Old Herborn University Monograph 27

Old Herborn University Seminar Monograph

27. PERSISTING CONSEQUENCES OF INTESTINAL INFECTION

EDITORS: PETER J. HEIDT DENNIS LANG MARK S. RIDDLE RICHARD WALKER VOLKER RUSCH

Old Herborn University Seminar Monograph 27

ISBN 3-923022-39-5 ISSN 1431-6579

COPYRIGHT © 2014 BY THE OLD HERBORN UNIVERSITY FOUNDATION. ALL RIGHTS RESERVED NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL, INCLUDING PHOTO- COPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER

EDITORS:

Prof. Dr. med. Peter J. Heidt, Richard Walker, PhD Foundation for Microbiology and Gnotobiology Enteric Vaccine Initiative Vlietweg 60A Vaccine Development Global Program PATH 2323 LE - Leiden 455 Massachusetts Ave NW, Suite 1000 The Netherlands Washington, DC 20001 USA

Dennis Lang, PhD Dr. rer. nat. Volker Rusch, Fogarty International Center Stiftung Old Herborn University National Institutes of Health Postfach 1765 Bethesda, MD 20814 D-35727 Herborn-Dill USA Germany

Mark S. Riddle, MD, PhD Enteric Diseases Department Naval Medical Research Center 503 Robert Grant Ave Silver Spring, MD 20910 USA

Old Herborn University Foundation Postfach 1765 D-35727 Herborn-Dill Germany Telephone: +49 - 2772 - 921100 Telefax: +49 - 2772 - 921101 Contents ———————————————————————————————————————

Participating authors V

I. EPITHELIAL STEM CELLS, SELF-RENEWAL AND DIFFERENTIATION IN THE INTESTINE (Laurens G. van der Flier and Hugo J. Snippert) 1 Summary ...... 1 Introduction ...... 1 Proliferating intestinal cells ...... 2 Wnt pathway...... 3 Intestinal stem cells ...... 4 Lgr5 is an intestinal stem cell marker ...... 5 Additional intestinal stem cell markers ...... 5 Neutral drift competition between multiple intestinal stem cells ...... 6 Intestinal organoids ...... 6 Intestinal stem cell niche ...... 7 Concluding remarks ...... 8 Literature ...... 8

II. IRRITABLE BOWEL SYNDROME: ROLE OF GUT BACTERIA AND BACTERIAL TOXINS (Christopher Chang, Gillian Barlow, Stacy Weitsman and Mark Pimentel) 13 Summary ...... 13 Introduction ...... 13 SIBO and IBS...... 14 Small bowel microbiome is altered in IBS...... 14 Treating SIBO in IBS ...... 15 Post-infectious IBS...... 15 Rat model of PI-IBS ...... 16 Antibiotic prophylaxis prevents the development of IBS-like characteristics in a rat model of C. jejuni infection ...... 16 Cytolethal distending toxin (Cdt) is required for expression of an IBS-like phenotype in the rat model ...... 17 Potential role of immunity in the development of PI-IBS ...... 17 Perspectives and future directions ...... 18 Literature ...... 19

I Contents (continued) ———————————————————————————————————————

III. FREQUENT SYMPTOMATIC OR ASYMPTOMATIC INFECTIONS MAY HAVE LONG-TERM CONSEQUENCES ON GROWTH AND COGNITIVE DEVELOPMENT (Benjamin J.J. McCormick on behalf of the MAL-ED Network Investigators) 23 The importance of diarrhoea ...... 23 Mortality from diarrhoea and entero-pathogens ...... 24 Morbidity from diarrhoea ...... 25 The MAL-ED Network ...... 26 Symptomatic infection ...... 26 Infection and growth ...... 28 Infection and cognition ...... 30 The legacy of diarrhoea ...... 31 Conclusions ...... 33 Acknowledgements ...... 34 Literature ...... 34

IV. UTILIZATION OF THE US DEPARTMENT OF DEFENSE MEDICAL SURVEILLANCE SYSTEM AND SERUM REPOSITORY FOR ASSESSING LONG-TERM HEALTH CONSEQUENCES OF ACUTE ENTERIC INFECTION (Chad K. Porter, Ashley N. Alcala, and Mark S. Riddle) 41 Introduction to DODSR and DMSS ...... 41 Population demographics ...... 46 Access to care ...... 47 Exposures of interest ...... 48 Outcomes of interest ...... 51 Limitations ...... 51 Utilization of serum repository ...... 54 Other large epidemiologic databases ...... 55 Conclusion ...... 56 Literature ...... 57

Contents (continued) ———————————————————————————————————————

V. AN ENTERIC TRIANGLE: PROTOZOAN INFECTIONS, LINKS TO ENVIRONMENTAL ENTEROPATHY AND THE POTENTIAL INFLUENCE OF THE INTESTINAL BACTERIAL MICROBIOME ON THIS INTERACTION AND OVERALL HEALTH IN THE DEVELOPING WORLD (Stacey L. Burgess and William A. Petri) 61 Summary...... 61 Enteric infections and environmental enteropathy ...... 61 Influence of commensal bacteria on man and mice ...... 63 Interactions between bacteria and protozoa ...... 65 Enteric infections, health and exploration of the microbiome in the developing world ...... 65 Final thoughts ...... 67 Literature ...... 67

VI. COELIAC DISEASE (Joseph A. Murray and Eric V. Marietta) 71 Summary ...... 71 Introduction ...... 71 Aetiology ...... 73 Immunopathogenesis ...... 77 Diagnosis ...... 78 Treatment ...... 79 Prognosis ...... 79 Future treatments ...... 80 Prevention ...... 80 Summary ...... 81 Literature ...... 81

VII. ACUTE ENTERIC INFECTIONS ALTER COMMENSAL MICROBIOTA: NEW MECHANISMS IN POST-INFECTIOUS INTESTINAL INFLAMMATORY DISORDERS (Andre G. Buret and Kristen Reti) 87 Summary ...... 87 Introduction ...... 87 Biology of the intestinal microbiota ...... 90 Campylobacter jejuni ...... 93 Giardia intestinalis ...... 96 Conclusions ...... 99 Acknowledgments ...... 99 Literature ...... 100

III

Contents (continued) ———————————————————————————————————————

VIII. THE ROLES OF PATHOGENIC AND COMMENSAL BACTERIA IN THE INTESTINAL AND BEHAVIOURAL MANIFESTATIONS OF FUNCTIONAL GI DISORDERS (Stephen M. Collins) 107 Summary ...... 107 Functional GI disorders ...... 107 Acknowledgements ...... 109 Literature ...... 110

IX. WHY BACTERIA MATTER: INSIGHTS FROM THE HYDRA HOLOBIONT (Thomas C.G. Bosch and Sebastian Fraune) 113 Summary ...... 113 From Karl August Möbius to the hologenome theory of evolution ...... 113 The Hydra holobiont ...... 115 The host actively shapes the colonizing microbiota ...... 116 How does the host control the microbiota in the context of specific developmental or environmental conditions? ...... 117 Antimicrobial peptides - key factors for host-bacteria co-evolution ...... 118 What are the microbes for? ...... 118 The holobiont in a constantly changing environment ...... 119 Concluding remarks ...... 120 Acknowledgements ...... 121 Literature ...... 121

X. PERSISTING CONSEQUENCE OF INTESTINAL INFECTION: SUMMARY OF THE SEMINAR (Mark Riddle and Richard Walker) 125 Introduction ...... 125 Epidemiology: defining the problem ...... 126 Post-infectious consequences: a developed world perspective ...... 127 Post-infectious consequences: a developing world perspective ...... 128 Pathogenic mechanisms: a growing framework of complexity ...... 130 The future: understanding complexity and exploring new frontiers ...... 132 Concepts and questions for further consideration ...... 134 Literature ...... 134

Participating Authors ———————————————————————————————————————

Thomas C.G. Bosch, Dr., Dr. h.c., Zoological Institute, Christian Albrechts University Kiel, Am Botanischen Garten 1-9, D-24118 Kiel, Germany. (E-mail: [email protected]).

Andre G. Buret, PhD, Department of Biological Sciences, University of Calgary, 2500 University Dr. N.W., Calgary (AB) T2N 1N4, Canada. (E-mail: [email protected]).

Christopher Chang, MD, PhD, GI Motility Program, Cedars-Sinai Medical Center, 8730 Alden Drive, Suite 225E, Los Angeles, California 90048, USA. (Email: [email protected]).

Stephen M. Collins, MBBS, FRCPC, GSK Chair in GI Research, Farncombe Family Digestive Health Research Institute, Faculty of Health Sciences, McMaster University, Room 2E16, Faculty of Health Sciences, McMaster University, 1200 Main Street West, Hamilton (ON) L8N 3Z5, Canada. (E-mail: [email protected]).

Benjamin J.J. McCormick, PhD, Fogarty International Center, National Institutes of Health, Building 16, Center Drive, Bethesda, MD, USA. (E-mail: [email protected]).

Joseph A. Murray, MD, Division of Gastroenterology and Hepatology, and Department of Immunology, Mayo Clinic, 200 1st St. S.W., Rochester, MN 55905, USA. (Email: [email protected]).

William A. Petri, Jr., MD, PhD, University of Virginia School of Medicine, P.O. Box 801340, Room 1709, Carter Harrison Research Building, Charlottesville, VA 22908-1340, USA. (Email: [email protected]).

Chad K. Porter, PhD, MPH, Enteric Diseases Department, Infectious Disease Directorate, Naval Medical Research Center, 503 Robert Grant Avenue, Silver Spring, Maryland 20910-7500, USA. (Email: [email protected]).

Mark S. Riddle, MD, PhD, Naval Medical Research Center, 503 Robert Grant Ave, Silver Spring, MD 20910, USA. (E-mail: [email protected]).

Laurens G. van der Flier, PhD, Hubrecht Institute, Royal Netherlands Academy of Arts and Sciences, University Medical Center Utrecht, Uppsalalaan 8, 3584 CT Utrecht, the Netherlands. (E-mail: [email protected]).

Richard Walker, PhD, Enteric Vaccine Initiative, Vaccine Development Global Program PATH, 455 Massachusetts Ave NW, Suite 1000, Washington, DC 20001, USA. (E-mail: [email protected]).

EPITHELIAL STEM CELLS, SELF-RENEWAL AND DIFFERENTIATION IN THE INTESTINE

LAURENS G. VAN DER FLIER1,2 and HUGO J. SNIPPERT1,3

1Hubrecht Institute, Royal Netherlands Academy of Arts and Sciences, University Medical Center Utrecht, Utrecht, the Netherlands; 2 Current address: SomantiX B.V., Utrecht, the Netherlands; 3 Current address: University Medical Center Utrecht, Utrecht, the Netherlands.

SUMMARY

The inner lining of the gut is a simple epithelium that completely self- renews every five days because of the high mechanical and chemical stress. At the basis of the epithelial homeostasis are intestinal stem cells that are located at the bottom of crypts. Progeny is generated on a daily basis to compensate for the loss of differentiated cells that cover the villi. Under normal conditions, active cycling stem cells produce daughter cells that compete with each other for residency in the stem cell niche. Upon niche displacement, a daughter cell will lose stem cell characteristics, migrates out of the crypts and matures into a terminally specialized differentiated cell type. Here, we describe and summarize the recent developments in the identification and characterization of intestinal stem cells.

INTRODUCTION

The primary function of the intestinal The majority of cells that are produced tract is the digestion and absorption of on a daily basis migrate out of the food. The gut is anatomically divided crypts and terminally differentiate into into the small intestine and the colon. one of the major three specialized cell The inner wall of the gut is covered types of the intestinal epithelium. with a simple columnar epithelium, These are the absorptive enterocytes, which performs the primary functions mucous-secreting goblet cells and hor- of (i) digestion via the secretion of en- mone-secreting entero-endocrine cells. zymes, (ii) water and nutrient absorp- Approximately three days after their tion and (iii) forms a barrier against gut terminal differentiation, the cells reach pathogens. In the small intestine, the the tip of the villus, undergo spontane- surface area is enlarged through epithe- ous apoptosis and are shed into the lu- lial protrusions called villi (Figure 1), men of the gut. Paneth cells are the while the colon has a flat surface fourth abundant cell type in the intes- epithelium. Proliferative cells reside in tine and the only differentiated cell the crypts of Lieberkühn, epithelial in- type that escapes the upward migration vaginations into the underlying connec- and instead settle at crypt bottoms (van tive tissue. These crypts harbour stem der Flier and Clevers, 2009). Paneth cells and transit amplifying cells that cells have a function in innate immun- are direct descendants of the stem cells. ity and antibacterial defence (Clevers

Figure 1: The intestine is lined with a single layer of epithelial cells. The H&E staining is showing the organization of the mouse small intestine in protruding villi and adjacent crypts. Crypts contain stem cells, progenitor cells and Paneth cells while the villus is covered with goblet cells, entero- endocrine cells and enterocytes.

and Bevins, 2013). Recently, it was role in constituting the intestinal stem shown that they also play an important cell niche (Sato et al., 2011b).

PROLIFERATING INTESTINAL CELLS

The intestinal epithelium is the most nal epithelium. However, the exact vigorously self-renewing tissue present location of these cells was unknown for in adult mammals. With the exception a long time since molecular markers of stem cells and Paneth cells, the inner were missing (Barker et al., 2008). In lining of the gut is completely renewed principal, there are only two functional every 5 days. It has been known for requirements that cells must fulfil in decades that multipotent stem cells fuel order to define them as stem cells, the proliferative activity of the intesti- commonly referred to as stemness.

Figure 2: Left panel; schematic representation of small intestinal crypt. Wnt signalling is the driving force behind proliferation and stem cell identity. Right panel; immunohistochemical staining for transcription factor Ascl2 in mouse small intestinal crypts. Ascl2 is specifically expressed in CBC cells (arrows).

First, stem cells are long-lived, prefera- cells are able to differentiate into all the bly throughout the life of an individual. specialized cell type(s) that are present Secondly, stem cells are mulipotent, in a tissue. which means that descendants of stem

WNT PATHWAY

The primary force that drives intestinal et al., 1997). Because of this intimate epithelial proliferation is the Wnt connection between Wnt signalling and signalling pathway. Mice that are mu- intestinal biology, we and others have tant for the intestine-specific Tcf4 tran- attempted to unravel the Wnt/Tcf4 tar- scription factor fail to establish get gene program activated in intestinal proliferative crypts (Korinek et al., crypts and colorectal tumours (van de 1998; van Es et al., 2012a), while Wetering et al., 2002; van der Flier et conditional deletion of ß-catenin (Fevr al., 2007; van Es et al., 2005). Wnt tar- et al., 2007; Ireland et al., 2004) as get gene expression has been identified well as transgenic expressing of the se- in three different crypt compartments. creted Wnt inhibitor Dickkopf-1 Most of the Wnt targets are expressed (Kuhnert et al., 2004; Pinto et al., by the rapidly dividing progenitor cells 2003) leads to disappearance of within the transit-amplifying compart- proliferative crypts in adult mice. ment that consist approximately the top Moreover, malignant transformation of two-third of the crypt. A second group intestinal epithelium is almost invaria- of genes is expressed in post-mitotic bly initiated by activating Wnt pathway Paneth cells that are located at the en- mutations (Korinek et al., 1997; Morin tire bottom of the crypt. The third

3 group contains only a very limited that are squeezed in between the Paneth number of genes (including Lgr5, cells at crypt bottoms. Already in the Ascl2, Rnf43 and EphB2) that are early ’70, they have been suggested, highly expressed in crypt base colum- based on their morphology and loca- nar (CBC) cells (Figure 2). CBC cells tion, to be intestinal stem cells (Cheng are small, undifferentiated cycling cells and Leblond, 1974).

