Fahey, Jr., K. A. Tappenden, and G. G. Freund. 2010. Sickness behavior inflammation through aryl hydrocarbon receptor signaling and regulation of mi- induced by endotoxin can be mitigated by the dietary soluble fiber, pectin, through croflora. Immunity 39: 386–399. up-regulation of IL-4 and Th2 polarization. Brain Behav. Immun. 24: 631–640. 75. Li, Y., S. Innocentin, D. R. Withers, N. A. Roberts, A. R. Gallagher, 95. Braniste, V., M. Al-Asmakh, C. Kowal, F. Anuar, A. Abbaspour, M. Toth, E. F. Grigorieva, C. Wilhelm, and M. Veldhoen. 2011. Exogenous stimuli maintain A. Korecka, N. Bakocevic, L. G. Ng, P. Kundu, et al. 2014. The gut microbiota intraepithelial lymphocytes via aryl hydrocarbon receptor activation. Cell 147: 629– 640. influences blood-brain barrier permeability in mice. Sci. Transl. Med. 6: 263ra158. 76. Park, J., M. Kim, S. G. Kang, A. H. Jannasch, B. Cooper, J. Patterson, and 96. Hsiao, E. Y., S. W. McBride, S. Hsien, G. Sharon, E. R. Hyde, T. McCue, C. H. Kim. 2015. Short-chain fatty acids induce both effector and regulatory T cells J. A. Codelli, J. Chow, S. E. Reisman, J. F. Petrosino, et al. 2013. Microbiota by suppression of histone deacetylases and regulation of the mTOR-S6K pathway. modulate behavioral and physiological abnormalities associated with neuro- Mucosal Immunol. 8: 80–93. developmental disorders. Cell 155: 1451–1463. 77. Berndt, B. E., M. Zhang, S. Y. Owyang, T. S. Cole, T. W. Wang, J. Luther, 97. Donaldson, G. P., S. M. Lee, and S. K. Mazmanian. 2016. Gut biogeography of the N. a. Veniaminova, J. L. Merchant, C. C. Chen, G. B. Huffnagle, and J. Y. Kao. bacterial microbiota. Nat. Rev. Microbiol. 14: 20–32. 2012. Butyrate increases IL-23 production by stimulated dendritic cells. Am. J. 98. Charbonneau, M. R., D. O’Donnell, L. V. Blanton, S. M. Totten, J. C. C. Davis, Physiol. Gastrointest. Liver Physiol. 303: G1384–G1392. M. J. Barratt, J. Cheng, J. Guruge, M. Talcott, J. R. Bain, et al. 2016. Sialylated 78. Angulo, P. 2002. Nonalcoholic fatty liver disease. N. Engl. J. Med. 346: 1221–1231. milk oligosaccharides promote microbiota-dependent growth in models of infant 79. Miele, L., V. Valenza, G. La Torre, M. Montalto, G. Cammarota, R. Ricci, undernutrition. Cell 164: 859–871. R. Masciana`, A. Forgione, M. L. Gabrieli, G. Perotti, et al. 2009. Increased in- 99. Zeevi, D., T. Korem, N. Zmora, D. Israeli, D. Rothschild, A. Weinberger, O. Ben- testinal permeability and tight junction alterations in nonalcoholic fatty liver disease. Yacov, D. Lador, T. Avnit-Sagi, M. Lotan-Pompan, et al. 2015. Personalized nu- Hepatology 49: 1877–1887. trition by prediction of glycemic responses. Cell 163: 1079–1094. Th eJournal of Brief Reviews Immunology

The Influence of the Microbiome on Allergic Sensitization to Food Catherine H. Plunkett* and Cathryn R. Nagler*,† The alarming increase in the incidence and severity of hood infections and altered the microbiota (11). Cesarean birth food allergies has coincided with lifestyle changes in and formula feeding have also disturbed nature’s coevolved Western societies, such as dietary modifications and in- strategy, altering founder microbial taxa and increasing sus- creased antibiotic use. These demographic shifts have ceptibility to diseases associated with this “Western” lifestyle profoundly altered the coevolved relationship between (12–14). Consumption of a highly processed modern diet, host and microbiota, depleting bacterial populations high in fat, low in fiber, and quite different from that of our critical for the maintenance of mucosal homeostasis. ancestors, has had profound consequences for the composition of the intestinal microbiota (15–17). Collectively, There is increasing evidence that the dysbiosis associ- Downloaded from ated with sensitization to food fails to stimulate protec- these studies suggest that environmental and lifestyle changes tive tolerogenic pathways, leading to the development have affected the relationship between the commensal micro- of the type 2 immune responses that characterize aller- biota and its human host and contributed to the increasing incidence of allergic disease. gic disease. Defining the role of beneficial allergy- The skin and all mucosal surfaces are populated by a site- protective members of the microbiota in the regulation specific microbiota (18). The microbes present can include http://www.jimmunol.org/ of tolerance to food has exciting potential for new in- bacteria, viruses, bacteriophages, Archaea, fungi, parasitic terventions to treat dietary allergies by modulation of worms, and protists (19, 20). Commensal bacteria are the best the microbiota. The Journal of Immunology, 2017, characterized, particularly in the gastrointestinal tract, where 198: 581–589. their density increases from an estimated 104 to 108 per milliliter of luminal contents in the small intestine to ∼1011 organisms per milliliter of luminal content in the colon, the he incidence of allergic diseases has increased dra- highest bacterial density of any environment analyzed (21). matically over the last 50 y, particularly in developed In addition to this large community of bacteria, the gas-

countries. The rise in prevalence occurred in such a short by guest on January 26, 2017 T trointestinal tract contains more immune cells than any time frame that genetics alone cannot explain it. This same time other organ. The two are in intimate communication; main- period was marked by improvements in sanitation, stark dietary tenance of homeostasis between these microbes and the changes, and increased vaccination and antibiotic use in Western immune system is essential to health. Exciting new research societies, all of which were linked to increasing susceptibility to is beginning to identify the mechanisms by which beneficial allergic and autoimmune diseases (1–4). What these lifestyle functions of the microbiota regulate tolerance to dietary changes have in common is their ability to alter the populations Ags (22, 23). of commensal microbes (the microbiota) that live in and on our In this review, we discuss the role of the microbiota in bodies (1). The hygiene hypothesis was the first to imply a link maintaining tolerance to food and examine how commensal between microbes and allergy by suggesting that the lower in- dysbiosis promotes the development of food allergy. Finally, cidence of allergic diseases in children with older siblings resulted we examine the clinical evidence for a role for the microbiota in from increased exposure to infectious disease in early childhood regulating food allergen sensitization and explore strategies for (5). This idea was expanded in subsequent epidemiological the development of microbiome-modulating therapeutics to studies that found that children raised in rural environments had prevent or treat food allergy. a lower incidence of allergic disease than those in urban settings and had greater environmental exposure to microbial products, Extending the hygiene hypothesis to the microbiota such as LPS (6–10). More recent work revised the original hy- The pathogenesis of food allergy involves an aberrant type giene hypothesis concept to include increased antibiotic use and 2immuneresponsetodietaryAgs.Themostcommon vaccination as other lifestyle changes that have reduced child- allergenic foods are tree nuts, peanuts, milk, eggs, shellfish, fish,

*Department of Pathology, The University of Chicago, Chicago, IL 60637; and Address correspondence and reprint requests to Dr. Cathryn R. Nagler, The University †Committee on Immunology, The University of Chicago, Chicago, IL 60637 of Chicago, Jules F. Knapp Medical Research Building R120, 924 East 57th Street, Chicago, IL 60637. E-mail address: [email protected] Received for publication July 22, 2016. Accepted for publication August 23, 2016. Abbreviations used in this article: DC, dendritic cell; fTh2, food Ag–speciﬁc Th2 cell; This work was supported by National Institutes of Health Grant R01 AI 106302, the GAP, goblet cell–associated Ag passage; ILC, innate lymphoid cell; LN, lymph node; Sunshine Foundation, The University of Chicago Center for Translational Medicine, mLN, mesenteric lymph node; RA, retinoic acid; SCFA, short-chain fatty acid; Treg, The University of Chicago Digestive Diseases Research Core Center (Grant DK42086), regulatory T cell. and Food Allergy Research and Education. Copyright Ó 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601266 582 BRIEF REVIEWS: FOOD ALLERGY AND THE MICROBIOME wheat, and soy, although a great number of other foods can various components of the microbiota influence immune elicit an allergic response (24). A healthy immune response to system development will inform therapeutic strategies aimed food Ags is a state of nonresponsiveness, referred to as oral at restoring the benefits conferred by particular microbial tolerance (25). When oral tolerance is not induced, food Ags communities. can instead evoke a response that is characterized by differentiation of naive T cells into food Ag–specific Th2 cells Oral tolerance to food Ags (fTh2s), which produce large amounts of IL-4 and IL-13 that The gastrointestinal tract is under constant bombardment by drive B cells to produce Ag-specific IgE (26). IgE binds to the microbial and food Ags. Therefore, of necessity, the healthy surface of mast cells and, upon re-exposure, Ag cross-linking intestinal immune system is geared toward a tolerogenic re- results in mast cell degranulation. The release of mast cell sponse characterized by the presence of large numbers of mediators, such as histamine, leads to the symptoms associ- regulatory T cells (Tregs). Originally, it was believed that oral ated with food allergy, including anaphylaxis (27). In support tolerance was primarily mediated by the generation of food of the epidemiological observations attributing a protective Ag–specific Tregs (37). Ag encountered in the lamina propria role to the microbiota in constraining allergic responses, is taken up by a population of CD103+ intestinal dendritic germ-free mice, which are born and raised in a sterile envi- cells (DCs) that migrate to the draining mesenteric lymph ronment, have an exaggerated systemic type 2 immune re- nodes (mLNs). Within the mLN, large amounts of retinoic sponse, which is characterized by high levels of IgE, and are acid (RA), a vitamin A derivative, and TGF-b produced by more susceptible to oral Ag–induced anaphylaxis than are CD103+ DCs and LN stromal cells instruct Ag-specific naive Downloaded from mice colonized with a diverse microbiota (28, 29). Early work T cells to differentiate into Tregs (38–40). Additionally, RA demonstrated that administration of LPS to germ-free mice is and TGF-b induce upregulation of the gut-homing receptors sufficient to restore oral tolerance (30). In keeping with this CCR9 and a4b7 on these newly differentiated Tregs to re- observation, mice unable to signal via TLR4, the receptor for cruit them back to the intestinal lamina propria (38–40), LPS, exhibit increased allergen-specific IgE and exacerbated where these Tregs are expanded by the production of IL-10 anaphylactic symptoms in response to repeated intragastric from resident CX3CR1 macrophages (41). Some of these http://www.jimmunol.org/ administration of peanut extract plus cholera toxin compared newly expanded Tregs may also enter into the circulation to with TLR4-sufficient mice (31). TLR4 is one of a group of mediate systemic tolerance to orally available Ags (42). In pattern recognition receptors that the immune system uses to support of this model, oral tolerance is abrogated in the ab- detect microbe-derived products, including LPS and DNA. sence of gut homing or in animals lacking CX3CR1 mac- Microbial sensing through TLRs is critical for maintaining rophages (41). Other work showed that oral Ag exposure intestinal homeostasis and limiting inflammation (32). Oral induces an allergic phenotype in mice with vitamin A defi- administration of a broad-spectrum antibiotic mixture evoked ciency (43). RA-deficient mLN DCs drive naive T cells to- food allergen sensitization in TLR4-sufficient mice, suggest- ward a pathogenic Th2 phenotype instead of Tregs (43). ing that intestinal bacteria were the source of the TLR4 ligand Collectively, these data support the concept that food Ag– by guest on January 26, 2017 (31). Recent work has begun to reveal an even more complex specific Tregs are critical for protection from dietary allergies role for LPS in allergic disease (33). Fecal samples were col- (41, 42). However, these studies only showed that Ag-specific lected during the first 3 y of life from genetically related, but Tregs, induced by oral administration of model food Ags, geographically separated, children at high (Finnish), low such as OVA, induce nonresponsiveness to subsequent pe- (Russian), and transitional (Estonian) risk for the develop- ripheral immunization (42). As such, it has not been clear ment of autoimmune and allergic disease, including food whether Treg development is a feature common to all food allergy. The investigators found that the low-risk Russian Ags found in a complex diet or whether these food Ag–specific children had higher proportions of Bifidobacterium, whereas Tregs also contribute to intestinal homeostasis. Recent work Finnish and Estonian children had increased abundance of using germ-free mice weaned onto an elemental diet void of Bacteroides. Surprisingly, metagenomic analysis revealed strik- Ags demonstrated that the majority of Tregs in the small ing differences in LPS synthesis between the Finnish and intestine are indeed induced in response to food Ags in a Russian cohorts. Russian children had LPS mostly originating complex diet (44). The induction of small intestinal Tregs from Escherichia coli, whereas the bulk of the LPS in Finnish occurs rapidly following the introduction of solid food and children originated from Bacteroides. Importantly, the LPS decreases over 4–6 wk following removal of food Ags (44). variant produced by Bacteroides was structurally and func- Mice weaned onto Ag-free diets had a greater proportion of tionally distinct from E. coli LPS. E. coli LPS is strongly Ag-specific T cells differentiating into inflammatory T cells immunostimulatory, and chronic exposure results in a re- rather than Tregs following oral Ag administration, suggesting fractory state known as endotoxin tolerance (34), which is that Tregs raised against dietary Ags limit proinflammatory thought to contribute to the protective effects of the micro- responses to model Ags, such as OVA. Therefore, food Ag– biota suggested by the hygiene hypothesis (35). Bacteroides- specific Tregs contribute to protection against allergic sensi- derived LPS was less immunostimulatory to primary human tization to dietary Ags, possibly by reinforcing a tolerogenic PBMCs than E. coli LPS, and when PBMCs were treated with environment in the small intestine. the two LPS variants mixed together, the high cytokine production elicited by E. coli LPS was abrogated. These findings Microbiota-mediated tolerance to food Ags raise the possibility that colonization early in life with a low- In addition to dietary Ags, the intestinal immune system must immunostimulatory microbiota can impair aspects of im- maintain a tolerogenic response to the microbiota resident mune education and predispose to inflammatory diseases, within the gut lumen. Components of the microbiota strongly such as food allergy (35, 36). A better understanding of how induce colonic Tregs, as demonstrated by a deficit in Tregs in The Journal of Immunology 583 the colonic lamina propria of germ-free mice, which increases previous literature that commensal bacteria induce a pop- in frequency following colonization (45, 46). Colonic Treg ulation of RORgt+ Tregs in the colon (48, 51). DCs from the induction was attributed to Clostridia, a class of mucosa- small intestine and colon migrate to anatomically different associated Firmicutes (45, 47). Spore-forming Clostridia iso- mLNs and induce immunologically distinct T cell responses lated from mouse and human feces strongly induce colonic (52, 53). Colonic and small intestinal DCs differ phenotyp- Tregs (45, 47). However, it was suggested more recently that ically and functionally, which may reflect the different anti- colonic Tregs can also be induced by other members of the genic burdens encountered by DCs at these physiologically microbiota (48). It is unclear whether there is a common distinct sites along the GI tract (52, 53). This finding may mode of action of Treg induction between these diverse help to explain why bacteria and food Ags induce distinct bacterial groups or whether they stimulate specific TLRs. It is populations of Tregs that differ in location and phenotype. also not known whether all (or most) Tregs induced by Evidence that bacteria-mediated Treg expansion can protect commensal bacteria bear bacteria-specific TCRs (49). More- from food allergen sensitization (22, 45, 47) suggests that over, there is evidence that bacteria-induced Tregs also con- food- and bacteria-derived Tregs work cooperatively to me- tribute to tolerance toward other Ags, including those from diate oral tolerance and protect from food allergen sensitiza- food. Indeed, Clostridia-induced Treg expansion was associ- tion (Fig. 1). It is possible that bacteria-specific Tregs migrate ated with protection from food allergen sensitization (22, 45, to the small intestinal lamina propria (perhaps through 47). Because there is heterogeneity within the intestinal Treg recirculation via the mLN) and secrete IL-10 to reinforce the population, it is possible that particular bacteria may induce tolerogenic environment (54). Indeed, a population of RORgt+ specific populations of Tregs that have different functions for Tregsispresentwithinthesmall intestine, albeit at a lower Downloaded from maintaining homeostasis (50). Kim et al. (44) observed that frequency than in the colon, and is reduced following anti- antibiotic treatment markedly reduces RORgt+ Tregs in the biotic treatment, supporting the idea that microbe-driven colonic lamina propria, suggesting that these Tregs are bac- Tregs can also localize at this site (44). In light of this, the teria dependent, whereas mice fed an Ag-free diet had a se- tolerogenic environment maintained by bacteria-derived Tregs 2 lective reduction in RORgt Tregs in the small intestine, likely contributes to the production of food Ag–specific Tregs, http://www.jimmunol.org/ indicating these are food Ag dependent. This finding supports rather than proallergic Th2 cells, upon subsequent exposure by guest on January 26, 2017

FIGURE 1. Induction of tolerance to food and bacterial Ags in the intestine: food Ag is taken up by DCs in the small intestine that migrate to the proximal mLN. TGF-b and RA, produced by LN stromal cells and DCs respectively, induce the differentiation of naive T cells to food Ag–specific Tregs (fTregs). RA and TGF-b also induce upregulation of gut-homing receptors on these newly differentiated Tregs to recruit them back to the lamina propria. TLR signaling by bacterial products induces the production of IL-10 by CX3CR1 macrophages resident in the lamina propria, supporting Treg expansion in this site. Bacterial products are also taken up by colonic DCs that migrate to the distal mLN and caudal LN to induce differentiation of bacterial-specific Tregs (bTregs). Although predominant in the colon, bTregs also migrate to the small intestine where they release IL-10 to maintain the tolerogenic immune environment. Fermentation of dietary fiber to SCFAs may enhance RA production by DCs and promote Treg differentiation. TLR signaling by bacterial products, such as LPS, induces a tolerogenic phenotype in colonic and small intestinal DCs that promotes differentiation of Tregs. 584 BRIEF REVIEWS: FOOD ALLERGY AND THE MICROBIOME to food Ags. In addition, RORgt+ Tregs are found in small Microbiota-mediated modulation of intestinal barrier function numbers in other body sites, including the lung, spleen, and It is remarkable to think that a single layer of epithelial cells is skin (55), suggesting that these cells disseminate throughout all that separates the enormous number and variety of food Ags the body. This may help to explain the role of the intestinal and microbes within the intestinal lumen from the underlying microbiota in protection from other allergic diseases, in- immune cell–rich lamina propria. Specialized epithelial sub- cluding asthma. populations have unique properties and functions that are Other studies are beginning to identify the signaling important for enforcing barrier protection to prevent un- pathways for bacteria-mediated Treg induction that are im- controlled access of Ags to the lamina propria (53). These portant for preventing allergic sensitization to food. Several include the production of antimicrobial peptides by Paneth reports implicated microbial sensing by host TLRs in the cells and of mucus by goblet cells, which together form a induction of colonic Tregs that maintain tolerogenic responses physical barrier that limits the access of bacteria to the epi- (46, 56). Mice lacking MyD88, an adaptor molecule for thelial surface (71). Additionally, intestinal epithelial cells downstream TLR signaling, develop more severe intestinal express tight junction proteins that seal off the underlying inflammation in response to epithelial damage induced by immune-rich lamina propria from excessive exposure to lu- dextran sodium sulfate, suggesting an important role for minal Ags (72). We found that increased intestinal per- microbial sensing in limiting inflammatory responses (32). meability contributes to food Ag sensitization. Stefka et al. TRAF6 is another important adaptor molecule for TLR observed that sensitization of mice to peanuts is associated signaling that acts downstream of MyD88 to activate tran- with an increased concentration of the major peanut allergens Downloaded from scription factors, such as NF-kB, to induce cytokine pro- Ara h 2 and Ara h 6 in the serum (22). Colonization with a duction (57). Mice with a specific deletion of TRAF6 in + consortium of mucosa-associated Clostridia reduced serum CD11c APCs (Traf6ΔCD11c)havereducednumbersof Tregs in their small intestine and present with a spontaneous Ara h 2 and Ara h 6 and ameliorated allergic sensitization type 2 inflammation at this site (58). Exacerbation of the to peanut, suggesting that commensal microbe–driven re- inflammatory phenotype in germ-free Traf6ΔCD11c mice inforcement of the epithelial barrier is important for protec- http://www.jimmunol.org/ further supports a role for microbe-induced Tregs in pre- tion against food allergy (22). The improved barrier function venting pathogenic type 2 intestinal inflammation (58, 59). observed after colonization by this Clostridia consortium was Interestingly, this phenotype was specific for the small intes- associated with an increase in goblet cell numbers and mucus tine, because no overt inflammation was observed in the co- production in the intestine, as well as upregulated expression lon. This may be due to the different physiological functions of the antimicrobial peptides Reg3b and Reg3g, suggesting a of the small and large intestine, because the large intestine broad effect on multiple intestinal epithelial cell types (22). functions primarily to reabsorb water, whereas the small in- Intestinal mucus and antimicrobial peptide production are testine encounters the bulk of the food Ags. As such, the regulated by the barrier-protective cytokine IL-22 (73). By tissue-selective inflammation observed in this model may in- using treatment with a neutralizing Ab for IL-22 or depletion by guest on January 26, 2017 dicate that the inflammatory response occurs as a result of a of innate lymphoid cells (ILCs), which produce IL-22, Stefka breakdown in tolerance to food Ags. et al. (22) demonstrated that Clostridia-induced IL-22 pro- In addition to microbial ligands that can be sensed via pattern duction by type 3 ILCs was necessary and sufficient to reduce recognition receptors, commensal bacteria release metabolites, intestinal permeability and prevent food Ag sensitization in such as short-chain fatty acids (SCFAs), upon fermentation of this model. This finding highlights the complex role of the insoluble dietary fibers (4). SCFAs, including acetate, propio- microbiota in modulating innate immune responses that, in nate, and butyrate, were demonstrated to have immunogenic turn, influence intestinal epithelial cell function and regulate activity locally, within the intestine, and systemically (23, 60– its barrier-protective properties (Fig. 2). 62). SCFAs are used by colonocytes as an energy source (63, Indeed, the microbiota has a profound influence on mucus 64), but they also have immunomodulatory properties by sig- production and goblet cell homeostasis (74). Germ-free mice naling through G protein–coupled receptors (65–67) and act to have altered mucus production compared with colonized inhibit histone deactelylases (68). One possible consequence of mice, with differences observed in the small and large intes- dysbiosis for the development of allergic responses may result tine (74). These mucus layers have unique features that limit from reduced levels of SCFAs (69). There is a body of literature access of bacteria to the intestinal epithelium (74). In the large supporting the idea that low levels of SCFAs are associated with intestine the mucus forms two layers: an inner sterile layer and an allergic phenotype and that increasing SCFA levels can a diffuse outer layer that is heavily colonized by bacteria that ameliorate disease (61, 63, 70). One mechanism by which are thought to feed on mucins and other proteins within the SCFAs protect against allergic disease is through the induction mucus layer (75). The permeability of the mucus layer differs of colonic Tregs (60, 62, 68). The addition of SCFAs to the along the length of the intestine. In the distal colon, the inner drinking water of germ-free mice resulted in an increased mucus layer is impenetrable to bacteria-sized beads, whereas abundance of colonic Tregs and protected against colonic in- some penetrability is observed in the proximal colon (76). flammation (62). Recent work showed that mice fed a high- In the small intestine, the mucus layer is diffuse and allows fiber diet had increased SCFA levels and were protected against limited penetration by bacteria; however, secretion of anti- food allergy, in part through enhanced induction of Tregs in microbial peptides and high gut motility (which removes this the mLN (23). DCs isolated from the mLN of the mice fed a easily detached mucus) helps to limit access to the epithelium high-fiber diet had increased retinal dehydrogenase activity (76). Within 5 wk postcolonization of germ-free mice, mucus compared with controls, suggesting a link between the pro- thickness, impenetrability, and detachment properties in the tective effects of dietary fiber and vitamin A metabolism (23). colon and small intestine resemble those seen in conventional The Journal of Immunology 585

A mice (74). Alteration in mucus properties postcolonization is associated with changes in the microbiota (74). Because treatment with some antibiotics also alters the mucus layer (77), this suggests that changes in the microbiota affect mucus-associated barrier function in the intestine. Stefka et al. (22) demonstrated that one mechanism by which Clostridia protect against allergic sensitization is by eliciting increased mucus production. Additionally, germ-free and antibiotic- treated mice have increased susceptibility to oral Ag sensitization (22, 29). Given the profound defect in mucus production in these mice, it is likely that the altered mucus production and subsequent defects in barrier function contribute to a breakdown in oral tolerance. However, goblet cells and mucus play a role in the maintenance of intestine homeostasis beyond their well-known contributions to barrier function. It was reported that culturing DCs in vitro with the mucus protein Muc2 induces a tolerogenic DC phenotype that induces greater Treg differentiation from naive T cells (78). Tolerance to OVA is impaired in Muc2-deficient mice; Downloaded from mucosal and systemic tolerance is restored when OVA is coadministered in the presence of Muc2 protein (78). Other work identified that goblet cells form goblet cell–associated Ag passages (GAPs) by taking up luminal Ag and delivering it to intestinal DCs (79–81). GAP formation is a regulated http://www.jimmunol.org/ process, induced by acetylcholine acting on goblet cells. GAPs form in the steady-state in the small intestine, but B goblet cell responsiveness to acetylcholine is inhibited by MyD88-dependent, goblet cell–intrinsic sensing of the gut microbiota in the colon (80). Deletion of MyD88 in goblet cells, or disruption of the microbiota by antibiotics, overrides the normal suppression by the microbiota and allows the formation of colonic GAPs, potentiating inflammation due to uncontrolled exposure to luminal contents (80, 81). More- by guest on January 26, 2017 over, altering GAP formation in early life results in persistent Th2 responses (R. Newberry, personal communication), providing an additional link between alterations in the gut microbiota in early life and a predisposition to food allergy. Intestinal epithelial cells also monitor the luminal environment and produce cytokines that direct immune responses in the underlying lamina propria (82). Stressed or damaged intestinal epithelial cells secrete the cytokines IL-33, TSLP, and IL-25, also collectively referred to as alarmins, to induce protective immunity and promote repair (82). These cytokines are associated with the initiation of the protective type 2 immune response that is responsible for the clearance of enteric helminths and repair of epithelial damage induced by infection (83–85). Epithelial alarmins activate type 2 ILCs to produce IL-13 and prime intestinal DCs to promote the differentiation of naive T cells to Th2 cells that mediate worm clearance (82, 86). Recently, the population of epithelial cells that produces IL-25 was revealed to be a specialized lineage known as tuft cells (83–85, 87). Howitt et al. showed that tuft FIGURE 2. The microbiota regulates protective and pathogenic barrier cells release IL-25 in response to activation by parasites via responses in the intestine. (A) A healthy microbiota induces a barrier- chemosensory receptors (84). Because some chemosensory protective response in the intestine, in part, through production of SCFAs receptors are G protein–coupled receptors, this raises the that are most likely to act on type 3 ILCs to produce IL-22. IL-22 induces antimicrobial peptide (AMP) production by Paneth cells and mucus production by goblet cells to reinforce barrier function, controlling the location and composition of the microbiota. This barrier-protective function prevents increased permeability to food Ags. Damaged or stressed epithelial cells release uncontrolled access to the lamina propria by food Ags to prevent allergic the alarmins IL-25, IL-33, and TSLP that activate type 2 ILCs to produce IL-4 sensitization. (B) Dysbiosis fails to induce these protective pathways. Dys- and IL-13, which promotes the development of allergic sensitization to food regulated epithelial barrier function and a compromised mucus layer allow Ags through the generation of fTh2s. 586 BRIEF REVIEWS: FOOD ALLERGY AND THE MICROBIOME interesting possibility that, in addition to helminths, these exhibited marked reductions in the abundance of four key receptors may also respond to microbe-derived signals, such as microbial genera (Faecalibacterium, Lachnospira, Rothia, and SCFAs, skewing the immune environment in the intestine. Veillonella) and in SCFA metabolites. Moreover, colonization Indeed, there is evidence that the microbiota can directly of germ-free mice with these genera protected from the de- regulate the expression of epithelial alarmins in the intestine. velopment of allergic airway inflammation (99). This pro- Intestinal expression of IL-25, IL-33, and TSLP is reduced in tection was also associated with increased levels of fecal germ-free mice and increases following colonization (88–90). butyrate, suggesting that one mechanism of protection was via Moreover, administration of IL-25 can alter the expression of SCFA signaling (99). Most studies have concluded that the antimicrobial peptides, changing the composition of the efficacy of conventional Lactobacillus-based probiotics is microbiota and demonstrating a role for IL-25 in the host- limited to infancy. Berni Canani et al. (107) showed that microbe cross-talk that is essential for homeostasis (91). In the treatment with a Lactobacillus GG–supplemented formula context of allergic inflammation, allergen exposure can in- accelerated the acquisition of tolerance in infants with cow’s duce the release of alarmins and drive the pathogenic type 2 milk allergy. Fecal samples collected from a small subset of in- immune response associated with disease (71). Overexpression fants from this study revealed stark differences in the micro- of IL-25 or IL-33 drives allergic responses to dietary Ags, biota of cow’s milk allergy patients compared with healthy suggesting that these epithelial-derived cytokines may con- age-matched controls (108). Twelve months of treatment tribute to the development of dietary allergies by skewing the with the tolerance-inducing formula was associated with an immune environment in the intestine from tolerogenic to increase in butyrate-producing bacteria and fecal butyrate proallergic (92, 93). A dysbiotic microbiome might also elicit levels, suggesting that one therapeutic effect of probiotic ad- Downloaded from epithelial alarmins and prime for allergic sensitization to ministration was the alteration of intestinal microbial com- food (94). In this setting, characterized by IL-4 production by munity structure (108). Preclinical mouse data also suggest mast cells and Th2 T cells, oral allergen exposure results in that dietary manipulation of the microbiota may be effective reprogramming of Tregs from a tolerogenic to a Th2-like for the treatment of allergic disease (23, 70). These studies are phenotype, further propagating the allergic response (95). providing exciting early evidence that the microbiota provides http://www.jimmunol.org/ IL-33 was shown to drive the expansion of GATA3+ Tregs protective and inductive signals to shape the intestinal im- in the colon (96), which are similar in phenotype to reprog- mune response and tip the balance in favor of tolerance over rammed Tregs implicated in the loss of tolerance to food Ags allergic sensitization. (95), supporting the idea that epithelial alarmin production Collectively, then, there is compelling evidence that allergic may alter the immunological environment in the intestine and phenotypes are associated with alterations in the intestinal contribute to allergic sensitization. In further support of this microbiome and that this dysbiosis may drive the allergic hypothesis, consumption of a low-fiber diet that exacerbates response. It is also possible that allergic inflammation itself food allergen sensitization is associated with increased ex- induces changes in the microbiota. Both scenarios may con- pression of IL-33 and TSLP in the intestine (23). Conversely, tribute to allergic sensitization at different times and stages of by guest on January 26, 2017 protection against allergic sensitization following administra- disease, highlighting the complex and dynamic nature of host- tion of a high-fiber diet alters the microbiota and is associated microbial interactions. As we learn more about the specific with reduced expression of IL-33 and TSLP. Taken together, immune-modulating effects of particular members of the these data suggest that a healthy microbiota may protect microbiota, we may be able to identify microbial signatures against allergic sensitization by reducing the expression of that are associated with proallergic responses, such as release of alarmins by intestinal epithelial cells. However, in conditions alarmins by epithelial cells, that contribute to allergic sen- of dysbiosis, the microbiota may induce elevated levels of sitization to food. Conversely, identification of a healthy these alarmins, resulting in aberrant Th2 responses toward microbiota that reinforces tolerance and barrier function in the dietary Ags by skewing the immune environment in the in- intestine will allow for better-targeted treatments to harness the testine toward a type 2, rather than a tolerogenic, response microbiota to restore health. (Fig. 2). Identifying how epithelial cell alarmin production is regulated may help to identify new targets to prevent allergic Conclusions sensitization to food. The marked increase in the incidence and severity of dietary allergies that has occurred in parallel with profound envi- Clinical considerations and potential therapeutics ronmental and lifestyle changes suggests a link between al- The findings described above outline the complex interplay terations in the microbiota and the rising prevalence of allergic between host immunity and the microbiota. Translational disease. Increasing knowledge of how the immune response is studies are beginning to explore a role for microbe-modulating influenced by the microbiota is revealing new approaches to therapeutics for diseases such as food allergy (97). Murine and treat diseases such as food allergy. Although there is already human studies suggested that dysbiosis early in life contrib- promising evidence in support of manipulating the microbiota utes to the development of allergic disease and that thera- during early life to prevent allergic sensitization, it is not yet peutic interventions that alter the microbial composition clear whether a stably established gut microbiota can be ef- during this time period may be most effective to prevent fectively manipulated to treat food allergy (109). We do know allergic sensitization (22, 29, 98–106). Recent analysis of that, in adults, the microbiota can be readily altered, even on one cohort of 319 children emphasized that changes in the daily timescales, by changes in components of the diet, par- microbiota in the first 100 d of life were most likely to be ticularly fiber (4, 15, 110). Moreover, the use of microbial associated with allergic disease (99). During this critical time metabolites to treat complex immune-mediated diseases is period, children at risk for developing allergic asthma starting to generate results (111). These observations lend The Journal of Immunology 587 promise to the vision that microbiome-modulating therapeu- 24. Sicherer, S. H., and H. A. Sampson. 2014. Food allergy: epidemiology, pathogenesis, diagnosis, and treatment. J. Allergy Clin. Immunol. 133: 291–307, quiz 308. tics will have efficacy later in life, particularly as an adjunctive 25. Iweala, O. I., and C. R. Nagler. 2006. Immune privilege in the gut: the estab- strategy to potentiate Ag-specific desensitization protocols, and lishment and maintenance of non-responsiveness to dietary antigens and commensal flora. Immunol. Rev. 213: 82–100. promote long-lasting tolerance (112). 26. Iweala, O. I., and A. W. Burks. 2016. Food allergy: our evolving understanding of its pathogenesis, prevention, and treatment. Curr. Allergy Asthma Rep. 16: 37. Acknowledgments 27. Brandt, E. B., R. T. Strait, D. Hershko, Q. Wang, E. E. Muntel, T. A. Scribner, N. Zimmermann, F. D. Finkelman, and M. E. Rothenberg. 2003. Mast cells are We thank P. Belda Ferre and E. Beltran for critical review of the manuscript. required for experimental oral allergen-induced diarrhea. J. Clin. Invest. 112: 1666–1677. 28. McCoy, K. D., N. L. Harris, P. Diener, S. Hatak, B. Odermatt, L. Hangartner, Disclosures B. M. Senn, B. J. Marsland, M. B. Geuking, H. Hengartner, et al. 2006. Natural IgE C.R.N. is president and cofounder of ClostraBio, Inc., a company developing production in the absence of MHC class II cognate help. Immunity 24: 329–339. microbiome-modulating therapeutics for the treatment of food allergies. The 29. Cahenzli, J., Y. Ko¨ller, M. Wyss, M. B. Geuking, and K. D. McCoy. 2013. In- testinal microbial diversity during early-life colonization shapes long-term IgE other author has no financial conflicts of interest. levels. Cell Host Microbe 14: 559–570. 30. Wannemuehler, M. J., H. Kiyono, J. L. Babb, S. M. Michalek, and J. R. McGhee. 1982. Lipopolysaccharide (LPS) regulation of the immune response: LPS converts References germfree mice to sensitivity to oral tolerance induction. J. Immunol. 129: 959–965. 1. Feehley, T., A. T. Stefka, S. Cao, and C. R. Nagler. 2012. Microbial regulation of 31. Bashir, M. E., S. Louie, H. N. Shi, and C. Nagler-Anderson. 2004. Toll-like re- allergic responses to food. Semin. Immunopathol. 34: 671–688. ceptor 4 signaling by intestinal microbes influences susceptibility to food allergy. 2. Gilbert, J. A., R. A. Quinn, J. Debelius, Z. Z. Xu, J. Morton, N. Garg, J. Immunol. 172: 6978–6987. J. K. Jansson, P. C. Dorrestein, and R. Knight. 2016. Microbiome-wide associa- 32. Rakoff-Nahoum, S., J. Paglino, F. Eslami-Varzaneh, S. Edberg, and R. Medzhitov. tion studies link dynamic microbial consortia to disease. Nature 535: 94–103. 2004. Recognition of commensal microflora by toll-like receptors is required for

3. Honda, K., and D. R. Littman. 2016. The microbiota in adaptive immune ho- intestinal homeostasis. Cell 118: 229–241. Downloaded from meostasis and disease. Nature 535: 75–84. 33. Vatanen, T., A. D. Kostic, E. d’Hennezel, H. Siljander, E. A. Franzosa, 4. Sonnenburg, J. L., and F. Ba¨ckhed. 2016. Diet-microbiota interactions as mod- M. Yassour, R. Kolde, H. Vlamakis, T. D. Arthur, A. M. Ha¨ma¨laïnen, et al; erators of human metabolism. Nature 535: 56–64. DIABIMMUNE Study Group. 2016. Variation in microbiome LPS immunogenicity 5. Strachan, D. P. 1989. Hay fever, hygiene, and household size. BMJ 299: 1259– contributes to autoimmunity in humans. [Published erratum appears in 2016 1260. Cell 165: 1551.] Cell 165: 842–853. 6. von Mutius, E., and D. Vercelli. 2010. Farm living: effects on childhood asthma 34. Watson, D. W., and Y. B. Kim. 1963. Modification of host responses to bacterial and allergy. Nat. Rev. Immunol. 10: 861–868. endotoxins. I. Specificity of pyrogenic tolerance and the role of hypersensitivity in 7. Riedler, J., W. Eder, G. Oberfeld, and M. Schreuer. 2000. Austrian children living on a pyrogenicity, lethality, and skin reactivity. J. Exp. Med. 118: 425–446. farm have less hay fever, asthma and allergic sensitization. Clin.Exp.Allergy30: 194–200. 35. Biswas, S. K., and E. Lopez-Collazo. 2009. Endotoxin tolerance: new mechanisms, http://www.jimmunol.org/ 8. Braun-Fahrla¨nder, C., M. Gassner, L. Grize, U. Neu, F. H. Sennhauser, molecules and clinical significance. Trends Immunol. 30: 475–487. H. S. Varonier, J. C. Vuille, and B. Wuthrich,€ Swiss Study on Childhood Allergy 36. Feehley, T., P. Belda-Ferre, and C. R. Nagler. 2016. What’s LPS got to do with it? and Respiratory Symptoms with Respect to Air Pollution. 1999. Prevalence of hay A role for gut LPS variants in driving autoimmune and allergic disease. Cell Host fever and allergic sensitization in farmer’s children and their peers living in the Microbe 19: 572–574. same rural community. SCARPOL team. Clin. Exp. Allergy 29: 28–34. 37. Faria, A. M., and H. L. Weiner. 1999. Oral tolerance: mechanisms and therapeutic 9. Braun-Fahrla¨nder, C., J. Riedler, U. Herz, W. Eder, M. Waser, L. Grize, applications. Adv. Immunol. 73: 153–264. S. Maisch, D. Carr, F. Gerlach, A. Bufe, et al; Allergy and Endotoxin Study Team. 38. Benson, M. J., K. Pino-Lagos, M. Rosemblatt, and R. J. Noelle. 2007. All-trans 2002. Environmental exposure to endotoxin and its relation to asthma in school- retinoic acid mediates enhanced T reg cell growth, differentiation, and gut homing age children. N. Engl. J. Med. 347: 869–877. in the face of high levels of co-stimulation. J. Exp. Med. 204: 1765–1774. 10. Waser, M., E. von Mutius, J. Riedler, D. Nowak, S. Maisch, D. Carr, W. Eder, 39. Sun, C. M., J. A. Hall, R. B. Blank, N. Bouladoux, M. Oukka, J. R. Mora, and Y. Belkaid. 2007. Small intestine lamina propria dendritic cells promote de novo

G. Tebow, R. Schierl, M. Schreuer, and C. Braun-Fahrla¨nder, ALEX Study team. by guest on January 26, 2017 2005. Exposure to pets, and the association with hay fever, asthma, and atopic generation of Foxp3 T reg cells via retinoic acid. J. Exp. Med. 204: 1775–1785. sensitization in rural children. Allergy 60: 177–184. 40. Coombes, J. L., K. R. Siddiqui, C. V. Arancibia-Ca´rcamo, J. Hall, C. M. Sun, 11. Bloomfield, S. F., G. A. Rook, E. A. Scott, F. Shanahan, R. Stanwell-Smith, and Y. Belkaid, and F. Powrie. 2007. A functionally specialized population of mucosal P. Turner. 2016. Time to abandon the hygiene hypothesis: new perspectives on CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-beta and retinoic acid- allergic disease, the human microbiome, infectious disease prevention and the role dependent mechanism. J. Exp. Med. 204: 1757–1764. of targeted hygiene. Perspect. Public Health 136: 213–224. 41. Hadis, U., B. Wahl, O. Schulz, M. Hardtke-Wolenski, A. Schippers, N. Wagner, 12. Dominguez-Bello, M. G., E. K. Costello, M. Contreras, M. Magris, G. Hidalgo, W. Muller,€ T. Sparwasser, R. Fo¨rster, and O. Pabst. 2011. Intestinal tolerance N. Fierer, and R. Knight. 2010. Delivery mode shapes the acquisition and requires gut homing and expansion of FoxP3+ regulatory T cells in the lamina structure of the initial microbiota across multiple body habitats in newborns. Proc. propria. Immunity 34: 237–246. Natl. Acad. Sci. USA 107: 11971–11975. 42. Pabst, O., and A. M. Mowat. 2012. Oral tolerance to food protein. Mucosal 13. Ba¨ckhed, F., J. Roswall, Y. Peng, Q. Feng, H. Jia, P. Kovatcheva-Datchary, Y. Li, Immunol. 5: 232–239. Y. Xia, H. Xie, H. Zhong, et al. 2015. Dynamics and stabilization of the human 43. Yokota-Nakatsuma, A., H. Takeuchi, Y. Ohoka, C. Kato, S. Y. Song, T. Hoshino, gut microbiome during the first year of life. [Published erratum appears in 2015 H. Yagita, T. Ohteki, and M. Iwata. 2014. Retinoic acid prevents mesenteric Cell Host Microbe 17: 852] Cell Host Microbe 17: 690–703. lymph node dendritic cells from inducing IL-13-producing inflammatory Th2 14. Guaraldi, F., and G. Salvatori. 2012. Effect of breast and formula feeding on gut cells. Mucosal Immunol. 7: 786–801. microbiota shaping in newborns. Front. Cell. Infect. Microbiol. 2: 94. 44. Kim, K. S., S. W. Hong, D. Han, J. Yi, J. Jung, B. G. Yang, J. Y. Lee, M. Lee, and 15. David, L. A., C. F. Maurice, R. N. Carmody, D. B. Gootenberg, J. E. Button, C. D. Surh. 2016. Dietary antigens limit mucosal immunity by inducing regula- B. E. Wolfe, A. V. Ling, A. S. Devlin, Y. Varma, M. A. Fischbach, et al. 2014. Diet tory T cells in the small intestine. Science 351: 858–863. rapidly and reproducibly alters the human gut microbiome. Nature 505: 559–563. 45. Atarashi, K., T. Tanoue, T. Shima, A. Imaoka, T. Kuwahara, Y. Momose, 16. Sonnenburg, E. D., S. A. Smits, M. Tikhonov, S. K. Higginbottom, G. Cheng, S. Yamasaki, T. Saito, Y. Ohba, et al. 2011. Induction of colonic N. S. Wingreen, and J. L. Sonnenburg. 2016. Diet-induced extinctions in the gut regulatory T cells by indigenous Clostridium species. Science 331: 337–341. microbiota compound over generations. Nature 529: 212–215. 46. Geuking, M. B., J. Cahenzli, M. A. Lawson, D. C. Ng, E. Slack, S. Hapfelmeier, 17. De Filippo, C., D. Cavalieri, M. Di Paola, M. Ramazzotti, J. B. Poullet, K. D. McCoy, and A. J. Macpherson. 2011. Intestinal bacterial colonization in- S. Massart, S. Collini, G. Pieraccini, and P. Lionetti. 2010. Impact of diet in duces mutualistic regulatory T cell responses. Immunity 34: 794–806. shaping gut microbiota revealed by a comparative study in children from Europe 47. Atarashi, K., T. Tanoue, K. Oshima, W. Suda, Y. Nagano, H. Nishikawa, and rural Africa. Proc. Natl. Acad. Sci. USA 107: 14691–14696. S. Fukuda, T. Saito, S. Narushima, K. Hase, et al. 2013. Treg induction by a 18. Cho, I., and M. J. Blaser. 2012. The human microbiome: at the interface of health rationally selected mixture of Clostridia strains from the human microbiota. Na- and disease. Nat. Rev. Genet. 13: 260–270. ture 500: 232–236. 19. The Human Microbiome Project Consortium. 2012. Structure, function and di- 48. Sefik, E., N. Geva-Zatorsky, S. Oh, L. Konnikova, D. Zemmour, A. M. McGuire, versity of the healthy human microbiome. Nature 486: 207–214. D. Burzyn, A. Ortiz-Lopez, M. Lobera, J. Yang, et al. 2015. Individual intestinal 20. Parfrey, L. W., W. A. Walters, and R. Knight. 2011. Microbial eukaryotes in the symbionts induce a distinct population of RORg+ regulatory T cells. Science 349: human microbiome: ecology, evolution, and future directions. Front. Microbiol. 2: 153. 993–997. 21. Walter, J., and R. Ley. 2011. The human gut microbiome: ecology and recent 49. Lathrop, S. K., S. M. Bloom, S. M. Rao, K. Nutsch, C. W. Lio, N. Santacruz, evolutionary changes. Annu. Rev. Microbiol. 65: 411–429. D. A. Peterson, T. S. Stappenbeck, and C. S. Hsieh. 2011. Peripheral education of 22.Stefka,A.T.,T.Feehley,P.Tripathi,J.Qiu,K.McCoy,S.K.Mazmanian, the immune system by colonic commensal microbiota. Nature 478: 250–254. M. Y. Tjota, G. Y. Seo, S. Cao, B. R. Theriault, et al. 2014. Commensal bacteria protect 50. Tanoue, T., K. Atarashi, and K. Honda. 2016. Development and maintenance of against food allergen sensitization. Proc.Natl.Acad.Sci.USA111: 13145–13150. intestinal regulatory T cells. Nat. Rev. Immunol. 16: 295–309. 23. Tan, J., C. McKenzie, P. J. Vuillermin, G. Goverse, C. G. Vinuesa, R. E. Mebius, 51. Ohnmacht, C., J. H. Park, S. Cording, J. B. Wing, K. Atarashi, Y. Obata, L. Macia, and C. R. Mackay. 2016. Dietary fiber and bacterial SCFA enhance oral V. Gaboriau-Routhiau, R. Marques, S. Dulauroy, M. Fedoseeva, et al. 2015. The tolerance and protect against food allergy through diverse cellular pathways. Cell microbiota regulates type 2 immunity through RORgammat(+) T cells. Science Reports 15: 2809–2824. 349: 989–993. 588 BRIEF REVIEWS: FOOD ALLERGY AND THE MICROBIOME

52. Houston, S. A., V. Cerovic, C. Thomson, J. Brewer, A. M. Mowat, and S. Milling. 78.Shan,M.,M.Gentile,J.R.Yeiser,A.C.Walland,V.U.Bornstein,K.Chen, 2016. The lymph nodes draining the small intestine and colon are anatomically B.He,L.Cassis,A.Bigas,M.Cols,etal.2013.Mucusenhancesguthomeo- separate and immunologically distinct. Mucosal Immunol. 9: 468–478. stasis and oral tolerance by delivering immunoregulatory signals. Science 342: 53. Mowat, A. M., and W. W. Agace. 2014. Regional specialization within the in- 447–453. testinal immune system. Nat. Rev. Immunol. 14: 667–685. 79. McDole, J. R., L. W. Wheeler, K. G. McDonald, B. Wang, V. Konjufca, 54. Cao, S., T. J. Feehley, and C. R. Nagler. 2014. The role of commensal bacteria in K. A. Knoop, R. D. Newberry, and M. J. Miller. 2012. Goblet cells deliver luminal the regulation of sensitization to food allergens. FEBS Lett. 588: 4258–4266. antigen to CD103+ dendritic cells in the small intestine. Nature 483: 345–349. 55. Lochner, M., L. Peduto, M. Cherrier, S. Sawa, F. Langa, R. Varona, 80. Knoop, K. A., K. G. McDonald, S. McCrate, J. R. McDole, and R. D. Newberry. D. Riethmacher, M. Si-Tahar, J. P. Di Santo, and G. Eberl. 2008. In vivo 2015. Microbial sensing by goblet cells controls immune surveillance of luminal equilibrium of proinﬂammatory IL-17+ and regulatory IL-10+ Foxp3+ ROR- antigens in the colon. Mucosal Immunol. 8: 198–210. gamma t+ T cells. J. Exp. Med. 205: 1381–1393. 81. Knoop, K. A., K. G. McDonald, D. H. Kulkarni, and R. D. Newberry. 2016. 56. Wang, S., L. M. Charbonnier, M. Noval Rivas, P. Georgiev, N. Li, G. Gerber, Antibiotics promote inﬂammation through the translocation of native commensal L. Bry, and T. A. Chatila. 2015. MyD88 adaptor-dependent microbial sensing by colonic bacteria. Gut 65: 1100–1109. regulatory T cells promotes mucosal tolerance and enforces commensalism. Im- 82. Hammad, H., and B. N. Lambrecht. 2015. Barrier epithelial cells and the control munity 43: 289–303. of type 2 immunity. Immunity 43: 29–40. 57. Ye, H., J. R. Arron, B. Lamothe, M. Cirilli, T. Kobayashi, N. K. Shevde, D. Segal, 83. Gerbe, F., E. Sidot, D. J. Smyth, M. Ohmoto, I. Matsumoto, V. Dardalhon, O. K. Dzivenu, M. Vologodskaia, M. Yim, et al. 2002. Distinct molecular P. Cesses, L. Garnier, M. Pouzolles, B. Brulin, et al. 2016. Intestinal epithelial tuft mechanism for initiating TRAF6 signalling. Nature 418: 443–447. cells initiate type 2 mucosal immunity to helminth parasites. Nature 529: 226– 58. Han, D., M. C. Walsh, P. J. Cejas, N. N. Dang, Y. F. Kim, J. Kim, L. Charrier- 230. Hisamuddin, L. Chau, Q. Zhang, K. Bittinger, et al. 2013. Dendritic cell ex- 84. Howitt, M. R., S. Lavoie, M. Michaud, A. M. Blum, S. V. Tran, J. V. Weinstock, pression of the signaling molecule TRAF6 is critical for gut microbiota-dependent C. A. Gallini, K. Redding, R. F. Margolskee, L. C. Osborne, et al. 2016. Tuft cells, immune tolerance. Immunity 38: 1211–1222. taste-chemosensory cells, orchestrate parasite type 2 immunity in the gut. Science 59. Han, D., M. C. Walsh, K. S. Kim, S. W. Hong, J. Lee, J. Yi, G. Rivas, C. D. Surh, 351: 1329–1333. and Y. Choi. 2015. Microbiota-independent ameliorative effects of antibiotics on 85. von Moltke, J., M. Ji, H. E. Liang, and R. M. Locksley. 2016. Tuft-cell-derived spontaneous Th2-associated pathology of the small intestine. [Published erratum IL-25 regulates an intestinal ILC2-epithelial response circuit. Nature 529: 221–

appears in 2015 PLoS One 10: e0133787.] PLoS One 10: e0118795. 225. Downloaded from 60. Arpaia, N., C. Campbell, X. Fan, S. Dikiy, J. van der Veeken, P. deRoos, H. Liu, 86. Hung, L. Y., I. P. Lewkowich, L. A. Dawson, J. Downey, Y. Yang, D. E. Smith, J. R. Cross, K. Pfeffer, P. J. Coffer, and A. Y. Rudensky. 2013. Metabolites and D. R. Herbert. 2013. IL-33 drives biphasic IL-13 production for noncanonical produced by commensal bacteria promote peripheral regulatory T-cell generation. type 2 immunity against hookworms. Proc. Natl. Acad. Sci. USA 110: 282–287. Nature 504: 451–455. 87. Gronke, K., and A. Diefenbach. 2016. Tuft cell-derived IL-25 activates and 61. Thorburn, A. N., C. I. McKenzie, S. Shen, D. Stanley, L. Macia, L. J. Mason, maintains ILC2. Immunol. Cell Biol. 94: 221–223. L. K. Roberts, C. H. Wong, R. Shim, R. Robert, et al. 2015. Evidence that asthma 88. Zaph, C., Y. Du, S. A. Saenz, M. G. Nair, J. G. Perrigoue, B. C. Taylor, is a developmental origin disease influenced by maternal diet and bacterial me- A. E. Troy, D. E. Kobuley, R. A. Kastelein, D. J. Cua, et al. 2008. Commensal- tabolites. Nat. Commun. 6: 7320. dependent expression of IL-25 regulates the IL-23–IL-17 axis in the intestine. 62. Smith, P. M., M. R. Howitt, N. Panikov, M. Michaud, C. A. Gallini, J. Exp. Med. 205: 2191–2198. http://www.jimmunol.org/ M. Bohlooly-Y, J. N. Glickman, and W. S. Garrett. 2013. The microbial me- 89. Mosconi, I., M. B. Geuking, M. M. Zaiss, J. C. Massacand, C. Aschwanden, tabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis. Science C. K. Kwong Chung, K. D. McCoy, and N. L. Harris. 2013. Intestinal bacteria 341: 569–573. induce TSLP to promote mutualistic T-cell responses. Mucosal Immunol. 6: 1157– 63. Tan, J., C. McKenzie, M. Potamitis, A. N. Thorburn, C. R. Mackay, and 1167. L. Macia. 2014. The role of short-chain fatty acids in health and disease. Adv. 90. De Salvo, C., X. M. Wang, L. Pastorelli, B. Mattioli, S. Omenetti, K. A. Buela, Immunol. 121: 91–119. S. Chowdhry, R. R. Garg, W. A. Goodman, A. Rodriguez-Palacios, et al. 2016. 64. Brestoff, J. R., and D. Artis. 2013. Commensal bacteria at the interface of host IL-33 drives eosinophil infiltration and pathogenic type 2 helper T-cell immune metabolism and the immune system. Nat. Immunol. 14: 676–684. responses leading to chronic experimental ileitis. Am. J. Pathol. 186: 885–898. 65. Brown, A. J., S. M. Goldsworthy, A. A. Barnes, M. M. Eilert, L. Tcheang, 91. Fricke, W. F., Y. Song, A. J. Wang, A. Smith, V. Grinchuk, E. Mongodin, C. Pei, D. Daniels, A. I. Muir, M. J. Wigglesworth, I. Kinghorn, N. J. Fraser, et al. 2003. B. Ma, N. Lu, J. F. Urban, Jr., et al. 2015. Type 2 immunity-dependent reduction The orphan G protein-coupled receptors GPR41 and GPR43 are activated by pro- of segmented filamentous bacteria in mice infected with the helminthic parasite by guest on January 26, 2017 pionate and other short chain carboxylic acids. J. Biol. Chem. 278: 11312–11319. Nippostrongylus brasiliensis. [Published erratum appears in 2015 Microbiome 3: 77.] 66. Singh, N., A. Gurav, S. Sivaprakasam, E. Brady, R. Padia, H. Shi, M. Thangaraju, Microbiome 3: 40. P. D. Prasad, S. Manicassamy, D. H. Munn, et al. 2014. Activation of Gpr109a, 92. Lee, J. B., C. Y. Chen, B. Liu, L. Mugge, P. Angkasekwinai, V. Facchinetti, receptor for niacin and the commensal metabolite butyrate, suppresses colonic C. Dong, Y. J. Liu, M. E. Rothenberg, S. P. Hogan, et al. 2016. IL-25 and CD4(+) inflammation and carcinogenesis. Immunity 40: 128–139. TH2 cells enhance type 2 innate lymphoid cell-derived IL-13 production, which 67. Thangaraju, M., G. A. Cresci, K. Liu, S. Ananth, J. P. Gnanaprakasam, promotes IgE-mediated experimental food allergy. J. Allergy Clin. Immunol. 137: D. D. Browning, J. D. Mellinger, S. B. Smith, G. J. Digby, N. A. Lambert, et al. 1216–1225.e5. 2009. GPR109A is a G-protein-coupled receptor for the bacterial fermentation 93. Chu, D. K., A. Llop-Guevara, T. D. Walker, K. Flader, S. Goncharova, product butyrate and functions as a tumor suppressor in colon. Cancer Res. 69: J. E. Boudreau, C. L. Moore, T. Seunghyun In, S. Waserman, A. J. Coyle, et al. 2826–2832. 2013. IL-33, but not thymic stromal lymphopoietin or IL-25, is central to mite 68. Furusawa, Y., Y. Obata, S. Fukuda, T. A. Endo, G. Nakato, D. Takahashi, and peanut allergic sensitization. J. Allergy Clin. Immunol. 131: 187–200.e8. Y. Nakanishi, C. Uetake, K. Kato, T. Kato, et al. 2013. Commensal microbe- 94. Noval Rivas, M., O. T. Burton, P. Wise, Y. Q. Zhang, S. A. Hobson, M. Garcia derived butyrate induces the differentiation of colonic regulatory T cells. Nature Lloret, C. Chehoud, J. Kuczynski, T. DeSantis, J. Warrington, et al. 2013. A 504: 446–450. microbiota signature associated with experimental food allergy promotes allergic 69. Thorburn, A. N., L. Macia, and C. R. Mackay. 2014. Diet, metabolites, and sensitization and anaphylaxis. J. Allergy Clin. Immunol. 131: 201–212. “western-lifestyle” inflammatory diseases. Immunity 40: 833–842. 95. Noval Rivas, M., O. T. Burton, P. Wise, L. M. Charbonnier, P. Georgiev, 70. Trompette, A., E. S. Gollwitzer, K. Yadava, A. K. Sichelstiel, N. Sprenger, H. C. Oettgen, R. Rachid, and T. A. Chatila. 2015. Regulatory T cell reprog- C. Ngom-Bru, C. Blanchard, T. Junt, L. P. Nicod, N. L. Harris, and ramming toward a Th2-cell-like lineage impairs oral tolerance and promotes food B. J. Marsland. 2014. Gut microbiota metabolism of dietary fiber influences al- allergy. Immunity 42: 512–523. lergic airway disease and hematopoiesis. Nat. Med. 20: 159–166. 96. Schiering, C., T. Krausgruber, A. Chomka, A. Fro¨hlich, K. Adelmann, 71. Peterson, L. W., and D. Artis. 2014. Intestinal epithelial cells: regulators of barrier E. A. Wohlfert, J. Pott, T. Griseri, J. Bollrath, A. N. Hegazy, et al. 2014. The function and immune homeostasis. Nat. Rev. Immunol. 14: 141–153. alarmin IL-33 promotes regulatory T-cell function in the intestine. Nature 513: 72. Marchiando, A. M., W. V. Graham, and J. R. Turner. 2010. Epithelial barriers in 564–568. homeostasis and disease. Annu. Rev. Pathol. 5: 119–144. 97. Lemon, K. P., G. C. Armitage, D. A. Relman, and M. A. Fischbach. 2012. 73. Sabat, R., W. Ouyang, and K. Wolk. 2014. Therapeutic opportunities of the IL- Microbiota-targeted therapies: an ecological perspective. Sci. Transl. Med. 4: 22–IL-22R1 system. Nat. Rev. Drug Discov. 13: 21–38. 137rv5. 74. Johansson, M. E., H. E. Jakobsson, J. Holmeń-Larsson, A. Schutte,€ A. Ermund, 98. Olszak, T., D. An, S. Zeissig, M. P. Vera, J. Richter, A. Franke, J. N. Glickman, A. M. Rodrı´guez-Pin˜eiro, L. Arike, C. Wising, F. Svensson, F. Ba¨ckhed, and R. Siebert, R. M. Baron, D. L. Kasper, and R. S. Blumberg. 2012. Microbial G. C. Hansson. 2015. Normalization of host intestinal mucus layers requires long- exposure during early life has persistent effects on natural killer T cell function. term microbial colonization. Cell Host Microbe 18: 582–592. Science 336: 489–493. 75. Johansson, M. E., M. Phillipson, J. Petersson, A. Velcich, L. Holm, and 99. Arrieta, M. C., L. T. Stiemsma, P. A. Dimitriu, L. Thorson, S. Russell, S. Yurist- G. C. Hansson. 2008. The inner of the two Muc2 mucin-dependent mucus layers Doutsch, B. Kuzeljevic, M. J. Gold, H. M. Britton, D. L. Lefebvre, et al; CHILD in colon is devoid of bacteria. Proc. Natl. Acad. Sci. USA 105: 15064–15069. Study Investigators. 2015. Early infancy microbial and metabolic alterations affect 76. Ermund, A., A. Schutte,€ M. E. Johansson, J. K. Gustafsson, and G. C. Hansson. risk of childhood asthma. Sci. Transl. Med. 7: 307ra152. 2013. Studies of mucus in mouse stomach, small intestine, and colon. I. Gastroin- 100. Bjo¨rksteń, B., P. Naaber, E. Sepp, and M. Mikelsaar. 1999. The intestinal mi- testinal mucus layers have different properties depending on location as well as over croflora in allergic Estonian and Swedish 2-year-old children. Clin. Exp. Allergy 29: the Peyer’s patches. Am. J. Physiol. Gastrointest. Liver Physiol. 305: G341–G347. 342–346. 77. Wlodarska, M., B. Willing, K. M. Keeney, A. Menendez, K. S. Bergstrom, N. Gill, 101. Thompson-Chagoyan, O. C., M. Fallani, J. Maldonado, J. M. Vieites, S. Khanna, S. L. Russell, B. A. Vallance, and B. B. Finlay. 2011. Antibiotic treatment alters the C. Edwards, J. Dore´, and A. Gil. 2011. Faecal microbiota and short-chain fatty colonic mucus layer and predisposes the host to exacerbated Citrobacter rodentium- acid levels in faeces from infants with cow’s milk protein allergy. Int. Arch. Allergy induced colitis. Infect. Immun. 79: 1536–1545. Immunol. 156: 325–332. The Journal of Immunology 589

102. Nakayama, J., T. Kobayashi, S. Tanaka, Y. Korenori, A. Tateyama, N. Sakamoto, selection for management of children with cow’s milk allergy influences the C. Kiyohara, T. Shirakawa, and K. Sonomoto. 2011. Aberrant structures of fecal rate of acquisition of tolerance: a prospective multicenter study. J. Pediatr. bacterial community in allergic infants profiled by 16S rRNA gene pyrosequenc- 163: 771–7.e1. ing. FEMS Immunol. Med. Microbiol. 63: 397–406. 108. Berni Canani, R., N. Sangwan, A. T. Stefka, R. Nocerino, L. Paparo, R. Aitoro, 103. Kallioma¨ki, M., S. Salminen, H. Arvilommi, P. Kero, P. Koskinen, and E. Isolauri. A. Calignano, A. A. Khan, J. A. Gilbert, and C. R. Nagler. 2016. Lactobacillus 2001. Probiotics in primary prevention of atopic disease: a randomised placebo- rhamnosus GG-supplemented formula expands butyrate-producing bacterial strains controlled trial. Lancet 357: 1076–1079. in food allergic infants. ISME J. 10: 742–750. 104. Azad, M. B., T. Konya, D. S. Guttman, C. J. Field, M. R. Sears, K. T. HayGlass, 109. Wesemann, D. R., and C. R. Nagler. 2016. The microbiome, timing, and barrier P. J. Mandhane, S. E. Turvey, P. Subbarao, A. B. Becker, et al; CHILD Study function in the context of allergic disease. Immunity 44: 728–738. Investigators. 2015. Infant gut microbiota and food sensitization: associations in 110. David, L. A., A. C. Materna, J. Friedman, M. I. Campos-Baptista, the first year of life. Clin. Exp. Allergy 45: 632–643. M. C. Blackburn, A. Perrotta, S. E. Erdman, and E. J. Alm. 2014. Host lifestyle 105. Ling, Z., Z. Li, X. Liu, Y. Cheng, Y. Luo, X. Tong, L. Yuan, Y. Wang, J. Sun, affects human microbiota on daily timescales. [Published erratum appears in 2016 L. Li, and C. Xiang. 2014. Altered fecal microbiota composition associated with Genome Biol. 17: 117.] Genome Biol. 15: R89. food allergy in infants. Appl. Environ. Microbiol. 80: 2546–2554. 111. Olle, B. 2013. Medicines from microbiota. Nat. Biotechnol. 31: 309–315. 106. Tamburini, S., N. Shen, H. C. Wu, and J. C. Clemente. 2016. The microbiome in 112. Vickery, B. P., J. P. Berglund, C. M. Burk, J. P. Fine, E. H. Kim, J. I. Kim, early life: implications for health outcomes. Nat. Med. 22: 713–722. C. A. Keet, M. Kulis, K. G. Orgel, R. Guo, et al. 2016. Early oral immunotherapy 107.BerniCanani,R.,R.Nocerino,G.Terrin,T.Frediani,S.Lucarelli,L.Cosenza, in peanut-allergic preschool children is safe and highly effective. J. Allergy Clin. A. Passariello, L. Leone, V. Granata, M. Di Costanzo, et al. 2013. Formula Immunol. doi: 10.1016/j.jaci.2016.05.027. Downloaded from http://www.jimmunol.org/ by guest on January 26, 2017 Th eJournal of Brief Reviews Immunology