INTESTINAL STEM CELLS

The exact location and identity of cells repopulate the entire blood. A intestinal stem cells has been the sub- critical note by this kind of assay is that ject of an intense scientific debate for the ability of the cell population to years. In addition to the CBC cell as function as stem cells is tested in a candidate intestinal stem cell, another challenged/stressed system. Therefore, popular postulation was that the cells cell transplantations between individu- that were physically located on top of als or species primarily test the ability the Paneth cell compartment, the so- of cells to adapt and grow in a foreign called +4 location (cell position milieu, rather than testing the actual counted from the crypt bottom), func- behaviour of cells prior to experimenta- tioned as the intestinal stem cells. Ra- tion. Nevertheless, transplantation as- tional for this hypothesis was that these says are a powerful tool to test stem- cells are label retaining (Potten et al., ness potential, but it is likely that the 1974) and extremely radiation sensitive population of cells with stemness (Potten, 1977). DNA label retention by potential is larger than the number of the +4 cells was postulated to be the cells that actually do function as stem result of asymmetric segregation of old cells in normal homeostasis. and new DNA strands (Potten, 1977; An alternative method to test stem- Potten et al., 2002) in order to protect ness is based on genetic marking and is their genome from damage. Asymmet- commonly referred to as lineage trac- ric DNA segregation was suggested to ing or fate mapping. Here, the DNA of be a stem cell specific phenomenon. specific (candidate) stem cells is However, definitive prove for stem- marked at any specific moment in time. ness needs to be demonstrated by Subsequently, over time all descend- experimental assays, rather than ants of the marked population inherit association based on morphology, loca- the mutation. Moreover, the genetic tion, marker expression or any other marking can be visualized via straight- specifically assigned cell-characteristic. forward laboratory techniques and al- An experimental assay to prove stem- lows experimental testing of the two ness is a transplantation experiment. functional descriptions of stemness. Hereby a cell population is isolated First, over time the genetic marking based of specific presence (or absence) needs to be detectable in all the cell of markers. Once isolated from the do- types that are present in the tissue nor, the cells are transplanted into a re- (multipotency). Second, the presence of cipient and tested for their ability to the genetic marking needs to be main- give rise to a complete "recovery" ef- tained within the tissue for the entire fect. For example in bone marrow life-time of the organism (self-re- transplantation experiments, trans- newal). The stemness of a specific cell planted donor haematopoietic stem population is tested without the pres-

4 ence of stress or other injury responses. only a limited number of cells within Therefore, lineage tracing assays test the marked group are the real stem actual stemness, i.e. it identifies cells cells. It is likely that the population of that function as stem cells in normal cells that contain actual stemness fall homeostasis. However, also with trac- within the larger population of cells ing experiments care needs to be taken that consists stemness potential. into account since it is possible that

LGR5 IS AN INTESTINAL STEM CELL MARKER

Lineage tracing was used to prove that intestine (Barker et al., 2007). the Lgr5 gene is specifically expressed Lineage tracing has also been perin intestinal stem cells and can be used formed for several markers that were as a stem cell specific marker. In situ claimed to be +4-cell specific, includ- hybridization experiments (mapping of ing: Bmi-1 (Sangiorgi and Capecchi, mRNA expression patterns) as well as 2008), Hopx (Takeda et al., 2011), different knock-in alleles revealed mTert (Montgomery et al., 2011) and expression of Lgr5 in CBC cells Lrig1 (Powell et al., 2012). However, (Barker et al., 2007; Tian et al., 2011). robust expression of most of these Most importantly, the genetic inducible markers was also detected in Lgr5+ activation of LacZ expression in Lgr5+ CBC cells (Munoz et al., 2012). A re- CBC cells (the genetic clonal marking) cent study shows that genetic lineage was inherited over time by all the tracing of DNA label-retaining cells differentiated cell types in the intestinal identified a rare, non-dividing secretory epithelium, thereby experimentally precursor that co-expresses Lgr5 and linking Lgr5+ CBC cells to multipo- all +4 makers. These cells are located tency. Moreover, once LacZ expression near crypt bottoms where they undergo was activated in CBC cells, the mark- terminal differentiation over periods of ing could be found during the rest of weeks towards Paneth cell lineage. the lifespan of the animals in the However upon tissue damage, prema- intestinal epithelium, showing that ture secretory precursors can revert Lgr5 expressing CBC cells represent their fate back into a cycling, Lgr5+ long-lived multipotent stem cells of the CBC cell (Buczacki et al., 2013).

ADDITIONAL INTESTINAL STEM CELL MARKERS

Isolation of Lgr5 expressing CBC cells a basic helix-loop-helix transcription using FACS allowed global gene factor. Transgenic overexpression of expression analysis (Munoz et al., Ascl2 throughout the intestinal epithe- 2012; van der Flier et al., 2009). One lium induces crypt hyperplasia and de of the genes enriched in the Lgr5+ cells novo crypt formation on villi. The turned out to be the CBC restricted opposite experiment, induced deletion Wnt target Ascl2 (Jubb et al., 2006; of Ascl2 from the intestine, results in Sansom et al., 2004; van der Flier et rapid loss of CBC stem cells. The com- al., 2007). Ascl2 is one of the mamma- bined results from these genetic gain lian homologous of the Drosophila and loss of function studies in the Achaete-scute complex genes encoding intestinal epithelium show that Ascl2

5 controls the intestinal stem cell fate receptors (Koo et al., 2012) and EphB2, (van der Flier et al., 2009). Other stem a cell surface marker which expression cell specific Wnt target genes are pattern was utilized to isolate human Rnf43, an ubiquitin ligase for Wnt intestinal stem cells (Jung et al., 2011).

NEUTRAL DRIFT COMPETITION BETWEEN MULTIPLE INTESTINAL STEM CELLS

On average, there are around 14 Lgr5+ cell or transit amplifying cell fates CBC cells at the bottom of intestinal based on their relative positioning to- crypts that are sandwiched in between wards the Paneth cells (niche cells). similar numbers of post-mitotic Paneth Due to the limited number of Paneth cells (Snippert et al., 2010). To be able cells, CBC stem cells compete in a neu- to map the fate of neighbouring stem tral fashion for limited niche space cells simultaneously, a multicolour re- (Snippert et al., 2010). As long as porter mouse was generated (R26R- Lgr5+ CBC (daughter) stem cells are in Confetti). Upon activation of R26R- close proximity to Paneth cells they Confetti, each cell inherits one out of will maintain stemness. In the case that four possible fluorescent markers via a (daughter) stem cell is displaced from DNA recombination, i.e. green, yellow, Paneth cells, i.e. loss of direct contact, red or blue. This ‘colour picking’ is they will be primed for differentiation completed in a random fashion, is spe- (Sato et al., 2011b). This stochastic cific per individual cell and inherited stem cell model in which cell fate is by daughter cells. Using the R26R- determined by presence in the Paneth Confetti mouse, the individual behav- cell niche fits also well with recent data iour of multiple intestinal stem cells in that demonstrated that progenitor cells the same crypt was followed in time. that initially lost their stemness, can The study revealed that Lgr5+ CBC revert towards their stem cell fate upon cells double their numbers each day by stress and injury responses (Buczacki et symmetric divisions, after which the al., 2013; Tian et al., 2011; van Es et daughter cells stochastically adopt stem al., 2012b).

INTESTINAL ORGANOIDS

Most of the above described stem cell Most cultures are named after the tissue work is based on genetic mouse mod- they were derived from, for example, els. However, it is essential to translate neurospheres (Reynolds and Weiss, those scientific findings from the 1992), mammospheres (Shackleton et mouse to the human situation, for in- al., 2006; Stingl et al., 2006) or colon- stance via transplantation assays or in spheres. A few years ago, another ex vitro culturing assays. The culture of vivo cell culture system has been devel- individual clones of cells that are de- oped for initially wild-type small rived from various types of progenitor intestinal epithelia. It allows long-term populations has become a widely used growth of so-called mini-guts or method to define the identity and intestinal organoids (Sato et al., 2009) behaviour of (human) stem cell types. (Figure 3). In contrast to sphere

Figure 3: Bright field image of mouse small intestinal organoid. Scale bar is 50 µm.

cultures, organoids are three-dimen- partmentalization between progenitor sional asymmetric cell-cultures that can and differentiated zones and correct be grown from primary cells that are positioning and migration of cell types isolated from mouse and human intes- along the crypt-villus axes. Since tines (Sato et al., 2011a). Organoids are intestinal organoid cultures do not con- cultured using matrixgel and a medium tain mesenchymal cells, the intestinal that contains a cocktail of growth fac- stem cell niche is provided by Paneth tors such as R-spondin, EGF and Nog- cells (secreting Wnt, EGF and Notch gin. Important, these organoids share a ligands) and the limited number of lot of characteristics with the normal added growth factors (Sato et al., intestinal epithelium, such as the pres- 2011b). ence of all the different cell types, com-

INTESTINAL STEM CELL NICHE

By using neonatal tissues, an alterna- maintained once organoid cultures are tive long-term multi-lineage intestinal initiated from human embryonic stem epithelial culture has been established cells which differentiation is guided in which mesenchymal parts are pre- towards intestinal fate (Spence et al., sent as well (Ootani et al., 2009). 2011). Interesting, the mesenchymal Mesenchymal niche architecture is also layer develops and differentiates along

7 with the epithelium with the same phenotype, while organoid cultures kinetics, suggesting an intimate signal- from these mice fail to develop. To ling crosstalk between mesenchyme compensate for the lack of endogenous and epithelium. Wnt3a production, Wnt3A mutant Active Wnt signalling is essential organoids can be rescued by addition of for organoid cultures. In the absence of Wnt3A or by growing them on primary mesenchymal tissue, Wnt ligands are mesenchymal cells that produce Wnt either produced endogenously by the proteins (Farin et al., 2012). Above epithelium (Wnt3A in Paneth cells of results illustrate that the mesenchyme mouse small intestine) or need to be surrounding crypts, as well as Paneth added exogenously in the medium. cells, function together as the niche by Interestingly however, Wnt3a knock- providing the essential signals to the out mice don’t have an intestinal intestinal epithelial stem cells.

CONCLUDING REMARKS

The intestinal epithelial (stem cell) cell types, cell fate specification and field has developed very rapidly since cellular behaviour. In the past, genetic the identification in 2007 of Lgr5 as the mouse studies have created a wealth of first definitive stem cell marker. The new insights on the biology of the intestinal epithelium represents a intestinal epithelium. The recently unique model to study adult stem cell established long term culture condi- biology and lineage specification. The tions of intestinal epithelium, espe- combination of a rapid self-renewing cially mini-organs from human origin, tissue, evident compartmentalization of will probably boost the research field proliferating and differentiated cell for the next generation by providing a types and a relative simple, repetitive variety of possibilities for research and tissue architecture, is ideal for the therapeutic applications of intestinal visualization and identification of stem biology.

LITERATURE

Barker, N., van de Wetering, M., and Clevers, (2013). H.: The intestinal stem cell. Genes Dev. Cheng, H. and Leblond, C.P.: Origin, 22, 1856-1864 (2008). differentiation and renewal of the four Barker, N., van Es, J.H., Kuipers, J., Kujala, P., main epithelial cell types in the mouse van den Born, M., Cozijnsen, M., small intestine. V. Unitarian Theory of the Haegebarth, A., Korving, J., Begthel, H., origin of the four epithelial cell types. Am. Peters, P.J., and Clevers, H.: Identification J. Anat. 141, 537-561 (1974). of stem cells in small intestine and colon Clevers, H.C. and Bevins, C.L.: Paneth cells: by marker gene Lgr5. Nature 449, 1003- maestros of the small intestinal crypts. 1007 (2007). Ann. Rev. Physiol. 75, 289-311 (2013). Buczacki, S.J., Zecchini, H.I., Nicholson, Farin, H.F., Van Es, J.H., and Clevers, H.: A.M., Russell, R., Vermeulen, L., Kemp, Redundant sources of Wnt regulate R., and Winton, D.J. Intestinal label- intestinal stem cells and promote formation retaining cells are secretory precursors of Paneth cells. Gastroenterology 143, expressing Lgr5. Nature 495, 65-69 1518-1529 (2012).

8 Fevr, T., Robine, S., Louvard, D., and intestine and colon revealed by adenoviral Huelsken, J.: Wnt/beta-catenin is essential expression of Dickkopf-1. Proc. Natl. for intestinal homeostasis and maintenance Acad. Sci. USA 101, 266-271 (2004). of intestinal stem cells. Mol. Cell. Biol. 27, Montgomery, R.K., Carlone, D.L., Richmond, 7551-7559 (2007). C.A., Farilla, L., Kranendonk, M.E., Ireland, H., Kemp, R., Houghton, C., Howard, Henderson, D.E., Baffour-Awuah, N.Y., L., Clarke, A.R., Sansom, O.J., and Ambruzs, D.M., Fogli, L.K., Algra, S., and Winton, D.J.: Inducible Cre-mediated Breault, D.T.: Mouse telomerase reverse control of gene expression in the murine transcriptase (mTert) expression marks gastrointestinal tract: effect of loss of beta- slowly cycling intestinal stem cells. Proc. catenin. Gastroenterology 126, 1236-1246 Natl. Acad. Sci. USA 108, 179-184 (2011). (2004). Morin, P.J., Sparks, A.B., Korinek, V., Barker, Jubb, A.M., Chalasani, S., Frantz, G.D., Smits, N., Clevers, H., Vogelstein, B., and R., Grabsch, H.I., Kavi, V., Maughan, N.J., Kinzler, K.W.: Activation of beta-catenin- Hillan, K.J., Quirke, P., and Koeppen, H.: Tcf signaling in colon cancer by mutations Achaete-scute like 2 (ascl2) is a target of in beta-catenin or APC. Science 275, Wnt signalling and is upregulated in 1787-1790 (1997). intestinal neoplasia. Oncogene 25, 3445- Munoz, J., Stange, D.E., Schepers, A.G., van 3457 (2006). de Wetering, M., Koo, B.K., Itzkovitz, S., Jung, P., Sato, T., Merlos-Suarez, A., Barriga, Volckmann, R., Kung, K.S., Koster, J., F.M., Iglesias, M., Rossell, D., Auer, H., Radulescu, S., Myant, K., Versteeg, R., Gallardo, M., Blasco, M.A., Sancho, E., Sansom, O.J., van Es, J.H., Barker, N., van Clevers, H., and Batlle, E.: Isolation and in Oudenaarden, A., Mohammed, S., Heck, vitro expansion of human colonic stem A.J., and Cklevers, H.: The Lgr5 intestinal cells. Nat. Med. 17, 1225-1227 (2011). stem cell signature: robust expression of Koo, B.K., Spit, M., Jordens, I., Low, T.Y., proposed quiescent '+4' cell markers. Stange, D.E., van de Wetering, M., van Es, EMBO J. 31, 3079-3091 (2012). J.H., Mohammed, S., Heck, A.J., Maurice, Ootani, A., Li, X., Sangiorgi, E., Ho, Q.T., M.M., and Clevers, H.: Tumour suppressor Ueno, H., Toda, S., Sugihara, H., RNF43 is a stem-cell E3 ligase that Fujimoto, K., Weissman, I.L., Capecchi, induces endocytosis of Wnt receptors. M.R., and Kuo, C.J.: Sustained in vitro Nature 488, 665-669 (2012). intestinal epithelial culture within a Wnt- Korinek, V., Barker, N., Moerer, P., van dependent stem cell niche. Nat. Med. 15, Donselaar, E., Huls, G., Peters, P.J., and 701-706 (2009). Clevers, H.: Depletion of epithelial stem- Pinto, D., Gregorieff, A., Begthel, H., and cell compartments in the small intestine of Clevers, H.: Canonical Wnt signals are mice lacking Tcf-4. Nat. Genet. 19, 379- essential for homeostasis of the intestinal 383 (1998). epithelium. Genes Dev. 17, 1709-1713 Korinek, V., Barker, N., Morin, P.J., van (2003). Wichen, D., de Weger, R., Kinzler, K.W., Potten, C.S.: Extreme sensitivity of some Vogelstein, B., and Clevers, H.: intestinal crypt cells to X and gamma Constitutive transcriptional activation by a irradiation. Nature 269, 518-521 (1977). beta-catenin-Tcf complex in APC-/- colon Potten, C.S., Kovacs, L., and Hamilton, E.: carcinoma. Science 275, 1784-1787 Continuous labelling studies on mouse (1997). skin and intestine. Cell Tissue Kinet. 7, Kuhnert, F., Davis, C.R., Wang, H.T., Chu, P., 271-283 (1974). Lee, M., Yuan, J., Nusse, R., and Kuo, Potten, C.S., Owen, G., and Booth, D.: C.J.: Essential requirement for Wnt Intestinal stem cells protect their genome signaling in proliferation of adult small by selective segregation of template DNA