The Influence of the Microbiome on Type 1 Diabetes Alexandra Paun,*,1 Christopher Yau,*,†,1 and Jayne S. Danska*,†,‡ Type 1 diabetes (T1D) is characterized by the autoim- similarities with the human disease both in terms of patho- mune destruction of pancreatic b cells. The rapid rise genesis, the autoantigens being recognized, and genetic suscep- in T1D incidence during the past 50 y suggests envi- tibility loci (reviewed in Ref. 3). Genetic and DNA sequence ronmental factors contribute to the disease. The trillion analysis in humans and the NOD mouse model demonstrated symbiotic microorganisms inhabiting the mammalian that the most potent inherited determinant of susceptibility to gastrointestinal tract (i.e., the microbiota) influence nu- T1D were variants in the MHC class II genes encoding DQ merous aspects of host physiology. In this study we review b-chain and I-A b-chain (4, 5). Functional studies in mouse the evidence linking perturbations of the gut microbiome models then provided direct evidence that the T1D-associated to pancreatic autoimmunity. We discuss data from rodent class II variants impact the presentation of islet-derived peptides models demonstrating the essential role of the gut micro- to T cells. Although MHC haplotypes are key contributors to Downloaded from biota on the development and function of the host’s T1D risk, they clearly act in concert with multiple non-MHC mucosal and systemic immune systems. Furthermore, we genetic factors. Genome-wide association studies have identified review findings from human longitudinal cohort studies over 50 human T1D susceptibility regions that control disease examining the influence of environmental and lifestyle risk (6, 7). Similar studies in the NOD mouse model uncovered multiple non-MHC susceptibility genes, several of which were

factors on microbiota composition and pancreatic auto- http://www.jimmunol.org/ also associated with diabetes risk in humans (e.g., CTLA-4, immunity. Taken together, these data underscore the re- PTPN22, IL2RA) (8), supporting the relevance of the NOD quirement for mechanistic studies to identify bacterial mouse model for investigation of dysregulated immune path- components and metabolites interacting with the innate ways in the disease. and adaptive immune system, which would set the basis What the identiﬁcation of human genetic risk variants does for preventative or therapeutic strategies in T1D. The not explain is the sharp rise in T1D incidence and change in Journal of Immunology, 2017, 198: 590–595. age at onset over the past 50 y. Evidence of increasing prevalence of childhood T1D emerged from European data beginning after

he recent recognition that an individual’s intestinal the Second World War (9–11). Studies confirm an annual in- by guest on January 26, 2017 microbial community may modify risk for type 1 diabetes crease averaging 3% in both low and high incidence countries T (T1D) has emerged from a rich history of investigation between 1960 and 1996 (12). Equally striking is the increased into the multifactorial causes of the disease. The acute clinical frequency of T1D diagnosed in very early childhood (10). consequences of diabetes mellitus have been documented in the Recent reports from international prospective birth cohort medical literature for centuries. Following the discovery of in- studies demonstrate the appearance of anti-islet Abs in chil- sulin at the University of Toronto in the early 1920s, treatment dren carrying high risk HLA haplotypes during the first year of T1D patients with the purified hormone dramatically impro- of life (13, 14). Rapid change in disease incidence among ved the primary disease symptoms. However, it would take genetically stable populations points to recent, dynamic en- another half-century of work to firmly establish that a disease vironmental factors acting in concert with heritable T1D risk. characterized by profound metabolic dysfunction resulted Given the autoimmune basis of the disease, these factors ul- from an autoimmune destruction of insulin-producing cells. timately tune immune responses, particularly in early post- A new, and at inception unorthodox, focus on associations natal life, increasing the probability of responses to islet Ags. with HLA polymorphisms was crucial to building the par- These environmental factors are not yet defined. In this study, adigm that T1D results from a T cell mediated autoimmune we discuss the burgeoning evidence in rodent models and in attack on pancreatic b cells (1, 2). prospective studies of at-risk children that the composition of The most widely used animal model for the study of T1D is the intestinal microbial community has likely been impacted by the NOD mouse model. The NOD model displays significant improved public health, and the use of antibiotics in agriculture

*Program in Genetics and Genome Biology, The Hospital for Sick Children, Toronto, Address correspondence and reprint requests to Dr. Jayne S. Danska, The Hospital for Ontario M5G 0A4, Canada; †Department of Immunology, University of Toronto, Sick Children, Peter Gilgan Research Tower, 686 Bay Street, Toronto, ON M5G, Toronto, Ontario M5S 1A8, Canada; and ‡Department of Medical Biophysics, Univer- Canada. E-mail address: [email protected] sity of Toronto, Toronto, Ontario M5G 1L7, Canada Abbreviations used in this article: ABX, antibiotic; CR, cathelicidin-related; GF, germ 1A.P. and C.Y. contributed equally to this work. free; SCFA, short chain fatty acid; SPF, speciﬁc pathogen free; T1D, type 1 diabetes; TEDDY, The Environmental Determinants of Diabetes in the Young. Received for publication August 31, 2016. Accepted for publication September 30, 2016. Copyright Ó 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00 This work was supported by the Canadian Institutes of Health Research (Grant 272636 to J.S.D.) and the Juvenile Diabetes Research Foundation International (Grants 17-2011-520 and 2-SRA-2015-307-Q-R to J.S.D).

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601519 The Journal of Immunology 591 and for treatment of childhood illnesses over the past seven ribosomal rDNA showed that MyD88 deficiency resulted in decades. Elegant studies in rodent models demonstrate that the altered gut microbiota compared with NOD animals, and gut microbiota have essential effects on the development and colonization of GF NOD mice by cohousing with SPF NOD. 2 2 function of the host’s mucosal and systemic immune systems MyD88 / animals resulted in reduced insulitis, suggesting 2 2 (reviewed in Ref. 15). The understanding of this fundamental colonization by NOD.MyD88 / -associated microbiota was interaction between commensal microbes and human immunity, T1D protective. coupled with the rapid advancement in approaches to charac- These data provided the first clear links between innate terize the genomic composition and begin to predict the function sensing of the microbiota, the resulting alterations in micro- of these organisms, provide the framework necessary to address biota composition, and T1D development in the NOD the role of the gut microbiome in T1D development. mouse (22). Analysis of several TLR signaling–knockout NOD animals under GF conditions helped to refine the possible roles The influence of the microbiome in animal models of T1D of TLRs in immune–microbial cross-talk and T1D. TLR4 and Rodent models, particularly the NOD mouse, have provided a TLR2 microbiota-dependent signaling were found to mediate wealth of data regarding the complex genetic etiology and T1D protection and increased susceptibility, respectively (23). breakdown of immunological tolerance in T1D. NOD is an It is still unclear if TLR signaling is directly T1D protective, inbred mouse strain with a high incidence of spontaneous or if it results in microbiota modulation, which has down- autoimmune diabetes. Diabetes in the NOD mouse shares stream or feedback effects on the host. Subsequent studies further explored potential mechanisms of microbiota modu- many features with multiple aspects of human T1D, including Downloaded from genetic susceptibility, immunopathogenesis, and responsiveness lation of host physiology, and downstream effects on T1D to environmental influences. immunopathology. Like human T1D, T1D in NOD mice is polygenic; over 20 insulin-dependent diabetes loci have been identified in the Sex-microbiome interaction in T1D NOD mouse model of T1D (16). Several of these loci, and Many human autoimmune diseases display a strong female

the pathways in which they act, are shared with human T1D bias, whereas isolated T1D does not. The mechanisms that http://www.jimmunol.org/ risk variants. In both NOD mice and humans, the MHC control these sex-dependent differences in auto-immunity are haplotype, particularly the class II genes I-A in the mouse and poorly understood. Although a sex bias is not observed in most DR/DQ in humans, contribute the greatest component of human T1D, NOD mice exhibit a sex bias (greater in females genetic risk for T1D, but are not sufficient to confer the disease than males) (20) in T1D incidence. Under GF conditions, (5, 17). male NOD mice have similar T1D incidence to females, sug- The immunopathology seen in NOD T1D mirrors key gesting that the sex bias is microbiome dependent (24). In aspects of T1D in humans: prior to T1D onset, APC NOD mice, microbiota composition is similar between males and lymphocyte infiltration of the islets, termed insulitis (18), and females until divergence around puberty. Surprisingly, can be observed. Furthermore, islet auto-Ab production (19), the transfer of adult male (M→F) but not adult female (F→F) by guest on January 26, 2017 an important preclinical phenotype in T1D pathogenesis in intestinal microbiota into weanling-age female NOD mice humans, is also present in NOD mice, with multiple shared by oral gavage resulted in durable changes in recipient gut specificities. microbiota composition and, significantly, protected them Interestingly, T1D incidence in the NOD mouse is highly against T1D. M→F NOD mice also showed increased serum dependent on environmental exposure. Early in the study of testosterone levels and alterations in other serum metabolites. the NOD model, researchers noted that T1D incidence in NOD Furthermore, T cell adoptive transfers into NOD.SCID mice mice was strongly affected by the colony hygiene status (20, 21), showed T cells from M→F NOD mice were delayed in their and exposure to a wide range of microbes and microbially ability to induce T1D, suggesting changes in the microbiota derived products could suppress T1D. In light of inconclusive could alter T cell pathogenicity. Interestingly, all the effects associations between hygiene and T1D incidence in the human of male microbiome transfer were testosterone dependent, as population, these observations prompted studies into the effects treatment of recipients with an androgen receptor antagonist of microbial exposure and colonization on T1D in the NOD abrogated changes in metabolites and protection against anti- model. islet autoimmunity. These data suggested that there is a window TLRs detect exposure to pathogen-associated molecular in early life for durable modification of the gut microbial com- patterns and regulate host immune responses. MyD88 is a munity, and induced changes in the gut microbiota composition common adaptor protein for multiple TLRs, as well as IL-1R, can result in changes in host hormones, leading to downstream making it a crucial molecule for innate immune sensing, and metabolite modulation, altered immune cell function, and T1D likely important in immune–microbial cross-talk. To inves- protection. tigate the role of microbial exposure and innate immune A subsequent study confirmed that microbiota composition sensing in T1D, NOD mice genetically deficient in MyD88 diverges near puberty, and that the sex differences were absent 2 2 (NOD.MyD88 / ) were generated and raised under specific when comparing castrated adult NOD males to females (25). pathogen free (SPF) or germ-free (GF) conditions (22). Moreover, colonization of GF NOD mice by specific bacterial 2 2 NOD.MyD88 / mice were protected from T1D under SPF taxa (Enterobacteriacae and Segmented Filamentous Bacteria) but, surprisingly, not under GF conditions. These results that were enriched in SPF males versus females induced an suggested that MyD88-dependent signaling was important for elevation in testosterone levels. These data further confirmed a T1D development, and that the protective effect of MyD88 hormone-dependent window of microbiome development, deficiency was dependent on the presence of microbiota. Fur- and the importance of microbial colonization in modulating thermore, high-throughput sequencing of gut bacterial 16S host hormone levels. 592 BRIEF REVIEWS: GUT MICROBIOME AND TYPE 1 DIABETES

These studies demonstrate that transfer of microbiota in and immune cells in the pancreatic compartment, resulting in young prepubescent mice can result in durable changes in control of the immunopathology of T1D. It will be of interest microbiota composition and T1D protection, suggesting an to determine whether physiological levels of SCFA in systemic early life window for the microbiome modulation of the im- circulation act through CR antimicrobial peptides, and other mune system and T1D susceptibility. As we will discuss in the antimicrobial peptides, to regulate the function of regulatory context of human cohort studies, microbiome is heavily T cell and macrophages in pancreatic autoimmunity. influenced in early life by a number of environmental factors, Therefore, evidence from rodent models of T1D provide a including mode of delivery, maternal and offspring diet, and clear framework suggesting microbiota colonization events in antibiotic treatment. early life can be modulated by environmental factors, and that manipulation or disruption of these events can have down- Antibiotic perturbation of the microbiota in T1D stream hormonal and metabolic consequences. Furthermore, Antibiotic (ABX) use in agriculture and medicine has increased these changes in host physiology can lead to alterations in both substantially over the past 50 y and is implicated in the host innate and adaptive immunity, as well as b-cell biology. concurrent rise in diseases, including obesity and susceptibility These rodent models also provide a unique opportunity to to enteric pathogens such as Clostridium difficile (26, 27). NOD further interrogate the interaction between genetic and microbe- pups born to mothers treated from conception with a broad- associated T1D risk factors. The recent availability of gene- spectrum ABX (streptomycin, colistin, and ampicillin) or editing techniques such as CRISPR-Cas will allow investigation of genetic variants of T1D risk genes in vivo, and the human- vancomycin alone displayed an increase in T1D incidence in Downloaded from adulthood (28). 16S rDNA sequencing found major ABX- ization of mouse models by selective editing of orthologous induced alterations of gut microbiota composition, with the ex- genes. Humanization is also possible on the microbial front, as pansion of the genera Escherichia, Lactobacillus and Sutterella and the isolation and culturing of human-derived microbiota con- loss of Clostridiales, Lachnospiraceae, Prevotellaceae and Rikenellaceae. tinues. Studies in other animal models of human disease have Thesechangesinthemicrobialcommunity were correlated with shown that colonization of GF animals with human-derived microbes can recapitulate disease phenotypes. New strains of loss of IL-17 producing T cells in the gut lamina propria in http://www.jimmunol.org/ ABX-treated mice, suggesting alterations in the microbiome resulted genetically and microbially humanized mice will be valuable tools in corresponding changes in the mucosal immune system. for the future study of the interaction between human genetic Interestingly, ABX treatment need not be long term to pro- variants and commensal organisms, and could be used to test the duce similar effects on T1D incidence, microbiota composition, impact of specific organisms or consortia on T1D susceptibility and immune cell phenotypes. Pulsed ABX treatment with the and pathogenesis. macrolide tylosin in early life resulted in increased insulitis and T1D incidence in male NOD mice (29). Furthermore, pulsed Human studies link gut microbiota composition to T1D risk ABX treatment altered the gene expression associated with sterol The gut microbiome goes through a period of intense metabolism in male mice, as well as innate immunity and T cell remodeling from birth until 3 y of age when it transitions to an by guest on January 26, 2017 differentiation, corroborating the previous studies that suggested adult-like composition (31). The infant’s microbiome de- microbiome modulation of host hormone production, and in- pends on delivery mode (32) and is influenced during infancy nate and adaptive immunity. and early childhood by breastfeeding (33), the introduction of Clearly, robust evidence supports microbiome modulation solid foods (31), and administration of antibiotics (34). Given of the mucosal immune system, and the connection between the continuous cross-talk between the microbiome and the microbiome and T1D immunopathology. However, the signals — mucosal and systemic immunity, investigating the impact of molecular, cellular or otherwise — that mediate the connection these disruptions on the developing immune and endocrine between the microbiome and extraintestinal sites are still poorly systems is paramount. Longitudinal cohorts have been estab- understood. Antimicrobial peptides, expressed primarily in the lished to examine metabolic, immunologic, and microbiome- gut, control commensal microbes and have immunomodu- related parameters from an individual across multiple time latory effects in the mucosa. Recently, antimicrobial peptides points. These studies are designed to address issues such as the have been implicated in modulating pancreatic autoimmunity microbiota development under the influence of various envi- in the NOD model. Levels of cathelicidin-related (CR) anti- ronmental, dietary, and lifestyle factors and the impact it has microbial peptide produced by pancreatic b cells were found to on pancreatic autoimmunity. be greater in C56BL/6 compared with NOD mice (30). Ex- Although it is difficult to pinpoint individual causal factors, ogenous systemic administration of CR antimicrobial peptide the early childhood period of dynamic changes of the microbiota suppressed T1D incidence in NOD mice and was associated and immune system coincides with the first measurements of with augmented frequencies of islet-associated regulatory autoantibodies associated with T1D as evidenced by prospective macrophages and T cells. CR antimicrobial peptide production birth cohort studies of children with high risk HLA haplotypes by b-cells was correlated with short chain fatty acid (SCFA) (13, 35–37). T1D is a T cell-mediated disease; however, the concentration in the blood and feces of the mice, and ad- development of autoantibodies is a major event in the pro- ministration of supra-physiological levels of SCFA resulted in gression toward T1D in a susceptible individual (38). T1D risk increased CR antimicrobial peptide production and decreased is calculated based on multiple variables, including family his- T1D incidence. Furthermore, perturbations of the microbiome tory and genetic risk, age, and the presence of autoantibodies by antibiotic treatment or M→F gut microbiota transfer were against one or more islet Ag groups. Identifying at-risk indi- associated with alterations in islet CR antimicrobial peptide viduals before disease onset is key for establishing longitudinal concentration and regulatory cell phenotypes. The authors study cohorts, which are essential for uncovering the environ- suggested that microbial metabolites regulate b-cell function mental determinants to autoimmunity. The Journal of Immunology 593

Longitudinal cohort studies of microbiome composition and T1D risk that future microbiome-based therapeutic strategies should be T1D incidence varies widely across the globe with Scandi- tailored based on geographical location. navian countries displaying some of the highest incidences in A longitudinal gut microbiome analysis from 4 to 26 mo of the world (39). The .10-fold difference in diabetes incidence age in 76 children from the Type 1 Diabetes Prediction and across Europe is only partly explained by the differences in Prevention project from Turku (Finland) found an increased high-risk HLA allele distributions (40). For example, there are abundance of Bacteroides dorei in the children who seroconver- striking the differences in the rate of T1D in the genetically ted to display autoantibodies compared with nonconverters similar populations of Estonia, Finland and Russian Karelia, (50). The abundance of B. dorei peaked at 7.6 mo of age in which has prompted comparative population studies to inves- autoantibody-positive children and preceded the appearance of tigate the environmental triggers of T1D discussed below. the first anti-islet autoantibodies (50). The same authors later Two dietary intervention studies were established in Finland examined the B. dorei genome methylation status in stool to test the hypothesis that bovine insulin present in cow milk samples from one case and one control Type 1 Diabetes protein is a diabetogenic factor (41). In the TRIGR (Trial to Prediction and Prevention subject containing high abundances Reduce IDDM in the Genetically at Risk) study (42) expo- of this bacterium (51). DNA adenosine methylation of the 59- sure to foreign dietary proteins was postponed until 6–8 mo GATC-39 motif in bacterial genomes regulates gene expres- of age and the development of autoantibodies was monitored sion and has been associated with increased bacterial virulence until the age of 6 y. Results from the FINDIA (Finnish Di- (52). The B. dorei genome from the control subject included no DNA methylation sites, whereas in the seroconverted child etary Intervention Trial for the Prevention of Type 1 Dia- Downloaded from betes) pilot study indicate the use of a bovine insulin-free more than 20,000 methylated sites were found (51). These formula during the first 6 mo of life decreased the appearance preliminary findings suggest that in addition to taxonomic of b-cell autoantibodies at the age of 3 y (43). A cross-sectional composition, microbial gene expression needs to be examined study based on these longitudinal cohorts compared the gut to begin to address microbiome function even when there is microbiomes of 18 children who tested positive for autoanti- no evidence of differential abundance of taxonomic groups. The DIABIMMUNE cohort study was initiated in 2008 body and 18 autoantibodies-negative controls (44). An in- http://www.jimmunol.org/ creased abundance of Bacteroides, decreased Bifidobacterium to examine the environmental causes underlying the 6–7-fold species and a lower abundance of lactate- and butyrate- higher incidence of autoimmune and allergic diseases in Finland producing bacteria characterized the microbiotas of children compared with neighboring Russian Karelia (53). This project is who seroconverted compared with nonconverters (44). designed to test the hygiene hypothesis in T1D in communities Certain high-risk HLA haplotypes are shared between celiac with different public health standards. DIABIMMUNE has disease and T1D patients (45). The German BABYDIET recruited newborn infants with defined HLA risk haplotypes as- cohort was initiated to investigate whether delayed exposure sociated with autoimmunity from Finland, Estonia, and Russian to dietary gluten could delay development of b-cell autoantibodies Karelia and collected longitudinal blood and stool samples from by guest on January 26, 2017 (46). Although no protective effect was found, a longitudinal 1moto3yofagealongwithextensiveclinicalmetadata(53). analysis of the gut microbiota of 22 autoantibody-positive and An initial analysis of the gut microbiome in the DIABIMMUNE 22 autoantibody-negative BABYDIET study children revealed cohort focused on Finnish and Estonian subjects (54). The gut altered bacterial interaction networks at the ages of 6 and microbiomes of 11 children who developed islet autoanti- 24mo(47).IncontrasttotheFinnishstudies,however,nodif- bodies, four of whom became diabetic during the study pe- ferences were observed in fecal microbiota diversity or com- riod, were analyzed together with 22 children who remained position between cases and controls. This contrast in outcomes autoantibody-free using 16S rRNA sequencing as well may indicate that similar to the population stratification issue as microbial metagenomics. A decrease in microbiome encountered by genome-wide association studies, other vari- diversity and reduced bacterial gene content were found in ables such as geographical location or ethnicity may preclude autoantibody-positive children during progression to T1D (54). the identification of predisposing or protective bacteria across Specifically, a decrease in Lachnospiraceae and Veillonellaceae was multiple populations. observed in children who became diabetic, accompanied by an The Environmental Determinants of Diabetes in the Young increase in Streptococcus, Blautia,andRuminococcus genera. (TEDDY) cohort (48) has enrolled over 8000 children with Functional analysis predicted from bacterial gene content high-risk HLA haplotypes from the U.S., Germany, Sweden, revealed that bacterial metabolism in autoantibody+ subjects was and Finland, and is well positioned to address the question of characterized by a greater abundance of genes involved in sugar microbiome heterogeneity and its impact on pancreatic au- transport and fewer genes for amino acid biosynthesis (54). toimmunity. Indeed, a pilot study of 90 high-risk, prediabetic Correlation between fecal bacterial abundance and host serum infants from the TEDDY cohort revealed strong geographical metabolites suggested that the microbiota of children who de- influences on the microbiome composition (49). The early- veloped T1D may be linked to an inflammatory environment life microbiomes of children from Finland and Colorado were conducive to autoimmunity (54). significantly less diverse than subjects from Sweden, Germany, A subsequent DIABIMMUNE cohort study reported meta- and Washington state (49). Bifidobacterium dominated the genomic characterization of the fecal microbiomes of 222 microbiota of infants up to 10 mo of age from Sweden and Finish, Estonian, and Russian children. The goal was to identify Washington, whereas subjects from the states of Georgia and factors associated with the risk of atopy and islet autoimmunity Florida and from Germany displayed greater abundances of (14). In this study, there were no correlations between islet Clostridium, Bifidobacterium,andVeillonella (49). These dif- autoantibodies or T1D status with microbiome composition. ferences in diversity and composition remained significant after The authors reported associations between microbiome taxo- adjustment for early-life and lifestyle variables and suggest nomic composition and gene content with geographical location 594 BRIEF REVIEWS: GUT MICROBIOME AND TYPE 1 DIABETES and considered this a proxy for higher (Finland and Estonia) or to autoimmune disease. Increasingly the focus is shifting to lower (Russia) predisposition to immune-mediated diseases. analysis of the microbial genomes, transcripts, metabolites, In the early-life microbiome, there were differences between and their antigenic potential to elucidate how the presence of the three locations in abundance of genes associated with the detrimental microbes or absence of beneficial ones promotes utilization of human milk oligosaccharides from breast milk. autoimmunity in tissues distant from the intestine. Finnish and Estonian infant human milk oligosaccharides Taken together, the findings from human cohorts support were mainly associated with Bacteroidetes whereas this func- the existing hypothesis emerging from animal models, i.e., that tion was associated with Bifidobacteria in the Russian children a continuous dialog exists between the resident gut microbiota (14). This difference in bacterial colonization of the infant and the immune system (56). Beginning with the gestational gut was associated with microbiome composition later in life, environment and continuing with infancy and early child- leading the authors to suggest that the higher probability of hood, the multiple microbial cues promote the development atopy/autoimmunity in Finnish and Estonian compared with of the immune system, readying it to cope with a more com- Russian children is associated with an increased abundance of plex microbiota and potential pathogens (57). In the context of Bacteroides species in infancy and early childhood (14). this intimate relationship, a microbiome imbalance can lead to Whether the presence of Bifidobacteria conferred protection autoimmunity in a genetically susceptible host, unable to pre- from immune-mediated diseases in Russian children is still vent the development of autoreactive cells. unknown, however, the Bifidobacterium species is one of the main components of probiotic preparations available. A re- Conclusions Downloaded from cent report from the TEDDY study group revealed that in Recent findings underscore the role of the gut microbiota as children with the high-risk HLA-DR3/4 genotype, adminis- a critical factor involved in the development of immune- tration of probiotics during the first 27 d of life was associated mediated diseases. Longitudinal studies in humans suggest with a 60% reduction in the autoimmunity risk (55). that individuals progressing to T1D exhibit a decrease in micro- In terms of predicted microbial function, the DIABIMMUNE biota diversity and intestinal dysbiosis characterized by loss of beneficial organisms. However, the mechanisms by which al-

study authors focused on the geographical disparities in terms http://www.jimmunol.org/ of LPS biosynthesis (14). In Finnish and Estonian children terations in the gut microbiota mediate tissue-specific auto- Bacteroides were predicted to be the main source of LPS, immunity are not yet understood. Future studies are needed to whereas in Russian subjects LPS may be mainly derived from investigate how microbially derived signals are broadcast from Escherichia coli. Exposure of healthy donor peripheral blood the mucosal environment to islet-specific reactivity. These studies cells to these two sources of LPS revealed that E. coli–derived should address the interaction of microbial components and LPS was more immunostimulatory and induced endotoxin metabolites with the adaptive immune system resulting in tolerance compared with Bacteroides LPS. The authors suggested T cell reactivity to islet Ags. that early exposure to LPS tunes immune responses to subsequent Ag exposures such that signals from stimulatory microbes, Disclosures by guest on January 26, 2017 like E. coli, dampen the immune response to autoantigens and The authors have no financial conflicts of interest. protect from autoimmunity (14). This is an interesting idea that will be pursued as the study cohort continues to mature. References Recognition of LPS and other bacterial components by 1. Nerup, J., P. Platz, O. O. Andersen, M. Christy, J. Lyngsoe, J. E. Poulsen, pathogen recognition receptors is one aspect of the complex L. P. Ryder, L. S. Nielsen, M. Thomsen, and A. Svejgaard. 1974. HL-A antigens and diabetes mellitus. Lancet 2: 864–866. cross-talk between the innate immune system and the gut 2. Cudworth, A. G., and J. C. Woodrow. 1974. Letter: HL-A antigens and diabetes microbiota. Previous studies in NOD mice with defects in mellitus. Lancet 2: 1153. various innate immunity–sensing pathways provide mecha- 3. Thayer, T. C., S. B. Wilson, and C. E. Mathews. 2010. Use of nonobese diabetic mice to understand human type 1 diabetes. Endocrinol. Metab. Clin. North Am. 39: 541–561. nistic evidence that abnormal sensing of certain members of 4. Acha-Orbea, H., and H. O. McDevitt. 1987. The first external domain of the the microbiota may promote the development of T1D (22, nonobese diabetic mouse class II I-A beta chain is unique. Proc. Natl. Acad. Sci. USA 84: 2435–2439. 23). To elucidate how changes in microbiome composition 5. Todd, J. A., J. I. Bell, and H. O. McDevitt. 1987. HLA-DQ beta gene contributes and function affect diabetes risk in genetically predisposed in- to susceptibility and resistance to insulin-dependent diabetes mellitus. Nature 329: 599–604. dividuals, future human studies should investigate the cellular 6. Barrett, J. C., D. G. Clayton, P. Concannon, B. Akolkar, J. D. Cooper, players responding to bacterial signals and the mechanisms by H. A. Erlich, C. Julier, G. Morahan, J. Nerup, C. Nierras, et al. 2009. Genome- which they promote autoimmunity in a distant organ. wide association study and meta-analysis find that over 40 loci affect risk of type 1 diabetes. Nat. Genet. 41: 703–707. A great advantage of prospective longitudinal studies is that 7. Concannon, P., S. S. Rich, and G. T. Nepom. 2009. Genetics of type 1A diabetes. they allow the investigation of whether the changes in micro- N. Engl. J. Med. 360: 1646–1654. 8. Wicker, L. S., J. Clark, H. I. Fraser, V. E. Garner, A. Gonzalez-Munoz, B. Healy, biome composition precede, accompany, or follow the devel- S. Howlett, K. Hunter, D. Rainbow, R. L. Rosa, et al. 2005. Type 1 diabetes genes opment of T1D. However, establishing a causal relationship and pathways shared by humans and NOD mice. J. Autoimmun. 25(Suppl): 29–33. 9. Bingley, P. J., and E. A. Gale. 1989. Rising incidence of IDDM in Europe. Diabetes between microbiome alterations and autoimmunity in human Care 12: 289–295. cohort studies is challenged by the complexity of the microbiota- 10. Kurtz, Z., C. S. Peckham, and A. E. Ades. 1988. Changing prevalence of juvenile- immune system interaction. Recent findings from hetero- onset diabetes mellitus. Lancet 2: 88–90. 11. Gale, E. A. 2002. The rise of childhood type 1 diabetes in the 20th century. Diabetes geneous populations recruited in multinational studies such as 51: 3353–3361. TEDDY (49) observed significant differences in microbiota 12. Onkamo, P., S. Vaä¨na¨nen, M. Karvonen, and J. Tuomilehto. 1999. Worldwide increase in incidence of Type I diabetes–the analysis of the data on published in- composition in children at-risk for T1D from the different cidence trends. Diabetologia 42: 1395–1403. study sites. Furthermore, the taxonomic composition of the gut 13. Krischer, J. P., K. F. Lynch, D. A. Schatz, J. Ilonen, A˚. Lernmark, W. A. Hagopian, M. J. Rewers, J. X. She, O. G. Simell, J. Toppari, et al. 2015. The 6 year incidence microbiota is an important component of these human studies but of diabetes-associated autoantibodies in genetically at-risk children: the TEDDY provides limited mechanistic insight into microbial contributions study. Diabetologia 58: 980–987. The Journal of Immunology 595

14. Vatanen, T., A. D. Kostic, E. d’Hennezel, H. Siljander, E. A. Franzosa, M. Yassour, 36. Ziegler, A. G., and E. Bonifacio; BABYDIAB-BABYDIET Study Group. 2012. R. Kolde, H. Vlamakis, T. D. Arthur, A. M. Ha¨ma¨laïnen, et al. 2016. Variation in Age-related islet autoantibody incidence in offspring of patients with type 1 dia- microbiome LPS immunogenicity contributes to autoimmunity in humans. [Pub- betes. Diabetologia 55: 1937–1943. lished erratum appears in 2016 Cell 165: 1551.] Cell 165: 842–853. 37. Parikka, V., K. Na¨nto¨-Salonen, M. Saarinen, T. Simell, J. Ilonen, H. Hyo¨ty, 15. Hooper, L. V., D. R. Littman, and A. J. Macpherson. 2012. Interactions between R. Veijola, M. Knip, and O. Simell. 2012. Early seroconversion and rapidly in- the microbiota and the immune system. Science 336: 1268–1273. creasing autoantibody concentrations predict prepubertal manifestation of type 1 16. Todd, J. A., and L. S. Wicker. 2001. Genetic protection from the inflammatory diabetes in children at genetic risk. Diabetologia 55: 1926–1936. disease type 1 diabetes in humans and animal models. Immunity 15: 387–395. 38. Ziegler, A. G., and G. T. Nepom. 2010. Prediction and pathogenesis in type 1 17. Wicker, L. S., J. A. Todd, and L. B. Peterson. 1995. Genetic control of autoimmune diabetes. Immunity 32: 468–478. diabetes in the NOD mouse. Annu. Rev. Immunol. 13: 179–200. 39. DIAMOND Project Group. 2006. Incidence and trends of childhood Type 1 di- 18. Ghosh, S., S. M. Palmer, N. R. Rodrigues, H. J. Cordell, C. M. Hearne, abetes worldwide 1990-1999. Diabet. Med. 23: 857–866. R. J. Cornall, J. B. Prins, P. McShane, G. M. Lathrop, L. B. Peterson, et al. 1993. 40. Rønningen, K. S., N. Keiding, and A. Green; Genomic Marker Contributors; Polygenic control of autoimmune diabetes in nonobese diabetic mice. Nat. Genet. 4: EURODIAB ACE Study Group. 2001. Correlations between the incidence of 404–409. childhood-onset type I diabetes in Europe and HLA genotypes. Diabetologia 44: 19. Miao, D., L. Yu, and G. S. Eisenbarth. 2007. Role of autoantibodies in type 1 B51–B59. diabetes. Front. Biosci. 12: 1889–1898. 41. Vaarala, O. 2006. Is it dietary insulin? Ann. N. Y. Acad. Sci. 1079: 350–359. 20. Pozzilli, P., A. Signore, A. J. Williams, and P. E. Beales. 1993. NOD mouse col- 42. Knip, M., S. M. Virtanen, K. Seppa¨, J. Ilonen, E. Savilahti, O. Vaarala, onies around the world–recent facts and figures. Immunol. Today 14: 193–196. A. Reunanen, K. Teramo, A. M. Ha¨ma¨laïnen, J. Paronen, et al. 2010. Dietary 21. Bach, J. F. 2002. The effect of infections on susceptibility to autoimmune and intervention in infancy and later signs of beta-cell autoimmunity. N. Engl. J. Med. allergic diseases. N. Engl. J. Med. 347: 911–920. 363: 1900–1908. 22. Wen, L., R. E. Ley, P. Y. Volchkov, P. B. Stranges, L. Avanesyan, A. C. Stonebraker, 43. Vaarala, O., J. Ilonen, T. Ruohtula, J. Pesola, S. M. Virtanen, T. Ha¨rko¨nen, C. Hu, F. S. Wong, G. L. Szot, J. A. Bluestone, et al. 2008. Innate immunity and M. Koski, H. Kallioinen, O. Tossavainen, T. Poussa, et al. 2012. Removal of bovine intestinal microbiota in the development of Type 1 diabetes. Nature 455: 1109–1113. insulin from cow’s milk formula and early initiation of beta-cell autoimmunity in 23. Burrows, M. P., P. Volchkov, K. S. Kobayashi, and A. V. Chervonsky. 2015. the FINDIA pilot study. Arch. Pediatr. Adolesc. Med. 166: 608–614. Microbiota regulates type 1 diabetes through Toll-like receptors. Proc. Natl. Acad. 44. de Goffau, M. C., K. Luopaja¨rvi, M. Knip, J. Ilonen, T. Ruohtula, T. Ha¨rko¨nen,

Sci. USA 112: 9973–9977. L. Orivuori, S. Hakala, G. W. Welling, H. J. Harmsen, and O. Vaarala. 2013. Fecal Downloaded from 24. Markle, J. G., D. N. Frank, S. Mortin-Toth, C. E. Robertson, L. M. Feazel, microbiota composition differs between children with b-cell autoimmunity and U. Rolle-Kampczyk, M. von Bergen, K. D. McCoy, A. J. Macpherson, and those without. Diabetes 62: 1238–1244. J. S. Danska. 2013. Sex differences in the gut microbiome drive hormone- 45. Smigoc Schweiger, D., A. Mendez, S. Kunilo Jamnik, N. Bratanic, N. Bratina, dependent regulation of autoimmunity. Science 339: 1084–1088. T. Battelino, J. Brecelj, and B. Vidan-Jeras. 2016. High-risk genotypes HLA-DR3- 25. Yurkovetskiy, L., M. Burrows, A. A. Khan, L. Graham, P. Volchkov, L. Becker, DQ2/DR3-DQ2 and DR3-DQ2/DR4-DQ8 in co-occurrence of type 1 diabetes D. Antonopoulos, Y. Umesaki, and A. V. Chervonsky. 2013. Gender bias in au- and celiac disease. Autoimmunity 49: 240–247. toimmunity is inﬂuenced by microbiota. Immunity 39: 400–412. 46. Gruhn, M., P. Rosenbaum, H. P. Bollhagen, and A. Bueschges. 2011. Studying the 26. Loo, V. G., A. M. Bourgault, L. Poirier, F. Lamothe, S. Michaud, N. Turgeon, neural basis of adaptive locomotor behavior in insects. J. Vis. Exp. (50):2629.