9 strands. J. Cell Sci. 115, 2381-2388 Stingl, J., Smyth, G.K., Asselin-Labat, (2002). M.L., Wu, L., Lindeman, G.J., and Powell, A.E., Wang, Y., Li, Y., Poulin, E.J., Visvader, J.E.: Generation of a functional Means, A.L., Washington, M.K., mammary gland from a single stem cell. Higginbotham, J.N., Juchheim, A., Prasad, Nature 439, 84-88 (2006). N., Levy, S.E., Guo, Y., Shyr, Y., Aronow, Snippert, H.J., van der Flier, L.G., Sato, T., van B.J., Haigis, K.M., Franklin, J.L., and Es, J.H., van den Born, M., Kroon- Coffey, R.J.: The pan-ErbB negative Veenboer, C., Barker, N., Klein, A.M., van regulator Lrig1 is an intestinal stem cell Rheenen, J., Simons, B.D., and Clevers, marker that functions as a tumor H.: Intestinal crypt homeostasis results suppressor. Cell 149, 146-158 (2012). from neutral competition between Reynolds, B.A. and Weiss, S.: Generation of symmetrically dividing Lgr5 stem cells. neurons and astrocytes from isolated cells Cell 143, 134-144 (2010). of the adult mammalian central nervous Spence, J.R., Mayhew, C.N., Rankin, S.A., system. Science 255, 1707-1710 (1992). Kuhar, M.F., Vallance, J.E., Tolle, K., Sangiorgi, E. and Capecchi, M.R.: Bmi1 is Hoskins, E.E., Kalinichenko, V.V., Wells, expressed in vivo in intestinal stem cells. S.I., Zorn, A.M., Shroyer, N.F., and Wells, Nat. Genet. 40, 915-920 (2008). J.M.: Directed differentiation of human Sansom, O.J., Reed, K.R., Hayes, A.J., Ireland, pluripotent stem cells into intestinal tissue H., Brinkmann, H., Newton, I.P., Batlle, in vitro. Nature 470, 105-109 (2011). E., Simon-Assmann, P., Clevers, H., Stingl, J., Eirew, P., Ricketson, I., Shackleton, Nathke, I.S., Clarke, A.R., and Winton, M., Vaillant, F., Choi, D., Li, H.I., and D.J.: Loss of Apc in vivo immediately Eaves, C.J.: Purification and unique perturbs Wnt signaling, differentiation, and properties of mammary epithelial stem migration. Genes Dev. 18, 1385-1390 cells. Nature 439, 993-997 (2006). (2004). Takeda, N., Jain, R., LeBoeuf, M.R., Wang, Q., Sato, T., Stange, D.E., Ferrante, M., Vries, Lu, M.M., and Epstein, J.A.: R.G., Van Es, J.H., Van den Brink, S., Van Interconversion between intestinal stem Houdt, W.J., Pronk, A., Van Gorp, J., cell populations in distinct niches. Science Siersema, P.D., and Clevers, H.: Long- 334, 1420-1424 (2011). term expansion of epithelial organoids Tian, H., Biehs, B., Warming, S., Leong, K.G., from human colon, adenoma, Rangell, L., Klein, O.D., and de Sauvage, adenocarcinoma, and Barrett's epithelium. F.J.: A reserve stem cell population in Gastroenterology 141, 1762-1772 (2011a). small intestine renders Lgr5-positive cells Sato, T., van Es, J.H., Snippert, H.J., Stange, dispensable. Nature 478, 255-259 (2011). D.E., Vries, R.G., van den Born, M., van de Wetering, M., Sancho, E., Verweij, C., Barker, N., Shroyer, N.F., van de de Lau, W., Oving, I., Hurlstone, A., van Wetering, M., and Clevers, H.: Paneth der Horn, K., Batlle, E., Coudreuse, D., cells constitute the niche for Lgr5 stem Haramis, A.P., Tjon-Pon-Fong, M., cells in intestinal crypts. Nature 469, 415- Moerer, P., van den Born, M., Soete, G., 418 (2011b). Pals, S., Eilers, M., Medema, R., and Sato, T., Vries, R.G., Snippert, H.J., van de Clevers, H.: The beta-catenin/TCF-4 Wetering, M., Barker, N., Stange, D.E., complex imposes a crypt progenitor van Es, J.H., Abo, A., Kujala, P., Peters, phenotype on colorectal cancer cells. Cell P.J., and Clevers, H.: Single Lgr5 stem 111, 241-250 (2002). cells build crypt-villus structures in vitro van der Flier, L.G., and Clevers, H.: Stem cells, without a mesenchymal niche. Nature 459, self-renewal, and differentiation in the 262-265 (2009). intestinal epithelium. Ann. Rev. Physiol. Shackleton, M., Vaillant, F., Simpson, K.J., 71, 241-260 (2009).

10 Van der Flier, L.G., Sabates-Bellver, J., Oving, Tcf4 in adult intestinal homeostatic self- I., Haegebarth, A., De Palo, M., Anti, M., renewal. Mol. Cell. Biol. 32, 1918-1927 Van Gijn, M.E., Suijkerbuijk, S., Van de (2012a). Wetering, M., Marra, G., and Clevers, H.: van Es, J.H., Jay, P., Gregorieff, A., van Gijn, The Intestinal Wnt/TCF Signature. M.E., Jonkheer, S., Hatzis, P., Thiele, A., Gastroenterology 132, 628-632 (2007). van den Born, M., Begthel, H., Brabletz, van der Flier, L.G., van Gijn, M.E., Hatzis, P., T., Taketo, M.M., and Clevers, H.: Wnt Kujala, P., Haegebarth, A., Stange, D.E., signalling induces maturation of Paneth Begthel, H., van den Born, M., Guryev, V., cells in intestinal crypts. Nat. Cell Biol. 7, Oving, I., van Es, J.H., Barkeer, N., Peters, 381-386 (2005). P.J., van de Wetering, M., and Clevers, H.: van Es, J.H., Sato, T., van de Wetering, M., Transcription factor achaete scute-like 2 Lyubimova, A., Nee, A.N., Gregorieff, A., controls intestinal stem cell fate. Cell 136, Sasaki, N., Zeinstra, L., van den Born, M., 903-912 (2009). Korving, J., Matens, A.C., Barker, N., van van Es, J.H., Haegebarth, A., Kujala, P., Oudenaarden, A., and Clevers, H.: Dll1+ Itzkovitz, S., Koo, B.K., Boj, S.F., secretory progenitor cells revert to stem Korving, J., van den Born, M., van cells upon crypt damage. Nat. Cell Biol. Oudenaarden, A., Robine, S., and Clevers, 14, 1099-1104 (2012b). H.: A critical role for the Wnt effector

IRRITABLE BOWEL SYNDROME: ROLE OF GUT BACTERIA AND BACTERIAL TOXINS

CHRISTOPHER CHANG, GILLIAN BARLOW, STACY WEITSMAN, and MARK PIMENTEL

GI Motility Program, Cedars-Sinai Medical Center, Los Angeles, California 90048, USA

SUMMARY

Irritable bowel syndrome (IBS) is a common condition affecting 10-15% of the population. While the precise aetiology remains unknown, our group has focused on the contributions of altered gut flora to IBS, including links between acute gastroenteritis and post-infectious IBS (PI-IBS) and the role of small intestinal bacterial overgrowth (SIBO). Breath testing and culture of small intestinal aspirates indicate that SIBO is present in over one-third of IBS subjects and almost two-thirds of diarrhoea-pre- dominant IBS (D-IBS) subjects, but only one-tenth of non-IBS subjects. The success of antibiotic therapies in impacting IBS phenotypes also sup- ports a bacterial role in IBS. Four large-scale multi-centre trials indicate that rifaximin significantly reduces SIBO and non-constipation IBS with little bacterial resistance or reduction of benefit in successive retreatments. To elucidate the mechanisms underlying PI-IBS, we developed a rat model using Campylobacter jejuni, the most common identified cause of acute gastroenteritis. This model mirrors many findings in human IBS subjects, including altered stool form more than three months after clearing initial infection, SIBO, increased intraepithelial lymphocytes (IELs), reductions in deep muscular plexus interstitial cells of Cajal (DMP- ICCs), required for normal intestinal motility, and increases in specific mucosal defence mediators. Common to all pathogens causing gastroenteritis is cytolethal distending toxin (Cdt). We developed a rat model using a mutant C. jejuni lacking Cdt, which demonstrated significantly ameliorated bowel phenotypes including SIBO and IELs and no reduction in DMP-ICCs, strongly implicating Cdt in the development of IBS phenotypes. Further, early exposure to C. jejuni appears to mitigate the acute effects of second infections, which suggested immunity might play a role in IBS development. Our recent data indicating molecular mimicry between Cdt and neural elements of the gut support this hypothesis and may form the basis for future biomarkers and potential therapies directed at the underlying causes of IBS.

INTRODUCTION

Irritable bowel syndrome (IBS) is the health care costs (The Lewin Group, most common chronic medical condi- 2001; Thompson, 1994) and affecting tion in the U.S. accounting for nearly 10-15% of the population (Drossman et 30% of all gastroenterology related al., 1982; Thompson and Heaton,

1980). This condition does not dis- anti-depressants were used as treatment criminate by age and the prevalence of for IBS. Our group has validated mi- this condition is equally common crobe-mediated theories in the patho- among young adults. These facts make physiology of post-infectious IBS (PI- IBS a very important disease state. IBS) and small intestinal bacterial However, the aetiology of IBS has re- overgrowth (SIBO) and spent the last mained unknown. 10 years characterizing alterations in Historically, IBS was treated as gut flora that may contribute to IBS. largely a psychological disorder and

SIBO AND IBS

Over the last decade, we have devel- spite continued controversy, a key oped a new hypothesis in IBS that fo- meta-analysis by our group has secured cused on bloating as a universal com- that breath testing is important in IBS plaint in IBS subjects. Through a series and that the odds of having a positive of studies, we demonstrated that SIBO, breath test in IBS compared to healthy which results from reduced gut motil- controls is far greater (Shah et al., ity, is more common in IBS. This was 2010). Since then, two studies have initially controversial as we used breath conclusively demonstrated through cul- testing as a surrogate of SIBO. While ture of the small bowel that IBS sub- we recognized the limitations of this jects have higher coliform (colon bac- technique, it was apparent that the teria) levels than healthy persons breath test was frequently abnormal in (Posserud et al., 2007; Pyleris et al., IBS compared to healthy controls. De- 2012).

SMALL BOWEL MICROBIOME IS ALTERED IN IBS

In another aspect of our on-going work jects were found to have SIBO (Pyleris to identify the mechanisms responsible et al., 2012). More recently, we have for IBS, we have begun to explore how begun to characterize the specific mi- changes in gut microbial populations crobial populations present in the small contribute to IBS. Since September intestine in IBS vs. control subjects. 2009, subjects undergoing upper en- Preliminary deep sequencing analyses doscopy have been recruited to a repos- suggest that the microbial profile (mi- itory developed by our collaborator Dr. crobiome) in the small intestine of IBS Evangelos Giamarellos-Bourboulis in subjects is markedly different from that Athens, Greece. Samples and data ob- in healthy controls, with significantly tained include serum, genetic material, reduced microbial diversity and subject phenotypes, and aspirates from overrepresentation of a few specific the small intestine (Pyleris et al., genera (Chang et al., 2013). The over- 2012). Conventional cultures of these growth of these specific microbial small intestinal samples demonstrated populations, possibly in concert with that 39.3% of IBS subjects had SIBO the loss of other microbes with im- (defined as >103 cfu/ml), compared to portant protective effects, may contrib- 11.3% of non-IBS subjects (subjects ute to the gut symptoms of IBS sub- with GI complaints that did not have jects. IBS). Moreover, 60% of D-IBS sub-

TREATING SIBO IN IBS

While the uncovered connection be- (Pimentel et al., 2011a, 2006). The tween IBS and SIBO was important, it most recent of these were the comple- was equally important to identify treat- tion of the Target 1 and 2 trials pub- ment options for those who suffer from lished singularly in the New England IBS. Based on our growing evidence Journal of Medicine. This final phase for the connection to SIBO, we initi- III study conclusively showed that ated a series of studies to test the use of rifaximin is effective in IBS without antibiotics in IBS. The first was the use constipation and that a 2-week treat- of neomycin in the first double blind ment had benefits lasting a minimum of antibiotic study (Pimentel et al., 2003). 12 weeks. No previous IBS therapy had Although neomycin successfully im- this kind of effect, suggesting that we proved IBS in the study, neomycin was impacted a potential mechanism of a poor antibiotic for SIBO. Thus we IBS. We also examined the effects of began looking for an alternate non-ab- retreatment with rifaximin in subjects sorbed antibiotic. Rifaximin had all the with non-constipated IBS, and found properties that made it ideal for IBS. It that more than 75% of subjects who was almost completely non-absorbed, initially responded to rifaximin also had no issues with bacterial resistance, responded to subsequent treatments, was highly effective for SIBO, was not with no significant reduction in benefit related to conventional antibiotics (thus for up to 5 successive retreatments would not create cross-resistance) and (Pimentel et al., 2011b). Furthermore, did not change stool flora. Since identi- there was no change in the duration of fying this antibiotic, we have con- benefit (median time between treat- ducted four large-scale multicentre tri- ments) for successive retreatments als to examine the efficacy of rifaximin (Pimentel et al., 2011b).

POST-INFECTIOUS IBS

Over the last decade, it has been estab- tremely common, so-called PI-IBS may lished that intestinal pathogens play a in fact constitute a large proportion of significant role in the development of IBS cases. Thus, reducing risk factors IBS. Numerous studies have shown for IBS development after acute gastro- that IBS can be precipitated by an epi- enteritis may have an impact on the sode of acute gastroenteritis, and that incidence of IBS. Although the mecha- up to 57% of subjects who otherwise nisms of PI-IBS remain unclear, inves- had normal bowel function may con- tigators have identified certain risk fac- tinue to have altered bowel function for tors for the development of IBS after at least 6 years after recovering from gastroenteritis. The two most signifi- the initial acute illness (Neal et al., cant of these are duration/severity of 2002). Based on two recent meta-anal- gastroenteritis and female sex (Gwee et yses of this research, approximately al., 1999; Neal et al., 1997). Stress, 10% of subjects who have documented manifest as recent traumatic life events, acute gastroenteritis develop IBS, with and a neurotic personality trait were a summary odds ratio of 6 to 7 for PI- also predictors of PI-IBS (Gwee et al., IBS (Halvorson et al., 2006; Thabane 1999). Evidence of low-grade inflam- et al., 2007). As gastroenteritis is ex- mation is evident in PI-IBS patients.

Rectal biopsies demonstrate mildly ele- sigmoid and terminal ileum biopsies vated intraepithelial lymphocytes and (Wang et al., 2004). Thus, acute gastro- entero-endocrine cells that persisted 12 enteritis may increase the risk of devel- months after infection with Campylo- oping IBS in a susceptible individual bacter jejuni (Spiller et al., 2000), through persistent low-grade activation which is the most common cause of of the gut immune system, or possibly acute gastroenteritis in the US (Tauxe, through establishment of an intestinal 1992). Increased rectal lymphocytes dysbiosis, defined as an alteration of also occur in general IBS patients, but the composition of the gut flora. Ani- to a lesser degree (Dunlop et al., 2003). mal infection models of PI-IBS will Elevated expression of pro-inflamma- play a key role in characterizing the tory cytokine IL-1β was detected in C. mechanistic pathways and underlying jejuni PI-IBS rectal biopsies (Gwee et alterations in this process. al., 2003) and in Shigella PI-IBS recto-

RAT MODEL OF PI-IBS

To study the underlying mechanisms of jal (DMP-ICCs) (Jee et al., 2010). The PI-IBS, we developed and validated a latter is particularly significant as ICCs rat model (Pimentel et al., 2008) using and myenteric nerves are known to be the human pathogen C. jejuni as the required for normal intestinal motility, infective agent. Over the past 8 years, including phase III of interdigestive we have characterized and validated motor activity (Nieuwenhuijs et al., this model, which exhibits phenotypes 1998). Further, evidence suggests that a that closely mimic those seen in human deficiency of phase III motor activity, patients with PI-IBS. These include: which is known to induce SIBO altered stool consistency that persists (Nieuwenhuijs et al., 1998; Vantrappen three months after the clearance of the et al., 1977), is associated with the de- acute infection and the development of velopment of SIBO in IBS SIBO (Pimentel et al., 2008), as well as (Nieuwenhuijs et al., 1998). We have increased rectal intraepithelial lympho- also confirmed alterations in mucosal cytes and reduced numbers of deep defence mediators such as TNF-α in muscular plexus interstitial cells of Ca- post-infectious rats (Sung et al., 2013).

ANTIBIOTIC PROPHYLAXIS PREVENTS THE DEVELOPMENT OF IBS-LIKE CHARACTERISTICS IN A RAT MODEL OF C. JEJUNI INFECTION

Antibiotic therapy (Lembo, 2008; Pi- proaches may be effective in prevent- mentel et al., 2003, 2006; Sharara et ing IBS. To determine the effect of al., 2006) and the C. jejuni vaccine prophylactic antibiotic therapy on the (Monteiro et al., 2009) mitigate the ef- development of post-infectious IBS- fects of gastroenteritis in humans like symptoms in rats, we gavaged rats (Lembo, 2008; Pimentel et al., 2003, with either rifaximin (200mg rifaximin 2006; Sharara et al., 2006). Since du- daily for 3 days) and C. jejuni 81-176 ration of gastroenteritis is a risk factor (5 x 108 cfu, given on day 2) (C+/R+), for IBS (Neal et al., 1997), these ap- or with C. jejuni alone (C+/R-). The

two groups were then compared for (Pimentel et al., 2011c). Further, we both acute C. jejuni colonization, and found that after 3 months, the C+/R- for the development of post-infectious rats had a greater persistent variability IBS-like symptoms three months later. in stool % wet weight than C+/R+ rats We found that rats that received rifaxi- (p<0.01), and that the average stool min prophylaxis (C+/R+) cleared the consistency was closer to normal in initial C. jejuni infection significantly C+/R+ rats than in the C+/R- rats faster than the C+/R- group (10.3±7.1 (Pimentel et al., 2011c). days vs. 12.6±5.9 days, p<0.01)

CYTOLETHAL DISTENDING TOXIN (CDT) IS REQUIRED FOR EXPRESSION OF AN IBS-LIKE PHENOTYPE IN THE RAT MODEL

The Cdt toxin is common to all bacte- the variance was 8.4±6.4, compared to rial organisms that cause post-infec- 4.2±2.4 for the Cdt- rats (p<0.001) tious IBS, and thus a potential candi- (Morales et al., 2011; Pokkunuri et al., date toxin mediator of IBS. To evaluate 2012). the effect of Cdt on the post-infectious To evaluate the effect of Cdt on gut IBS-like symptoms in rats, rats exposed histology, DMP-ICC were also exam- to a Cdt knockout (Cdt-) mutant of C. ined in Cdt- rats. Three months after jejuni were evaluated for stool form clearance of the initial C. jejuni infec- three months after clearing the initial tion from their stool, the ileum was re- C. jejuni infection, and compared to sected and CD117 immunostaining results obtained in rats infected with (anti c-kit) was performed on cross- wild type C. jejuni (Cdt+). Stool sam- sections of mucosa. The results re- ples were collected and evaluated over vealed normal DMP-ICC staining in a 3-day period to obtain an average the ilea of these rats, which is in stool form. Stool form at 3 months marked contrast to the reduced num- post-infection was significantly better bers and altered appearance of DMP- in the Cdt- group. Furthermore, 42% of ICC seen in sections from rats exposed the Cdt+ rats had altered stool form 2 to wild type C. jejuni (Morales et al., out of 3 days, as compared to 18% of 2011; Pokkunuri et al., 2012). Taken the Cdt- rats (p=0.028, Fisher’s exact together, these data strongly implicate test). In addition, the stool wet weight Cdt as directly contributing to the de- in the Cdt+ rats demonstrated an alter- velopment and severity of PI-IBS phe- nating pattern of wet and dry, such that notypes.

POTENTIAL ROLE OF IMMUNITY IN THE DEVELOPMENT OF PI-IBS

Epidemiologic studies from developing immunity might play a role in PI-IBS. countries indicate that the prevalence To begin to determine whether immun- of PI-IBS in developing countries is ity was involved in the effects of C. lower than or similar to that in devel- jejuni and Cdt in the gut, we infected oped countries, despite much higher 50 rats with C. jejuni 81-176 as juve- incidence of Campylobacter enteritis, niles, and then re-infected them with which suggested to us that acquired the same strain two months later, and

compared the results to those from an- against the development of SIBO, the other 50 rats exposed for the first time second infection was better tolerated as adults. Fewer rats infected for a se- and resulted in less severe stool pheno- cond time (juvenile then adult (J+/A+)) types, perhaps suggesting the develop- had detectable stool C. jejuni compared ment of tolerance to the SIBO. Moreo- to their first infection (p<0.05) and ver, for rats who were first exposed as compared to the 50 rats that only re- infants and then reinfected, we found ceived C. jejuni as an adult (J-/A+) that those who had diarrhoea during the (p<0.05). In addition, the number of first exposure were more likely to de- days of loose stool was less for a se- velop SIBO than those who did not cond infection than the first (1.58±0.16 (p=0.02) (Sung et al., 2013). This result vs. 2.50±0.29 days, p<0.05) and the mirrors findings in humans that suggest overall duration of the second infection a relationship between the severity of was shorter than the first (p=0.001). acute illness and the likelihood of sub- These data suggest that exposure to C. sequent development of IBS (Halvor- jejuni early in life mitigated the acute son et al., 2006; Thabane et al., 2007), effects of a second infection (Sung et and led to our hypothesis that immunity al., 2013). might play a significant role in the de- We next compared the development velopment and severity of PI-IBS phe- of SIBO and stool phenotypes during notypes. Recent data generated in our the post-infectious period. Interest- laboratory support this hypothesis, and ingly, 47% (23/49) of J+/A+ rats devel- further suggest that the effects of im- oped SIBO, as compared to 26% munity are the result of molecular (13/50) of J-/A+ rats (p=0.019), but J- mimicry between Cdt and endogenous /A+ rats that developed SIBO had neural elements of the gut (Pimentel et greater alterations in stool consistency al., submitted). These results may form than J+/A+ rats that developed SIBO the basis for future biomarkers and po- (p<0.01) (Sung et al., 2013). These data tential therapies that could be directed suggest that while rats exposed to C. at the cause of IBS, rather than just jejuni early in life were not protected treating SIBO.

PERSPECTIVES AND FUTURE DIRECTIONS

In summary, our work to date indicates DMP-ICC which are required for gut that alterations in gut microbial popula- motility as well as increased IELs and tions contribute to IBS and the devel- immune response mediators, and re- opment of IBS phenotypes. Specifi- sults in altered stool phenotypes and cally, we have shown that SIBO, which the development of SIBO in the post- is caused by reduced gut motility, is infectious phenotypes. Further, we present in human IBS subjects, particu- have shown that these are likely poten- larly D-IBS subjects, and can be ame- tiated by the bacterial toxin Cdt. Lastly, liorated using antibiotics such as rifaxi- recent work by our group indicates that min. Further, acute gastroenteritis the small intestinal microbiome is caused by C. jejuni and similar patho- markedly different in IBS subjects, gens can precipitate PI-IBS. Using a rat with significantly reduced microbial model, we have shown that acute C. diversity and overrepresentation of a jejuni infection exhibits altered gut his- few specific genera, and that immunity tology including reductions in the may play a significant role in the devel-

opment of IBS, due to molecular mim- omarkers and targeted therapies for icry between Cdt and neural gut moie- IBS based on immune targets. The re- ties. Future work will include charac- sults will allow us to significantly im- terization of the small intestinal micro- pact the diagnosis and treatment of bial signatures in IBS vs. healthy con- IBS. trols, and the development of bi-

LITERATURE

Chang, C., Funari, V., Giamarellos-Bourboulis, Soffer, E., Conklin, J.L., and Pimentel, M.: E.J., Pyleris, E., Pistiki, K., Tang, J., Lee, ICC density predicts bacterial overgrowth C., Harkins, T., Kim, G., Weitsman, S., in a rat model of post-infectious IBS. Barlow, G. M., Pimentel, M.; Deep World J. Gastroenterol. 16, 3680-3686 sequencing reveals that the microbiome of (2010). the human duodenum is unique and unre- Lembo, A.: Rifaximin for the treatment of diar- lated to stool bacterial profiling. Gastro- rhea-associated irritable bowel syndrome: enterology 144, Supplement 1, S- short term treatment leading to long term 908. (2013). sustained response, efficacy of co-primary Drossman, D.A., Sandler, R.S., McKee, D.C., endpoints: Adequate relief/control of and Lovitz, A.J.: Bowel patterns among global IBS symptoms and IBS-associated subjects not seeking health care. Use of a bloating. In: Digestive Disease Week, San questionnaire to identify a population with Diego, CA, USA (2008) bowel dysfunction. Gastroenterology 83, Monteiro, M.A., Baqar, S., Hall, E.R., Chen, 529-534 (1982). Y.H., Porter, C.K., Bentzel, D.E., Apple- Dunlop, S.P., Jenkins, D., and Spiller, R.C.: bee, L., and Guerry, P.: Capsule polysac- Distinctive clinical, psychological, and his- charide conjugate vaccine against diarrheal tological features of postinfective irritable disease caused by Campylobacter jejuni. bowel syndrome. Am. J. Gastroenterol. 9, Infect. Immun. 77, 1128-1136 (2009). 1578-1583 (2003). Morales, W., Pimentel, M., Hwang, L., Kunkel, Gwee, K.A., Leong, Y.L., Graham, C., D., Pokkunuri, V., Basseri, B., Low, K., McKendrick, M.W., Collins, S.M., Wal- Wang, H., Conklin, J.L., and Chang, C.: ters, S.J., Underwood, J.E., and Read, Acute and chronic histological changes of N.W.: The role of psychological and bio- the small bowel secondary to C. jejuni logical factors in postinfective gut dys- infection in a rat model for post-infectious function. Gut 44, 400-406 (1999). IBS. Dig. Dis. Sci. 56, 2575-2584 (2011). Gwee, K.A., Collins, S.M., Read, N.W., Rajna- Neal, K.R., Hebden, J., and Spiller, R.: Preva- kova, A., Deng, Y., Graham, J.C., lence of gastrointestinal symptoms six McKendrick, M.W., and Moochhala, S.M.: months after bacterial gastroenteritis and Increased rectal mucosal expression of risk factors for development of the irritable interleukin 1beta in recently acquired post- bowel syndrome: postal survey of patients. infectious irritable bowel syndrome. Gut Bmj 314, 779-782 (1997). 52, 523-526 (2003). Neal, K.R., Barker, L., and Spiller, R.C.: Prog- Halvorson, H.A., Schlett, C.D., and Riddle, nosis in post-infective irritable bowel syn- M.S.: Postinfectious irritable bowel syndrome: a six year follow up study. Gut 51, drome – a meta-analysis. Am. J. Gastroen- 410-413 (2002). terol. 101, 1894-1899 (2006). Nieuwenhuijs, V.B., Verheem, A., van Jee, S.R., Morales, W., Low, K., Chang, C., Duijvenbode-Beumer, H., Visser, M.R., Zhu, A., Pokkunuri, V., Chatterjee, S., Verhoef, J., Gooszen, H.G., and Akker-

mans, L.M.: The role of interdigestive of post-infectious irritable bowel syn- small bowel motility in the regulation of drome. J. Neurogastroenterol. Motil. 18, gut microflora, bacterial overgrowth, and 434-442 (2012). bacterial translocation in rats. Ann. Surg. Posserud, I., Stotzer, P.O., Bjornsson, E.S., 228, 188-193 (1998). Abrahamsson, H., and Simren, M.: Small Pimentel, M., Chow, E.J., and Lin, H.C.: Nor- intestinal bacterial overgrowth in patients malization of lactulose breath testing cor- with irritable bowel syndrome. Gut 56, relates with symptom improvement in irri- 802-808 (2007). table bowel syndrome. A double-blind, Pyleris, E., Giamarellos-Bourboulis, E.J., Tzi- randomized, placebo-controlled study. vras, D., Koussoulas, V., Barbatzas, C., Am. J. Gastroenterol. 98, 412-419 (2003). and Pimentel, M.: The Prevalence of over- Pimentel, M., Park, S., Mirocha, J., Kane, S.V., growth by aerobic bacteria in the small in- and Kong, Y. The effect of a nonabsorbed testine by small bowel culture: relationship oral antibiotic (rifaximin) on the symptoms with irritable bowel syndrome. Dig. Dis. of the irritable bowel syndrome: a random- Sci. 57, 1321-1329 (2012). ized trial. Ann. Intern. Med. 145, 557-563 Shah, E.D., Basseri, R.J., Chong, K., and (2006). Pimentel, M.: Abnormal breath testing in Pimentel, M., Chatterjee, S., Chang, C., Low, IBS: a meta-analysis. Dig. Dis. Sci. 55, K., Song, Y., Liu, C., Morales, W., Ali, L., 2441-2449 (2010). Lezcano, S., Conklin, J., and Finegold, S.: Sharara, A.I., Aoun, E., Abdul-Baki, H., Moun- A new rat model links two contemporary zer, R., Sidani, S., and Elhajj, I.: A ran- theories in irritable bowel syndrome. Dig. domized double-blind placebo-controlled Dis. Sci. 53, 982-989 (2008). trial of rifaximin in patients with ab- Pimentel, M., Lembo, A., Chey, W.D., Zakko, dominal bloating and flatulence. Am. J. S., Ringel, Y., Yu, J., Mareya, S.M., Shaw, Gastroenterol. 101, 326-333(2006). A.L., Bortey, E., and Forbes, W.P.: Rifaxi- Spiller, R.C., Jenkins, D., Thornley, J.P., Heb- min therapy for patients with irritable den, J.M., Wright, T., Skinner, M., and bowel syndrome without constipation. N. Neal, K.R.: Increased rectal mucosal Engl. J. Med. 364, 22-32 (2011a). enteroendocrine cells, T lymphocytes, and Pimentel, M., Morales, W., Chua, K., Barlow, increased gut permeability following acute G., Weitsman, S., Kim, G., Amichai, Campylobacter enteritis and in post-dysen- M.M., Pokkunuri, V., Rook, E., Mathur, teric irritable bowel syndrome. Gut 47, R., and Marsh, Z.: Effects of rifaximin 804-811 (2000). treatment and retreatment in nonconsti- Sung, J., Morales, W., Kim, G., Pokkunuri, V., pated IBS subjects. Dig. Dis. Sci. 56, Weitsman, S., Rooks, E., Marsh, Z., Bar- 2067-2072 (2011b). low, G.M., Chang, C., and Pimentel, M.: Pimentel, M., Morales, W., Jee, S.R., Low, K., Effect of repeated Campylobacter jejuni Hwang, L., Pokkunuri, V., Mirocha, J., infection on gut flora and mucosal defense Conklin, J., and Chang, C.: Antibiotic in a rat model of post infectious functional Prophylaxis Prevents the Development of a and microbial bowel changes. Neurogas- Post-Infectious Phenotype in a New Rat troenterol. Motil. 25, 529-537 (2013). Model of Post-Infectious IBS. Dig. Dis. Tauxe, R.T.: Epidemiology of Campylobacter Sci. 56, 1962-1966 (2011c). jejuni infections in the United States and Pokkunuri, V., Pimentel, M., Morales, W., Jee, other industrialized nations. Current Status S.R., Alpern, J., Weitsman, S., Marsh, Z., and Future Trends. In: Campylobacter je- Low, K., Hwang, L., Khoshini, R., Barlow, juni: current status and future trends (Na- G.M., Wang, H., and Chang, C.: Role of chamkin, I., Blaser, M.J., and Tompkins, cytolethal distending toxin in altered stool L.S., Eds.). American Society for Micro- form and bowel phenotypes in a rat model biology, Washington, DC, USA, 9-19

(1992). Thompson, W.G. and Heaton, K.W.: Func- Thabane, M., Kottachchi, D.T., and Marshall, tional bowel disorders in apparently J.K.: Systematic review and meta-analysis: healthy people. Gastroenterology 79, 283- the incidence and prognosis of post-infec- 288 (1980). tious irritable bowel syndrome. Aliment. Vantrappen, G., Janssens, J., Hellemans, J., and Pharmacol. Ther. 26, 535-44 (2007). Ghoos, Y.: The interdigestive motor com- The Lewin Group: The burden of gastero- plex of normal subjects and patients with intestinal diseases. American Gastro- bacterial overgrowth of the small intestine. enterological Association, Bethesda, MD, J. Clin. Invest. 59, 1158-1166 (1977). USA (2001). Wang, L.H., Fang, X.C., and Pan, G.Z.: Bacil- Thompson, W.G.: The functional gastero- lary dysentery as a causative factor of irri- intestinal bowel disorders. In: The func- table bowel syndrome and its pathogene- tional gasterointestinal disorders (Dross- sis. Gut 53, 1096-1101 (2004). man, D.A., Ed.). Little, Brown, Boston, USA, 117-134 (1994)