B. Toye, A. Beaudoin, E. H. Frost, R. Gilca, et al. 2011. Host and pathogen factors 47. Endesfelder, D., W. zu Castell, A. Ardissone, A. G. Davis-Richardson, http://www.jimmunol.org/ for Clostridium difficile infection and colonization. N. Engl. J. Med. 365: 1693–1703. P. Achenbach, M. Hagen, M. Pflueger, K. A. Gano, J. R. Fagen, J. C. Drew, et al. 27. Hensgens, M. P., E. C. Keessen, M. M. Squire, T. V. Riley, M. G. Koene, E. de 2014. Compromised gut microbiota networks in children with anti-islet cell auto- Boer, L. J. Lipman, and E. J. Kuijper; European Society of Clinical Microbiology immunity. Diabetes 63: 2006–2014. and Infectious Diseases Study Group for Clostridium difficile (ESGCD). 2012. 48. Hagopian, W. A., H. Erlich, A. Lernmark, M. Rewers, A. G. Ziegler, O. Simell, Clostridium difficile infection in the community: a zoonotic disease? Clin. Microbiol. B. Akolkar, R. Vogt, Jr., A. Blair, J. Ilonen, et al. 2011. The environmental de- Infect. 18: 635–645. terminants of diabetes in the young (TEDDY): genetic criteria and international 28. Candon, S., A. Perez-Arroyo, C. Marquet, F. Valette, A. P. Foray, B. Pelletier, diabetes risk screening of 421 000 infants. Pediatr. Diabetes 12: 733–743. C. Milani, M. Ventura, J. F. Bach, and L. Chatenoud. 2015. Antibiotics in early life 49. Kemppainen, K. M., A. N. Ardissone, A. G. Davis-Richardson, J. R. Fagen, alter the gut microbiome and increase disease incidence in a spontaneous mouse K. A. Gano, L. G. Leoń-Novelo, K. Vehik, G. Casella, O. Simell, A. G. Ziegler, model of autoimmune insulin-dependent diabetes. [Published erratum appears in et al. 2015. Early childhood gut microbiomes show strong geographic differences 2016 PLoS One 11: e0147888.] PLoS One 10: e0125448. among subjects at high risk for type 1 diabetes. Diabetes Care 38: 329–332. 29. Livanos, A. E., T. U. Greiner, P. Vangay, W. Pathmasiri, D. Stewart, S. McRitchie, 50. Davis-Richardson, A. G., A. N. Ardissone, R. Dias, V. Simell, M. T. Leonard, by guest on January 26, 2017 H. Li, J. Chung, J. Sohn, S. Kim, et al. 2016. Antibiotic-mediated gut microbiome K. M. Kemppainen, J. C. Drew, D. Schatz, M. A. Atkinson, B. Kolaczkowski, et al. perturbation accelerates development of type 1 diabetes in mice. Nat. Microbiol. 1: 2014. Bacteroides dorei dominates gut microbiome prior to autoimmunity in 16140. Finnish children at high risk for type 1 diabetes. Front. Microbiol. 5: 678. 30. Sun, J., L. Furio, R. Mecheri, A. M. van der Does, E. Lundeberg, L. Saveanu, 51. Leonard, M. T., A. G. Davis-Richardson, A. N. Ardissone, K. M. Kemppainen, Y. Chen, P. van Endert, B. Agerberth, and J. Diana. 2015. Pancreatic b-cells limit J. C. Drew, J. Ilonen, M. Knip, O. Simell, J. Toppari, R. Veijola, et al. 2014. The autoimmune diabetes via an immunoregulatory antimicrobial peptide expressed methylome of the gut microbiome: disparate Dam methylation patterns in intestinal under the influence of the gut microbiota. Immunity 43: 304–317. Bacteroides dorei. Front. Microbiol. 5: 361. 31. Koenig, J. E., A. Spor, N. Scalfone, A. D. Fricker, J. Stombaugh, R. Knight, 52. Heusipp, G., S. Fa¨lker, and M. A. Schmidt. 2007. DNA adenine methylation and L. T. Angenent, and R. E. Ley. 2011. Succession of microbial consortia in the bacterial pathogenesis. Int. J. Med. Microbiol. 297: 1–7. developing infant gut microbiome. Proc. Natl. Acad. Sci. USA 108(Suppl 1): 4578– 53. Kondrashova, A., T. Seiskari, J. Ilonen, M. Knip, and H. Hyo¨ty. 2013. The ‘Hygiene 4585. hypothesis’ and the sharp gradient in the incidence of autoimmune and allergic dis- 32. Dominguez-Bello, M. G., E. K. Costello, M. Contreras, M. Magris, G. Hidalgo, eases between Russian Karelia and Finland. APMIS 121: 478–493. N. Fierer, and R. Knight. 2010. Delivery mode shapes the acquisition and structure 54.Kostic,A.D.,D.Gevers,H.Siljander,T.Vatanen,T.Hyo¨tylaïnen, A.-M. Ha¨ma¨laïnen, of the initial microbiota across multiple body habitats in newborns. Proc. Natl. Acad. A. Peet, V. Tillmann, P. Po¨ho¨, I. Mattila, et al. 2015. The dynamics of the human infant Sci. USA 107: 11971–11975. gut microbiome in development and in progression toward type 1 diabetes. Cell Host 33. Palmer, C., E. M. Bik, D. B. DiGiulio, D. A. Relman, and P. O. Brown. 2007. Microbe 17: 260–273. Development of the human infant intestinal microbiota. PLoS Biol. 5: e177. 55. Uusitalo, U., X. Liu, J. Yang, C. A. Aronsson, S. Hummel, M. Butterworth, 34. Cox, L. M., S. Yamanishi, J. Sohn, A. V. Alekseyenko, J. M. Leung, I. Cho, A˚ . Lernmark, M. Rewers, W. Hagopian, J. X. She, et al 2016. Association of early S. G. Kim, H. Li, Z. Gao, D. Mahana, et al. 2014. Altering the intestinal micro- exposure of probiotics and islet autoimmunity in the TEDDY study. JAMA Pediatr. biota during a critical developmental window has lasting metabolic consequences. 170: 20–28. Cell 158: 705–721. 56. Hooper, L. V., and A. J. Macpherson. 2010. Immune adaptations that maintain 35. Kimpima¨ki, T., A. Kupila, A. M. Ha¨ma¨laïnen, M. Kukko, P. Kulmala, K. Savola, homeostasis with the intestinal microbiota. Nat. Rev. Immunol. 10: 159–169. T. Simell, P. Keskinen, J. Ilonen, O. Simell, and M. Knip. 2001. The first signs of 57. Gomez de Aguero,€ M., S. C. Ganal-Vonarburg, T. Fuhrer, S. Rupp, Y. Uchimura, beta-cell autoimmunity appear in infancy in genetically susceptible children from H. Li, A. Steinert, M. Heikenwalder, S. Hapfelmeier, U. Sauer, et al. 2016. The the general population: the finnish Type 1 diabetes prediction and prevention study. maternal microbiota drives early postnatal innate immune development. Science J. Clin. Endocrinol. Metab. 86: 4782–4788. 351: 1296–1302. Th eJournal of Brief Reviews Immunology

Influence of the Gut Microbiome on Autoimmunity in the Central Nervous System Sara L. Colpitts and Lloyd H. Kasper Autoimmune disorders of the CNS have complex path- zygotic twins highlights the importance of environmental in- ogeneses that are not well understood. In multiple fluence on disease onset (5, 6). It has been postulated that sclerosis and neuromyelitis optica spectrum disorders, bacterial and viral infections acquired from the environment T cells destroy CNS tissue, resulting in severe disabil- can trigger the development of disease by molecular mimicry ities. Mounting evidence suggests that reducing inflam- and bystander activation mechanisms. The microbiota that mation in the CNS may start with modulation of the colonize the intestine (referred to herein as the gut micro- gut microbiome. The lymphoid tissues of the gut are biome), although located within the confines of the body, are foreign organisms that have evolved to live in symbiosis with specialized for the induction of regulatory cells, which Downloaded from are directly responsible for the suppression of CNS- their human host. Thus, the composition of the gut micro- damaging autoreactive T cells. Whether cause or effect, biome has the potential to influence MS pathogenesis. Indeed, the onset of dysbiosis in the gut of patients with multiple germ-free (GF), or gnotobiotic, mice devoid of all commensal gut flora have dramatically attenuated susceptibility to experi- sclerosis and neuromyelitis optica provides evidence of mental autoimmune (or allergic) encephalomyelitis (EAE), a communication along the gut–brain axis. Thus, current mouse model of human MS (7, 8). Conversely, the existence of http://www.jimmunol.org/ and future therapeutic interventions directed at microbiome CNS disease has the potential to impact the homeostasis of the modulation are of considerable appeal. The Journal of gut. Indeed, mice with EAE have increased permeability of the Immunology, 2017, 198: 596–604. intestinal mucosa allowing for leakage of luminal contents into the body (9). In patients with MS, the existence of a gut–brain utoimmune disorders of the CNS have debilitating connection is further evidenced by increases in constipation, consequences in afflicted patients. In multiple scle- fecal incontinence, and gut permeability (10, 11). Intestinal bowel A rosis (MS), autoreactive T cells attack the myelin disease is also more common in MS patients and their families sheath surrounding nerves in the brain and spinal cord by (12, 13). However, gut microbes themselves play an important secreting a variety of inflammatory mediators that result in role in maintaining the integrity of the intestinal epithelium (14). by guest on January 26, 2017 demyelination of both white and gray matter. Although disease Furthermore, the gut microbiota can influence the permeability symptoms are heterogeneous, such destruction can result in of the blood–brain barrier (BBB) by modulating expression of fatigue, numbness or tingling in the extremities, muscle weak- tight junction proteins in endothelial tissues (15). ness, dizziness and vertigo, bladder and bowel problems, changes in cognitive function, and emotional instability. Neuromyelitis IL-17–producing Th17 cells drive pathology in CNS autoimmune optica (NMO) is an autoimmune disease within the larger family disease of NMO spectrum disorders that is similar to MS and exhibits The generation of Th17 cells is intimately linked to the gut some overlapping symptoms, but NMO is more often associated microbiome. GF mice have reduced numbers of Th17 cells, but with astrocyte injury and damage to the optic nerve and spinal reconstitution of GF mice with a single commensal called cord that can cause pain and vision loss in addition to muscle- segmented filamentous bacteria can induce Th17 cells and re- related deficiencies. store susceptibility to EAE (8, 16). The importance of the gut The etiology of MS is complex, involving both genetic and microbiota in the development of Th17 cells and their role in environmental factors. Although the highly polymorphic HLA CNS autoimmunity have been reviewed in detail by others (17, genes were the first to be identified and remain the most studied, 18). Briefly, Th17 cells are a subset of CD4+ effector T cells technological advances have allowed for the identification of that express the transcription factor retinoic acid (RA)–related additional single nucleotide polymorphisms that correlate with orphan receptor gt and can produce a variety of cytokines such disease (most of which are associated with immune function) as IL-17A, IL-17F, IL-22, and GM-CSF. The differentiation of (1–4). However, the low concordance rate between homo- Th17 cells is dependent on TGF-b, IL-6, IL-21, and IL-1b

Department of Microbiology and Immunology, Geisel School of Medicine, Dartmouth Abbreviations used in this article: AQP4, aquaporin-4; BBB, blood–brain barrier; Breg, College, Hanover, NH 03755 regulatory B cell; DC, dendritic cell; EAE, experimental autoimmune encephalomyelitis; GF, germ-free; IEC, intestinal epithelial cell; LN, lymph node; MS, multiple sclerosis; Received for publication August 18, 2016. Accepted for publication October 12, 2016. NMO, neuromyelitis optica; PSA, polysaccharide A; RA, retinoic acid; SCFA, short- This work was supported by grants from the National Institutes of Health (AI110170) chain fatty acid; Treg, regulatory T cell. and the National Multiple Sclerosis Society (RG4662A2/1) (to L.H.K.). Ó Address correspondence and reprint requests to Dr. Sara L. Colpitts, Geisel School of Copyright 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00 Medicine, 706 Remsen, 66 College Street, Hanover, NH 03755. Email address: sara.l. [email protected]

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601438 The Journal of Immunology 597

(present in various combinations) whereas IL-23 is required for tight junctions between intestinal epithelial cells (IECs) (28). their stability and maintenance. In addition to cytokine-derived DC probing of the lumen is dependent on MyD88 signaling in signals, engagement of the aryl hydrocarbon receptor, known to neighboring IECs, which suggests an essential role for IEC–DC interact with environmental toxins, drives the generation of cross-talk in detection of intestinal microbiota during steady- Th17 cells and can exacerbate EAE disease (19). Extracellular state and dysbiosis (29). Specialized IECs called M cells can ATP is yet another factor that can promote Th17 development also transport material across the epithelial barrier and deliver (20). Ligands of the aryl hydrocarbon receptor and ATP can luminal content to DCs located in the underlying lymphoid both be produced by the microbiota, thus affecting the devel- tissue (30). Although IECs have the ability to influence opment of Th17 cells (17). immune cell behavior (reviewed in Ref. 31), their role in CNS Th17 cells are essential for host defense against bacterial and autoimmunity is poorly defined. However, Kusu et al. (32) fungal pathogens, but in the context of autoimmunity, their have shown that IECs can directly limit commensal-dependent presence is often destructive. Even before Th17 cells had been ATP levels in the small intestine through expression of identified and characterized, increased IL-17 mRNA was noted ectonucleoside triphosphate diphosphohydrolase-7, thus reducing in the blood and CSF of MS patients (21). Increased Th17 Th17 cells and disease severity in EAE. cells and IL-17 protein were subsequently found in the brain The ability of DCs to promote tolerance is highly dependent of patients with MS (22). Similar observations regarding in- on their ability to produce the anti-inflammatory cytokine IL- creased Th17 cells and IL-17 were noted in patients with 10. The presence of IL-10 in the cytokine milieu at the time of NMO (23–25). In both diseases, the highest levels of IL-17 a DC–T cell interaction will direct the T cell toward a reg- Downloaded from were associated with clinical relapse compared with remission ulatory fate. In addition to IL-10, production of TGF-b is (21, 24). Importantly, Kebir et al. (26) showed that IL-17 essential for the tolerance-inducing potential of DCs (33). A receptor expression was upregulated in MS lesions, allowing specialized subset of nonlymphoid CD103-expressing DCs is IL-17 to increase the permeability of the BBB and promote present in the GALT that are developmentally related to migration of CD4+ lymphocytes into the CNS. In this review, conventional CD8a+ DCs requiring expression of ID2, IFN we focus on the regulatory mechanisms that antagonize the regulatory factor 8, and Batf3 (34, 35). In particular, CD103+ http://www.jimmunol.org/ pathogenic activity of Th17 cells during EAE and MS. DCs in the small intestine can express high levels of the enzyme aldehyde dehydrogenase, allowing them to metabolize The immune system coordinates anti-inflammatory responses along the dietary vitamin A to RA and preferentially drive regulatory gut–brain axis Tcell(Treg) generation (36–38). Although it has previously been The GALT is a unique immune compartment that is associated shown that treatment with exogenous RA or a synthetic RA with the induction of regulatory cells. The GALT comprises receptor agonist is sufficient to significantly reduce the severity of 80% of the body’s immune system and includes, but is not EAE (39, 40), a recent study suggests that RA may also reduce limited to, the mesenteric lymph nodes (LNs), Peyer’s patches, disease by directly inhibiting IL-17 production not only from and the lamina propria of the small and large intestine. Due to CD4+ T cells but also gd T cells (41). Increased expression of by guest on January 26, 2017 the constant onslaught of foreign material arriving in the gut as IDO is yet another mechanism used by CD103+ DCs to pro- a result of daily food consumption, it is essential for the gut to mote regulatory cell development and tolerance (42), and IDO- have extensive mechanisms in place that promote tolerance to deficient mice develop exacerbated EAE (43). food-related Ags. It is not surprising that the microflora that Note that macrophages can also adopt a phenotype that colonizes the gut has taken advantage of these mechanisms to promotes the induction of immunosuppressive cells. These promote its survival by interacting directly with the GALT to macrophages, termed alternatively activated macrophages, influence the emergence of regulatory cells. Even though these have a unique phenotype but also produce IL-10. Whereas a regulatory cells are induced in the gut, they have the potential role for alternatively activated macrophages has been dem- to suppress inflammation at bodily sites far distal from the onstrated in EAE and MS (reviewed in Ref. 44), their role in gastrointestinal tract such as the CNS. Cells of both the innate communication along the gut–brain axis is less defined. and adaptive immune systems have essential roles in commu- Regulatory T cells. The primary role of Tregs is to maintain nication along this gut–brain axis. peripheral tolerance. Most autoreactive T cells are removed Dendritic cells. In general, dendritic cells (DCs) are the sentinels from the T cell repertoire by deletion in the thymus in a of the immune system. As professional APCs, they patrol the process termed central tolerance. However, low frequencies body, including the GALT, taking in both self and foreign of T cells specific for self proteins escape deletion and enter matter. DCs are specialized for the processing and presentation the peripheral circulation. Tregs have the capacity to counteract of peptide Ags on their surface in the context of MHC class I the proinflammatory activity of autoreactive T cells, including the and MHC class II molecules, which dictate their interac- production of IL-17. During EAE, Tregs migrate to the CNS + + tion with CD8 and CD4 T cells, respectively. The primary where they suppress inflammation (45). Tregs are most notably mechanism by which DCs recognize foreign microbes is the identified by the expression of CD25 (the high-affinity 2 expression of pattern recognition receptors specific for evo- receptor for IL-2) and Foxp3, but populations of Foxp3 lutionarily conserved microbe-associated molecular patterns. cells with regulatory function have been described (46). Tregs TLRs are one family of pattern recognition receptors that use mediate the suppression of autoreactive T cells through the a common intracellular signaling protein called MyD88. In expression of inhibitory molecules such as CTLA-4 and particular, engagement of TLR2, which recognizes a variety GITR and cytokine production (IL-10 and TGF-b). A sub- of microbe-associated molecular patterns, has been associated set of Tregs expresses the ectonucleotidase CD39 (also known with the induction of tolerogenic DCs (27). DCs can directly as ectonucleoside triphosphate diphosphohydrolase-1) (47, sample luminal content by extending processes through the 48). CD39 acts in concert with CD73 to break down ATP to 598 BRIEF REVIEWS: CNS AUTOIMMUNITY AND THE GUT MICROBIOME adenosine. Because ATP has proinflammatory properties whereas and MS (62, 74, 75), immunomodulatory approaches that adenosine promotes anti-inflammatory IL-10 production, partic- shift that balance between these two populations by reducing ularly during EAE (49, 50), CD39 expression promotes regula- inflammatory B cells while promoting regulatory populations tory function by T cells. Interestingly, CD4+ T cells have proven to could prove therapeutic (76, 77). Furthermore, B cells may be somewhat plastic regarding their differentiation potential such also prove beneficial based on their ability to enhance the Treg thatTh17cells,eventhosespecificforCNSAgs,canacquire population (78). regulatory properties specifically within the gut (51). Similarly, sequestration of pathogenic Th17 cells in the intestine can Microbial dysbiosis in patients with CNS autoimmunity significantly reduce CNS inflammation (52), but the role of the Several recent studies have addressed the following critical microbiota has not been examined in these processes. question: are there significant differences in the microbial Unfortunately, although Tregs are present in MS patients, contents of the gut between patients with CNS autoimmunity they exhibit inferior functional capacity compared with and healthy controls? Although authors have found subtle healthy controls. Viglietta et al. (53) were the first to describe differences in the exact composition of gut microflora within the reduced ability of Tregs isolated from MS patients to the patient versus control populations (as would be expected suppress activated T cells, which was later confirmed by others considering differences in geographical location), the over- + (54, 55). Indeed, FOXP3 expression is reduced in CD25 whelming conclusion is that, indeed, microbial dysbiosis is Tregs isolated from MS patients (55–57). The reduction in the present in the intestine of MS patients. functional capacity of Tregs from patients with MS could also In a cohort of 31 patients with relapsing-remitting MS Downloaded from be the result of reduced IL-10 production (58, 59). Fur- compared with 36 healthy controls, Chen et al. (79) found thermore, Fletcher et al. (60) identified a reduction in the significant differences in microbiota structure between pa- + frequency of CD39 Tregs, which were also impaired in their tients with MS and healthy controls. There was no difference ability to suppress IL-17 production from activated T cells. in overall species richness (a diversity) between healthy con- Importantly, both FOXP3 expression and IL-10 production trols and MS patients, but within the MS patient cohort, were restored in patients treated with IFN-b (57). However, there was a trend toward reduced species richness in patients http://www.jimmunol.org/ studies also suggest that differences may exist in the sup- with active disease whereas patients in remission were similar pressive potential of Tregs in patients with relapsing-remitting to the healthy controls. Such changes in a diversity could MS versus secondary progressive MS (55, 57, 58, 60). suggest a role for the gut microbiome in disease exacerbation, Regulatory B cells. It is now appreciated that regulatory B cells but future longitudinal studies are needed to establish cor- (Bregs) also play a significant role in immune suppression relation. At the community level, relapsing-remitting MS during EAE (61, 62). Although phenotypically diverse, Bregs patients exhibited an enrichment of Pseudomonas and Myco- can be loosely defined as any B cell capable of producing IL- plana (Proteobacteria), Blautia and Dorea (Firmicutes), and 10 (63). IL-35 has also been implicated in the regulatory Pedobacter (Bacteroidetes) with a decreased abundance of by guest on January 26, 2017 activity of B cells during EAE (64). Similar to Tregs, B cells Adlercreutzia and Collinsella (Actinobacteria), Lactobacillus can also use the CD39/CD73 axis to regulate inflammation (Firmicutes), and Parabacteroides (Bacteroidetes). In a second through the reduction of ATP levels, but this has not been study with a cohort of 60 MS patients and 43 healthy con- tested in the EAE model (65). Unlike Tregs, studies suggest trols, Methanobrevibacter (Euryarchaeota) and Akkermansia that Bregs assert their immunosuppressive activity locally (in (Verrucomicrobia) were identified as increased and Butyr- the draining LN) but not within the CNS itself (66). However, icimonas (Bacteroidetes) was decreased in MS patients (80). a recent study showed that the adoptive transfer of in vitro– Interestingly, Jangi et al. (80) performed secondary analysis activated pro–B cells (bone marrow B cells stimulated with separating treated versus untreated MS patients as indepen- CpG-B) could significantly reduce EAE symptoms when dent cohorts (n = 32 and 28, respectively). They found certain transferred therapeutically (67). The authors found that genera such as Prevotella (Bacteroidetes) and Sutterella (Pro- these cells matured into IL-10–producing Bregs that were teobacteria) that were reduced in untreated patients but re- able to traffic to the spinal cord. Bregs are capable of directly stored to normal levels with treatment. Furthermore, they responding to PAMPs via TLRs, and their ability to reduce identified Sarcina (Firmicutes) as being reduced only in EAE is dependent on TLR2/4 expression (68). Furthermore, treated patients, which highlights the potential of MS thera- despite normal numbers of Bregs in MyD88 knockout mice pies to influence the gut microbiome as well. Prevotella and (based on extracellular phenotype), the absence of MyD88 Sutterella species were also significantly reduced in a Japanese signaling significantly reduced the overall regulatory function cohort of MS patients (81). However, in this study, 14 of the of B cells based on cytokine production (69). Taken together, 19 species with reduced prevalence were located in Clostridia these findings suggest that direct TLR engagement on B cells, cluster XIVa or IV with three additional species identified potentially deriving from the microbiome, is essential for Breg from the genus Bacteroides. Thus, although dysbiosis is clearly induction and function during EAE. evident in MS patients, particularly those naive to treatment, The contribution of Bregs to the suppression of autoimmune the cause-and-effect relationship between gut dysbiosis and disease in patients with MS and their role in current and novel CNS autoimmunity is still unclear. Changes in the micro- therapeutics are actively being explored. A corresponding biome (i.e., the environment) could play a role in predis- population of Bregs has been identified in humans capable of posistion to the development of MS and/or act as a trigger for suppressing activated T cell proliferation (70, 71). In MS initiating disease in genetically predisposed individuals, but patients, Bregs are deficient in their ability to produce IL-10 additional studies are required to determine whether the al- when stimulated in vitro (72, 73). Although inflammatory tered microbiome drives changes in immunity or the onset of B cells have also been implicated in the pathogenesis of EAE immunological disease induces modifications in the microflora. The Journal of Immunology 599

Both MS and NMO are driven by pathogenic Th17 cells total number of CD5+CD1d+ B cells (86). Importantly, reactive against self-proteins. Unlike MS, in patients with antibiotic-induced Bregs had potent immunosuppressive ac- NMO, the target of autoreactive T cells has been identified as tivity in vivo that was significantly greater than Bregs isolated aquaporin-4 (AQP4), a water channel protein that transports from control treated animals, despite similar levels of IL-10 water across cell membranes. AQP4 is expressed by astrocytes production in vitro. Importantly, note that a second study in the brain. The dominant peptide epitope of AQP4 rec- found conflicting results showing that antibiotic treatment ognized by T cells shares significant sequence homology with significantly reduced the number and frequency of Bregs as an ATP-binding cassette transporter permease from Clostrid- measured by IL-10 production, but these differences could be ium perfringens, a human gut commensal, and AQP4-specific the result of the different antibiotic mixtures used in the two T cells cross-react with C. perfringens (82). When comparing studies, different vendors, housing, and/or diet (87). Re- the microbiome of patients with NMO to healthy controls, gardless, additional preclinical studies are required to deter- principal component analysis revealed significant composi- mine whether microbiome-altering therapeutics are effective tional differences. C. perfringens was the second most enriched in the absence of Bregs. taxon and was overabundant compared with either healthy Several other studies have described the beneficial effects of controls or patients with MS (83). Indeed, others have found oral antibiotic treatment. Yokote et al. (88) used a unique that C. perfringens type A is significantly reduced in MS pa- mixture of antibiotics to significantly reduce EAE disease. tients compared with healthy controls (84). Importantly, the Similar to our studies, they found an increase in total IL-10 increase in C. perfringens could not be attributed to the use of production from the mesenteric LN, but no increase in the + Downloaded from immune-modulating therapy because a subset of patients in frequency of Foxp3 Tregs. Alternatively, they found that the both the NMO and MS groups were treated with rituximab. success of oral antibiotic treatment was dependent on a subset Similar to MS and EAE, it is unknown whether the over- of invariant NK T cells. Minocycline, an oral tetracycline representation of C. perfringens in the gut of NMO patients is antibiotic commonly used for the treatment of acne, has also the cause or effect of autoimmune disease. The presence of been used in a rat model of EAE to reduce disease severity C. perfringens in the gut has the potential to act as a molecular both prophylactically and therapeutically (89). Subsequently, http://www.jimmunol.org/ mimic. Furthermore, other bacteria, such as Fibrobacteres, minocycline has been used in three clinical trials alone or in were also enriched in NMO patients and could contribute to combination with current therapies to treat patients with MS. the overall disease state independently or in collaboration with When administered to patients in combination with glatir- C. perfringens. amer acetate, minocycline showed a trend toward reduced CNS deterioration (90), but when combined with IFN-b,therewas Emergence of microbiome-directed therapies for EAE and MS no significant difference with placebo-treated controls (91). Current therapies available for patients with MS can be cat- However, the therapeutic window for minocycline may be egorized as immune-modulating drugs and immunosuppres- earlier in disease rather than later because treatment with sants (i.e., corticosteroids). In general, the goal of the former is minocycline significantly reduced conversion to MS when by guest on January 26, 2017 to slow nervous system degeneration whereas the latter can treatment was initiated at the time of the first clinical demye- often improve symptoms to help maintain quality of life. There linating event (92). Importantly, note that whereas minocycline are currently eight Food and Drug Administration–approved may promote autoimmune suppression by modulation of the immune-modulating drugs (some available in multiple forms) intestinal microflora to repair the dysbiosis observed in MS that can be used alone or in combination to treat MS. Re- patient, it can also have multiple direct immune-modulating gardless, there is a great need for novel approaches to combat effects that could contribute to the observed protection. Al- CNS destruction. Considering the now well-defined link though minocycline may or may not ultimately prove beneficial between the gut microbiota and brain physiology, it is not in the context of MS, other oral antibiotics or combinations of surprising that new therapies are being developed that target antibiotics could be considered or developed in the future to the gut microbiome. promote anti-inflammatory communication along the gut–brian Oral antibiotics. In our laboratory, we have shown that mod- axis. Indeed, vancomycin was shown to improve the symptoms ulation of the gut microbiota using orally administered broad- of autism in 8 out of 10 children studied (93). spectrum antibiotics is sufficient to provide significant protection Probiotic usage. By definition, a probiotic is any live microor- against EAE (85). This treatment regimen was effective when the ganism that confers a significant health benefit on the host. drugs were given orally (by gavage or in the drinking water) but This term can refer to both commensal microbes that normally + not i.p. The frequency and total number of Foxp3 Tregs was reside in the gut and exogenous, possibly food-borne, microbes significantly increased in both mesenteric and CNS-draining that travel through the intestine following consumption. LNs when mice with EAE were treated with antibiotics, and Numerous bacterial strains given alone or in combination there was a corresponding increase in IL-10 production as have been shown to improve CNS inflammation, including well.Furthermore,CD103+ DCs isolated from the GALT of Lactobacillus species, Pediococcus acidilactici, Bifidobacterium antibiotic-treated mice were superior in their ability to induce bifidum, Bifidobacterium animalis,andStreptococcus thermophiles Tregs from naive T cells in vitro. These findings provide strong (94–97). Genetically engineered bacterial strains such as evidence in support of anti-inflammatory communication along Lactococcus lactis expressing heat shock protein 65 from the gut–brain axis following modulation of the microbiome. Mycobacterium leprae have also been used to reduce clinical We have also previously shown a connection between symptoms of EAE (98, 99). Even one type of yeast, Candida modulation of the gut microbiome and the induction of Bregs. kefyr, commonly found in fermented foods, significantly reduced When mice were treated orally with broad-spectrum antibi- EAE disease (100). In almost all of the studies in which otics, there was a significant increase in the frequency and probiotics significantly improved EAE disease, the reduction in 600 BRIEF REVIEWS: CNS AUTOIMMUNITY AND THE GUT MICROBIOME