FREQUENT SYMPTOMATIC OR ASYMPTOMATIC INFECTIONS MAY HAVE LONG-TERM CONSEQUENCES ON GROWTH AND COGNITIVE DEVELOPMENT

BENJAMIN J.J. MCCORMICK, on behalf of the MAL-ED Network Investigators

Fogarty International Center, National Institutes of Health, Bethesda, MD, USA

THE IMPORTANCE OF DIARRHOEA

Diarrhoeal disease represents a contin- eases (Keusch et al. 2006): acute diar- uing public health challenge. This is rhoeal disease (<7 days), prolonged (7≤ based on two combined issues: in terms days <14) and persistent (≥14 days) of mortality it ranks as the second most (Moore et al. 2010). Mortality and common cause of death in children un- morbidity are associated with higher der 5 years old (Liu et al., 2012) and in rates of diarrhoeal disease episodes and terms of morbidity it has been associ- in particular with persistent diarrhoea. ated with long-term deficits in physical In general, the burden of diarrhoeal dis- (Checkley et al. 2008) and cognitive ease is highest in low income countries, development (Fischer Walker et al., though it remains a universal health 2012). challenge (Black et al., 2003). Diarrhoea as a clinical symptom is de- Research into diarrhoeal disease as fined as increased stool mass due to a non-specific syndrome has a long his- excess fluid in stool (Field, 2003). tory and correspondingly has produced Most commonly this is assessed by the non-specific health interventions (Sedg- frequency of loose stools (by conven- wick and Macnutt, 1910; Blaise et al., tion ≥3 loose stools per day), where 2007). Considerable progress has been “loose” is when a stool takes the shape made in establishing the role of specific of its container. Disruption to normal enteric pathogens as causes of diar- functioning of the gut, where water rhoeal symptoms, but perhaps more would normally be reabsorbed (Woods, important is the lack of detail regarding 1990; Field, 2003), results in loss of entero-pathogens as a cause of the fluids in stool and hence risk of dehy- long-term morbidity in the absence of dration and death. Whilst there are overt diarrhoea. many causes of diarrhoea the most Here we present preliminary find- common – oft assumed cause – is en- ings based on incomplete data from teric infection (WHO, 1988; Thapar The Etiology, Risk Factors and Interac- and Sanderson, 2004). In fact, enteric tions of Enteric Infections and Malnu- pathogens were deemed to be responsi- trition and the Consequences for Child ble for between 12.5-94.3% of the Health and Development Project moderate to severe diarrhoea (Kotloff (MAL-ED). Through the analysis of et al. 2012) in the first year of life in both symptomatic and asymptomatic the recent Global Enteric Multicenter stool samples we are beginning to Study (Kotloff et al. 2013). quantify the role of asymptomatic in- Diarrhoeal symptoms that continue fection as a driver of the long-term over a number of days is classed into morbid consequence of the diarrhoeal one of three syndromic diarrhoeal dis- disease syndrome.

Figure 1: The "vicious cycle" of enteric infection-malnutrition and long-term consequences.

MORTALITY FROM DIARRHOEA AND ENTERO-PATHOGENS

A key detail of the studies that estimate diarrhoea. Inevitably with collated cer- the worldwide burden of diarrhoeal tificates of death there is considerable mortality is that they are based, not on variability in the robustness of report- measures of diarrhoea as the name of ing and specificity of the cause of death the syndrome might imply, but on en- (Murray et al., 2007). For example, one teric infection (Black et al., 2010). classification included in the estimation International Classification of Diseases of the burden of mortality is “presumed certificates (WHO, 1990) are restricted infectious origin” and upon closer in- just to the range of codes correspond- spection, some certificates are based on ing to ‘intestinal infectious disease’ ‘verbal autopsy’ without laboratory (A00-A09). Intestinal infection is one, confirmation, which has been estimated albeit probably the first, among many to have a positive predictive value of possible causes of diarrhoea, but other just 0.39 for diarrhoeal disease (Setel et causes of diarrhoeal symptoms are (for al., 2006). Diarrhoea then, in some these purposes) ignored. The estimated cases at least, is presumed to be the re- burden of mortality is, therefore, spe- sult of enteric infection in the absence cifically an estimate of the mortality of laboratory confirmation. To make attributable to enteric infection. full use of the estimates of mortality, Identifying an enteric infection is diarrhoea must be assumed to be a considerably harder than observing a reliable indicator of enteric infection. visible physiological symptom like

Figure 2: The location of the 8 cohorts of children who make up the MAL-ED study population. PEL Iquitos, Peru; BRF Fortaleza, Brazil; SAV Limpopo, S. Africa; TZH Haydom, Tanzania; PKN Naushahro Feroze, Pakistan; INV Vellore, India; NEB Bhaktapur, Nepal; BGD Mirpur, Bangladesh.

MORBIDITY FROM DIARRHOEA

Only in the most severe cases does and Kaufmann, 2007), thereby impair- diarrhoea lead to dehydration and risk ing a child’s ability to resist (a) recur- of death (Snyder and Merson, 1982; rent infections (Moore et al., 2010) and Glass et al., 1991). Far more common illness (DeBoer et al., 2012) from (g) are self-limiting episodes of diarrhoea increasingly ‘risky’ behaviours/envi- (Baqui et al., 1991). The importance of ronments. Indeed, it is the exploration diarrhoeal disease is not, therefore diar- of this cycle that has formed the essen- rhoea per se, so much as more in- tial hypotheses of the MAL-ED study: sidious and cryptic morbidity that is 1. Infection with specific entero-patho- considered so common as to be gens contributes to stunting, wast- overlooked (de Wit et al., 2001; Mac- ing, and/or micronutrient deficien- Dougall et al., 2008). A considerable cies causing intestinal inflammation body of evidence suggests that enteric and/or by altering the barrier and ad- infection in children can result in a sorptive functions of the gut; and, ‘vicious cycle of poverty’ (Guerrant et 2. The combination of enteric infec- al., 2008) (Figure 1): (a) infection with tions and malnutrition results in enteric pathogens is associated with (b) growth and cognitive impairments impaired gut-function(Salazar-Lindo et in young children and may lead to al., 2004; Petri et al., 2008; Viswana- impaired immunity as measured by than et al., 2009; Costa et al., 2011) responses to childhood vaccines. that can then (c) exacerbate malnutri- This cycle is predicated on the role of tion (Guerrant et al., 2008) and restrict enteric pathogens. Enteric infection re- the processing of nutrients necessary sults both in physical damage to the gut for (d) physical (Mondal et al., 2012) and to a diversion of metabolic activity and (e) cognitive development (Guer- away from growth and towards com- rant et al., 1999; Niehaus et al., 2002; bating infection(s). What is unclear is Lorntz et al., 2006) on top of (f) sup- the relative degree to which infections pressing immune responses (Schaible are reflected by diarrhoeal symptoms.

Table 1: Diagnostic accuracy of diarrhoea as an indicator of enteric infection, using data from all 8 field sites and children from 0-6 months of age ——————————————————————————————————————— Mean CI N ——————————————————————————————————————— Sensitivity 0.18 (0.15,0.21) 6,203 Specificity 0.91 (0.89,0.93) 5,966 Accuracy 0.61 (0.58,0.64) 12,169 ———————————————————————————————————————

THE MAL-ED NETWORK

The Malnutrition and Enteric Infections and assessments, including those for (MAL-ED) network (“Mal-ED”) was gut function, monthly anthropometry, formed to investigate the aetiology, risk nutritional intake, cognitive develop- factors, and interactions of enteric in- ment, micronutrient status, environ- fections and under-nutrition and their mental assessment, and vaccine re- consequences for child health and de- sponse. Follow-up will cease in Febru- velopment (Lang, 2011; The MAL-ED ary 2014, when the last child enrolled Network, 2013a). This prospective at each site will have turned two years field, clinical and laboratory-based ob- of age. servational study follows cohorts of The data presented here are from children from (approximately) birth to the early stages of the MAL-ED project 24 months. The population studied – at the time of writing, data and sam- come from eight sites in countries with ple collection has not yet ended – and a historic high incidence of under- the analyses here represent tentative nutrition and diarrhoeal disease (Figure preliminary findings based on incom- 2). plete data and should be viewed in that Enrolment of new-borns began in context. Given the staggered enrolment November 2009 and ended in February and substantial quantity of data being 2012. Each MAL-ED Network site has generated through longitudinal surveil- enrolled more than 200 study subjects lance, new data are still being added at regular intervals from near birth and at different rates depending both on (prior to 17 days after birth) to 24 the nature of the data and on the site. months of age. To reach each site’s One early finding from MAL-ED is the goal of following 200 children over considerable heterogeneity in the bio- two years and allowing for dropouts, logical data among the field sites – a each site enrolled approximately 10-12 finding that is consistent across many children per month. Active surveillance aspects of the study and requires con- for illness symptoms is accomplished siderable attention and caution both in by visiting each home two times per analysis and translation into general week. Using harmonized protocols, actionable conclusions. each site performs a number of tests

SYMPTOMATIC INFECTION

The assumption that diarrhoea is a use- surveillance. Diarrhoea is logistically ful indicator of enteric infection is im- easy to identify, though with some plicit in the majority of public health variability as to what rate of loose

Monthly Diarrhea BGD 31263126 17291729 INV 31093109 831831 PKN 32463246 24742474 NEB 36833683 11221122 BRF 18011801 5454 PEL 30433043 17161716 SAV 37023702 164164 TZH 31143114 175175 2 1 0 1 2 1.5 0.5 0.5 1.5 Average pathogens/stool

Figure 3: The average number of positive tests for pathogens per stool. Monthly "control" stools are shown in grey and reflected in the hashed bars that overlay the diarrhoeal stools in red. The number of stools tested is given in numbers at the end of each bar. Each bar corresponds to a MAL-ED site (see Figure 1 for site names). This figure is based on stools from children aged 0-12 months.

stools constitutes ‘unusual’ (Kosek et the utility of diarrhoea as a diagnostic al., 2003). What is consistent, though, test for the presence of enteric infection is the use of diarrhoea as an indicator (Sternberg and Hadgu, 2001; Mura- for the pernicious morbidity associated kami, 2010). To simplify this assess- with the syndrome. ment, all the various tests including Experience to date from the MAL- ELISA, microscopy and PCR confir- ED network suggests that diarrhoea is, mation were treated as a binary varia- at best, a poor substitute for laboratory ble such that any test being positive confirmed enteric infection. Stools are was used to indicate infection. What collected: 1) on a monthly schedule to has emerged from a preliminary assess- represent background carriage of path- ment in cohort children from 0-6 ogenic gut flora in the absence of overt months of age is a sobering observation symptoms, and 2) ad hoc when a that the sensitivity and specificity of mother declares that a child has had diarrhoea as an indicator of enteric in- diarrhoea, defined as ≥3 loose stools fection were 0.18 and 0.91 respectively within 24 hours as recommended by (Table 1). The low sensitivity reflects a WHO (2006). large number of asymptomatic infec- A panel of some 57 enteric patho- tions that would be classed as false gens is tested in the MAL-ED study negatives (i.e. no diarrhoeal symptoms including common viruses, bacteria despite at least one positive laboratory and parasites (The MAL-ED Network, test for an enteric pathogen). However, 2013b). Using a generalized estimating the high specificity is the result of the equation to account for repeated sam- small number of cases with a diarrhoeal pling of the same children and to con- (indeed any) stool that did not have a trol for such factors as breast-feeding corresponding positive test for a patho- and age, these tests were used to assess gen.

Indeed the paucity of diarrhoea as a tion. One possible reason for this is that diagnostic indicator of enteric infection symptoms are a product not of the mere is evident in the average number of presence of pathogens – a limitation pathogens detected in stools that have imposed by traditional testing methods been tested so far in MAL-ED (Figure – but of the absolute number of any 3). The number of pathogens detected given pathogen or the quantitative is remarkably similar between monthly combinations of pathogens present. and diarrhoeal stools, with an overall With regard to this latter case, emerg- average 1.7 pathogens identified per ing technologies offer exciting possibil- diarrhoeal stool and 1.3 pathogens per ities in quantifying pathogen burden monthly “control” stool. based on quantitative PCR (Operario Diarrhoea consequently appears to and Houpt, 2011; Platts-Mills et al., be an under-estimate of enteric infec- 2012).

INFECTION AND GROWTH

Enteric pathogens theoretically have a we are assessing the role of those path- twofold impact on growth and develop- ogens that, to date, have been detected ment: first they damage the intestine by in >0.5% of all stools collected (either inducing inflammation, decreasing bar- normal monthly or diarrhoeal samples) rier function, decreasing absorptive ca- amongst the total MAL-ED study pop- pacity or combinations of all three ulation (Figure 4). (Field, 2003); and second, they can di- The change in a child’s linear vert metabolic energy towards an im- growth can then be calculated over a mune response and away from optimal defined period. Lee et al. (2013) re- growth. Evidence of an association be- ported a 9-month period in their final tween enteric infection and deficits in model, though we are analysing our growth or cognition are somewhat am- data over a range of from 2 to 18 biguous – more so considering that months. Over the corresponding inter- many studies have used diarrhoea as a val, the numbers of positive pathogen surrogate of infection. tests are summed. This model, thereby, In a recent study, Lee et al., (2013) attributes a portion of the change in present a simple model of the growth length over a period of time to infec- of a child in relation to Campylobacter tions with a particular enteric pathogen. infection. Specifically, they distinguish In the case of infections that result in between those infections that were transient perturbation in growth, with from symptomatic (diarrhoeal) stools growth deficits that can readily be and those from regularly collected caught up, the effect of infection is asymptomatic (‘control’) stools. This is likely to be noted over short duration an interesting analysis, not least be- lags. Pathogens with longer, more per- cause they make a rare attempt to ex- sistent effects or repeated infections plicitly differentiate the two types of with one or more pathogens are more carriage (another example being likely to be identified in longer inter- changes of weight in relation to Cryp- vals over which their cumulative im- tosporidium infection (Checkley et al., pact becomes apparent. 1997)). Individually, each pathogen is com- We are developing a similar model paratively rare and consequently the though rather than just Campylobacter, impact on the change in height over 9

Figure 4: The percentage of stools with positive test results for pathogens in the MAL-ED study. The horizontal line distinguishes those pathogens (in red) that were found in >0.5% of all stool samples and those that were not (in grey). This figure is based on microbiology data from children 0-24 months old, aggregating the currently available data from all 8 field sites.

months was both negligible and statisti- spite the fact that the effect is greater in cally non-significant (at 5%). In pre- symptomatic infections, the overall im- liminary analyses, there appear to be pact of asymptomatic infection is likely four exceptions: (i) STEC is negatively to be greater as they are considerably associated with a change in height more frequently represented in the data. when detected in diarrhoeal samples (- Similarly, aggregating all infections, 0.34cm/infection/9 months) as are (ii) regardless of provenance, appears to be symptomatic infections with giardia (- statistically significantly associated 0.05cm/infection/9 months), while (iii) with negative growth and is just under EIEC and (iv) Campylobacter in the mean of the effect of the two stool asymptomatic stools appear to be asso- types. ciated with positive (+0.14 cm/infec- It is surprising how few analyses tion/9 months) and negative (-0.03 cm/ have explicitly considered asympto- infection/9 months) growth respective- matic infection in relation to growth, ly. particularly given the number of infec- When pathogens are pooled, how- tions that are unlikely to result in ever, both symptomatic (-0.03 cm/in- symptoms (Hellard et al., 2000) or con- fection/9 months) and asymptomatic tinued infection once symptoms have (-0.01 cm/infection/9 months) infec- resolved (Tangermann et al., 1991). tions were statistically significantly as- Despite this, a model of early MAL-ED sociated with decreased growth. De- data that ignores the type of stool in

favour of a binary ‘infected or not’, is the co-incidence of pathogens may the most informative model based on have greater impact than pathogens in the Akaike Information Criterion (AIC) isolation. From the MAL-ED data, goodness of fit (Akaike, 1974) suggest- aggregating across the field sites, some ing that it is the presence of pathogen 37% of diarrhoeal stools and 17% of and not symptoms that are of greatest the monthly stools have had more than importance to growth. one pathogen detected (most of which What remains to be assessed in this had two positive tests, but a maximum type of analysis is the consequence of of 9 pathogens have been detected in simultaneous co-infections. In theory the same stool so far).