CNS inflammation was attributed to the induction of Tregs and/or significant reduction in disease progression compared with IL-10 production (95–97, 100). Furthermore, when healthy uninfected patients as measured by multiple parameters (115). volunteers were fed Bifidobacterium infantis 35624 for 8 wk, These studies implicated increased IL-10 from both Tregs and + the frequency of FOXP3 CD4 T cells in the blood was signif- Bregs in disease attenuation (115, 116). icantly increased compared with pretreatment measurements (101). Considering the regulatory potential of helminth infections, Probiotics also have an effect on Bregs. Mercadante et al. multiple laboratories have similarly shown that in vivo in- (102) have shown that L. lactis can promote the generation of fection with helminths can reduce the severity of EAE by IL-10–producing B cells that mediate tolerance in a model of inducing a combination of IL-10, Tregs, and Bregs (117–121). graft-versus-host disease. Clostridium butyricum, when given Furthermore, exposure to parasite products and/or Ags can in combination with specific immunotherapy, significantly reduce EAE disease, suggesting that live infection is not a increases the frequency of IL-10+ B cells in both mice and requirement for immunosuppression similar to probiotics humans (103, 104). (114, 122, 123). However, in a follow-up to the aforemen- In some cases, efficacy has been observed using heat-killed tioned study, when anti-helminth treatment was initiated in 4 bacteria (97, 100). Although heat-killed bacteria do not of the 12 patients due to the onset of parasitosis symptoms, classify as probiotics (because they are not alive), it suggests the parasites were eliminated, but the severity of MS disease that bacterial-derived products can nonetheless have thera- quickly progressed to mirror the uninfected patient cohort peutic potential. Indeed, Bacteroides fragilis functions as a (124). Regardless, the preclinical and correlation studies have probiotic because it significantly reduces the severity of EAE encouraged the development of helminth-based therapeutics in mice (105), but outer membrane vesicles from B. fragilis to treat MS. One strategy has been to infect MS patients with Downloaded from can also induce immunomodulatory effects and prevent in- eggs from Trichuris suis, a helminth unable to establish long- flammation (106). Furthermore, we found that the ability of lived infection in humans. Unfortunately, two small studies B. fragilis to function as a probiotic was dependent on ex- have presented conflicting results, although both found no pression of a single capsular polysaccharide (polysaccharide A adverse symptoms associated with treatment (125, 126). [PSA]) (105). We found that PSA purified from B. fragilis can Larger studies are required to determine whether T. suis egg http://www.jimmunol.org/ both prophylactically and therapeutically reduce EAE disease therapy will prove successful in significantly reducing the when administered to mice via oral gavage (107). Both progression of MS disease. Alternatively, the identification B. fragilis and PSA can condition DCs to generate Tregs in a and development of new helminth-derived Ags has the po- TLR2-dependent manner (105–109). In our studies with tential to shift the balance from a proinflammatory to an anti- human PBMCs, we have shown that naive CD4 T cells can inflammatory milieu, thus reducing CNS autoimmunity. + + acquire a CD39 FOXP3 regulatory phenotype when cocul- Dietary modification. In the absence of any exogenous manip- tured with DCs and purified PSA in vitro (110). This aug- ulation or ingested therapeutics, modulation of the gut mentation of Treg function suggests supplementation with microbiome can occur simply by changing one’s diet. Studies PSA could be an attractive novel therapeutic for patients with have shown that the gut microbiome is significantly different by guest on January 26, 2017 MS. Importantly, the ability of gut microbes and microbial between obese and lean individuals (127), but microbial products to influence CNS activity applies not only in the composition can change in as little as 1–2 d following dietary context of autoimmunity but also in murine models of de- intervention (128). In EAE, consumption of a calorie-restricted pression and autism spectrum disorder, further supporting the diet can improve disease symptoms whereas a high-salt diet importance of the gut–brain axis (111–113). exacerbates disease by promoting Th17 differentiation (129– Helminth therapy. Whereas the gut microbiome usually refers to 131). A recent study also found a positive correlation between those microbes that exist in a symbiotic relationship with their salt intake and both exacerbation rates and radiological activity human host, certain microbes maintain a parasitic relationship in patients with relapsing-remitting MS (132). Interestingly, with their host, many of which exist in the intestine. The dietary Ags can impact immunity independent of the gut presence or absence of parasites can thus contribute to the onset microbiome based on studies in GF mice fed an elemental of autoimmune disease as suggested in the hygiene hypothesis. diet (133). The hygiene hypothesis proposes a negative correlation be- Overall, increased microbial diversity is associated with an tween the decrease in parasitic infections and the increase in increase in fiber-rich foods (134). In particular, a high-fiber autoimmune diseases in the developed world as a result of diet promotes specific species of microbes within the Firmi- increased hygiene. cutes and Bacteroidetes phyla. These microbes are responsible Parasitic infections are often long-lasting chronic infections for the breakdown of nondigestible fiber in the colon and that require a certain degree of immunosuppression to pro- produce short-chain fatty acids (SCFAs) as part of the fer- mote and maintain their longevity. Parasites, in particular mentation process. SCFAs, including propionate, acetate, and helminths or worms, accomplish this feat by driving the in- butyrate, play a critical role in suppressing inflammation by duction of Th2 cells that produce anti-inflammatory cytokines, inducing Tregs (135). Conversely, a diet rich in long-chain including IL-4, IL-10, IL-13, and TGF-b. By reducing the fatty acids can promote Th17 differentiation and exacerbate production of Th2 cytokines in vivo, the balance is shifted disease (136). SCFAs also play an important role in the toward the activity of Th1/Th17 responses that drive auto- maintenance of CNS integrity. As mentioned above, GF mice immunity. This has significant bearing on MS, as evidenced exhibit increased permeability in the BBB (15). However, by a negative correlation between high rates of helminth in- when GF mice were monocolonized with either Clostridium fection and high MS prevalence worldwide (114). In one tyrobutyricum or Bacteroides thetaiotaomicron, both of which study, the authors followed 12 MS patients actively infected produce SCFAs, the defects in BBB permeability were re- with intestinal helminths during a 4-y period and found a stored. Interestingly, both MS and NMO patients have a The Journal of Immunology 601 distinct urinary metabolic signature compared with healthy 10. Yacyshyn, B., J. Meddings, D. Sadowski, and M. B. Bowen-Yacyshyn. 1996. Multiple sclerosis patients have peripheral blood CD45RO+ B cells and increased controls (137). Intermediates involved in propionate meta- intestinal permeability. Dig. Dis. Sci. 41: 2493–2498. bolism were significantly decreased in patients with MS, 11. Nusrat, S., E. Gulick, D. Levinthal, and K. Bielefeldt. 2012. Anorectal dysfunction in multiple sclerosis: a systematic review. ISRN Neurol. 2012: 376023. which could be influenced by changes in the gut microbiota. 12. Kimura, K., S. F. Hunter, M. S. Thollander, E. V. Loftus, Jr., L. J. Melton, III, Thus, the presence or absence of SCFA-producing microbes is P. C. O’Brien, M. Rodriguez, and S. F. Phillips. 2000. Concurrence of inflam- important to consider when assessing the dysbiosis of auto- matory bowel disease and multiple sclerosis. Mayo Clin. Proc. 75: 802–806. 13. Gupta, G., J. M. Gelfand, and J. D. Lewis. 2005. Increased risk for demyelinating immune patients and also when developing probiotic strate- diseases in patients with inflammatory bowel disease. Gastroenterology 129: 819– gies. Furthermore, recent advances in personalized nutrition 826. 14. Kozakova, H., M. Schwarzer, L. Tuckova, D. Srutkova, E. Czarnowska, I. Rosiak, (i.e., the development of individual diet plans based on gut T. Hudcovic, I. Schabussova, P. Hermanova, Z. Zakostelska, et al. 2016. Colo- microbiota and other parameters) have been used to control nization of germ-free mice with a mixture of three lactobacillus strains enhances blood glucose levels, which has significant applications in the the integrity of gut mucosa and ameliorates allergic sensitization. Cell. Mol. Immunol. 13: 251–262. prevention and/or treatment of type II diabetes (138). It 15. Braniste, V., M. Al-Asmakh, C. Kowal, F. Anuar, A. Abbaspour, M. To´th, would be of distinct appeal to use a similar approach to not A. Korecka, N. Bakocevic, L. G. Ng, P. Kundu, et al. 2014. The gut microbiota influences blood-brain barrier permeability in mice. Sci. Transl. Med. 6: 263ra158. only correct microbial dysbiosis through diet but to also es- 16. Ivanov, I. I., Rde. L. Frutos, N. Manel, K. Yoshinaga, D. B. Rifkin, R. B. Sartor, tablish a long-term dietary approach to reduce inflammatory B. B. Finlay, and D. R. Littman. 2008. Specific microbiota direct the differentiation of IL-17-producing T-helper cells in the mucosa of the small intestine. Cell activity and relapses in patients with MS (139, 140). Host Microbe 4: 337–349. 17. Chewning, J. H., and C. T. Weaver. 2014. Development and survival of Th17 cells within the intestines: the influence of microbiome- and diet-derived signals. J.

Conclusions Immunol. 193: 4769–4777. Downloaded from The gut–brain axis provides for the critical exchange of bio- 18. Dos Passos, G. R., D. K. Sato, J. Becker, and K. Fujihara. 2016. Th17 cells pathways in multiple sclerosis and neuromyelitis optica spectrum disorders: logic information that affects both the physiology and im- pathophysiological and therapeutic implications. Mediators Inﬂamm. 2016: munology of the host. The bidirectional activity between the 5314541. gut microbiome and the GALT allows for the establishment 19.Veldhoen,M.,K.Hirota,A.M.Westendorf,J.Buer,L.Dumoutier, J. C. Renauld, and B. Stockinger. 2008. The aryl hydrocarbon receptor links of a systemic homeostatic balance. Immune regulation in CNS TH17-cell-mediated autoimmunity to environmental toxins. Nature 453: 106–109.

demyelinating diseases reflects a balance between those cells http://www.jimmunol.org/ 20. Atarashi, K., J. Nishimura, T. Shima, Y. Umesaki, M. Yamamoto, M. Onoue, driving disease, such as Th17 cells, and regulatory cells from H. Yagita, N. Ishii, R. Evans, K. Honda, and K. Takeda. 2008. ATP drives lamina both T and B cell lineages that are influenced by the micro- propria TH17 cell differentiation. Nature 455: 808–812. biome. Improved balance of the dysregulated immune function 21. Matusevicius, D., P. Kivisa¨kk, B. He, N. Kostulas, V. Ozenci, S. Fredrikson, and H. Link. 1999. Interleukin-17 mRNA expression in blood and CSF mononuclear in experimental CNS demyelinating disease can be achieved by cells is augmented in multiple sclerosis. Mult. Scler. 5: 101–104. a variety of approaches that alter colonization of the gut mi- 22. Tzartos, J. S., M. A. Friese, M. J. Craner, J. Palace, J. Newcombe, M. M. Esiri, and L. Fugger. 2008. Interleukin-17 production in central nervous system-infiltrating croflora. Thus, novel approaches to treating human MS, as well T cells and glial cells is associated with active disease in multiple sclerosis. Am. J. as other autoimmune conditions, aimed at modulation of the Pathol. 172: 146–155. 23. Ishizu, T., M. Osoegawa, F. J. Mei, H. Kikuchi, M. Tanaka, Y. Takakura, microbiota may represent a major paradigm shift in how we M. Minohara, H. Murai, F. Mihara, T. Taniwaki, and J. Kira. 2005. Intrathecal by guest on January 26, 2017 approach treating human disease. activation of the IL-17/IL-8 axis in opticospinal multiple sclerosis. Brain 128: 988– 1002. 24. Wang, H. H., Y. Q. Dai, W. Qiu, Z. Q. Lu, F. H. Peng, Y. G. Wang, J. Bao, Disclosures Y. Li, and X. Q. Hu. 2011. Interleukin-17-secreting T cells in neuromyelitis optica and multiple sclerosis during relapse. J. Clin. Neurosci. 18: 131321317. The authors have no financial conflicts of interest. 25. Li, Y., H. Wang, Y. Long, Z. Lu, and X. Hu. 2011. Increased memory Th17 cells in patients with neuromyelitis optica and multiple sclerosis. J. Neuroimmunol. 234: 155–160. References 26. Kebir, H., K. Kreymborg, I. Ifergan, A. Dodelet-Devillers, R. Cayrol, M. Bernard, 1. Sawcer, S. 2008. The complex genetics of multiple sclerosis: pitfalls and prospects. F. Giuliani, N. Arbour, B. Becher, and A. Prat. 2007. Human TH17 lymphocytes Brain 131: 3118–3131. promote blood-brain barrier disruption and central nervous system inflammation. 2. Sawcer, S., G. Hellenthal, M. Pirinen, C. C. Spencer, N. A. Patsopoulos, Nat. Med. 13: 1173–1175. L. Moutsianas, A. Dilthey, Z. Su, C. Freeman, S. E. Hunt, et al.; International 27. Manicassamy, S., R. Ravindran, J. Deng, H. Oluoch, T. L. Denning, S. P. Kasturi, Multiple Sclerosis Genetics Consortium.; Wellcome Trust Case Control Con- K. M. Rosenthal, B. D. Evavold, and B. Pulendran. 2009. Toll-like receptor sortium. 2011. Genetic risk and a primary role for cell-mediated immune mech- 2-dependent induction of vitamin A-metabolizing enzymes in dendritic cells pro- anisms in multiple sclerosis. Nature 476: 214–219. motes T regulatory responses and inhibits autoimmunity. Nat. Med. 15: 401–409. 3. Beecham, A. H., N. A. Patsopoulos, D. K. Xifara, M. F. Davis, A. Kemppinen, 28. Rescigno, M., M. Urbano, B. Valzasina, M. Francolini, G. Rotta, R. Bonasio, C. Cotsapas, T. S. Shah, C. Spencer, D. Booth, A. Goris, et al.; International F. Granucci, J. P. Kraehenbuhl, and P. Ricciardi-Castagnoli. 2001. Dendritic cells Multiple Sclerosis Genetics Consortium. 2013. Analysis of immune-related loci express tight junction proteins and penetrate gut epithelial monolayers to sample identifies 48 new susceptibility variants for multiple sclerosis. Nat. Genet. 45: bacteria. Nat. Immunol. 2: 361–367. 1353–1360. 29. Chieppa, M., M. Rescigno, A. Y. Huang, and R. N. Germain. 2006. Dynamic 4. Lill, C. M., F. Luessi, A. Alcina, E. A. Sokolova, N. Ugidos, B. de la Hera, imaging of dendritic cell extension into the small bowel lumen in response to L. Guillot-Noe¨l, S. Malhotra, E. Reinthaler, B. M. Schjeide, et al. 2015. Genome- epithelial cell TLR engagement. J. Exp. Med. 203: 2841–2852. wide significant association with seven novel multiple sclerosis risk loci. J. Med. 30. Ohno, H. 2016. Intestinal M cells. J. Biochem. 159: 151–160. Genet. 52: 848–855. 31. Peterson, L. W., and D. Artis. 2014. Intestinal epithelial cells: regulators of barrier 5. Mumford, C. J., N. W. Wood, H. Kellar-Wood, J. W. Thorpe, D. H. Miller, and function and immune homeostasis. Nat. Rev. Immunol. 14: 141–153. D. A. Compston. 1994. The British Isles survey of multiple sclerosis in twins. 32. Kusu, T., H. Kayama, M. Kinoshita, S. G. Jeon, Y. Ueda, Y. Goto, R. Okumura, Neurology 44: 11–15. H. Saiga, T. Kurakawa, K. Ikeda, et al. 2013. Ecto-nucleoside triphosphate 6. Willer, C. J., D. A. Dyment, N. J. Risch, A. D. Sadovnick, and G. C. Ebers, diphosphohydrolase 7 controls Th17 cell responses through regulation of luminal Canadian Collaborative Study Group. 2003. Twin concordance and sibling re- ATP in the small intestine. J. Immunol. 190: 774–783. 33. Chen, W., W. Jin, N. Hardegen, K. J. Lei, L. Li, N. Marinos, G. McGrady, and currence rates in multiple sclerosis. Proc. Natl. Acad. Sci.USA 100: 12877–12882. 2 7. Berer, K., M. Mues, M. Koutrolos, Z. A. Rasbi, M. Boziki, C. Johner, H. Wekerle, S. M. Wahl. 2003. Conversion of peripheral CD4+CD25 naive T cells to CD4+ and G. Krishnamoorthy. 2011. Commensal microbiota and myelin autoantigen CD25+ regulatory T cells by TGF-b induction of transcription factor Foxp3. J. cooperate to trigger autoimmune demyelination. Nature 479: 538–541. Exp. Med. 198: 1875–1886. 8. Lee, Y. K., J. S. Menezes, Y. Umesaki, and S. K. Mazmanian. 2011. Proin- 34. Ginhoux, F., K. Liu, J. Helft, M. Bogunovic, M. Greter, D. Hashimoto, J. Price, flammatory T-cell responses to gut microbiota promote experimental autoimmune N. Yin, J. Bromberg, S. A. Lira, et al. 2009. The origin and development of encephalomyelitis. Proc. Natl. Acad. Sci. USA 108(Suppl. 1): 4615–4622. nonlymphoid tissue CD103+ DCs. J. Exp. Med. 206: 3115–3130. 9. Nouri, M., A. Bredberg, B. Westro¨m, and S. Lavasani. 2014. Intestinal barrier 35. Edelson, B. T., W. Kc, R. Juang, M. Kohyama, L. A. Benoit, P. A. Klekotka, dysfunction develops at the onset of experimental autoimmune encephalomyelitis, C. Moon, J. C. Albring, W. Ise, D. G. Michael, et al. 2010. Peripheral CD103+ and can be induced by adoptive transfer of auto-reactive T cells. PLoS One 9: dendritic cells form a unified subset developmentally related to CD8a+ conven- e106335. tional dendritic cells. J. Exp. Med. 207: 823–836. 602 BRIEF REVIEWS: CNS AUTOIMMUNITY AND THE GUT MICROBIOME

36. Coombes, J. L., K. R. Siddiqui, C. V. Arancibia-Ca´rcamo, J. Hall, C. M. Sun, pathogenic Th17 cells and are impaired in multiple sclerosis. J. Immunol. 183: Y. Belkaid, and F. Powrie. 2007. A functionally specialized population of mucosal 7602–7610. CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-b and retinoic acid- 61. Fillatreau, S., C. H. Sweenie, M. J. McGeachy, D. Gray, and S. M. Anderton. dependent mechanism. J. Exp. Med. 204: 1757–1764. 2002. B cells regulate autoimmunity by provision of IL-10. Nat. Immunol. 3: 944– 37. Mucida, D., Y. Park, G. Kim, O. Turovskaya, I. Scott, M. Kronenberg, and 950. H. Cheroutre. 2007. Reciprocal TH17 and regulatory T cell differentiation me- 62. Matsushita, T., K. Yanaba, J. D. Bouaziz, M. Fujimoto, and T. F. Tedder. 2008. diated by retinoic acid. Science 317: 256–260. Regulatory B cells inhibit EAE initiation in mice while other B cells promote 38. Sun, C. M., J. A. Hall, R. B. Blank, N. Bouladoux, M. Oukka, J. R. Mora, and disease progression. J. Clin. Invest. 118: 3420–3430. Y. Belkaid. 2007. Small intestine lamina propria dendritic cells promote de novo 63. Tedder, T. F. 2015. B10 cells: a functionally defined regulatory B cell subset. J. generation of Foxp3 T reg cells via retinoic acid. J. Exp. Med. 204: 1775–1785. Immunol. 194: 1395–1401. 39. Klemann, C., B. J. Raveney, A. K. Klemann, T. Ozawa, S. von Ho¨rsten, K. Shudo, 64. Shen, P., T. Roch, V. Lampropoulou, R. A. O’Connor, U. Stervbo, E. Hilgenberg, S. Oki, and T. Yamamura. 2009. Synthetic retinoid AM80 inhibits Th17 cells and S. Ries, V. D. Dang, Y. Jaimes, C. Daridon, et al. 2014. IL-35-producing B cells ameliorates experimental autoimmune encephalomyelitis. Am. J. Pathol. 174: are critical regulators of immunity during autoimmune and infectious diseases. 2234–2245. Nature 507: 366–370. 40. Zhan, X. X., Y. Liu, J. F. Yang, G. Y. Wang, L. Mu, T. S. Zhang, X. L. Xie, 65. Kaku, H., K. F. Cheng, Y. Al-Abed, and T. L. Rothstein. 2014. A novel mech- J. H. Wang, Y. M. Liu, Q. F. Kong, et al. 2013. All-trans-retinoic acid ameliorates anism of B cell-mediated immune suppression through CD73 expression and experimental allergic encephalomyelitis by affecting dendritic cell and monocyte adenosine production. J. Immunol. 193: 5904–5913. development. Immunology 138: 333–345. 66. Matsumoto, M., A. Baba, T. Yokota, H. Nishikawa, Y. Ohkawa, H. Kayama, 41. Raverdeau, M., C. J. Breen, A. Misiak, and K. H. Mills. 2016. Retinoic acid A. Kallies, S. L. Nutt, S. Sakaguchi, K. Takeda, et al. 2014. Interleukin-10- suppresses IL-17 production and pathogenic activity of gd T cells in CNS auto- producing plasmablasts exert regulatory function in autoimmune inflammation. immunity. Immunol. Cell Biol. 94: 763–773. Immunity 41: 1040–1051. 42. Matteoli, G., E. Mazzini, I. D. Iliev, E. Mileti, F. Fallarino, P. Puccetti, 67. Korniotis, S., C. Gras, H. Letscher, R. Montandon, J. Me´gret, S. Siegert, S. Ezine, M. Chieppa, and M. Rescigno. 2010. Gut CD103+ dendritic cells express indo- P. G. Fallon, S. A. Luther, S. Fillatreau, and F. Zavala. 2016. Treatment of on- leamine 2,3-dioxygenase which influences T regulatory/T effector cell balance and going autoimmune encephalomyelitis with activated B-cell progenitors maturing oral tolerance induction. Gut 59: 595–604. into regulatory B cells. Nat. Commun. 7: 12134.

43. Yan, Y., G. X. Zhang, B. Gran, F. Fallarino, S. Yu, H. Li, M. L. Cullimore, 68. Lampropoulou, V., K. Hoehlig, T. Roch, P. Neves, E. CalderońGo´mez, Downloaded from A. Rostami, and H. Xu. 2010. IDO upregulates regulatory T cells via tryptophan C. H. Sweenie, Y. Hao, A. A. Freitas, U. Steinhoff, S. M. Anderton, and catabolite and suppresses encephalitogenic T cell responses in experimental auto- S. Fillatreau. 2008. TLR-activated B cells suppress T cell-mediated autoimmunity. immune encephalomyelitis. J. Immunol. 185: 5953–5961. J. Immunol. 180: 4763–4773. 44. Jiang, Z., J. X. Jiang, and G. X. Zhang. 2014. Macrophages: a double-edged sword 69. Yanaba, K., J. D. Bouaziz, T. Matsushita, T. Tsubata, and T. F. Tedder. 2009. in experimental autoimmune encephalomyelitis. Immunol. Lett. 160: 17–22. The development and function of regulatory B cells expressing IL-10 (B10 cells) 45. McGeachy, M. J., L. A. Stephens, and S. M. Anderton. 2005. Natural recovery requires antigen receptor diversity and TLR signals. J. Immunol. 182: 7459–7472. and protection from autoimmune encephalomyelitis: contribution of CD4+CD25+ 70. Iwata, Y., T. Matsushita, M. Horikawa, D. J. Dilillo, K. Yanaba, G. M. Venturi, regulatory cells within the central nervous system. J. Immunol. 175: 3025–3032. P. M. Szabolcs, S. H. Bernstein, C. M. Magro, A. D. Williams, et al. 2011. 46. Roncarolo, M. G., S. Gregori, M. Battaglia, R. Bacchetta, K. Fleischhauer, and Characterization of a rare IL-10-competent B-cell subset in humans that parallels http://www.jimmunol.org/ M. K. Levings. 2006. Interleukin-10-secreting type 1 regulatory T cells in rodents mouse regulatory B10 cells. Blood 117: 530–541. and humans. Immunol. Rev. 212: 28–50. 71. Kessel, A., T. Haj, R. Peri, A. Snir, D. Melamed, E. Sabo, and E. Toubi. 2012. 47. Deaglio, S., K. M. Dwyer, W. Gao, D. Friedman, A. Usheva, A. Erat, J. F. Chen, Human CD19+CD25high B regulatory cells suppress proliferation of CD4+ T cells K. Enjyoji, J. Linden, M. Oukka, et al. 2007. Adenosine generation catalyzed by and enhance Foxp3 and CTLA-4 expression in T-regulatory cells. Autoimmun. CD39 and CD73 expressed on regulatory T cells mediates immune suppression. Rev. 11: 670–677. J. Exp. Med. 204: 1257–1265. 72. Duddy, M., M. Niino, F. Adatia, S. Hebert, M. Freedman, H. Atkins, H. J. Kim, 48. Borsellino, G., M. Kleinewietfeld, D. Di Mitri, A. Sternjak, A. Diamantini, and A. Bar-Or. 2007. Distinct effector cytokine profiles of memory and naive R. Giometto, S. Ho¨pner, D. Centonze, G. Bernardi, M. L. Dell’Acqua, et al. 2007. human B cell subsets and implication in multiple sclerosis. J. Immunol. 178: 6092– Expression of ectonucleotidase CD39 by Foxp3+ Treg cells: hydrolysis of extra- 6099. cellular ATP and immune suppression. Blood 110: 1225–1232. 73. Hirotani, M., M. Niino, T. Fukazawa, S. Kikuchi, I. Yabe, S. Hamada, Y. Tajima, 49. Tsutsui, S., J. Schnermann, F. Noorbakhsh, S. Henry, V. W. Yong, B. W. Winston, and H. Sasaki. 2010. Decreased IL-10 production mediated by Toll-like receptor 9 by guest on January 26, 2017 K. Warren, and C. Power. 2004. A1 adenosine receptor upregulation and activation in B cells in multiple sclerosis. J. Neuroimmunol. 221: 95–100. attenuates neuroinflammation and demyelination in a model of multiple sclerosis. 74. Miyazaki, Y., R. Li, A. Rezk, H. Misirliyan, C. Moore, N. Farooqi, M. Solis, J. Neurosci. 24: 1521–1529. L. G. Goiry, O. de Faria Junior, V. D. Dang, et al; CIHR/MSSC New Emerging 50. Yao, S. Q., Z. Z. Li, Q. Y. Huang, F. Li, Z. W. Wang, E. Augusto, J. C. He, Team Grant in Clinical Autoimmunity; MSSRF Canadian B cells in MS Team. X. T. Wang, J. F. Chen, and R. Y. Zheng. 2012. Genetic inactivation of the 2014. A novel microRNA-132-sirtuin-1 axis underlies aberrant B-cell cytokine adenosine A2A receptor exacerbates brain damage in mice with experimental au- regulation in patients with relapsing-remitting multiple sclerosis [corrected]. toimmune encephalomyelitis. J. Neurochem. 123: 100–112. [Published erratum appears in 2014 PLoS One 9: e109041.] PLoS One 9: e105421. 51. Esplugues, E., S. Huber, N. Gagliani, A. E. Hauser, T. Town, Y. Y. Wan, 75. Li, R., A. Rezk, Y. Miyazaki, E. Hilgenberg, H. Touil, P. Shen, C. S. Moore, W. O’Connor, Jr., A. Rongvaux, N. Van Rooijen, A. M. Haberman, et al. 2011. L. Michel, F. Althekair, S. Rajasekharan, et al; Canadian B cells in MS Team. Control of TH17 cells occurs in the small intestine. Nature 475: 514–518. 2015. Proinflammatory GM-CSF-producing B cells in multiple sclerosis and B cell 52. Berer, K., M. Boziki, and G. Krishnamoorthy. 2014. Selective accumulation of depletion therapy. Sci. Transl. Med. 7: 310ra166. pro-inflammatory T cells in the intestine contributes to the resistance to autoim- 76. Chen, D., M. Blazek, S. Ireland, S. Ortega, X. Kong, A. Meeuwissen, A. Stowe, mune demyelinating disease. PLoS One 9: e87876. L. Carter, Y. Wang, R. Herbst, and N. L. Monson. 2014. Single dose of glyco- 53. Viglietta, V., C. Baecher-Allan, H. L. Weiner, and D. A. Hafler. 2004. Loss of engineered anti-CD19 antibody (MEDI551) disrupts experimental autoimmune functional suppression by CD4+CD25+ regulatory T cells in patients with multiple encephalomyelitis by inhibiting pathogenic adaptive immune responses in the bone sclerosis. J. Exp. Med. 199: 971–979. marrow and spinal cord while preserving peripheral regulatory mechanisms. J. 54. Haas, J., A. Hug, A. Vieho¨ver, B. Fritzsching, C. S. Falk, A. Filser, T. Vetter, Immunol. 193: 4823–4832. L. Milkova, M. Korporal, B. Fritz, et al. 2005. Reduced suppressive effect of CD4+ 77. Li, R., A. Rezk, L. M. Healy, G. Muirhead, A. Prat, J. L. Gommerman, and CD25high regulatory T cells on the T cell immune response against myelin oli- A. Bar-Or, MSSRF Canadian B Cells in MS Team. 2016. Cytokine-defined B cell godendrocyte glycoprotein in patients with multiple sclerosis. Eur. J. Immunol. 35: responses as therapeutic targets in multiple sclerosis. Front. Immunol. 6: 626. 3343–3352. 78. Ray, A., S. Basu, C. B. Williams, N. H. Salzman, and B. N. Dittel. 2012. A novel 55. Venken, K., N. Hellings, K. Hensen, J. L. Rummens, R. Medaer, M. B. D’hooghe, IL-10-independent regulatory role for B cells in suppressing autoimmunity by B. Dubois, J. Raus, and P. Stinissen. 2006. Secondary progressive in contrast to maintenance of regulatory T cells via GITR ligand. J. Immunol. 188: 3188–3198. relapsing-remitting multiple sclerosis patients show a normal CD4+CD25+ regu- 79. Chen, J., N. Chia, K. R. Kalari, J. Z. Yao, M. Novotna, M. M. Soldan, latory T-cell function and FOXP3 expression. J. Neurosci. Res. 83: 1432–1446. D. H. Luckey, E. V. Marietta, P. R. Jeraldo, X. Chen, et al. 2016. Multiple 56. Huan, J., N. Culbertson, L. Spencer, R. Bartholomew, G. G. Burrows, Y. K. Chou, sclerosis patients have a distinct gut microbiota compared to healthy controls. Sci. D. Bourdette, S. F. Ziegler, H. Offner, and A. A. Vandenbark. 2005. Decreased Rep. 6: 28484. FOXP3 levels in multiple sclerosis patients. J. Neurosci. Res. 81: 45–52. 80. Jangi, S., R. Gandhi, L. M. Cox, N. Li, F. von Glehn, R. Yan, B. Patel, 57. Venken, K., N. Hellings, M. Thewissen, V. Somers, K. Hensen, J. L. Rummens, M. A. Mazzola, S. Liu, B. L. Glanz, et al. 2016. Alterations of the human gut R. Medaer, R. Hupperts, and P. Stinissen. 2008. Compromised CD4+ CD25high microbiome in multiple sclerosis. Nat. Commun. 7: 12015. regulatory T-cell function in patients with relapsing-remitting multiple sclerosis is 81. Miyake, S., S. Kim, W. Suda, K. Oshima, M. Nakamura, T. Matsuoka, correlated with a reduced frequency of FOXP3-positive cells and reduced FOXP3 N. Chihara, A. Tomita, W. Sato, S. W. Kim, et al. 2015. Dysbiosis in the gut expression at the single-cell level. Immunology 123: 79–89. microbiota of patients with multiple sclerosis, with a striking depletion of species 58. Soldan, S. S., A. I. Alvarez Retuerto, N. L. Sicotte, and R. R. Voskuhl. 2004. belonging to clostridia XIVa and IV clusters. PLoS One 10: e0137429. Dysregulation of IL-10 and IL-12p40 in secondary progressive multiple sclerosis. 82. Varrin-Doyer, M., C. M. Spencer, U. Schulze-Topphoff, P. A. Nelson, R. M. Stroud, J. Neuroimmunol. 146: 209–215. B. A. Cree, and S. S. Zamvil. 2012. Aquaporin 4-specific T cells in neuromyelitis 59. Astier, A. L., G. Meiffren, S. Freeman, and D. A. Hafler. 2006. Alterations in optica exhibit a Th17 bias and recognize Clostridium ABC transporter. Ann. Neurol. CD46-mediated Tr1 regulatory T cells in patients with multiple sclerosis. J. Clin. 72: 53–64. Invest. 116: 3252–3257. 83. Cree, B. A., C. M. Spencer, M. Varrin-Doyer, S. E. Baranzini, and S. S. Zamvil. 60. Fletcher, J. M., R. Lonergan, L. Costelloe, K. Kinsella, B. Moran, C. O’Farrelly, 2016. Gut microbiome analysis in neuromyelitis optica reveals over-abundance of N. Tubridy, and K. H. Mills. 2009. CD39+Foxp3+ regulatory T cells suppress Clostridium perfringens. Ann. Neurol. 80: 443–447. The Journal of Immunology 603