INFECTION AND COGNITION

As well as being required for physical interactions and the immediate environ- growth, nutrients and growth factors ment (Caldwell and Bradley 1984). regulate brain development during foe- Alongside these child metrics, maternal tal and early postnatal life. Over this reasoning (Raven and Court, 1996) and period, the brain demonstrates its great- depressive symptoms (Beusenberg and est plasticity, but also its greatest vul- Orley 1994) are also assessed. nerability to nutrient insufficiency, In much the same fashion as the hence early brain development is cru- physiological stages of the vicious cy- cial for later cognitive development cle, there are conceptual routes by and learning (Rice and Barone, 2000; which symptomatic infection (diar- Thompson and Nelson, 2001). Over rhoea) might influence opportunities 200 million children (worldwide) under for cognitive development beyond the five years old have been estimated to direct interference with nutrition (Mur- fail to reach their full cognitive poten- ray-Kolb and Beard, 2009; Armony- tial, which cascades into poorer school Sivan et al., 2010; Lima et al., 2013) performance and a subsequent reduc- since diarrhoea may also alter infant- tion in economic productivity (Gran- caregiver interactions in terms of both tham-McGregor et al., 2007). Many quality and quantity (Lamb et al., factors contribute to this period of cog- 1999). In addition, nutritional status nitive development including, and most itself may feedback into the infant- relevant here, are nutrient deficiencies caregiver interaction (Gardner et al., and infectious disease (Walker et al., 2005). 2007). Cognition is hard to assess, because Through a suite of tests, MAL-ED it is difficult to define a specific meas- aims to assess various aspects of cogni- urement(s) (Siegler, 1989; Richardson tive development at several time points and Richardson, 2002). This is made over the first two years of life (The more complex because conceptually, MAL-ED Network, 2013c). The devel- cognitive development is a function of opmental assessments are performed at other factors that are themselves am- periodic intervals to capture the pro- biguous constructs (Walker et al., gression of child development across a 2007) – for example parental engage- range of domains including: a general ment, the child’s environment and nu- cognitive assessment (Bayley, 2005), trition. Though intuitively important, memory, motor development, language these constructs are hard to analyse be- (Fenson et al., 2006), learning, temper- cause they are troublesome to quantify ament (Wachs and Desai, 1993), social consistently and to pin down to specific

observable metrics. A solution that is Working Group, 1986), both appear often used is to combine observable significantly related to the Bayley measurements (for example, individual score. items on surveys or responses to ques- The illness and infection variables tionnaires) as multi-dimensional repre- have not yet been demonstrated to be sentations of unobserved ‘latent’ varia- statistically related to the cognitive bles (Hoyle and Smith, 1994; Mac- score, with the exception that the total Callum and Austin, 2000). days of maternally reported fever that An example of one solution to ex- may be associated with a diminished amining these thematic variables is Bayley score. That these variables were given by Amorim et al. (2010). Individ- not related to cognition does not neces- ual measurements were combined into sarily mean that they are not important themes: the sum of sections from the factors. In the context of a structural Home Observation for the Measure- equation (as used by Amorim et al., ment of the Environment (HOME) 2010), measured variables explicitly Scale (Caldwell and Bradley, 1984) feed into latent variables, thus specify- represent “parenting style” and the ing the direct and indirect relationships child’s “environment” and anthropo- between variables. In this particular metric measures as surrogates of “nu- model, there are likely to be strong cor- tritional status”. These unobservable relations between the non-significant constructs were then used to estimate illness factors and the highly signifi- the observed cognitive score measured cant nutrition factors (for example see as the sum of the cognitive and lan- the above association between height guage subscales from the Bayley scales and infection). As such, there are puta- of infant development, 3rd edition tive indirect, though not shown here, (BSID-III) (Bayley, 2005). associations between infection and cog- Amorim et al. (2010) found the nu- nition. It is perhaps noteworthy that the tritional scores to be statistically sig- mean estimate of the association with nificant predictors of the cognitive de- cognition is stronger for the sympto- velopment score. This relationship ap- matic infections, which again may pears also to hold in early MAL-ED (though not demonstrated here) reflect analyses; the enrolment weight and the additional associations expected length-for-age, which is an indicator of between diarrhoeal symptoms and care- long-term nutritional status (WHO giver interactions.

THE LEGACY OF DIARRHOEA

Despite the weakness of diarrhoea as well as loose stools. This confusion al- an indicator of enteric infection and the lows the persistence of language that lack of studies quantifying the impact suggests that diarrhoea (the symptom) of asymptomatic infection, there re- is a useful indicator of the syndrome mains a persistent literature that dis- despite the lack of evidence that “loose cusses diarrhoea as a ‘cause’ of physi- stools” per se are a robust metric of cal and cognitive shortfalls. The reason enteric infection. Indeed this language for this apparent paradox may lie in the under-estimates the importance of the naming of the syndrome. Diarrhoea is largely asymptomatic syndrome. used as a short-hand for both the over- An additive Bayesian network (abn) all “diarrhoeal disease syndrome” (in- (Heckerman et al., 1995; Lewis and dicative of serious gut dysfunction) as McCormick, 2012) using the earliest

Figure 5: The statistical relationships between variables using an additive Bayesian network (abn). MAL-ED data from all 8 field sites for children at 6 months of age were used, with random effects for the field site. Line width indicates the strength of statistical association.

available data was constructed in an the abn avoids the temptation to falla- effort to agnostically identify relation- ciously combine the results of individ- ships among variables being studied in ual regression models that have a single MAL-ED. The abn was used for struc- (presumed independent response varia- ture discovery, identifying a single ble) (Hand et al., 1997). Furthermore it globally optimal directed acyclic graph better reflects the co-dependency of the (DAG) that describes the data (Koivisto many interacting components of a sys- and Sood, 2004; McCormick et al., tem that is by definition a collection of 2013). The abn is a multivariate model, related variables. This allows a statisti- treating each variable as both a poten- cal structure to emerge from the data tial predictor of every other variable themselves that identifies, not only the included in the analysis as well as a immediate relationship between a suite response. Unlike traditional models of variables and a single outcome, but (such as a generalized linear model), also the relationships between the

many variables that are both directly ble data at the time of the analysis, and indirectly associated within a sys- were measures of cognitive develop- tem (Lewis and Ward, 2013). ment and nutrition. The latter would What emerges from MAL-ED data appear central to the manifestation of of the first 6 months of life is that, the diarrhoeal disease – the assumption within these preliminary data, there ap- is that the syndrome impacts the effi- pear to be no direct associations be- cient processing of nutritional input, for tween diarrhoeal symptoms and anthro- example through physical damage to pometric measures (Figure 5). This the intestine, the rapid egestion of food makes sense in the context of the anal- before nutrients are absorbed. Such yses above. What also makes sense is data are being collected within the the presence of indirect statistical asso- MAL-ED study, however, and when ciations between diarrhoeal symptoms, available, will be added to this emerg- enteric pathogens (not distinguishing ing model. Based on other studies, it is between symptomatic and asympto- perhaps unlikely that the impacts on matic) and anthropometry (mean WAZ growth become detectable (above the and maximum length by 6 months old). between-site heterogeneity) until chil- This putative pathway is particularly dren are older (Checkley et al., 1998; interesting because it reflects what is Lima et al., 2000; Moore et al., 2001). widely expected from the diarrhoeal As additional data become available it disease syndrome even if the disease will be possible to reassess this early symptoms are not reliable indicators. observation. What were missing from the availa-

CONCLUSIONS

There is a considerable body of evi- risk of dehydration leading to hospitali- dence supporting a conceptual model in zation or death, symptomatic diarrhoea which growth and cognitive develop- is clearly useful. However, preliminary ment are negatively affected by under- evidence from this longitudinal pro- nutrition. A prime cause, aside from spective study suggests that any enteric poor quality or low quantity of food, is infection– symptomatic or not – con- the interference with optimal gut func- tributes to decreased growth. The im- tion by enteric infection. Within this pact of each infection is relatively cycle, enteric infection is frequently small and slow to emerge, with the sig- referred to as diarrhoeal disease, how- nificance of each infection becoming ever this belies the importance of more apparent as the lag since the in- asymptomatic infection. fection increases. These results may Enteric infection resulting in diar- change when examining the complete rhoea is relatively uncommon com- data from MAL-ED and in the broader pared to an overwhelming number of context of growth outcomes over the asymptomatic infections or continued entire 24 months of the study are mod- carriage (whether or not deleterious) elled. observed in the settings studied in Likewise, any association between MAL-ED. The question is whether or enteric infection and cognitive develop- not symptoms are important contribu- ment may become clearer as more data tors to growth outcomes? As indicators become available from older children. of gut dysfunction and to indicate the Even so, early data from MAL-ED hint

that the role that entero-pathogens play (the symptom) is a reliable indicator of is indirect - acting on nutrition and gut infection and/or dysfunction. In thereby influencing cognitive devel- fact, enteric infection is what is com- opment. Such a hierarchy of interac- monly understood when referring to tions is likely to be most clearly identi- “diarrhoea” and that seems to, based on fied by accounting for the many inter- preliminary observations noted here, linked associations in the wider syn- have long-term consequences for drome. growth and cognitive development, The diarrhoeal disease syndrome is whether or not those infections are something of a misnomer that has dis- associated with overt diarrhoea. torted the expectations that diarrhoea

ACKNOWLEDGMENTS

The Etiology, Risk Factors and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development Project (MAL-ED) is carried out as a collaborative project supported by the Bill and Melinda Gates Foundation, the Foundation for the NIH and the National Institutes of Health/Fogarty International Center. The authors thank the staff and participants of the MAL-ED Network Project for their important contributions.

LITERATURE

Akaike, H.: A new look at the statistical model Beusenberg, M. and Orley, J.H.: Users guide to identification IEEE Trans. Autom. Contr, the self reporting questionnaire (SRQ). 19, 716-723 (1974). WHO, Geneva, Switzerland (1994). Amorim, L.D., Fiaccone, R.L., Santos, C.A., Black, R.E., Cousens, S., Johnson, H.L., Lawn, Santos, T.N., Moraes, L.T., Oliveira, N.F., J.E., Rudan, I., Bassani, D.G., Jha, P., Barbosa, S.O., Santos, D.N., Santos, L.M., Campbell, H., Walker, C.F., and Cibulskis, Matos, S.M., and Barreto, M.L.: Structural R.: Global, regional, and national causes of equation modeling in epidemiology. Cad. child mortality in 2008: a systematic anal- Saúde Pública, 26, 2251-2262 (2010). ysis. Lancet, 375, 1969-1987 (2010). Armony-Sivan, R., Kaplan-Estrin, M., Jacob- Black, R.E., Morris, S.S., and Bryce, J.: Where son, S.W., and Lozoff, B.: Iron-deficiency and why are 10 million children dying anemia (IDA) in infancy and mother-infant every year? Lancet, 361, 2226-2234 interaction during feeding. J. Dev. Behav. (2003). Pediatr. 31, 326-332 (2010). Blaise, H., Yongsi, N., and Dovie, D.B.K.: Di- Baqui, A.H., Black, R.E., Yunus, M.D., Hoque, arrheal diseases in the history of public A.R.., Chowdhury, H.R., and Sack, R.B.: health. Arch. Med. Res. 38, 159-163 Methodological issues in diarrhoeal dis- (2007). eases epidemiology: definition of diar- Caldwell, B.M. and Bradley, R.H.: Home ob- rhoeal episodes. Int. J. Epidemiol., 20, servation for measurement of the environ- 1057-1063 (1991). ment. University of Arkansas, Little Rock, Bayley, N.: Bayley Scales of Infant and Tod- AR, USA (1984). dler Development - Third edition: Admin- Checkley, W., Buckley, G., Gilman, R.H., As- istration Manual. San Antonio, TX, USA. sis, A.M.O., Guerrant, R.L., Morris, S.S., Harcourt Assessment (2005). Molbak, K., Valentiner-Branth, P., Lanata,

C.F., Black, R.E., and Childhood Malnutri- Fischer Walker, C.L., Lamberti, L., Adair, L., tion and Infection Network: Multi-country Guerrant, R.L., Lescano, A.G., Martorell, analysis of the effects of diarrhoea on R., Pinkerton, R.C., and Black, R.E.: Does childhood stunting. Int. J. Epidemiol. 37, childhood diarrhea influence cognition be- 816-830 (2008). yond the diarrhea-stunting pathway? PLoS Checkley, W., Epstein, L.D., Gilman, R.H., ONE 7, e47908 (2012). Black, R.E., Cabrera, L. and Sterling, C.R.: Gardner, J.M.M., Powell, C.A., Baker-Hen- Effects of Cryptosporidium parvum infec- ningham, H., Walker, S.P., Cole, T.J., and tion in Peruvian children: growth faltering Grantham-McGregor, S.M.: Zinc supple- and subsequent catch-up growth. Am. J. mentation and psychosocial stimulation: Epidemiol. 148, 497-506 (1998). effects on the development of undernour- Checkley, W., Gilman, R.H., Epstein, L.D., ished Jamaican children. Am. J. Clin. Nutr. Suarez, M., Diaz, J.F., Cabrera, L., Black, 82, 399-405 (2005). R.E., and Sterling, C.R.: Asymptomatic Glass, R.I., Lew, J.F., Gangarosa, R.E., Le- and symptomatic cryptosporidiosis: their Baron, C.W., and Ho, M.S.: Estimates of acute effect on weight gain in Peruvian morbidity and mortality rates for diarrheal children. Am. J. Epidemiol. 145, 156-163 diseases in American children. J. Pediatr. (1997). 118, S27-S33 (1991). Costa, L.B., JohnBull, E.A., Reeves, J.T., Se- Grantham-McGregor, S., Cheung, Y.B., Cueto, villeja, J.E., Freire, R.S., Hoffman, P.S., S., Glewwe, P., Richter, L., and Strupp, B.: Lima, A.A.M., Oriá, R.B., Roche, J.K., Developmental potential in the first 5 years Guerrant, R.L., and Warren, C.A.: Cryp- for children in developing countries. Lan- tosporidium-malnutrition interactions: mu- cet 369, 60-70 (2007). cosal disruption, cytokines, and TLR sig- Guerrant, D.I., Moore, S.R., Lima, A.A., Pat- naling in a weaned murine model. J. Para- rick, P.D., Schorling, J.B., and Guerrant, sitol 97, 1113-1120 (2011). R.L.: Association of early childhood diar- DeBoer, M.D., Lima, A.A.M., Oría, R.B., rhea and cryptosporidiosis with impaired Scharf, R.J., Moore, S.R., Luna, M.A., and physical fitness and cognitive function Guerrant, R.L.: Early childhood growth four-seven years later in a poor urban com- failure and the developmental origins of munity in northeast Brazil. Am. J. Trop. adult disease: do enteric infections and Med. Hyg. 61, 707-713 (1999). malnutrition increase risk for the metabolic Guerrant, R.L., Oriá, R.B., Moore, S.R., Oriá, syndrome? Nutr. Rev. 70, 642-653 (2012). M.O., and Lima, A.A.: Malnutrition as an de Wit, M.A., Kortbeek, L.M., Koopmans, enteric infectious disease with long-term M.P., de Jager, C.J., Wannet, W.J., Bar- effects on child development. Nutr. Rev. telds, A.I., and van Duynhoven, Y.T.: A 66, 487-505 (2008). comparison of gastroenteritis in a general Hand, D.J., Mcconway, K.J., and Stanghellini, practice-based study and a community- E.: Graphical models of applicants for based study. Epidemiol. Infect. 127, 389- credit. IMA J. Manag. Math. 8, 143-155 397 (2001). (1997). Fenson, L., Marchman, V.A., Thal, D., Bates, Heckerman, D., Geiger, D., and Chickering, E., Dale, P.S., and Reznik, J.S.: The Mac- D.M.: Learning Bayesian networks: the Arthur-Bates communicative development combination of knowledge and statistical inventories user’s guide and technical data. Machine Learning 20, 197-243 manual, 2nd revised edition. Brookes Pub- (1995). lishing Co., Baltimore, MD, USA (2006). Hellard, M.E., Sinclair, M.I., Hogg, G.G., and Field, M.: Intestinal ion transport and the path- Fairley, C.K.: Prevalence of enteric patho- ophysiology of diarrhea. J. Clin. Invest. gens among community based asympto- 111, 931-943 (2003). matic individuals. J. Gastroenterol. Hepa-