84. Rumah, K. R., J. Linden, V. A. Fischetti, and T. Vartanian. 2013. Isolation of 106. Shen, Y., M. L. Giardino Torchia, G. W. Lawson, C. L. Karp, J. D. Ashwell, and Clostridium perfringens type B in an individual at first clinical presentation of S. K. Mazmanian. 2012. Outer membrane vesicles of a human commensal multiple sclerosis provides clues for environmental triggers of the disease. PLoS mediate immune regulation and disease protection. Cell Host Microbe 12: 509– One 8: e76359. 520. 85. Ochoa-Repa´raz, J., D. W. Mielcarz, L. E. Ditrio, A. R. Burroughs, D. M. Foureau, 107. Ochoa-Repa´raz, J., D. W. Mielcarz, Y. Wang, S. Begum-Haque, S. Dasgupta, S. Haque-Begum, and L. H. Kasper. 2009. Role of gut commensal microflora in D. L. Kasper, and L. H. Kasper. 2010. A polysaccharide from the human com- the development of experimental autoimmune encephalomyelitis. J. Immunol. 183: mensal Bacteroides fragilis protects against CNS demyelinating disease. Mucosal 6041–6050. Immunol. 3: 487–495. 86. Ochoa-Repa´raz, J., D. W. Mielcarz, S. Haque-Begum, and L. H. Kasper. 2010. 108. Wang, Y., K. M. Telesford, J. Ochoa-Repa´raz, S. Haque-Begum, M. Christy, Induction of a regulatory B cell population in experimental allergic encephalo- E. J. Kasper, L. Wang, Y. Wu, S. C. Robson, D. L. Kasper, and L. H. Kasper. myelitis by alteration of the gut commensal microflora. Gut Microbes 1: 103–108. 2014. An intestinal commensal symbiosis factor controls neuroinflammation via 87. Rosser, E. C., K. Oleinika, S. Tonon, R. Doyle, A. Bosma, N. A. Carter, TLR2-mediated CD39 signalling. Nat. Commun. 5: 4432. + K. A. Harris, S. A. Jones, N. Klein, and C. Mauri. 2014. Regulatory B cells are 109. Round, J. L., and S. K. Mazmanian. 2010. Inducible Foxp3 regulatory T-cell induced by gut microbiota-driven interleukin-1b and interleukin-6 production. development by a commensal bacterium of the intestinal microbiota. Proc. Natl. Nat. Med. 20: 1334–1339. Acad. Sci. USA 107: 12204–12209. 88. Yokote, H., S. Miyake, J. L. Croxford, S. Oki, H. Mizusawa, and T. Yamamura. 110. Telesford, K. M., W. Yan, J. Ochoa-Reparaz, A. Pant, C. Kircher, M. A. Christy, 2008. NKT cell-dependent amelioration of a mouse model of multiple sclerosis by S. Begum-Haque, D. L. Kasper, and L. H. Kasper. 2015. A commensal symbiotic + + altering gut flora. Am. J. Pathol. 173: 1714–1723. factor derived from Bacteroides fragilis promotes human CD39 Foxp3 T cells and 89. Popovic, N., A. Schubart, B. D. Goetz, S. C. Zhang, C. Linington, and I. D. Duncan. Treg function. Gut Microbes 6: 234–242. 2002. Inhibition of autoimmune encephalomyelitis by a tetracycline. Ann. Neurol. 51: 111. Desbonnet, L., L. Garrett, G. Clarke, J. Bienenstock, and T. G. Dinan. 2008. The 215–223. probiotic Bifidobacteria infantis: an assessment of potential antidepressant prop- 90. Metz, L. M., D. Li, A. Traboulsee, M. L. Myles, P. Duquette, J. Godin, erties in the rat. J. Psychiatr. Res. 43: 164–174. M. Constantin, and V. W. Yong, GA/Minocycline Study Investigators. 2009. 112. Desbonnet, L., L. Garrett, G. Clarke, B. Kiely, J. F. Cryan, and T. G. Dinan. Glatiramer acetate in combination with minocycline in patients with relapsing– 2010. Effects of the probiotic Bifidobacterium infantis in the maternal separation remitting multiple sclerosis: results of a Canadian, multicenter, double-blind, model of depression. Neuroscience 170: 1179–1188. placebo-controlled trial. Mult. Scler. 15: 1183–1194. 113. Hsiao, E. Y., S. W. McBride, S. Hsien, G. Sharon, E. R. Hyde, T. McCue, Downloaded from 91. Sørensen, P. S., F. Sellebjerg, J. Lycke, M. Fa¨rkkila¨,A.Creánge, C. G. Lund, J. A. Codelli, J. Chow, S. E. Reisman, J. F. Petrosino, et al. 2013. Microbiota M. Schluep, J. L. Frederiksen, E. Stenager, C. Pfleger, et al; RECYCLINE Study modulate behavioral and physiological abnormalities associated with neuro- Investigators. 2016. Minocycline added to subcutaneous interferon b-1a in mul- developmental disorders. Cell 155: 1451–1463. tiple sclerosis: randomized RECYCLINE study. Eur. J. Neurol. 23: 861–870. 114. Peoń, A. N., and L. I. Terrazas. 2016. Immune-regulatory mechanisms of classical 92. Metz, L. M., D. Li, A. Traboulsee, P. Duquette, V. W. Yong, M. Eliasziw, and experimental multiple sclerosis drugs: a special focus on helminth-derived G. Cerchiaro, J. Greenfield, A. Riddehough, M. Yeung, et al. 2015. Minocycline treatments. Curr. Med. Chem. 23: 1152–1170. reduces the relative risk of multiple sclerosis in people experiencing their first 115. Correale, J., and M. Farez. 2007. Association between parasite infection and immune responses in multiple sclerosis. Ann. Neurol. 61: 97–108.

clinical demyelinating event by 44.6%: results of a phase III double-blind placebo http://www.jimmunol.org/ controlled Canadian multicentre clinical trial. Mult. Scler. 21 (Supp. 11): 780–781 116. Correale, J., M. Farez, and G. Razzitte. 2008. Helminth infections associated with (Abstr. 227). multiple sclerosis induce regulatory B cells. Ann. Neurol. 64: 187–199. ¨ ¨ 93. Sandler, R. H., S. M. Finegold, E. R. Bolte, C. P. Buchanan, A. P. Maxwell, 117. La Flamme, A. C., K. Ruddenklau, and B. T. Backstrom. 2003. Schistosomiasis decreases central nervous system inflammation and alters the progression of ex- M. L. Vaïsa¨nen, M. N. Nelson, and H. M. Wexler. 2000. Short-term benefit from perimental autoimmune encephalomyelitis. Infect. Immun. 71: 4996–5004. oral vancomycin treatment of regressive-onset autism. J. Child Neurol. 15: 429– 118. Walsh, K. P., M. T. Brady, C. M. Finlay, L. Boon, and K. H. Mills. 2009. In- 435. fection with a helminth parasite attenuates autoimmunity through TGF-b-medi- 94. Ezendam, J., A. de Klerk, E. R. Gremmer, and H. van Loveren. 2008. Effects of ated suppression of Th17 and Th1 responses. J. Immunol. 183: 1577–1586. Bifidobacterium animalis administered during lactation on allergic and autoim- 119. Wu, Z., I. Nagano, K. Asano, and Y. Takahashi. 2010. Infection of non- mune responses in rodents. Clin. Exp. Immunol. 154: 424–431. encapsulated species of Trichinella ameliorates experimental autoimmune en- 95. Kwon, H. K., G. C. Kim, Y. Kim, W. Hwang, A. Jash, A. Sahoo, J. E. Kim, cephalomyelitis involving suppression of Th17 and Th1 response. Parasitol. Res. J. H. Nam, and S. H. Im. 2013. Amelioration of experimental autoimmune en- 107: 1173–1188. cephalomyelitis by probiotic mixture is mediated by a shift in T helper cell im- by guest on January 26, 2017 120. Wilson, M. S., M. D. Taylor, M. T. O’Gorman, A. Balic, T. A. Barr, K. Filbey, mune response. Clin. Immunol. 146: 217–227. + hi S. M. Anderton, and R. M. Maizels. 2010. Helminth-induced CD19 CD23 96. Lavasani, S., B. Dzhambazov, M. Nouri, F. Fa˚k, S. Buske, G. Molin, H. Thorlacius, B cells modulate experimental allergic and autoimmune inflammation. Eur. J. Alenfall, B. Jeppsson, and B. Westro¨m. 2010. A novel probiotic mixture exerts a J. Immunol. 40: 1682–1696. therapeutic effect on experimental autoimmune encephalomyelitis mediated by IL-10 121. Reyes, J. L., A. F. Espinoza-Jimeńez, M. I. Gonza´lez, L. Verdin, and L. I. Terrazas. producing regulatory T cells. PLoS One 5: e9009. 2011. Taenia crassiceps infection abrogates experimental autoimmune encephalo- 97. Takata, K., M. Kinoshita, T. Okuno, M. Moriya, T. Kohda, J. A. Honorat, myelitis. Cell. Immunol. 267: 77–87. T. Sugimoto, A. Kumanogoh, H. Kayama, K. Takeda, et al. 2011. The lactic acid 122. Sewell, D., Z. Qing, E. Reinke, D. Elliot, J. Weinstock, M. Sandor, and Z. Fabry. bacterium Pediococcus acidilactici suppresses autoimmune encephalomyelitis by 2003. Immunomodulation of experimental autoimmune encephalomyelitis by inducing IL-10-producing regulatory T cells. PLoS One 6: e27644. helminth ova immunization. Int. Immunol. 15: 59–69. 98. Ochoa-Repa´raz, J., C. Riccardi, A. Rynda, S. Jun, G. Callis, and D. W. Pascual. 123. Zheng, X., X. Hu, G. Zhou, Z. Lu, W. Qiu, J. Bao, and Y. Dai. 2008. Soluble egg 2007. Regulatory T cell vaccination without autoantigen protects against experi- antigen from Schistosoma japonicum modulates the progression of chronic pro- mental autoimmune encephalomyelitis. J. Immunol. 178: 1791–1799. gressive experimental autoimmune encephalomyelitis via Th2-shift response. 99. Rezende, R. M., R. P. Oliveira, S. R. Medeiros, A. C. Gomes-Santos, A. C. Alves, J. Neuroimmunol. 194: 107–114. F. G. Loli, M. A. Guimara˜es, S. S. Amaral, A. P. da Cunha, H. L. Weiner, et al. 124. Correale, J., and M. F. Farez. 2011. The impact of parasite infections on the course 2013. Hsp65-producing Lactococcus lactis prevents experimental autoimmune en- + + of multiple sclerosis. J. Neuroimmunol. 233: 6–11. cephalomyelitis in mice by inducing CD4 LAP regulatory T cells. J. Autoimmun. 125. Benzel, F., H. Erdur, S. Kohler, M. Frentsch, A. Thiel, L. Harms, K. P. Wandinger, 40: 45–57. and B. Rosche. 2012. Immune monitoring of Trichuris suis egg therapy in multiple 100. Takata, K., T. Tomita, T. Okuno, M. Kinoshita, T. Koda, J. A. Honorat, sclerosis patients. J. Helminthol. 86: 339–347. M. Takei, K. Hagihara, T. Sugimoto, H. Mochizuki, et al. 2015. Dietary yeasts 126. Voldsgaard, A., P. Bager, E. Garde, P. A˚keson, A. M. Leffers, C. G. Madsen, reduce inflammation in central nerve system via microflora. Ann. Clin. Transl. C. Kapel, A. Roepstorff, S. M. Thamsborg, M. Melbye, et al. 2015. Trichuris suis Neurol. 2: 56–66. ova therapy in relapsing multiple sclerosis is safe but without signals of beneficial 101. Konieczna, P., D. Groeger, M. Ziegler, R. Frei, R. Ferstl, F. Shanahan, effect. Mult. Scler. 21: 1723–1729. E. M. Quigley, B. Kiely, C. A. Akdis, and L. O’Mahony. 2012. Bifidobacterium 127. Turnbaugh, P. J., M. Hamady, T. Yatsunenko, B. L. Cantarel, A. Duncan, infantis 35624 administration induces Foxp3 T regulatory cells in human peripheral R. E. Ley, M. L. Sogin, W. J. Jones, B. A. Roe, J. P. Affourtit, et al. 2009. A core blood: potential role for myeloid and plasmacytoid dendritic cells. Gut 61: 354–366. gut microbiome in obese and lean twins. Nature 457: 480–484. 102. Mercadante, A. C., S. M. Perobelli, A. P. Alves, T. Gonçalves-Silva, W. Mello, 128. David, L. A., C. F. Maurice, R. N. Carmody, D. B. Gootenberg, J. E. Button, A. C. Gomes-Santos, A. Miyoshi, V. Azevedo, A. M. Faria, and A. Bonomo. 2014. B.E.Wolfe,A.V.Ling,A.S.Devlin,Y.Varma,M.A.Fischbach,etal.2014. Oral combined therapy with probiotics and alloantigen induces B cell–dependent Diet rapidly and reproducibly alters the human gut microbiome. Nature 505: long-lasting specific tolerance. J. Immunol. 192: 1928–1937. 559–563. 103. Liao, H. Y., L. Tao, J. Zhao, J. Qin, G. C. Zeng, S. W. Cai, Y. Li, J. Zhang, and 129. Piccio, L., J. L. Stark, and A. H. Cross. 2008. Chronic calorie restriction attenuates H. G. Chen. 2016. Clostridium butyricum in combination with specific immu- experimental autoimmune encephalomyelitis. J. Leukoc. Biol. 84: 940–948. notherapy converts antigen-specific B cells to regulatory B cells in asthmatic pa- 130. Kleinewietfeld, M., A. Manzel, J. Titze, H. Kvakan, N. Yosef, R. A. Linker, tients. Sci. Rep. 6: 20481. D. N. Muller, and D. A. Hafler. 2013. Sodium chloride drives autoimmune 104. Shi, Y., L. Z. Xu, K. Peng, W. Wu, R. Wu, Z. Q. Liu, G. Yang, X. R. Geng, J. Liu, disease by the induction of pathogenic TH17 cells. Nature 496: 518–522. Z. G. Liu, et al. 2015. Specific immunotherapy in combination with Clostridium 131. Wu, C., N. Yosef, T. Thalhamer, C. Zhu, S. Xiao, Y. Kishi, A. Regev, and butyricum inhibits allergic inflammation in the mouse intestine. Sci. Rep. 5: 17651. V. K. Kuchroo. 2013. Induction of pathogenic TH17 cells by inducible salt-sensing 105. Ochoa-Repa´raz, J., D. W. Mielcarz, L. E. Ditrio, A. R. Burroughs, S. Begum- kinase SGK1. Nature 496: 513–517. Haque, S. Dasgupta, D. L. Kasper, and L. H. Kasper. 2010. Central nervous 132. Farez, M. F., M. P. Fiol, M. I. Gaitań, F. J. Quintana, and J. Correale. 2015. system demyelinating disease protection by the human commensal Bacteroides Sodium intake is associated with increased disease activity in multiple sclerosis. fragilis depends on polysaccharide A expression. J. Immunol. 185: 4101–4108. J. Neurol. Neurosurg. Psychiatry 86: 26–31. 604 BRIEF REVIEWS: CNS AUTOIMMUNITY AND THE GUT MICROBIOME

133. Kim, K. S., S. W. Hong, D. Han, J. Yi, J. Jung, B. G. Yang, J. Y. Lee, M. Lee, and 137. Gebregiworgis, T., H. H. Nielsen, C. Massilamany, A. Gangaplara, J. Reddy, C. D. Surh. 2016. Dietary antigens limit mucosal immunity by inducing regula- Z. Illes, and R. Powers. 2016. A urinary metabolic signature for multiple sclerosis tory T cells in the small intestine. Science 351: 858–863. and neuromyelitis optica. J. Proteome Res. 15: 659–666. 134. Cotillard, A., S. P. Kennedy, L. C. Kong, E. Prifti, N. Pons, E. Le Chatelier, 138. Zeevi, D., T. Korem, N. Zmora, D. Israeli, D. Rothschild, A. Weinberger, M. Almeida, B. Quinquis, F. Levenez, N. Galleron, et al; ANR MicroObes O. Ben-Yacov, D. Lador, T. Avnit-Sagi, M. Lotan-Pompan, et al. 2015. Person- Consortium. 2013. Dietary intervention impact on gut microbial gene richness. alized nutrition by prediction of glycemic responses. Cell 163: 1079–1094. Nature 500: 585–588. 139. Riccio, P., R. Rossano, M. Larocca, V. Trotta, I. Mennella, P. Vitaglione, 135. Smith, P. M., M. R. Howitt, N. Panikov, M. Michaud, C. A. Gallini, M. Ettorre, A. Graverini, A. De Santis, E. Di Monte, and M. G. Coniglio. 2016. M. Bohlooly-Y, J. N. Glickman, and W. S. Garrett. 2013. The microbial me- Anti-inﬂammatory nutritional intervention in patients with relapsing-remitting tabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis. Science 341: and primary-progressive multiple sclerosis: a pilot study. Exp. Biol. Med. (May- 569–573. wood) 241: 620–635. 136. Haghikia, A., S. Jo¨rg, A. Duscha, J. Berg, A. Manzel, A. Waschbisch, A. Hammer, 140. Bisht, B., W. G. Darling, R. E. Grossmann, E. T. Shivapour, S. K. Lutgendorf, D. H. Lee, C. May, N. Wilck, et al. 2015. Dietary fatty acids directly impact L. G. Snetselaar, M. J. Hall, M. B. Zimmerman, and T. L. Wahls. 2014. A central nervous system autoimmunity via the small intestine. [Published erratum multimodal intervention for patients with secondary progressive multiple sclerosis: appears in 2016 Immunity. 44:951–953] Immunity 43: 817–829. feasibility and effect on fatigue. J. Altern. Complement. Med. 20: 347–355. Downloaded from http://www.jimmunol.org/ by guest on January 26, 2017 Th eJournal of Brief Reviews Immunology

Inside Out: HIV, the Gut Microbiome, and the Mucosal Immune System Jay Liu,*,1 Brett Williams,†,1 Daniel Frank,* Stephanie M. Dillon,* Cara C. Wilson,* and Alan L. Landay‡ The components of the human gut microbiome have coupled with Norman Pace’s application (3) in the 1980s of been found to inﬂuence a broad array of pathologic con- then-new molecular biological techniques such as PCR and ditions ranging from heart disease to diabetes and even DNA/RNA sequencing to studies of microbial ecology ushered to cancer. HIV infection upsets the delicate balance in in the current era of culture-independent microbiology (3–9). the normal host-microbe interaction both through alter- High-throughput sequencing of marker genes (most notably the ations in the taxonomic composition of gut microbial 16S and 18S rRNA genes of bacteria/archaea and eukaryotes,

communities as well as through disruption of the normal respectively), deep shotgun sequencing of mixed community Downloaded from host response mechanisms. In this article we review the DNA/RNA, metabolomics, and other ’omics technologies now current methods of gut microbiome analysis and the allow the study of complex microbial ecosystems associated with resulting data regarding how HIV infection might change the human body in much greater detail and without the need to the balance of commensal bacteria in the gut. Addition- ﬁrst propagate axenic cultures (10–12). The goal of rRNA gene analysis is centered on identifying ally, we cover the various effects gut microbes have on host

and enumerating the types of organisms occurring in a microbial http://www.jimmunol.org/ immune homeostasis and the preliminary but intriguing community (“who is there”), whereas the other ’omics tech- data on how HIV disrupts those mechanisms. Finally, we nologies seek to understand the functional capacity and mo- briefly describe some of the important biomolecules pro- lecular outputs of the community (“what are they doing”) (10, duced by gut microbiota and the role that they may play 13). Small subunit (SSU) gene analyses were generally applied in maintaining host immune homeostasis with and with- first and most frequently to questions of dysbiosis and HIV. out HIV infection. The Journal of Immunology,2017, Several highly conserved regions are found scattered across the 198: 605–614. SSU gene, permitting the design of broad-range PCR primer sets with universal (i.e., three domain) or pan-bacterial specifici- by guest on January 26, 2017 oncerted work during the last three decades has ties. These primers allow one to PCR amplify the sequence genes demonstrated that the trillions of microbes (bacterial, from a broad range of microorganisms. Following SSU gene C viral, and fungal) that call the human body home are sequencing, significant bioinformatics processes are required to not just inert passengers, but are active participants in the generate interpretable data. Although insertion into a well- immune processes affecting our overall health. Our microbial supported phylogenetic tree is the gold standard for inferring tenants have a profound influence on host biology as well as on thetypeoforganismfromwhichanSSUsequencewasobtained, disease processes, impacting many fields of medicine, including the sheer volume of sequences generated by next-generation se- infectious disease, gastroenterology, pulmonology, immunol- quencing platforms raises significant computational challenges ogy, rheumatology, cardiology, endocrinology, geriatrics, and in this regard. Therefore, most modern SSU analysis pipelines even oncology. This review focuses on alterations of the gut employ heuristic approaches to assign taxonomic names to microbiome (termed “dysbiosis”) that occur with HIV infec- sequences or closely related groups of sequences. In any case, tion and the associated disruption of gut homeostasis and sub- the size of the reference database used, as well as the length and sequent clinical consequences. position of the SSU fragment sequenced (most next-generation sequencers cannot generate full-length sequences, but instead focus on particular variable regions), will place fundamental Technologies to study the microbiome limitations on the accuracy of the sequence classification pro- The study of commensal bacteria traditionally has been limited cess: some regions of the SSU gene may not encode sufficient to culturable organisms. However, the revolution in molecular information to infer species-level classifications. Furthermore, phylogenetics initiated in the 1970s by Woese and Fox (1, 2) microbial species names assigned on the basis of classical

*Division of Infectious Disease, Department of Medicine, Anschutz Medical Campus, Address correspondence and reprint requests to Dr. Alan L. Landay, Rush University University of Colorado, Aurora, CO 80045; †Division of Infectious Disease, Depart- Medical Center, Professional Ofﬁce Building No. 1, 1725 West Harrison Street, Room ment of Medicine, Rush Medical College, Chicago, IL 60612; and ‡Department of 306, Chicago, IL 60612. E-mail address: [email protected] Immunology and Microbiology, Rush Medical College, Chicago, IL 60612 Abbreviations used in this article: ART, antiretroviral therapy; FMT, fecal microbiota 1J.L. and B.W. contributed equally to this work. transplant; LPMC, lamina propia mononuclear cell; MSM, men who have sex with men; OTU, operational taxonomic unit; sCD14, soluble CD14; SCFA, short-chain fatty acid; ORCID: 0000-0003-1715-4930 (B.W.). SSU, small subunit; TMAO, trimethylamine-N-oxide; Treg, regulatory T cell. Received for publication August 4, 2016. Accepted for publication October 4, 2016. Copyright Ó 2017 by The American Association of Immunologists, Inc. 0022-1767/17/$30.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601355 606 BRIEF REVIEWS: HIV AND THE GUT MICROBIOME biochemical and morphological measurements may not map with early onset frailty, physical debility, and neurocognitive precisely onto the tree of life, which is derived from SSU se- dysfunction (25). Finally, the human gut is a critical site of host- quences. Consequently, molecular sequence data often are used microbe interaction, and any process that perturbs either side of to define relatedness groups termed operational taxonomic units that equation will have profound clinical and gut homeostatic (OTUs) on the basis of sequence dissimilarity–based clustering implications. HIV infection may affect both sides by altering the (numerical OTUs) or taxonomic binning (taxonomic OTUs). components of the gut microbiome and by changing the host In addition to enumerating OTUs, sequencing data can be immune response to gut microbes. used to describe the biodiversity of one or more microbial communities (14). Alpha diversity describes the biodiversity Microbiome alterations associated with HIV-1 infection within an individual microbial community (i.e., a single The human gut bacterial microbiome consists primarily of four specimen) and can be measured by different indices: richness phyla: Firmicutes, Bacteroidetes, Actinobacteria, and Proteo- simply counts the number of OTUs in a community, whereas bacteria (26). Composition of the gut microbiome varies sig- its overall complexity can be quantified by Shannon or Simp- nificantly depending on socioeconomic factors, age, geography, son diversity indices. In contrast, b diversity measures describe and diet (27, 28). For example, in children from Burkina Faso, the overlap in distributions of OTUs between two or more Africa, a diet high in fiber (normal for that region) appears to communities and are commonly presented as the dissimilarity enrich the microbiome at the phylum level, with Bacteroidetes between the microbiomes of two samples. Indices such as Bray- enriched at the expense of Firmicutes and Enterobacteriaceae Curtis, Jaccard, and UniFrac can all be used to infer b diversity

compared with European children (27). Enterotype, or the Downloaded from (14, 15). UniFrac, for instance, measures the distances between general makeup of the gut microbiome, appears to be shared communities based on lineages they contain; it can be used as a among people sharing the same dwelling (29). At the genus distance metric and can be visualized on a principal coordinate level there is a relatively robust association between the plant- plot. However, each index differs with respect to its weighting and carbohydrate-based diet of agrarian cultures and an of rare versus abundant taxa, presence/absence versus abun- enterotype with a low Bacteroides/Prevotella ratio (28, 30). dance measures, and consideration of phylogenetic differences Conversely, an enterotype with a relatively high Bacteroides/ http://www.jimmunol.org/ among taxa. Furthermore, b diversity indices form the basis of sta- Prevotella ratio is associated with the animal product–based tistical tests, such as permutation-based ANOVA (PERMANOVA) Western diet, which is high in protein and saturated fat. tests, that assess whether a potential predictor variable (e.g., The extent to which HIV infection and resultant immune disease occurrence) is statistically significant (16). More explor- compromise alter the gut microbiome remains controversial. atory techniques not described in this review involving ordina- Controlling for the myriad factors that influence gut micro- tion (e.g., nonmetric multidimensional scaling), hierarchical biota, including the factors mentioned above and other likely clustering, and a variety of other machine learning algorithms confounders such as antibiotic use, antiretroviral drugs also can be applied to OTU data, as well as to other ’omics themselves, and sexual practices, is quite difficult given the datasets. relatively small sample sizes generally available for study of the by guest on January 26, 2017 human microbiome. No well-controlled studies to date have Importance of the gut microbiome evaluated the gut microbiota of the same individuals before and The gut microbiota plays a critical role in gut homeostasis, after infection with HIV. Studies using SIV-infected nonhu- which in turn influences a wide variety of clinical outcomes. In man primates have far more control over these factors and are general, gut bacteria help in the processing of nutrients either able to obtain samples before and after experimental infection through digestion of insoluble fibers, breaking down of nu- with SIV. Wild chimpanzees, which develop progressive im- trients, or by providing important metabolites that influence munodeficiency with SIV infection, develop instability of their nutrition. On a micromolecular level, some of the microbial gut microbiota following SIV infection (31). This is in con- organisms of the gut produce important precursors to small trast to wild gorillas, which maintain a stable gut microbiome molecules such as trimethylamine-N-oxide (TMAO), which following SIV infection, although it is unknown whether go- has been associated with cardiovascular disease (17), and rillas develop progressive CD4 depletion from SIV. Therefore, short-chain fatty acids (SCFAs), which are not only a source this discrepancy may be related to loss of intestinal immune of energy for enterocytes but also may have important immu- surveillance in chimpanzees but not gorillas (32). However, nomodulatory effects on the gut mucosal immune system rhesus macaques have repeatedly been shown to have an enteric (18). The taxonomic composition of the microbiome is im- microbiome that is not substantially altered by SIV infection, portant in and of itself, as commensal organisms have a role in except temporarily during acute infection (33–35). Despite regulating the development and activation of the lamina pro- long-term stability of the gastrointestinal microbiome in ma- pria T cell compartments (19–22). Additionally, certain com- caques, microbial translocation in advanced disease occurs with ponents of the Bacteroidetes and Proteobacteria phyla can be the phylum Proteobacteria, which was found in lymph nodes considered as pathobionts (defined as commensal bacteria that and liver and was overrepresented in the stool in the acute phase have pathogenic potential in the appropriate clinical context) of SIV infection (levels diminish with chronic infection) (36). (23). On a clinical level, some cross-sectional studies have Studies involving HIV-infected human cohorts, despite their demonstrated differences in the gut microbiota in obese indi- inherent limitations, have found a number of reproducible viduals compared with lean individuals, whereas a case control differences in the gut microbiome between HIV-infected and study identified differences at the phylum level between in- uninfected individuals. Note that the literature is heteroge- dividuals who developed diabetes compared with healthy neous with regard to the type of subjects studied (treated versus controls (reviewed in Ref. 24). In older populations, cross- untreated), type of sampling method used (rectal sponge, stool sectional studies have associated differences in the gut microbiome swab, stool, tissue site biopsy), and the level of bacterial The Journal of Immunology 607 classification examined. The microbiome differences observed In the study by Dinh et al. (45), Proteobacteria and several using various sampling methods have been investigated by subtaxa, including Enterobacteriaceae, which contains many several groups. Early work by Zoetendal et al. (37) in HIV- common pathogens, were overrepresented in HIV-positive negative individuals found that the bacteria associated with individuals and were associated with immune activation. mucosal biopsy samples differed greatly from those found This association between HIV infection and increased abun- in feces from the same individual. Dillon et al. (38) likewise dance of Proteobacteria, particularly in mucosal samples, may observed differences in HIV-associated changes in taxa be more meaningful than the Prevotella/Bacteroides shift given abundance depending on the sampling method. Specifically, the tendency of Protoebacteria to translocate in the nonhu- an increased abundance of the phylum Proteobacteria in man primate model (36). HIV-infected individuals was only noted with mucosal sam- Supporting the role of HIV-1 in driving dysbiosis is the ples, suggesting that luminal samples may miss potentially finding that individuals on virally suppressive ART tend to pathogenic alterations. However, HIV-associated changes in have a microbiome shifted closer to that of uninfected controls the relative abundance of the genera and families in the compared with individuals with untreated HIV infection, but phylum Bacteroidetes were generally similar across all sam- ART was not associated with complete normalization of the pling platforms. Furthermore, abundance of certain taxa (e.g., microbiome in stool and rectal sponge samples (40, 46). The Prevotella sp.) in mucosal samples strongly associated with microbiome findings in treated versus untreated HIV-infected markers of mucosal and systemic inflammation, whereas individuals suggest that ongoing mucosal inflammation Prevotella abundance in fecal samples did not. This suggests driven by unchecked viral replication may largely drive the Downloaded from that the mucosa-associated microbiome may more closely observed microbiome changes. Conversely, the partial “nor- interact with and potentially influence the immune system. malization” of the microbiome observed with virus suppres- The gut microbiome also seems to vary depending on seg- sion on ART would tend to suggest that as the mucosal ment of the gut, with samples from the terminal ileum and immune system recovers, it is better able to manage the gut right colon best differentiating between HIV-positive subjects environment and allow for a more normal microbiome. and HIV-negative controls (39). Unfortunately, there does However, not all studies have found a stool microbiome shift http://www.jimmunol.org/ not appear to be consensus regarding which type of sampling toward HIV-negative controls following ART (47). Although should be used routinely in studies (see Table I), and this published data are limited, antiretrovirals themselves may limits the ability to compare findings across studies. have a direct effect on commensal bacteria and/or the phage In developed nations, where the great majority of human viruses that infect them, which may in part explain persistent microbiome studies have been performed, studies have variously microbiome differences between long-term–treated HIV- reported that HIV-infected individuals have significantly greater infected individuals versus seronegative controls (36). abundance of Prevotella and/or significantly less Bacteroides at Loss of fecal microbial diversity has been noted in the the genus level than do uninfected controls (38–44). This settings of Clostridium difficile colitis and inflammatory bowel finding was observed in cohorts of treated and untreated HIV- disease (51, 52). Several authors have found a similar loss of by guest on January 26, 2017 infected persons and in both stool and mucosal biopsy samples. diversity associated with HIV infection status (39, 46–48, However, not all microbiome studies of treated and untreated 53). However, an association between HIV infection and HIV-infected individuals observed this phenomenon in rectal decreased microbial diversity has not been consistently ob- sponge specimens or stool (45–47). Although Nowak et al. served; in fact, several studies reported no loss of diversity, and (47) did not find overabundance of the genus Prevotella in their one found increased diversity in varying HIV-infected patient Swedish HIV-positive cohort at baseline, they did note that cohorts (38, 40, 43, 45). Importantly, MSM sexual behavior Prevotella decreased following antiretroviral therapy (ART). In itself was associated with increased microbial diversity, a the largest study to date examining the gut microbiome of finding that suggests that failure to control for sexual behav- HIV-infected individuals, Noguera-Julian et al. (48) found, at iors could confound the diversity analyses in some studies. the genus level, that a high Prevotella/low Bacteroides enterotype in stool specimens was highly associated with men who have Expansion of the virome during HIV infection sex with men (MSM) behavior regardless of HIV-1 infection Anelloviruses, which are ssDNA viruses, are highly prevalent in status, which may explain the perceived association between stool and blood in the general population but have not been this enterotype and HIV infection status in prior studies that implicated as causative agents of disease (54). Early virome did not control for sexual behavior. A study of stool samples studies found that plasma levels of anellovirus DNA increase from Uganda, where HIV transmission is predominantly het- in AIDS and other immunocompromised conditions and erosexual, seems to support the hypothesis that a Prevotella- then decline again following immunologic recovery (54). predominant enterotype is driven by sexual behavior, as it did More recently, the potential role of these viruses in the not find Prevotella predominance to be dependent on HIV-1 pathogenesis of immune activation has been investigated. serostatus or level of immunosuppression (49). However, note Anelloviruses such as torque teno virus have been associated that the HIV-1–seronegative Ugandans had Prevotella-predominant with progression of liver disease in HIV/hepatitis C virus stool microbiota at baseline, possibly due to dietary differences, so coinfection but do not appear to be driving chronic T cell the study may have been underpowered to detect a small HIV- activation in HIV (55, 56). In a nonhuman primate model, associated increase in Prevotella abundance. Handley et al.(34) found a .10-fold increase in stool viral However, enrichment of the phylum Proteobacteria and sequences following pathogenic SIV infection including ade- reciprocal diminishment of the phylum Firmicutes in HIV- noviruses and picornaviruses. Adenoviruses in particular were infected subjects, particularly in mucosal samples, appears to associated with enteritis and progressive immune dysfunction be more consistently reported in the literature (39, 43, 44, 50). (34). Nonpathogenic SIV infection in African green monkeys 608 BRIEF REVIEWS: HIV AND THE GUT MICROBIOME did not result in expansion of the stool virome (34). A human and massive depletion of lamina propia CD4 cells as soon as study from Uganda found that both anelloviruses and ade- 7 d after infection (73, 74). In humans, .50% of lamina noviruses were increased in those with low CD4 counts (57, propria CD4 cells are depleted in early and acute HIV in- 58). Future studies should investigate what role, if any, these fection (75), with a selective targeting of Th CD4 cells that viruses play in disrupting the epithelial barrier integrity of the produce IL-17 and IL-22. The mechanism of this depletion is gut, thereby contributing to microbial translocation. likely cell death of productively infected cells via apoptosis as well as of bystander cells via pyroptosis (76, 77). The density HIV infection and its disruption of host-microbe interactions of mucosal CD4 cells does not always fully recover despite HIV infection is associated with a chronic inflammatory state effective ART and viral suppression (78, 79). Poor CD4 cell as represented by increased circulating soluble TNFRs 1 and 2, recovery leads to important functional consequences for the IL-6 (59), and markers of T cell activation (CD38 and HLA- gut mucosal barrier, especially failure to protect against in- DR expression) (60). Additionally, HIV infection is associated vading pathogens as well as loss of cytokines necessary to with increased plasma markers for microbial translocation/ support normal barrier function (80, 81). Poor CD4 recon- monocyte activation (LPS and soluble CD14 [sCD14]) and stitution has also been found to correlate with tight junction epithelial barrier damage (e.g., intestinal fatty acid–binding dysfunction between the epithelial cells of the mucosal barrier protein) (61). Immune activation and gut barrier disruption are (82). Additionally, immunohistochemistry of gut biopsies in highest in acute infection and fall with chronic infection, but HIV-infected immune nonresponders with poor CD4 recovery despite ART revealed decreased epithelial cell prolif-