tol. 15, 290-293 (2000). M.J., Akinsola, A., Mandomando, I., Hoyle, R.H. and Smith, G.T.: Formulating clin- Nhampossa, T., Acácio, S., Biswas, K., ical research hypotheses as structural equa- O’Reilly, C.E., Mintz, E.D., Berkeley, tion models: a conceptual overview. J. L.Y., Muhsen, K., Sommerfelt, H., Rob- Consult. Clin. Psychol. 62, 429-440 ins-Browne, R.M., and Levine, M.M.: Bur- (1994). den and aetiology of diarrhoeal disease in Keusch, G.T., Fontaine, O., Bhargava, A., infants and young children in developing Boschi-Pinto, C., Bhutta, Z.A., Gotuzzo, countries (the Global Enteric Multicenter E., Rivera, J., Chow, J., Shahid-Salles, S., Study, GEMS): a prospective, case-control and Laxminarayan, R.: Diarrheal Diseases. study. Lancet 382, 209-222 (2013). In: Disease Control Priorities in Develop- Lamb, M., Hwang, C., Ketterlinus, R., and Fra- ing Countries, 2nd edition (Jamison, D.T., casso, M.: Parent-child relationships: de- Breman, J.G., Measham, A.R., Alleyne, velopment in the context of the family. In: G., Claeson, M., Evans, D.B., Jha, P., Developmental psychology: an advanced Mills A., and Musgrove, P., Eds). World textbook, 4th edition (Bornstein, M.H. and Bank, Washington, DC, USA, 371-387 Lamb, M.E., Eds.). Lawrence Erlbaum As- (2006). sociates, Mahwah, NJ, USA, 411-450 Koivisto, M. and Sood, K.: Exact Bayesian (1999). structure discovery in Bayesian networks. Lang, D.: The Mal-Ed Project: Deciphering the Journal of Machine Learning Research 5, relationships among normal gut flora, en- 549-573 (2004). teric infection and malnutrition and their Kosek, M., Bern, C., and Guerrant, R.L.: The association with immune response to vac- global burden of diarrhoeal disease, as esti- cines. In: Development of strategies to mated from studies published between overcome barriers to effective mucosal im- 1992 and 2000. Bull. World Health Org. munization of infants in developing coun- 81, 197-204 (2003). tries. Old Herborn University Monograph Kotloff, K.L., Blackwelder, W.C., Nasrin, D., 24 (Heidt, P.J., Walker, R.I., and Rusch, Nataro, J.P., Farag, T.H., van Eijk, A., V., Eds.). Old Herborn University Founda- Adegbola, R.A., Alonso, P.L., Breiman, tion, Herborn, Germany, 73-82 (2011). R.F., Golam Faruque, A.S., Saha, D., Sow, Lee, G., Pan, W., Peñataro Yori, P., Paredes S.O., Sur, D., Zaidi, A.K.M., Biswas, K., Olortegui, M., Tilley, D., Gregory, M., Panchalingam, S., Clemens, J.D., Cohen, Oberhelman, R., Burga, R., Chavez, C.B., D., Glass, R.I., Mintz, E.D., Sommerfelt, and Kosek, M.: Symptomatic and asymp- H., and Levine, M.M.: The Global Enteric tomatic Campylobacter infections associ- Multicenter Study (GEMS) of diarrheal ated with reduced growth in Peruvian chil- disease in infants and young children in dren. PLoS Negl. Trop. Dis. 7, e2036 developing countries: epidemiologic and (2013). clinical methods of the case/control study. Lewis, F.I. and McCormick, B.J.J.: Revealing Clin. Infect. Dis. 55, S232-S245 (2012). the complexity of health determinants in Kotloff, K.L., Nataro, J.P., Blackwelder, W.C., resource-poor settings. Am. J. Epidemiol. Nasrin, D., Farag, T.H., Panchalingam, S., 176, 1051-1059 (2012). Wu, Y., Sow, S.O., Sur, D., Breiman, R.F., Lewis, F. and Ward, M.: Improving epidemio- Faruque, A.S., Zaidi, A.K., Saha, D., logic data analyses through multivariate re- Alonso, P.L., Tamboura, B., Sanogo, D., gression modelling. Emerg. Themes Epi- Onwuchekwa, U., Manna, B., Rama- demiol. 10, 4 (2013). murthy, T., Kanungo, S., Ochieng, J.B., Lima, A.A.M., Kvalsund, M.P., de Souza, Omore, R., Oundo, J.O., Hossain, A., Das, P.P.E., Figueiredo, I.L., Soares, A.M., S.K., Ahmed, S., Qureshi, S., Quadri, F., Mota, R.M.S., Lima, N.L., Pinkerton, Adegbola, R.A., Antonio, M., Hossain, R.C., Patrick, P.P., Guerrant, R.L., and

Oria, R.B.: Zinc, vitamin A, and glutamine ternal malnutrition to infant malnutrition in supplementation in Brazilian shantytown Bangladesh. Clin. Infect. Dis. 54, 185-192 children at risk for diarrhea results in sex- (2012). specific improvements in verbal learning. Moore, S.R., Lima, A.A.M., Conaway, M.R., Clinics 68, 351-358 (2013). Schorling, J.B., Soares, A.M., and Guer- Lima, A.A., Moore, S.R., Barboza, M.S., Jr, rant, R.L.: Early childhood diarrhoea and Soares, A.M., Schleupner, M.A., Newman, helminthiases associate with long-term lin- R.D., Sears, C.L., Nataro, J.P., Fedorko, ear growth faltering. Int. J. Epidemiol. 30, D.P., Wuhib, T., Schorling, J.B., and Guer- 1457-1464 (2001). rant, R.L.: Persistent diarrhea signals a Moore, S.R., Lima, N.L., Soares, A.M., Oria, critical period of increased diarrhea bur- R.B., Pinkerton, R.C., Barrett, L.J., Guer- dens and nutritional shortfalls: a prospec- rant, R.L., and Lima, A.A.M.: Prolonged tive cohort study among children in north- episodes of acute diarrhea reduce growth eastern Brazil. J. Infect. Dis. 181, 1643- and increase risk of persistent diarrhea in 1651 (2000). children. Gastroenterology 139, 1156-1164 Liu, L., Johnson, H.L., Cousens, S., Perin, J., (2010). Scott, S., Lawn, J.E., Rudan, I., Campbell, Murakami, P.: Bayesian estimation of H., Cibulskis, R., Li, M., Mathers, C., and sensitivity and specificity. URL Black, R.E.: Global, regional, and national http://www.biostat.jhsph.edu/~pmurakam/ causes of child mortality: an updated sys- epi_tests.html (2010). tematic analysis for 2010 with time trends Murray, C.J.L., Lopez, A.D., Barofsky, J.T., since 2000. Lancet 379, 2151-2161 (2012). Bryson-Cahn, C., and Lozano, R.: Estimat- Lorntz, B., Soares, A.M., Moore, S.R., Pinker- ing population cause-specific mortality ton, R., Gansneder, B., Bovbjerg, V.E., fractions from in-hospital mortality: vali- Guyatt, H., Lima, A.M., and Guerrant, dation of a new method. PLoS Med. 4, R.L.: Early childhood diarrhea predicts im- e326 (2007). paired school performance. Pediatr. Infect. Murray-Kolb, L.E. and Beard, J.L.: Iron defi- Dis. J. 25, 513-520 (2006). ciency and child and maternal health. Am. MacCallum, R.C. and Austin, J.T.: Applica- J. Clin. Nutr. 89, 946S-950S (2009). tions of structural equation modeling in Niehaus, M.D., Moore, S.R., Patrick, P.D., psychological research. Annu. Rev. Psy- Derr, L.L., Lorntz, B., Lima, A.A., and chol. 51, 201-226 (2000). Guerrant, R.L.: Early childhood diarrhea is MacDougall, L., Majowicz, S., Dore, K., Flint, associated with diminished cognitive func- J., Thomas, K., Kovacs, S., and Sockett, tion 4 to 7 years later in children in a P.: Under-reporting of infectious gastro- northeast Brazilian shantytown. Am. J. intestinal illness in British Columbia, Can- Trop. Med. Hyg. 66, 590-593 (2002). ada: who is counted in provincial com- Operario, D.J. and Houpt, E.: Defining the municable disease statistics? Epidemiol. causes of diarrhea. Curr. Opin. Infect. Dis. Infect. 136, 248-256 (2008). 24, 464-471 (2011). Mal-ED: URL http://mal-ed.fnih.org/ Petri, W.A., Miller, M., Binder, H.J., Levine, McCormick, B.J.J., Sanchez-Vazquez, M.J., M.M., Dillingham, R., and Guerrant, R.L.: and Lewis, F.I.: Using Bayesian networks Enteric infections, diarrhea, and their im- to explore the role of weather as a potential pact on function and development. J. Clin. determinant of disease in pigs. Prev. Vet. Invest. 118, 1277-1290 (2008). Med. 110, 54-63 (2013). Platts-Mills, J.A., Operario, D.J., and Houpt, Mondal, D., Minak, J., Alam, M., Liu, Y., Dai, E.R.: Molecular diagnosis of diarrhea: cur- J., Korpe, P., Liu, L., Haque, R., and Petri, rent status and future potential. Curr. In- W.A. Jr.: Contribution of enteric infection, fect. Dis. Rep. 14, 41-46 (2012). altered intestinal barrier function, and ma- Raven, J.C. and Court, J.H.: Manual for Ra-

ven’s progressive matrices and vocabulary ificity and coverage properties of the con- scales: standard progressive matrices. Ox- fidence intervals. Stat. Med. 20, 1529-1539 ford Psychologists Press, Oxford, UK (2001). (1996). Tangermann, R.H., Gordon, S., Wiesner, P., Rice, D. and Barone, S.: Critical periods of and Kreckman, L.: An outbreak of cryptos- vulnerability for the developing nervous poridiosis in a day-care center in Georgia. system: evidence from humans and animal Am. J. Epidemiol. 133, 471-476 (1991). models. Environ. Health Perspect. 108, Thapar, N. and Sanderson, I.R.: Diarrhoea in 511-533 (2000). children: an interface between developing Richardson, D.K. and Richardson, K.: Models and developed countries. Lancet 363, 641- of cognitive development. Psychology 653 (2004). Press, Hove, UK (2002). The MAL-ED Network. (2013a) The MAL-ED Salazar-Lindo, E., Allen, S., Brewster, D.R., Project: A multinational and multidiscipli- Elliott, E.J., Fasano, A., Phillips, A.D., nary approach to understand the relation- Sanderson, I.R., Tarr, P.I., and Latin ship between malnutrition, enteric infec- American Society for Pediatric Gastro- tions gut physiology and immune re- enterology, Hepatology and Nutrition: sponses in infants/children in resource Intestinal infections and environmental en- poor environments. Manuscript in prepara- teropathy: Working Group report of the se- tion. cond World Congress of Pediatric Gastro- The MAL-ED Network. (2013b) Microbiologic enterology, Hepatology, and Nutrition. J. methods utilized in the MAL-ED cohort. Pediatr. Gastroenterol. Nutr. 39, Suppl. 2, Manuscript in preparation. S662-669 (2004). The MAL-ED Network. (2013c) The MAL-ED Schaible, U.E. and Kaufmann, S.H.E.: Mal- Research Network: Methods and lessons nutrition and infection: complex mecha- learned when assessing early child devel- nisms and global impacts. PLoS Med. 4, opment and caregiving mediators in infants e115 (2007). and young children in 8 low- and middle- Sedgwick, W.T. and Macnutt, J.S.: On the income countries. Manuscript in prepara- Mills-Reincke phenomenon and Hazen’s tion. theorem concerning the decrease in mortal- Thompson, R.A. and Nelson, C.A.: Develop- ity from diseases other than typhoid fever mental science and the media. Early brain following the purification of public water- development. Am. Psychol. 56, 5-15 supplies. J. Infect. Dis. 7, 489-564 (1910). (2001). Setel, P.W., Whiting, D.R., Hemed, Y., Chan- Viswanathan, V.K., Hodges, K., and Hecht, G.: dramohan, D., Wolfson, L.J., Alberti, Enteric infection meets intestinal function: K.G.M.M., and Lopez, A.D.: Validity of how bacterial pathogens cause diarrhoea. verbal autopsy procedures for determining Nat. Rev. Microbiol. 7, 110-119 (2009). cause of death in Tanzania. Trop. Med. Int. Wachs, T.D. and Desai, S.: Parent-report Health 11, 681-696 (2006). measures of toddler temperament and at- Siegler, R.S.: Mechanisms of cognitive devel- tachment: their relation to each other and opment. Annu. Rev. Psychol. 40, 353-379 to the social microenvironment. Infant Be- (1989). hav. Dev. 16, 391-396 (1993). Snyder, J.D. and Merson, M.H.: The magnitude Walker, S.P., Wachs, T.D., Meeks Gardner, J., of the global problem of acute diarrhoeal Lozoff, B., Wasserman, G.A., Pollitt, E., disease: a review of active surveillance and Carter, J.A.: Child development: risk data. Bull. World Health Organ. 60, 605- factors for adverse outcomes in developing 613 (1982). countries. Lancet 369, 145-157 (2007). Sternberg, M.R. and Hadgu, A.: A GEE ap- WHO: Persistent diarrhoea in children in de- proach to estimating sensitivity and spec- veloping countries: memorandum from a

WHO meeting. Bull. World Health Org. WHO Working Group. Use and interpretation 66, 709-717 (1988). of anthropometric indicators of nutritional WHO. International Classification of Diseases status. Bull. World Health Org. 64, 929- (ICD). URL http://www.who.int/ 941 (1986). classifications/icd/en/ Woods, T.A.: Diarrhea. In: Clinical methods: WHO: WHO Child Growth Standards: meth- the history, physical, and laboratory exam- ods and Development. Length/height-for- inations, 3rd edition (Walker, H.K., Hall, age, weight-for-age, weight-for-length, W.D., and Hurst, J.W., Eds.). Butter- weight-for-height and body mass index- worths, Boston, MA, USA (1990). for-age. World Health Organization, Ge- neva, Switzerland (2006).

UTILIZATION OF THE US DEPARTMENT OF DEFENSE MEDICAL SURVEILLANCE SYSTEM AND SERUM REPOSITORY FOR ASSESSING LONG-TERM HEALTH CONSEQUENCES OF ACUTE ENTERIC INFECTION

CHAD K. PORTER, ASHLEY N. ALCALA, and MARK S. RIDDLE

Enteric Diseases Department, Naval Medical Research Center, Silver Spring, MD, USA

INTRODUCTION TO DODSR AND DMSS

Second only to randomized controlled The active duty US military represents trials (RCTs), prospectively designed a large, uniquely healthy, young, active cohort studies represent the best way to subset of the general US population explore the association between expo- and is one amenable to epidemiologic sure and outcome in human popula- studies. In 1986, the US Army tions. However, given practical limita- established a data repository with the tions of sample size requirements, the stated purpose of supporting HIV high frequency of lost-to-follow-up, the screening, care and research (Ruber- relative rarity of some outcomes and tone et al., 2002). Seven years later, the other funding and logistical constraints, scope of this system expanded to in- such epidemiologic investigation is clude all illness and injuries of im- rarely feasible. In absence of this study portance for public health and/or the design, one begins to explore the po- DoD. Additionally, self-completed pre- tential for other, population-based data and post-deployment health surveys sets from which cohort studies can be were initiated in the 1990s to document reconstructed. One such available data exposures during deployment as well source is available within the confines as general changes in health subsequent of the United States Department of De- to deployment (Moore et al., 2010). fense (US DoD). The objective of this These data are linked within the De- paper is to review aspects and elements fense Medical Surveillance System by of the system maintained by the US social security number to longitudi- DoD and how they can be utilized to nal demographic, medical encounter, explore novel associations including vaccination and deployment data on all the persisting consequences of intesti- active duty service members. nal infection.

———————————————————————————————————— Disclaimer: The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, nor the U.S. Government. This is a US Government work. There are no restrictions on its use. There were no financial conflicts of interests among any of the authors. The study protocol was approved by the Naval Medical Research Center Institutional Review Board in compliance with all applicable Federal regulations governing the protection of human subjects. This study was conducted under support of the Military Infectious Disease Research Program.