ART decreases them further (61). Some authors, however, have Downloaded from noted that these plasma markers of inflammation remain eration and increased neutrophil infiltration (a surrogate for higher in treated, virally suppressed HIV patients (62, 63) or epithelial barrier dysfunction) (83). These changes were cor- spontaneous viremic suppression compared with HIV-negative related with increased circulating sCD14, a systemic marker controls (64, 65). However, some studies have demonstrated of LPS-induced monocyte activation and indirectly of mi- resolution of microbial translocation with long-term ART (66) crobial translocation. Th22 and Th17 cells along with a subset of innate lymphocyte and have reported normal levels of monocyte activation in http://www.jimmunol.org/ long-term nonprogressors (67). cells (i.e., type 3 innate lymphoid cells) are responsible for On a clinical level, elevated systemic markers for inflam- production of IL-22, a critical cytokine for epithelial barrier mation, microbial translocation, and epithelial barrier damage maintenance that induces stem cell–mediated epithelial cell have been linked to poor clinical outcomes even in treated proliferation in the gut (84). Disruption of IL-22 production populations (62). Sandler et al. (68) showed in 2011 that may worsen epithelial barrier dysfunction and increase micro- sCD14 was an independent predictor of mortality in HIV bial translocation. Indeed, compared with age-matched con- infection. Timmons et al. (69) analyzed levels of LPS and trols, multiple groups have shown depletion of Th22 cells in sCD14 in study participants with and without HIV infection gut mucosa of HIV-infected individuals, and one group and correlated them to body mass index, lipid panels, and showed an association of Th22 cell depletion with markers of by guest on January 26, 2017 insulin resistance. sCD14 in particular negatively correlated microbial translocation (50, 80, 81, 85). Note that not all in- with high-density lipoprotein, especially in those HIV- vestigators endorse the loss of Th22 cells as the primary cause infected individuals on ART (69). Note, however, that neg- of epithelial barrier dysfunction, instead pointing to type 3 ative correlation did not account for class of ART, which is innate lymphoid cells as most important for maintenance of known to affect lipid levels as well. Tenorio et al. (70) showed epithelial barrier integrity during HIV infection (80). To ad- that higher soluble markers of inflammation (soluble TNFR1, dress the effect of HIV-associated dysbiosis on Th22 expres- IL-6) correlated with non-AIDS–defining events such as sion, a recent small pilot study of fecal microbiota transplants stroke, cardiovascular disease, and non-AIDS cancers. Finally, (FMTs) from healthy macaques to SIV-infected macaques Hunt et al. (63) showed that epithelial barrier dysfunction as demonstrated increased frequencies of peripheral Th22 cells measured by peripheral blood levels of intestinal fatty acid– after transplantation and decreased gut mucosal inflammatory binding protein and zonulin-1 predicted mortality in HIV markers, although it did not comment on gut mucosal Th22 infection, even after adjustment for CD4 count. cell frequencies (86). From a pathophysiological perspective, there have been The components of the microbiome can directly contribute multiple explanations proposed for this phenomenon, in- to CD4 cell depletion in HIV infection. Several recent pub- cluding a direct effect of viral replication and immune acti- lications have reported on the use of in vitro modeling to vation caused by coinfections (such as CMV). However, one characterize interactions between commensal microbes and theory focuses on the effect of HIV infection and resultant primary human lamina propria immune cells with regard to dysbiosis on the gut mucosal barrier and immune system. The how they influence HIV-1 replication and depletion of lamina implications of a leaky gut barrier are 2-fold: systemic propria CD4 T cells. Exposure to heat-killed Escherichia coli inflammation is increased due to circulation of microbial (a pathobiont) resulted in enhanced activation, proliferation, components in the bloodstream (71), and there is an increase and HIV-1 infection of CD4+ T cells within lamina propia in the exposure of the resident gut mucosal T cell population mononuclear cell (LPMC) cultures (87). In subsequent studies, to new Ags. The etiology of this gut barrier dysfunction in Dillon et al. (88) were able to show that mucosal bacteria HIV infection may originate in the gut lamina propria and its that were altered in abundance in colon biopsies of HIV- resident CD4 T cells. Lamina propria T cells may be more infected individuals also enhanced HIV replication when susceptible to HIV infection due to high levels of activation cocultured with LPMCs in vitro. Increased CD4 infection and expression of HIV receptors such as CCR5 (72). Ex- following exposure to commensal bacteria appeared to be due periments in SIV nonhuman primate models revealed rapid in part to increased expression of the HIV-1 coreceptor CCR5, The Journal of Immunology 609 induced predominantly by exposure to Gram-negative bacteria. At broader taxonomic levels, the order Lactobacilliales and When highly expressed Gram-negative (LPS) and Gram- family Bacteroidaceae seemed to protect against CD4 activa- positive (lipoteichoic acid) cell wall components were tested, tion and depletion (43, 98). Other members of the family only LPS enhanced CD4 expression of CCR5 (88). The Bacteroidaceae or species-specific metabolic differences within mechanisms of death of gut CD4 cells by HIV and the impact the genus Bacteroides may account for the discrepant effects of of bacterial exposure on cell death have also been investigated. the genus Bacteroides versus family Bacteroidaceae. Studies us- Steele et al. (76) reported that pyroptosis was an important ing Lactobacillus-containing probiotics in HIV appear to sup- HIV-associated mechanism of bystander (non–productively port the role of Lactobacillus spp. in reducing CD4 activation infected) CD4 T cell killing in LPMCs in addition to the and depletion, although similar to much of the literature re- apoptosis caused by direct viral infection. Furthermore, apo- garding probiotics, results have been mixed (99–101). Conversely, ptotic death of CD4 T cells was increased in HIV-infected Prevotella and two subtaxa of the phylum Proteobacteria (the LPMC cultures following exposure to pathobiont bacteria. order Enterobacteriales and the family Enterobacteriaceae) have These findings may explain the enhanced inflammatory sig- been associated with CD4 T cell activation and depletion (38, 43, nature seen in HIV gut enteropathy, as pyroptosis ultimately 88, 102). Note that the association between Prevotella abundance results in release of cytoplasmic contents, including IL-1b, and mucosal T cell activation was only noted with microbiome which causes increased permeability in intestinal epithelial cell analysis from colon biopsy samples, not stool or fecal aspirates, tight junctions (89) and may ultimately contribute to epithelial suggesting that mucosa-adherent bacteria may be more important barrier dysfunction. Increased mucosal HIV-1 replication and in determining mucosal immune status than are luminal organ- Downloaded from apoptotic CD4 T cell death following bacterial exposure may isms (38). To our knowledge, only four works published to date help to explain how microbial translocation contributes to viral have examined microbiome specimens from gastrointestinal mu- pathogenesis in the early stages of HIV infection. cosal biopsy specimens from HIV-infected individuals (38, 39, In addition to their contribution to the selective targeting 43, 103). Future larger studies are needed to clarify whether and depletion of CD4 cells by HIV virus, the altered gut microbiome analyses of mucosal or luminal samples better microbiome may have further effects on gut Th subset–related correlate with systemic markers of inflammation. http://www.jimmunol.org/ immune homeostasis. This is because commensal bacteria As mentioned above, soluble markers of gut permeability, have a role in regulating the normal balance between regu- microbial translocation, and immune activation are clearly latory T cells (Tregs), which are anti-inflammatory in nature, linked to clinical outcomes of interest (63, 68, 70). Linking and Th17 cells, which produce the proinflammatory cytokine particular microbes to these markers has been more challeng- IL-17 and regulate host responses to commensal and patho- ing. Vujkovic-Cvijin et al. (43) found several soluble markers, genic microbiota. One study in germ-free mice demonstrated including IL-6, soluble TNFR2, kynurenine/tryptophan ratio, that segmented filamentous bacteria (a gut commensal most and IP-10, to be associated with a Proteobacteria-heavy cadre often found in the terminal ileum) were shown to be necessary of HIV-associated microbes. Dillon et al. (38) found an asso- for the development of Th17 cells (20). Additionally, Clostridium ciation between overall mucosal microbiome changes and by guest on January 26, 2017 species likely have a role in the accumulation of Tregs in mouse plasma LPS as well as with peripheral blood and mucosal T cell gut (19, 21). Bacteroides fragilis, a prominent anaerobic com- activation. Dinh et al. (45) found that abundances of the order mensal, is thought to inhibit CD4 differentiation into Th17 Enterobacteriales and family Enterobacteriaceae positively cor- and increase differentiation into Tregs in mouse gut (22). The related with levels of sCD14, IL-1b, and IFN-g. The class importance of HIV infection–related dysbiosis in depletion of Erysipelotrichi and genus Barnesiella were positively associated Th17 cells was highlighted (albeit indirectly) by studies in SIV- with IFN-a levels. We should be cautious when interpreting infected macaques that showed improvement of the frequency these results, as alterations in plasma markers may have been of polyfunctional Th17 cells (defined in the study as Th17 cells driven by HIV infection and the alterations in microbiota by capable of producing IL-17 and IFN-g) with probiotic/ behavioral differences. Variations in the subjects, sampling meth- prebiotic treatment plus IL-21 supplementation (90, 91). ods, and taxonomic levels investigated make it particularly difficult, These findings are currently being investigated in human at this stage, to draw any firm conclusions regarding which studies (AIDS Clinical Trials Group 5352), which will further microbes are most responsible for immune activation. Future elucidate this mechanism. studies, including in vitro and ex vivo experiments and direct Finally, gut microbiota may be associated with improvement evaluation of the metabolites found in stool and mucosal sam- in the CD4 count, which continues to be an important pro- ples, will likely improve our understanding of these interactions. gnostic indicator and predicts non-AIDS events and mortality in Direct identification of translocating bacteria may also be addition to AIDS-associated morbidity and mortality (57, 92– useful in determining what defines a proinflammatory micro- 96). As described above, poor CD4 recovery is linked to mi- biome. Klase et al. (36) found in an SIV infection model that crobial translocation, although it is unclear whether there is a bacterial DNA of the phylum Proteobacteria was overrepre- causal association and, if so, in which direction (97). Several sented in mesenteric lymph nodes and liver tissue of infected authors have investigated the association between various gut primates, suggesting that these organisms preferentially trans- bacterial taxa and peripheral CD4 recovery. As with previously locate. The order Enterobacteriales and the family Enter- noted microbiome associations, findings have not always been obacteriacea of the phylum Proteobacteria were significantly confirmed, and, importantly, note the taxonomic level at which more prevalent in both stool and gut mucosal samples of HIV- the association was found. Abundance of the genus Bacteroides infected individuals than in uninfected control participants in in stool and colonic biopsies was associated with lower pe- several studies (38, 41, 43, 45, 102). A small study from China ripheral CD4 recovery whereas Lactobacillales abundance was found that DNA from the order Pseudomonadales (phylum associated with a higher peripheral CD4 percentage (38, 98). Proteobacteria) and in particular the genera Moraxella and 610 BRIEF REVIEWS: HIV AND THE GUT MICROBIOME

Psychrobacter was found in the blood of treatment-naive sub- Importantly, in the context of HIV infection, butyrate- jects with advanced immunosuppression in significantly higher producing species have also been shown to be decreased in abundance than in HIV-negative controls (104). As sequencing abundance in enteric mucosa and stool (38, 39, 41, 46, 47), technology improves, species- and strain-specific data will be and high-dose butyrate may reduce bacteria-induced CD4 invaluable in determining which organisms, if any, preferen- T cell cytokine production, activation, and HIV infection tially translocate during HIV-1 infection. levels in an ex vivo intestinal cell model (S.M. Dillon, J. Kibbie, E.J. Lee, K. Guo, M.L. Santiago, G.L. Austin, Important bacterial metabolites S. Gianella, A.L. Landay, A.M Donovan, D.N. Frank, M.D. Metagenomics can be used to identify the genes for a particular McCarter, and C.C. Wilson, submitted for publication). In- metabolic pathway and allow us to bypass difficulties related to terestingly, although not yet shown in the gut, oral flora that the identification of microbiota at the species or strain level. produce SCFAs have been shown to enhance reactivation of Microbial translocation and resultant immune activation have latent virus, including HIV-1, through inhibition of histone been linked to microbial zeatin biosynthesis and the meta- deacetylase (122). This could ultimately benefit the host, as bolism of tryptophan to kynurenine (42, 43). The plasma latently infected cells can be a source of productive viral kynurenine/tryptophan ratio is elevated in HIV-infected per- replication and result in persistent life-long infection. sons and has been linked with poor CD4 recovery, intestinal barrier dysfunction, and mortality (63, 105–107). Organisms Conclusions

with the genes for tryptophan to kynurenine metabolism are It is clear that HIV infection results in severe damage to the Downloaded from overrepresented in HIV-infected individuals (43). Serrano- intestinal mucosal compartment with epithelial barrier damage Villar et al. (108) demonstrated that 3-hydroxyanthranilate, a and depletion of CD4 T cell subsets critical to maintenance of product of the kynurenine pathway, was elevated in stool intestinal homeostasis. Although many early pilot studies bacteria from HIV-infected individuals but was undetectable in reported HIV-associated changes in the enteric microbiome, controls and those with lupus or C. difﬁcile colitis. Note, both in composition and in diversity, more recent studies

however, that IDO1 is expressed by activated monocytes and suggest that confounding factors such as sexual behavior may http://www.jimmunol.org/ dendritic cells, and thus a portion of the tryptophan meta- explain some of those original findings rather than HIV in- bolism and kynurenine production that occurs in HIV infec- fection status per se. However, a reduction in overall microbial tion is derived from human cells and is likely driven by diversity is found in HIV-infected individuals even when inflammation (109). Lactobacillus spp. appear to inhibit IDO1 accounting for other lifestyle factors. Additionally, HIV in- and are selectively depleted in SIV-infected macaques, so this fection has been associated with increases in Proteobacteria, may be a mechanism by which Lactobacillus spp. prevent CD4 which contains several species that can be considered patho- activation and depletion (110). bionts in the appropriate clinical setting. By and large, despite Microbial metabolites other than those of tryptophan ca- the preponderance of early data gathered, firm conclusions on tabolism have also been associated with end-organ disease. Tang the exact nature of HIV-associated dysbiosis, including the by guest on January 26, 2017 and colleagues (17, 111, 112) showed that the production of impact of lifestyle, diet, comorbidities, and treatment effects TMAO, a small molecule that is closely linked to coronary across the spectrum of HIV-infected patient populations, await artery disease, mortality, and renal insufficiency is dependent larger well-controlled studies that include ethnically and on gut microbes. Studies of TMAO in HIV-infected cohorts geographically diverse study populations. Furthermore, con- have had conflicting results regarding whether TMAO is as- trolled, longitudinal studies of persons from acute infection sociated with higher risk of coronary events in HIV, and it does (and perhaps before infection if already part of an existing HIV not appear that levels of TMAO are particularly elevated in uninfected cohort) to chronic infection, although technically persons with HIV infection (113–116). challenging, would shed significant light on the evolution of Recently, SCFAs have received some attention as important microbiome changes with HIV infection. Despite these lin- biomolecules produced by gut microbes. SCFAs such as bu- gering questions, there is solid evidence of associations between tyrate are produced from bacterial fermentation of insoluble microbiome features (especially in the gut mucosa) and mu- fibers in the diet by certain components of the Firmicutes cosal and systemic measures of inflammation and immune phylum (117). Butyrate has proven to be critical in a number activation (summarized in Table I), although whether these of ways to the human gut. It functions as a direct energy associations are causal or simply correlative remains unclear. source for colonocytes, such that starvation of such in in- Intriguing mechanistic hypotheses regarding the direct and flammatory bowel disease is theorized to cause apoptosis and indirect effects of the altered microbiome on gut immune enhance epithelial barrier damage (118). Butyrate not only function have been proposed, yet due to the complexity of the nourishes colonocytes but also stimulates production of mu- microbial communities involved and our evolving under- cin, enhancing gut mucosal barrier integrity. Xiong et al. standing of the factors that influence them (in normal health, (119) showed in 2016 that butyrate directly upregulated host much less during HIV infection), specific studies drilling defense peptides in piglet gut epithelial cells and helped down to the exact mechanisms involved are still incomplete. ameliorate the pathogenic consequences of E. coli 0157:H7 Interventions designed to modify the gut microbiome and infection. Additionally, butyrate reportedly regulates and thereby reduce inflammation-associated comorbidities are ob- expands the colonic Treg population, promoting an anti- viously of great interest to the medical and scientific commu- inflammatory phenotype (120, 121). Gut dysbiosis in aging nities. Most interventions studied to date, in several different populations is characterized by decreased Firmicutes phyla disease settings, have focused on either directly repopulating species abundance and decreased production of SCFAs in the the gut with beneficial commensals (probiotics or direct fecal stool, as well as increased epithelial barrier dysfunction (25). transplant) or encouraging the growth of beneficial commensals The Journal of Immunology 611

with intake of substrate for their growth (prebiotics). Other

: interventions involve repletion of important biomolecules such , as introducing intracolonic butyrate to ameliorate autoimmune disease (118) or active infection (in piglets) (119). Furthermore, Blautia a Bacteroides still other preclinical study interventions are investigating directed exercise regimens to alter the gut microbiota for health beneﬁt (123). Early results of a pilot FMT trial in HIV were reported at the conference on retroviruses and opportunistic : serum TNF-

; Erysipelotrichi, infections in Boston in 2016 [NCT02256592 (124)]. Somsouk g ; Erysipelotrichi:

a et al. found, in a study of six subjects with treated HIV infec-

; Enterobacteriales/Enterobacteriaceae: tion, that with FMT the microbiomes of subjects shifted, : serum lipoteichoic acid; b

Ruminococcus transiently, toward those of their respective donors. They did not ﬁnd any changes in the serum kynurenine/tryptophan ratio, , sCD14, IFN- b lower EndoCAb a marker of immune dysfunction in HIV that has been linked Taxa Associated with higher plasma LTA associated with sCD14 Ruminococcus to mortality in HIV. Larger studies with potentially more in- Unfavorable Immune Markers

: higher plasma TNF- tensive interventions will be needed to determine whether FMT

Rhodobacteraceae, Halomonadaceae, is effective in reducing microbial translocation and immune activation. Downloaded from (left colon), Barnesiella peripheral CD8 activation; no taxa was signiﬁcantly

higher plasma IL-1 Studies of probiotics in HIV-infected individuals are extremely plasma kynurenine/tryptophan, IP-10; Micrococcaceae, Clostridia, Lachnospiraceae, heterogeneous, including varying probiotic organisms, formu- Pasteurellaceae: numerous markers in both serum and GALT Enterobacteriaceae: mucosal T cell activation; Staphylococcaceae: Enterobacteriales: duodenal mucosa CD4 depletion; Bacteroidales: lations, durations of treatment, and outcomes. The one common Bacteroides

:Prevotella colonic T cell and myeloid dendritic cell activation; finding in these studies has been that probiotics are generally safe and well tolerated (125). Studies of Bifidobacterium and : Lactobacillus probiotics have shown some improvement in CD4 http://www.jimmunol.org/ : count with supplementation, although changes are not always statistically significant (100, 101, 126, 127). Microbial trans-

Roseburia location and immune activation have also been the endpoints of (ileum): lower Faecalibacterium probiotic trials, and both yeast (Saccharomyces)– and bacterial : higher colon Th1

lower sCD14 (Bifidobacterium and Lactobacillus)–based probiotics have shown lower serum LPS Taxa Associated with promise in reducing some inflammatory markers, but results prevalence; Favorable Immune Markers activation; Rikenellaceae: Bacteroides mucosal IL-17 production Dialister have been mixed (99, 126) A probiotic/prebiotic (Lactobacillus, serum IL-6; and higher peripheral CD4% Bifidobacterium, Streptococcus plus inulin) mixture was shown to improve gut CD4 reconstitution in an SIV model and this by guest on January 26, 2017 multistrain probiotic is currently the subject of a randomized, placebo-controlled trial, that is, AIDS Clinical Trials Group A5350 (90, 128). Results from this large, well-designed, mul-

Samples ticenter study will help deﬁne whether these organisms play a role in reconstitution of the gut immune system and prevention left colon, and feces

immunohistochemistry of microbial translocation. A better understanding of the mech- terminal ileum, right colon, anistic interactions between gut microbes and their metabolic products and host mucosal immune cells in the setting of HIV- associated homeostatic disruption will likely lead to novel interventions that may ultimately reduce gut-derived inﬂammation

Location and its associated comorbidities. Aurora, CO Stool and mucosal biopsy Boston, MA Stool higher plasma IL-1 Gammaproteobacteria: Sacramento, CA Stool duodenal tissue; San Francisco, CA Rectal mucosa biopsy Bacteroidaceae: less T cell Disclosures 2 The authors have no ﬁnancial conﬂicts of interest. 2 controls Chicago, IL Mucosal samples of controls controls untreated, 2 2

+ References 2 1. Woese, C. R. 1987. Bacterial evolution. Microbiol. Rev. 51: 221–271. long-term

+ 2. Woese, C. R., and G. E. Fox. 1977. Phylogenetic structure of the prokaryotic virally suppressed controls virally suppressed, untreated, 14 age/

Participants domain: the primary kingdoms. Proc. Natl. Acad. Sci. USA 74: 5088–5090. , 22 HIV subjects prior to and + MSM initiating ART San Diego, CA Stool swab Lactobacillales: lower sCD14 + + + + + 3. Pace, N. R. 1997. A molecular view of microbial diversity and the biosphere. 1 HIV Science 276: 734–740. risk-matched controls nonprogressor, 9 HIV cases, 16 HIV 4. Relman, D. A., S. Falkow, P. E. LeBoit, L. A. Perkocha, K. W. Min, D. F. Welch, 18 HIV sex-matched HIV All male: 6 HIV after 9 mo of ART, 5 HIV

13 HIV and L. N. Slater. 1991. The organism causing bacillary angiomatosis, peliosis hepatis, and fever and bacteremia in immunocompromised patients. N. Engl. J. Med. 324: 1514. 5. Relman, D. A., J. S. Loutit, T. M. Schmidt, S. Falkow, and L. S. Tompkins. 1990. The agent of bacillary angiomatosis. An approach to the identiﬁcation of uncul-

Study tured pathogens. N. Engl. J. Med. 323: 1573–1580. 6. Relman, D. A., T. M. Schmidt, R. P. MacDermott, and S. Falkow. 1992. Identiﬁ- et al. (98) et al. (43) rez-Santiago

´ cation of the uncultured bacillus of Whipple’s disease. N. Engl. J. Med. 327: 293–301. Dinh et al. (45) 21 HIV Vujkovic-Cvijin Pe Ellis et al. (102) 10 HIV Dillon et al. (38) 18 HIV Mutlu et al. (39) 21 HIV 7. Wilson, K. H., R. Blitchington, R. Frothingham, and J. A. Wilson. 1991. Phy-

Table I. Taxa associated with immune markers in HIV logeny of the Whipple’s-disease-associated bacterium. Lancet 338: 474–475. 612 BRIEF REVIEWS: HIV AND THE GUT MICROBIOME

8. Wilson, K. H., R. Blitchington, B. Hindenach, and R. C. Greene. 1988. Species- 37. Zoetendal, E. G., A. von Wright, T. Vilpponen-Salmela, K. Ben-Amor, specific oligonucleotide probes for rRNA of Clostridium difficile and related species. A. D. Akkermans, and W. M. de Vos. 2002. Mucosa-associated bacteria in the J. Clin. Microbiol. 26: 2484–2488. human gastrointestinal tract are uniformly distributed along the colon and differ 9. Wilson, K. H., R. B. Blitchington, and R. C. Greene. 1990. Amplification of from the community recovered from feces. Appl. Environ. Microbiol. 68: 3401– bacterial 16S ribosomal DNA with polymerase chain reaction. J. Clin. Microbiol. 3407. 28: 1942–1946. 38. Dillon, S. M., E. J. Lee, C. V. Kotter, G. L. Austin, Z. Dong, D. K. Hecht, 10. Frank, D. N., and N. R. Pace. 2001. Molecular-phylogenetic analyses of human S. Gianella, B. Siewe, D. M. Smith, A. L. Landay, et al. 2014. An altered intestinal gastrointestinal microbiota. Curr. Opin. Gastroenterol. 17: 52–57. mucosal microbiome in HIV-1 infection is associated with mucosal and systemic 11. Frank, D. N., and N. R. Pace. 2008. Gastrointestinal microbiology enters the immune activation and endotoxemia. Mucosal Immunol. 7: 983–994. metagenomics era. Curr. Opin. Gastroenterol. 24: 4–10. 39. Mutlu, E. A., A. Keshavarzian, J. Losurdo, G. Swanson, B. Siewe, C. Forsyth, 12. Pace, N. R., J. Sapp, and N. Goldenfeld. 2012. Phylogeny and beyond: scientific, A. French, P. Demarais, Y. Sun, L. Koenig, et al. 2014. A compositional look at historical, and conceptual significance of the first tree of life. Proc. Natl. Acad. Sci. the human gastrointestinal microbiome and immune activation parameters in HIV USA 109: 1011–1018. infected subjects. PLoS Pathog. 10: e1003829. 13. Frank, D. N. 2011. Growth and development symposium: promoting healthier 40. Lozupone, C. A., M. Li, T. B. Campbell, S. C. Flores, D. Linderman, humans through healthier livestock: animal agriculture enters the metagenomics M. J. Gebert, R. Knight, A. P. Fontenot, and B. E. Palmer. 2013. Alterations in era. J. Anim. Sci. 89: 835–844. the gut microbiota associated with HIV-1 infection. Cell Host Microbe 14: 329– 14. Schloss, P. D., and J. Handelsman. 2007. The last word: books as a statistical 339. metaphor for microbial communities. Annu. Rev. Microbiol. 61: 23–34. 41. Sun, Y., Y. Ma, P. Lin, Y. W. Tang, L. Yang, Y. Shen, R. Zhang, L. Liu, J. Cheng, 15. Lozupone, C., and R. Knight. 2005. UniFrac: a new phylogenetic method for J. Shao, et al. 2016. Fecal bacterial microbiome diversity in chronic HIV-infected comparing microbial communities. Appl. Environ. Microbiol. 71: 8228–8235. patients in China. Emerg. Microbes Infect. 5: e31. 16. Anderson, M. J. 2001. A new method for non-parametric multivariate analysis of 42. Va´zquez-Castellanos, J. F., S. Serrano-Villar, A. Latorre, A. Artacho, M. L. Ferru´s, variance. Austral Ecol. 26: 32–46. N. Madrid, A. Vallejo, T. Sainz, J. Martıńez-Botas, S. Ferrando-Martıńez, et al. 17. Tang, W. H., Z. Wang, B. S. Levison, R. A. Koeth, E. B. Britt, X. Fu, Y. Wu, and 2015. Altered metabolism of gut microbiota contributes to chronic immune ac- S. L. Hazen. 2013. Intestinal microbial metabolism of phosphatidylcholine and tivation in HIV-infected individuals. Mucosal Immunol. 8: 760–772. cardiovascular risk. N. Engl. J. Med. 368: 1575–1584. 43. Vujkovic-Cvijin, I., R. M. Dunham, S. Iwai, M. C. Maher, R. G. Albright,

18. Rivie`re, A., M. Selak, D. Lantin, F. Leroy, and L. De Vuyst. 2016. Bifidobacteria M. J. Broadhurst, R. D. Hernandez, M. M. Lederman, Y. Huang, M. Somsouk, Downloaded from and butyrate-producing colon bacteria: importance and strategies for their stim- et al. 2013. Dysbiosis of the gut microbiota is associated with HIV disease pro- ulation in the human gut. Front. Microbiol. 7: 979. gression and tryptophan catabolism. Sci. Transl. Med. 5: 193ra91. 19. Ivanov, I. I., K. Atarashi, N. Manel, E. L. Brodie, T. Shima, U. Karaoz, D. Wei, 44. Yang, L., M. A. Poles, G. S. Fisch, Y. Ma, C. Nossa, J. A. Phelan, and Z. Pei. K. C. Goldfarb, C. A. Santee, S. V. Lynch, et al. 2009. Induction of intestinal 2016. HIV-induced immunosuppression is associated with colonization of the Th17 cells by segmented filamentous bacteria. Cell 139: 485–498. proximal gut by environmental bacteria. AIDS 30: 19–29. 20. Ivanov, I. I., Rde. L. Frutos, N. Manel, K. Yoshinaga, D. B. Rifkin, R. B. Sartor, 45. Dinh, D. M., G. E. Volpe, C. Duffalo, S. Bhalchandra, A. K. Tai, A. V. Kane, B. B. Finlay, and D. R. Littman. 2008. Specific microbiota direct the differenti- C. A. Wanke, and H. D. Ward. 2015. Intestinal microbiota, microbial translo- ation of IL-17-producing T-helper cells in the mucosa of the small intestine. Cell cation, and systemic inflammation in chronic HIV infection. J. Infect. Dis. 211: Host Microbe 4: 337–349. 19–27. http://www.jimmunol.org/ 21. Atarashi, K., T. Tanoue, T. Shima, A. Imaoka, T. Kuwahara, Y. Momose, 46. McHardy, I. H., X. Li, M. Tong, P. Ruegger, J. Jacobs, J. Borneman, P. Anton, G. Cheng, S. Yamasaki, T. Saito, Y. Ohba, et al. 2011. Induction of colonic and J. Braun. 2013. HIV Infection is associated with compositional and functional regulatory T cells by indigenous Clostridium species. Science 331: 337–341. shifts in the rectal mucosal microbiota. Microbiome 1: 26. 22. Round, J. L., S. M. Lee, J. Li, G. Tran, B. Jabri, T. A. Chatila, and 47. Nowak, P., M. Troseid, E. Avershina, B. Barqasho, U. Neogi, K. Holm, J. R. Hov, S. K. Mazmanian. 2011. The Toll-like receptor 2 pathway establishes colonization K. Noyan, J. Vesterbacka, J. Sva¨rd, et al. 2015. Gut microbiota diversity predicts by a commensal of the human microbiota. Science 332: 974–977. immune status in HIV-1 infection. AIDS 29: 2409–2418. 23. Chow, J., H. Tang, and S. K. Mazmanian. 2011. Pathobionts of the gastroin- 48. Noguera-Julian, M., M. Rocafort, Y. Guilleń, J. Rivera, M. Casadella`, P. Nowak, testinal microbiota and inflammatory disease. Curr. Opin. Immunol. 23: 473–480. F. Hildebrand, G. Zeller, M. Parera, R. Bellido, et al. 2016. Gut microbiota linked 24. Baothman, O. A., M. A. Zamzami, I. Taher, J. Abubaker, and M. Abu-Farha. to sexual preference and HIV infection. EBioMedicine 5: 135–146. 2016. The role of gut microbiota in the development of obesity and diabetes. 49. Monaco, C. L., D. B. Gootenberg, G. Zhao, S. A. Handley, M. S. Ghebremichael, Lipids Health Dis. 15: 108. E. S. Lim, A. Lankowski, M. T. Baldridge, C. B. Wilen, M. Flagg, et al. 2016. by guest on January 26, 2017 25. Claesson, M. J., I. B. Jeffery, S. Conde, S. E. Power, E. M. O’Connor, S. Cusack, Altered virome and bacterial microbiome in human immunodeficiency virus- H. M. Harris, M. Coakley, B. Lakshminarayanan, O. O’Sullivan, et al. 2012. Gut associated acquired immunodeficiency syndrome. Cell Host Microbe 19: 311–322. microbiota composition correlates with diet and health in the elderly. Nature 488: 50. Dillon, S. M., E. J. Lee, C. V. Kotter, G. L. Austin, S. Gianella, B. Siewe, 178–184. D. M. Smith, A. L. Landay, M. C. McManus, C. E. Robertson, et al. 2016. Gut 26. Ley, R. E., M. Hamady, C. Lozupone, P. J. Turnbaugh, R. R. Ramey, dendritic cell activation links an altered colonic microbiome to mucosal and sys- J. S. Bircher, M. L. Schlegel, T. A. Tucker, M. D. Schrenzel, R. Knight, and temic T-cell activation in untreated HIV-1 infection. Mucosal Immunol. 9: 24–37. J. I. Gordon. 2008. Evolution of mammals and their gut microbes. Science 320: 51. Chang, J. Y., D. A. Antonopoulos, A. Kalra, A. Tonelli, W. T. Khalife, 1647–1651. T. M. Schmidt, and V. B. Young. 2008. Decreased diversity of the fecal microbiome 27. De Filippo, C., D. Cavalieri, M. Di Paola, M. Ramazzotti, J. B. Poullet, in recurrent Clostridium difficile–associated diarrhea. J. Infect. Dis. 197: 435–438. S. Massart, S. Collini, G. Pieraccini, and P. Lionetti. 2010. Impact of diet in 52. Willing, B. P., J. Dicksved, J. Halfvarson, A. F. Andersson, M. Lucio, Z. Zheng, shaping gut microbiota revealed by a comparative study in children from Europe G. Jarnerot, C. Tysk, J. K. Jansson, and L. Engstrand. 2010. A pyrosequencing and rural Africa. Proc. Natl. Acad. Sci. USA 107: 14691–14696. study in twins shows that gastrointestinal microbial profiles vary with inflamma- 28. Yatsunenko, T., F. E. Rey, M. J. Manary, I. Trehan, M. G. Dominguez-Bello, tory bowel disease phenotypes. Gastroenterology 139: 1844–1854.e1. M. Contreras, M. Magris, G. Hidalgo, R. N. Baldassano, A. P. Anokhin, et al. 2012. 53. Yu, G., D. Fadrosh, B. Ma, J. Ravel, and J. J. Goedert. 2014. Anal microbiota Human gut microbiome viewed across age and geography. Nature 486: 222–227. profiles in HIV-positive and HIV-negative MSM. AIDS 28: 753–760. 29. Lax, S., D. P. Smith, J. Hampton-Marcell, S. M. Owens, K. M. Handley, 54. Thom, K., and J. Petrik. 2007. Progression towards AIDS leads to increased N. M. Scott, S. M. Gibbons, P. Larsen, B. D. Shogan, S. Weiss, et al. 2014. Torque teno virus and Torque teno minivirus titers in tissues of HIV infected Longitudinal analysis of microbial interaction between humans and the indoor individuals. J. Med. Virol. 79: 1–7. environment. Science 345: 1048–1052. 55. Li, L., X. Deng, A. C. Da Costa, R. Bruhn, S. G. Deeks, and E. Delwart. 2015. 30. Lozupone, C. A., J. I. Stombaugh, J. I. Gordon, J. K. Jansson, and R. Knight. Virome analysis of antiretroviral-treated HIV patients shows no correlation be- 2012. Diversity, stability and resilience of the human gut microbiota. Nature 489: tween T-cell activation and anelloviruses levels. J. Clin. Virol. 72: 106–113. 220–230. 56. Garcıá-A´lvarez, M., J. Berenguer, E. Alvarez, M. Guzmań-Fulgencio, J. Cosıń, 31. Moeller, A. H., M. Shilts, Y. Li, R. S. Rudicell, E. V. Lonsdorf, A. E. Pusey, P. Miralles, P. Catalań, J. C. Lo´pez, J. M. Rodrı´guez, D. Micheloud, et al. 2013. M. L. Wilson, B. H. Hahn, and H. Ochman. 2013. SIV-induced instability of the Association of torque teno virus (TTV) and torque teno mini virus (TTMV) with chimpanzee gut microbiome. Cell Host Microbe 14: 340–345. liver disease among patients coinfected with human immunodeficiency virus and 32. Moeller, A. H., M. Peeters, A. Ayouba, E. M. Ngole, A. Esteban, B. H. Hahn, and hepatitis C virus. Eur. J. Clin. Microbiol. Infect. Dis. 32: 289–297. H. Ochman. 2015. Stability of the gorilla microbiome despite simian immuno- 57. Guaraldi, G., G. Orlando, S. Zona, M. Menozzi, F. Carli, E. Garlassi, A. Berti, deficiency virus infection. Mol. Ecol. 24: 690–697. E. Rossi, A. Roverato, and F. Palella. 2011. Premature age-related comorbidities 33. Brenchley, J. M. 2013. Mucosal immunity in human and simian immunodefi- among HIV-infected persons compared with the general population. Clin. Infect. ciency lentivirus infections. Mucosal Immunol. 6: 657–665. Dis. 53: 1120–1126. 34. Handley, S. A., L. B. Thackray, G. Zhao, R. Presti, A. D. Miller, L. Droit, 58. Handley, S. A., C. Desai, G. Zhao, L. Droit, C. L. Monaco, A. C. Schroeder, P. Abbink, L. F. Maxfield, A. Kambal, E. Duan, et al. 2012. Pathogenic simian J. P. Nkolola, M. E. Norman, A. D. Miller, D. Wang, et al. 2016. SIV infection- immunodeficiency virus infection is associated with expansion of the enteric mediated changes in gastrointestinal bacterial microbiome and virome are associ- virome. Cell 151: 253–266. ated with immunodeficiency and prevented by accination. Cell Host Microbe 19: 35. McKenna, P., C. Hoffmann, N. Minkah, P. P. Aye, A. Lackner, Z. Liu, 323–335. C. A. Lozupone, M. Hamady, R. Knight, and F. D. Bushman. 2008. The macaque 59. Neuhaus, J., D. R. Jacobs, Jr., J. V. Baker, A. Calmy, D. Duprez, A. La Rosa, gut microbiome in health, lentiviral infection, and chronic enterocolitis. PLoS L. H. Kuller, S. L. Pett, M. Ristola, M. J. Ross, et al. 2010. Markers of inflam- Pathog. 4: e20. mation, coagulation, and renal function are elevated in adults with HIV infection. 36. Klase, Z., A. Ortiz, C. Deleage, J. C. Mudd, M. Quinõnes, E. Schwartzman, J. Infect. Dis. 201: 1788–1795. N. R. Klatt, L. Canary, J. D. Estes, and J. M. Brenchley. 2015. Dysbiotic bacteria 60. Deeks, S. G., C. M. Kitchen, L. Liu, H. Guo, R. Gascon, A. B. Narvaéz, P. Hunt, translocate in progressive SIV infection. Mucosal Immunol. 8: 1009–1020. J. N. Martin, J. O. Kahn, J. Levy, et al. 2004. Immune activation set point during The Journal of Immunology 613

early HIV infection predicts subsequent CD4+ T-cell changes independent of viral complexes feature late-treated individuals with suboptimal CD4+ T-cell recovery load. Blood 104: 942–947. upon virologically suppressive combination antiretroviral therapy. AIDS 30: 991– 61. Tincati, C., D. C. Douek, and G. Marchetti. 2016. Gut barrier structure, mucosal 1003. immunity and intestinal microbiota in the pathogenesis and treatment of HIV 83. Somsouk, M., J. D. Estes, C. Deleage, R. M. Dunham, R. Albright, infection. AIDS Res. Ther. 13: 19. J. M. Inadomi, J. N. Martin, S. G. Deeks, J. M. McCune, and P. W. Hunt. 2015. 62. Viskovic, K., S. Zidovec-Lepej, L. Gorenec, I. Grgic, D. Lukas, S. Zekan, Gut epithelial barrier and systemic inflammation during chronic HIV infection. A. Dragobratovic, and J. Begovac. 2014. Cardiovascular markers of inflammation AIDS 29: 43–51. and serum lipid levels in HIV-infected patients with undetectable viraemia. J. Int. 84. Lindemans, C. A., M. Calafiore, A. M. Mertelsmann, M. H. O’Connor, AIDS Soc. 17(4, Suppl. 3): 19548. J. A. Dudakov, R. R. Jenq, E. Velardi, L. F. Young, O. M. Smith, G. Lawrence, 63. Hunt, P. W., E. Sinclair, B. Rodriguez, C. Shive, B. Clagett, N. Funderburg, et al. 2015. Interleukin-22 promotes intestinal-stem-cell-mediated epithelial re- J. Robinson, Y. Huang, L. Epling, J. N. Martin, et al. 2014. Gut epithelial barrier generation. Nature 528: 560–564. dysfunction and innate immune activation predict mortality in treated HIV in- 85. Ko¨k, A., L. Hocqueloux, H. Hocini, M. Carrie`re, L. Lefrou, A. Guguin, fection. J. Infect. Dis. 210: 1228–1238. P. Tisserand, H. Bonnabau, V. Avettand-Fenoel, T. Prazuck, et al. 2015. Early 64.Krishnan,S.,E.M.Wilson,V.Sheikh,A.Rupert,D.Mendoza,J.Yang,R.Lempicki, initiation of combined antiretroviral therapy preserves immune function in the gut S. A. Migueles, and I. Sereti. 2014. Evidence for innate immune system activation in of HIV-infected patients. Mucosal Immunol. 8: 127–140. HIV type 1-infected elite controllers. J. Infect. Dis. 209: 931–939. 86. Hensley-McBain, T., A. S. Zevin, J. Manuzak, E. Smith, J. Gile, C. Miller, 65. Leoń, A., L. Leal, B. Torres, C. Lucero, A. Inciarte, M. Arnedo, M. Plana, J. Vila, B. Agricola, M. Katze, R. K. Reeves, C. S. Kraft, et al. 2016. Effects of fecal J. M. Gatell, and F. Garcıá. 2015. Association of microbial translocation bio- microbial transplantation on microbiome and immunity in simian immunodefi- markers with clinical outcome in controllers HIV-infected patients. AIDS 29: 675– ciency virus-infected macaques. J. Virol. 90: 4981–4989. 681. 87. Dillon, S. M., J. A. Manuzak, A. K. Leone, E. J. Lee, L. M. Rogers, 66. Kim, C. J., L. R. McKinnon, C. Kovacs, G. Kandel, S. Huibner, D. Chege, M. D. McCarter, and C. C. Wilson. 2012. HIV-1 infection of human intestinal K. Shahabi, E. Benko, M. Loutfy, M. Ostrowski, and R. Kaul. 2013. Mucosal lamina propria CD4+ T cells in vitro is enhanced by exposure to commensal Th17 cell function is altered during HIV infection and is an independent predictor Escherichia coli. J. Immunol. 189: 885–896. of systemic immune activation. J. Immunol. 191: 2164–2173. 88. Dillon, S. M., E. J. Lee, A. M. Donovan, K. Guo, M. S. Harper, D. N. Frank, 67. Ciccone, E. J., J. H. Greenwald, P. I. Lee, A. Biancotto, S. W. Read, M. A. Yao, M. D. McCarter, M. L. Santiago, and C. C. Wilson. 2016. Enhancement of HIV-1 + + J. N. Hodge, W. L. Thompson, S. B. Kovacs, C. L. Chairez, et al. 2011. CD4 infection and intestinal CD4 T cell depletion ex vivo by gut microbes altered Downloaded from T cells, including Th17 and cycling subsets, are intact in the gut mucosa of HIV-1- during chronic HIV-1 infection. Retrovirology 13: 5. infected long-term nonprogressors. J. Virol. 85: 5880–5888. 89. Al-Sadi, R., D. Ye, H. M. Said, and T. Y. Ma. 2011. Cellular and molecular 68. Sandler, N. G., H. Wand, A. Roque, M. Law, M. C. Nason, D. E. Nixon, mechanism of interleukin-1b modulation of Caco-2 intestinal epithelial tight C. Pedersen, K. Ruxrungtham, S. R. Lewin, S. Emery, et al; INSIGHT SMART junction barrier. J. Cell. Mol. Med. 15: 970–982. Study Group. 2011. Plasma levels of soluble CD14 independently predict mor- 90. Klatt, N. R., L. A. Canary, X. Sun, C. L. Vinton, N. T. Funderburg, tality in HIV infection. J. Infect. Dis. 203: 780–790. D. R. Morcock, M. Quinõnes, C. B. Deming, M. Perkins, D. J. Hazuda, et al. 69. Timmons, T., C. Shen, G. Aldrovandi, A. Rollie, S. K. Gupta, J. H. Stein, and 2013. Probiotic/prebiotic supplementation of antiretrovirals improves gastroin- M. P. Dube´. 2014. Microbial translocation and metabolic and body composition testinal immunity in SIV-infected macaques. J. Clin. Invest. 123: 903–907. measures in treated and untreated HIV infection. AIDS Res. Hum. Retroviruses 30: 91. Ortiz, A. M., Z. A. Klase, S. R. DiNapoli, I. Vujkovic-Cvijin, K. Carmack, http://www.jimmunol.org/ 272–277. M. R. Perkins, N. Calantone, C. L. Vinton, N. E. Riddick, J. Gallagher, et al. 70. Tenorio, A. R., Y. Zheng, R. J. Bosch, S. Krishnan, B. Rodriguez, P. W. Hunt, 2016. IL-21 and probiotic therapy improve Th17 frequencies, microbial translo- J. Plants, A. Seth, C. C. Wilson, S. G. Deeks, et al. 2014. Soluble markers of cation, and microbiome in ARV-treated, SIV-infected macaques. Mucosal Immu- inflammation and coagulation but not T-cell activation predict non-AIDS-defining nol. 9: 458–467. morbid events during suppressive antiretroviral treatment. J. Infect. Dis. 210: 92. Alcaide, M. L., A. Parmigiani, S. Pallikkuth, M. Roach, R. Freguja, M. Della 1248–1259. Negra, H. Bolivar, M. A. Fischl, and S. Pahwa. 2013. Immune activation in HIV- 71. Deeks, S. G., R. Tracy, and D. C. Douek. 2013. Systemic effects of inflammation infected aging women on antiretrovirals—implications for age-associated comor- on health during chronic HIV infection. Immunity 39: 633–645. bidities: a cross-sectional pilot study. PLoS One 8: e63804. 72. Lapenta, C., M. Boirivant, M. Marini, S. M. Santini, M. Logozzi, M. Viora, 93. Helleberg, M., G. Kronborg, C. S. Larsen, G. Pedersen, C. Pedersen, N. Obel, and F. Belardelli, and S. Fais. 1999. Human intestinal lamina propria lymphocytes are J. Gerstoft. 2013. CD4 decline is associated with increased risk of cardiovascular naturally permissive to HIV-1 infection. Eur. J. Immunol. 29: 1202–1208. disease, cancer, and death in virally suppressed patients with HIV. Clin. Infect. Dis. by guest on January 26, 2017 73. Guadalupe, M., E. Reay, S. Sankaran, T. Prindiville, J. Flamm, A. McNeil, and 57: 314–321. S. Dandekar. 2003. Severe CD4+ T-cell depletion in gut lymphoid tissue during 94. Rodger, A. J., R. Lodwick, M. Schechter, S. Deeks, J. Amin, R. Gilson, R. Paredes, primary human immunodeficiency virus type 1 infection and substantial delay in E. Bakowska, F. N. Engsig, A. Phillips, and E. S. G. Insight Smart, INSIGHT restoration following highly active antiretroviral therapy. J. Virol. 77: 11708– SMART, ESPRIT Study Groups. 2013. Mortality in well controlled HIV in the 11717. continuous antiretroviral therapy arms of the SMART and ESPRIT trials com- 74. Kewenig, S., T. Schneider, K. Hohloch, K. Lampe-Dreyer, R. Ullrich, N. Stolte, pared with the general population. AIDS 27: 973–979. C. Stahl-Hennig, F. J. Kaup, A. Stallmach, and M. Zeitz. 1999. Rapid mucosal 95. Takuva, S., M. Maskew, A. T. Brennan, L. Long, I. Sanne, and M. P. Fox. 2014. CD4+ T-cell depletion and enteropathy in simian immunodeficiency virus-infected Poor CD4 recovery and risk of subsequent progression to AIDS or death despite rhesus macaques. Gastroenterology 116: 1115–1123. viral suppression in a South African cohort. J. Int. AIDS Soc. 17: 18651. 75. Brenchley, J. M., T. W. Schacker, L. E. Ruff, D. A. Price, J. H. Taylor, 96. van Lelyveld, S. F., L. Gras, A. Kesselring, S. Zhang, F. De Wolf, A. M. Wensing, G. J. Beilman, P. L. Nguyen, A. Khoruts, M. Larson, A. T. Haase, and and A. I. Hoepelman, ATHENA national observational cohort study. 2012. Long- D. C. Douek. 2004. CD4+ T cell depletion during all stages of HIV disease occurs term complications in patients with poor immunological recovery despite viro- predominantly in the gastrointestinal tract. J. Exp. Med. 200: 749–759. logical successful HAART in Dutch ATHENA cohort. AIDS 26: 465–474. 76. Steele, A. K., E. J. Lee, J. A. Manuzak, S. M. Dillon, J. D. Beckham, 97. Merlini, E., F. Bai, G. M. Bellistrı`, C. Tincati, A. d’Arminio Monforte, and M. D. McCarter, M. L. Santiago, and C. C. Wilson. 2014. Microbial exposure G. Marchetti. 2011. Evidence for polymicrobic flora translocating in peripheral alters HIV-1-induced mucosal CD4+ T cell death pathways ex vivo. Retrovirology blood of HIV-infected patients with poor immune response to antiretroviral 11: 14. therapy. PLoS One 6: e18580. 77. Doitsh, G., N. L. Galloway, X. Geng, Z. Yang, K. M. Monroe, O. Zepeda, 98. Pe´rez-Santiago, J., S. Gianella, M. Massanella, C. A. Spina, M. Y. Karris, P. W. Hunt, H. Hatano, S. Sowinski, I. Munõz-Arias, and W. C. Greene. 2014. S. R. Var, D. Patel, P. S. Jordan, J. A. Young, S. J. Little, et al. 2013. Gut Lac- Cell death by pyroptosis drives CD4 T-cell depletion in HIV-1 infection. Nature tobacillales are associated with higher CD4 and less microbial translocation during 505: 509–514. HIV infection. AIDS 27: 1921–1931. 78. Mehandru, S., M. A. Poles, K. Tenner-Racz, P. Jean-Pierre, V. Manuelli, P. Lopez, 99. d’Ettorre, G., G. Ceccarelli, N. Giustini, S. Serafino, N. Calantone, G. De A. Shet, A. Low, H. Mohri, D. Boden, et al. 2006. Lack of mucosal immune Girolamo, L. Bianchi, V. Bellelli, T. Ascoli-Bartoli, S. Marcellini, et al. 2015. reconstitution during prolonged treatment of acute and early HIV-1 infection. Probiotics reduce inflammation in antiretroviral treated, HIV-infected individuals: [Published erratum appears in 2006 PLoS Med. 3: e546.] PLoS Med. 3: e484. results of the “Probio-HIV” clinical trial. PLoS One 10: e0137200. 79. Serrano-Villar, S., T. Sainz, Z. M. Ma, N. S. Utay, T. W. Chun, S. Mann, 100. Hummelen, R., J. Changalucha, N. L. Butamanya, T. E. Koyama, A. Cook, A. D. Kashuba, B. Siewe, A. Albanese, P. Troia-Cancio, et al. 2016. Effects of J. D. Habbema, and G. Reid. 2011. Effect of 25 weeks probiotic supplementation combined CCR5/integrase inhibitors-based regimen on mucosal immunity in on immune function of HIV patients. Gut Microbes 2: 80–85. HIV-infected patients naı¨ve to antiretroviral therapy: a pilot randomized trial. 101. Irvine, S. L., R. Hummelen, S. Hekmat, C. W. Looman, J. D. Habbema, and [Published erratum appears in 2016 PLoS Pathog. 12: e1005540] PLoS Pathog. 12: G. Reid. 2010. Probiotic yogurt consumption is associated with an increase of CD4 e1005381. count among people living with HIV/AIDS. J. Clin. Gastroenterol. 44: e201–e205. 80. Fernandes, S. M., A. R. Pires, C. Ferreira, R. B. Foxall, J. Rino, C. Santos, 102. Ellis, C. L., Z. M. Ma, S. K. Mann, C. S. Li, J. Wu, T. H. Knight, T. Yotter, L. Correia, J. Poças, H. Veiga-Fernandes, and A. E. Sousa. 2014. Enteric mucosa T. L. Hayes, A. H. Maniar, P. V. Troia-Cancio, et al. 2011. Molecular charac- integrity in the presence of a preserved innate interleukin 22 compartment in HIV terization of stool microbiota in HIV-infected subjects by panbacterial and order- type 1-treated individuals. J. Infect. Dis. 210: 630–640. level 16S ribosomal DNA (rDNA) quantification and correlations with immune 81. Kim, C. J., A. Nazli, O. L. Rojas, D. Chege, Z. Alidina, S. Huibner, S. Mujib, activation. J. Acquir. Immune Defic. Syndr. 57: 363–370. E. Benko, C. Kovacs, L. Y. Shin, et al. 2012. A role for mucosal IL-22 production 103. Paquin-Proulx, D., C. Ching, I. Vujkovic-Cvijin, D. Fadrosh, L. Loh, Y. Huang, and Th22 cells in HIV-associated mucosal immunopathogenesis. Mucosal Immu- M. Somsouk, S. V. Lynch, P. W. Hunt, D. F. Nixon, and D. SenGupta. 2016. nol. 5: 670–680. Bacteroides are associated with GALT iNKT cell function and reduction of mi- 82. Tincati, C., E. Merlini, P. Braidotti, G. Ancona, F. Savi, D. Tosi, E. Borghi, crobial translocation in HIV-1 infection. Mucosal Immunol. DOI: 10.1038/ M. L. Callegari, B. Mangiavillano, A. Barassi, et al. 2016. Impaired gut junctional mi.2016.34. 614 BRIEF REVIEWS: HIV AND THE GUT MICROBIOME

104. Li, S. K., R. K. Leung, H. X. Guo, J. F. Wei, J. H. Wang, K. T. Kwong, S. S. Lee, association with coronary stenosis and HIV serostatus. J. Acquir. Immune Defic. C. Zhang, and S. K. Tsui. 2012. Detection and identification of plasma bacterial Syndr. 72: 114–118. and viral elements in HIV/AIDS patients in comparison to healthy adults. Clin. 116. Srinivasa, S., K. V. Fitch, J. Lo, H. Kadar, R. Knight, K. Wong, S. Abbara, Microbiol. Infect. 18: 1126–1133. D. Gauguier, J. Capeau, F. Boccara, and S. K. Grinspoon. 2015. Plaque burden in 105. Bipath, P., P. F. Levay, and M. Viljoen. 2015. The kynurenine pathway activities HIV-infected patients is associated with serum intestinal microbiota-generated in a sub-Saharan HIV/AIDS population. BMC Infect. Dis. 15: 346. trimethylamine. AIDS 29: 443–452. 106. Byakwaga, H., Y. Boum, II., Y. Huang, C. Muzoora, A. Kembabazi, S. D. Weiser, 117. Mirande, C., E. Kadlecikova, M. Matulova, P. Capek, A. Bernalier-Donadille, J. Bennett, H. Cao, J. E. Haberer, S. G. Deeks, et al. 2014. The kynurenine E. Forano, and C. Be´ra-Maillet. 2010. Dietary fibre degradation and fermentation pathway of tryptophan catabolism, CD4+ T-cell recovery, and mortality among by two xylanolytic bacteria Bacteroides xylanisolvens XB1A and Roseburia intestinalis HIV-infected Ugandans initiating antiretroviral therapy. J. Infect. Dis. 210: 383– XB6B4 from the human intestine. J. Appl. Microbiol. 109: 451–460. 391. 118. Wa¨chtershaüser, A., and J. Stein. 2000. Rationale for the luminal provision of 107. Gaardbo, J. C., M. Trøsied, B. Stiksrud, Ø. Midttun, P. M. Ueland, H. Ullum, butyrate in intestinal diseases. Eur. J. Nutr. 39: 164–171. and S. D. Nielsen. 2015. Increased tryptophan catabolism is associated with in- 119. Xiong, H., B. Guo, Z. Gan, D. Song, Z. Lu, H. Yi, Y. Wu, Y. Wang, and H. Du. creased frequency of CD161+ Tc17/MAIT cells and lower CD4+ T-cell count in 2016. Butyrate upregulates endogenous host defense peptides to enhance disease HIV-1 infected patients on cART after 2 years of follow-up. J. Acquir. Immune resistance in piglets via histone deacetylase inhibition. Sci. Rep. 6: 27070. Defic. Syndr. 70: 228–235. 120. Furusawa, Y., Y. Obata, S. Fukuda, T. A. Endo, G. Nakato, D. Takahashi, 108. Serrano-Villar, S., D. Rojo, M. Martıńez-Martıńez, S. Deusch, J. F. Va´zquez- Y. Nakanishi, C. Uetake, K. Kato, T. Kato, et al. 2013. Commensal microbe- Castellanos, T. Sainz, M. Vera, S. Moreno, V. Estrada, M. J. Gosalbes, et al. 2016. derived butyrate induces the differentiation of colonic regulatory T cells. Nature HIV infection results in metabolic alterations in the gut microbiota different from 504: 446–450. those induced by other diseases. Sci. Rep. 6: 26192. 121. Smith, P. M., M. R. Howitt, N. Panikov, M. Michaud, C. A. Gallini, 109. Jenabian, M. A., M. El-Far, K. Vyboh, I. Kema, C. T. Costiniuk, R. Thomas, M. Bohlooly-Y, J. N. Glickman, and W. S. Garrett. 2013. The microbial me- J. G. Baril, R. LeBlanc, C. Kanagaratham, D. Radzioch, et al; Montreal Primary tabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis. Science Infection and Slow Progressor Study Groups. 2015. Immunosuppressive trypto- 341: 569–573. phan catabolism and gut mucosal dysfunction following early HIV infection. 122. Ye, F., and J. Karn. 2015. Bacterial short chain fatty acids push all the buttons J. Infect. Dis. 212: 355–366. needed to reactivate latent viruses. Stem Cell Epigenet. 2: e532.

110. Vujkovic-Cvijin, I., L. A. Swainson, S. N. Chu, A. M. Ortiz, C. A. Santee, 123. Allen, J. M., M. E. Berg Miller, B. D. Pence, K. Whitlock, V. Nehra, Downloaded from A. Petriello, R. M. Dunham, D. W. Fadrosh, D. L. Lin, A. A. Faruqi, et al. 2015. H. R. Gaskins, B. A. White, J. D. Fryer, and J. A. Woods. 2015. Voluntary and Gut-resident Lactobacillus abundance associates with IDO1 inhibition and Th17 forced exercise differentially alters the gut microbiome in C57BL/6J mice. J. Appl. dynamics in SIV-infected macaques. Cell Rep. 13: 1589–1597. Physiol. 118: 1059–1066. 111. Tang, W. H., Z. Wang, D. J. Kennedy, Y. Wu, J. A. Buffa, B. Agatisa-Boyle, 124. University of California, San Francisco. Fecal Microbiota Transplantation in HIV X. S. Li, B. S. Levison, and S. L. Hazen. 2015. Gut microbiota-dependent tri- (FMT-HIV). In: ClinicalTrials.gov [Internet]. Bethesda (MD): National Library methylamine N-oxide (TMAO) pathway contributes to both development of renal of Medicine (US). 2000- [cited 2016 September 19]. Available from: https:// insufﬁciency and mortality risk in chronic kidney disease. Circ. Res. 116: 448–455. clinicaltrials.gov/ct2/show/NCT02256592 NLM Identiﬁer: NCT02256592. 112. Wang, Z., E. Klipfell, B. J. Bennett, R. Koeth, B. S. Levison, B. Dugar, 125. Stadlbauer, V. 2015. Immunosuppression and probiotics: are they effective and

A. E. Feldstein, E. B. Britt, X. Fu, Y. M. Chung, et al. 2011. Gut flora metabolism safe? Benef. Microbes 6: 823–828. http://www.jimmunol.org/ of phosphatidylcholine promotes cardiovascular disease. Nature 472: 57–63. 126. Gonza´lez-Hernańdez, L. A., L. F. Jave-Suarez, M. Fafutis-Morris, K. E. Montes- 113. Haissman, J. M., A. Knudsen, H. Hoel, A. Kjær, U. S. Kristoffersen, R. K. Berge, Salcedo, L. G. Valle-Gutierrez, A. E. Campos-Loza, L. F. Enciso-Go´mez, and T. L. Katzenstein, A. Svardal, T. Ueland, P. Aukrust, et al. 2016. Microbiota- J. F. Andrade-Villanueva. 2012. Synbiotic therapy decreases microbial transloca- dependent marker TMAO is elevated in silent ischemia but is not associated with tion and inflammation and improves immunological status in HIV-infected pa- first-time myocardial infarction in HIV infection. J. Acquir. Immune Defic. Syndr. tients: a double-blind randomized controlled pilot trial. Nutr. J. 11: 90. 71: 130–136. 127. Gautam, N., R. Dayal, D. Agarwal, R. Kumar, T. P. Singh, T. Hussain, and 114. Knudsen, A., T. E. Christensen, K. Thorsteinsson, A. A. Ghotbi, P. Hasbak, S. P. Singh. 2014. Role of multivitamins, micronutrients and probiotics supple- A. M. Lebech, S. D. Nielsen, J. R. Hov, R. Berge, R. S. Ripa, et al. 2016. mentation in management of HIV infected children. Indian J. Pediatr. 81: 1315– Microbiota-dependent marker TMAO is not associated with decreased myocardial 1320. perfusion in well-treated HIV-infected patients as assessed by 82Rubidium PET/ 128. Kim, C. J., S. L. Walmsley, J. M. Raboud, C. Kovacs, B. Coburn, R. Rousseau, CT. J. Acquir. Immune Defic. Syndr. 72: e83–e85. R. Reinhard, R. Rosenes, and R. Kaul. 2016. Can probiotics reduce inflammation 115. Miller, P. E., S. A. Haberlen, T. T. Brown, J. B. Margolick, J. A. DiDonato, and enhance gut immune health in people living with HIV: study designs for the by guest on January 26, 2017 S. L. Hazen, M. D. Witt, L. A. Kingsley, F. J. Palella, Jr., M. Budoff, et al. 2016. probiotic visbiome for inflammation and translocation (PROOV IT) pilot trials. Brief report: intestinal microbiota-produced trimethylamine-N-oxide and its HIV Clin. Trials 17: 147–157.