Copyright Statement: Authors are employees of the U.S. Government and military service members. This work was prepared as part of official duties. Title 17 U.S.C. §105 provides that ‘Copy- right protection under this title is not available for any work of the United States Government.’ Title 17 U.S.C. §101 defines a U.S. Government work as a work prepared by a military service member or employee of the U.S. Government as part of that person’s official duties.

Figure 1: Estimated number of serum samples in DoD serum repository.

Almost simultaneously to the establish- sources with fairly limited access, it is ment of the serum repository, in 1985, useful to explore how these data and the US DoD implemented a Human samples have begun to be utilized to Immunodeficiency Virus (HIV) screen- explore the association between acute ing program in which remnants of sera infection and chronic gastrointestinal collected from recruits that tested nega- (GI) sequelae, what additional ques- tive for HIV was archived (Moore et tions are pending and how archives al., 2010). Subsequently, the sera were within the US DoD can be utilized to transported to a single facility and the answer some of these questions. Addi- serum repository was born. In the tionally, full acknowledgement and 1990s, in response to illnesses affiliated evaluation of the strengths and limita- with participation in the first Gulf War, tions of the DMSS and the Department the DoD expanded serologic screening of Defense serum Repository (DODSR) by implementing pre- and post-deploy- related to assessing the link between ment blood draws for serum archival to acute enteric infection and post-infec- identify potentially important expo- tious sequelae is necessary to identify sures during deployment. The serum knowledge gaps that need filling with repository has grown exponentially supplementary data and other data/re- since that time and at present houses pository systems available. The objec- over 56 million serum samples from tive of this report is to summarize the potential recruits, active duty and for- key findings to date utilizing the DMSS/ mer active duty service members (Fig- DODSR, to outline expanded areas of ure 1). These serum samples can be research in which they can be utilized linked (via social security number) to to explore novel hypotheses on infec- data within the Defense Medical Sur- tion and chronic GI sequelae and begin veillance System (DMSS). to frame the mechanism(s) by which While these are invaluable re- these studies could be conducted.

Table 1: List of gastrointestinal conditions that do not meet the standard for appointment, enlistment or induction with relevant ICD-9 and CPT codes (DOD instruction 6130.03) ——————————————————————————————————————————————————————————— Site/Condition: Exclusionary conditions: ——————————————————————————————————————————————————————————— Oesophageal Disease: (1) Current or history of oesophageal disease (530.0-530-9), including but not limited to ulceration, varices, fistula, or achalasia. (2) Gastro-Oesophageal Reflux Disease (GERD) (530.81), with complications. (a) Stricture or B-ring. (b) Dysphagia. (c) Recurrent symptoms or esophagitis despite maintenance medication. (d) Barrett’s esophagitis. (e) Extra-oesophageal complications; reactive airway disease; recurrent sinusitis or dental complications. (3) History of surgical correction (fundoplication or dilation) for GERD within 6 months (45.89). (4) Current or history of dysmotility disorders, to include diffuse oesophageal spasm, nutcracker oesophagus, non- specific motility disorder, and achalasia. (5) Eosinophilic oesophagitis. (6) Other oesophageal strictures, for example lye or other caustic ingestion ——————————————————————————————————————————————————————————— Stomach and Duodenum: (1) Current dyspepsia requiring medication; or history of dyspepsia lasting 3 or more consecutive months and requiring medication within the preceding 12 months. (2) Gastric or duodenal ulcers: (a) Current ulcer or history of treated ulcer within the last 3 months. (b) Recurrent or complicated by bleeding, obstruction, or perforation within preceding 5 years confirmed by endoscopy. (3) History of surgery for peptic ulceration or perforation (533.0-599.9). (4) History of gastroparesis. (5) History of bariatric surgery of any type (e.g., lap-band or gastric bypass surgery for weight loss). (6) History of gastric varices. ——————————————————————————————————————————————————————————— Small and Large Intestine: (1) Current or history of inflammatory bowel disease, including but not limited to indeterminate (558.9), Crohn’s disease (555), ulcerative colitis (556), or ulcerative proctitis (556.2). (2) Current infectious colitis not otherwise specified (009.1). (3) Current or history of intestinal malabsorption syndromes (579.9), including but not limited to celiac sprue, pancreatic insufficiency, post-surgical and idiopathic (579). Lactase deficiency does not meet the standard only if of sufficient severity to require frequent intervention, or to interfere with normal function. (4) Current or history of gastrointestinal functional and motility disorders within the past 2 years, including but not limited to pseudo-obstruction, megacolon, history of volvulus, or chronic constipation (564.0) and or diarrhoea (787.91), regardless of cause, persisting or symptomatic in the past 2 years. ———————————————————————————————————————————————————————————

Table 1 (continued): List of gastrointestinal conditions that do not meet the standard for appointment, enlistment or induction with relevant ICD-9 and CPT codes (DOD instruction 6130.03) ——————————————————————————————————————————————————————————— Site/Condition: Exclusionary conditions: ——————————————————————————————————————————————————————————— Small and Large Intestine: (5) History of gastrointestinal bleeding (578), including positive occult blood (792.1), if the cause has not been corrected. Meckel’s diverticulum (751.0), if surgically corrected more than 6 months prior DOES meet the standard. (6) Current or history of irritable bowel syndrome (564.1) of sufficient severity to require frequent intervention or prescription medication or to interfere with normal function. (7) History of bowel resection (CPT 44202-44203). (8) Current or history of symptomatic diverticular disease of the intestine (562). (9) Personal or family history of familial adenomatous polyposis syndrome or hereditary non-polyposis colon cancer syndrome. ——————————————————————————————————————————————————————————— Hepatic-Biliary Tract: (1) Current acute or chronic hepatitis, hepatitis carrier state (070), hepatitis in the preceding 6 months or persistence of symptoms after 6 months, or objective evidence of impairment of liver function. (2) Current or history of cirrhosis (571), hepatic cysts (573.8), abscess (572.0), or sequelae of chronic liver disease (571.3). (3) Current or history of symptomatic cholecystitis (575.10), unless successfully surgically corrected; postcholecystectomy syndrome; or other disorders of the gallbladder and biliary system (576). Cholecystectomy DOES meet the standard if performed more than 6 months prior to examination and patient remains asymptomatic. Endoscopic procedure to correct choledocholithiasis, if performed more than 6 months prior to examination and patient remains asymptomatic, MAY meet the standard. (4) History of sphincter of Oddi dysfunction. (5) Choledochocyst. (6) Primary biliary cirrhosis or primary sclerosing cholangitis. (7) Current or history of pancreatitis, acute (577.0) or chronic (577.1). (8) Pancreatic cyst. (9) History of pancreatic surgery. (10) Current or history of metabolic liver disease, including but not limited to hemochromatosis (275.0), Wilson’s disease (275.1), or alpha-1 anti-trypsin deficiency (273.4). Gilbert’s syndrome DOES meet the standard. (11) Current enlargement of the liver from any cause (789.1). ——————————————————————————————————————————————————————————— Anorectal: (1) Current anal fissure or anal fistula (565). (2) Current or history of anal or rectal polyp (569.0), prolapse (569.1), stricture (569.2), or faecal incontinence (787.6), within the last 2 years. History of removal of juvenile or inflammatory polyp DOES meet the standard. (3) Current haemorrhoid (internal or external), when large, symptomatic, or with a history of bleeding (455) within the last 60 days. ———————————————————————————————————————————————————————————

POPULATION DEMOGRAPHICS

One of the most unique attributes of the an effect modifier of the risk associated Department of Defense repository is with antecedent exposure and studies the population from which the data are of active duty personnel have included obtained. As of February 28, 2013, age as a matching variable between there were 1.39 million active duty ser- cases and controls. However, matching vice members with an average age of on this potential important covariate 28.8 years (Defense Manpower Data precludes any assessment of its impact Center; DMDC). This very healthy, on the association between infection physically active subset of the general and outcome among this uniquely population is one in which assessment young population. of the risk of chronic GI sequelae fol- The overwhelming majority (85.4% lowing from infection is beneficial as it as of Sept 2012) (DMDC) of active is generally free of oftentimes con- duty military personnel are male com- founding co-morbidities. Specifically, pared to an estimated 50.2% of the gen- all United States Department of De- eral US population (2010 US census). fense recruits are subject to detailed Importantly, being of female gender physical exams and medical history has been repeatedly associated with an prior to being processed for recruit increased risk of FGD including IBS, training. The specific requirements are functional dyspepsia and functional outlined in the Department of Defense constipation (Mearin, 2011). Incidence Medical Standards for Appointment, of these functional outcomes in mili- Enlistment, or Induction. The current tary populations based on these data is list of gastrointestinal conditions that lower than what has been reported else- do not meet the standard for appoint- where using similar methodology. As ment, enlistment or induction, updated such, studies should ensure matching 28 April 2010 under Department of De- on gender for assessment of most PI- fense Instruction 6130.03 are included FGD. Furthermore, stress has a role in in Table 1 and highlight the differences IBS risk and may confound or modify between this and the general popula- the effect of enteric infection on FGD tion. Conversely, studies in this popula- risk in a population that is frequently tion often lack external validity to the deployed into what are likely broadly general population which may limit considered to be stressful environments extrapolation of data to other dissimilar (Drossman, 2011). Additionally, gen- populations. der has been indicated as a risk factor Age is a known risk factor for for more pathological conditions such many of the chronic GI outcomes cur- as IBD and coeliac disease (Green and rently linked to IGE including IBS, Cellier, 2007; Karlinger et al., 2000). IBD and coeliac disease. With the Despite the potential lack of external overwhelming majority (approximately validity, numerous large cohort studies 80%) of active duty personnel aged 18- have elucidate novel findings of public 35 (DMDC), estimates of relative risks health importance; examples include may be confounded by age. Specifi- the Framingham Heart Study (Dawber cally, risk of PI-IBS has been shown to et al., 1951) and those of smoking and decrease with increasing age (Marshall mortality among a cohort of physicians et al., 2006; Neal et al., 1997). Studies (Doll and Hill, 1954; Doll and Pike, to date are too limited to assess 1972). whether age serves as a confounder or

Figure 2: Map of the Number of Military Treatment Facilities in the US and Puerto Rico.

ACCESS TO CARE

One of the factors limiting the utiliza- treatment facilities (MTFs) (Figure 2) tion of medical encounter databases in or the 31 locations internationally other populations is variability in ac- (http://www.tricare.mil/mtf/main1.aspx; cess to medical care. For example, accessed 5/10/13). In a reference pop- studies within large provider networks ulation followed for an average of 3.7 (Kaiser, Aetna, etc.) within the United years, the median number of outpatient States have frequently reported find- medical encounters per subject was 29 ings within their covered population; (interquartile range: 14, 56) for just un- however, it is unclear if the covered der an estimated 8 outpatient medical and non-covered populations in those encounters annually per active duty regions are comparable or whether cov- service member (Porter et al., 2012). ered individuals sought care outside of These encounters included general their provider network (Koebnick et al., health physicals, vaccinations, proce- 2012; Stephenson et al., 2005). In con- dures, illness and injuries handled in an trast, active duty military personnel ambulatory setting. As expected, the have equal and unfettered access to number of inpatient visits was much healthcare at any one of the 184 US- lower (annually <1 per service mem- (and Puerto Rico-) based military ber).

Figure 3: Number of medical encounters for infectious gastroenteritis among active duty US military personnel 2003-2012.

EXPOSURES OF INTEREST

The under-reporting phenomenon of gella and entero-haemorrhagic E. coli; acute gastroenteritis is well-appreciated thus an assessment of the estimated and has been documented by the US minimum number of available relevant Centers for Disease Control and Pre- exposures for these pathogens is war- vention (CDC) most recently in esti- ranted. As shown in Figure 4, medical mates of foodborne-related illness in encounters associated with these 4 the US (Scallan, 2011a,b). The DMSS pathogens have remained relatively sta- is no different from other passive re- ble over the past 10 years with a total porting systems and it is anticipated number of just under 3,000 total en- that estimates of infectious (and/or counters. Importantly, these docu- toxin-mediated) gastroenteritis rates mented medical encounters are likely significantly underestimate true inci- to greatly underestimate the actual dence. Nonetheless, over the past 10 population exposed to these pathogens years (2003-2012), there have been 1.8 (Porter et al., 2012a). million documented medical encoun- Importantly, the above events are ters for acute gastroenteritis of pre- limited to those occurring at a MTF sumed infectious/toxin-mediated origin and do not include the numerous ex- (Figure 3), the overwhelming majority posures that occur during military de- (99.2%) of which occurred in an outpa- ployment and/or assignment in regions tient setting. Current studies of the of the globe with high rates of travel- post-infectious consequences of acute lers’ diarrhoea. While deployment to infection have most frequently focused those regions may serve as a surrogate on specific bacterial pathogens includ- of exposure, prior studies have been ing Campylobacter, Salmonella, Shi- limited in their ability to link increased

Figure 4: Number of medical encounters for Salmonella, Shigella, Campylobacter and E. coli O157:H7 among active duty US military personnel 2003-2012.

sequelae risk with deployment to high with the serum specimens as well as TD regions (Porter et al., 2011a, demographic and medical encounter 2012b, 2013a). This may be due to a data. Data obtained as part of these potential healthy worker effect or non- forms include deployment information specific in the exposure (or in this case (e.g., country, duration), general the exposure surrogate) thus biasing assessments of in-theatre and post-de- effect estimates toward the null hy- ployment health and exposures of inter- pothesis of no association. est during deployment. While this form A potential mechanism to circum- has undergone several iterations, most vent this limitation is to utilize deploy- recently in SEPT 2012, the version pre- ment health assessments designed to ceding the current edition collected document exposures occurring during data on potential IGE. In 2011, Porter deployment. US Department of De- et al described an increased odds of fense policy mandates the collection self-reported diarrhoea/vomiting and maintenance of deployment health among FGD cases compared to data (DoDI 6490.03, Deployment matched controls both of whom were Health, 11 Aug. '06). One mechanism first-time deployers with only one de- utilized to facilitate these mandates is ployment during the surveillance pe- the pre- and post-deployment health riod (Porter et al., 2011b). While lack- assessments (DD 2795 and 2796, ing sensitivity and specificity, these respectively). These self-assessments data sources open the door for obtain- are completed electronically, main- ing important deployment-specific ex- tained by the AFHSC and can be linked posures.

Table 2: Epidemiologic studies of post-infectious sequelae conducted utilizing data obtained from DMSS ——————————————————————————————————————— Reference Exposure(s) Outcome(s) and estimated relative risk/odds ratio ——————————————————————————————————————— Curry et al., 2010 Infectious Reactive arthritis: 4.4 (2.2, 8.7) gastroenteritis Nonspecific arthropathy: 1.8 (1.5, 2.1)

Porter et al., 2008 Infectious Crohn’s disease: 1.5 (1.2, 2.0) gastroenteritis Ulcerative colitis: 1.4 (1.1, 1.7)

Porter et al., 2011a Infectious Constipation: 2.2 (2.0, 2.3) gastroenteritis Dyspepsia: 2.4 (2.1, 2.7) Functional diarrhea: 6.3 (4.4, 8.9) Irritable bowel syndrome: 3.7 (3.4, 4.1)

Porter et al., 2011b Diarrhea/vomiting Constipation: 1.9 (0.9, 3.9) during deployment Dyspepsia: 6.8 (2.9, 15.4) Irritable bowel syndrome: 6.3 (2.5, 15.4) Any FGD: 2.9 (1.8, 4.8)

Porter et al., 2012a Infectious Crohn’s disease 1.5 (0.4, 6.3) gastroenteritisa Ulcerative colitis 3.5 (1.4, 9.0)

Porter et al., 2012b Norovirus Constipation: 1.3 (0.9, 1.8) Dyspepsia: 1.4 (0.8, 2.5) IBS: 0.7 (0.3, 1.5) GERD: 1.4 (1.1, 1.8)

Porter et al., 2013c Campylobacter Constipation: 1.6 (1.3, 2.0) Salmonella Dyspepsia: 1.3 (1.0, 1.9) Shigella IBS: 2.9 (2.2, 3.8) Yersinia GERD: 1.6 (1.4, 1.8)

Riddle et al., 2012 Infectious Celiac disease: 2.1 (1.4, 3.0) gastroenteritis

Riddle et al., 2013 Campylobacterb Celiac disease: 3.5 (0.7, 18.0) ——————————————————————————————————————— IBS=Irritable bowel syndrome; FGD=Functional gastrointestinal disorder a Among personnel with IBS b No increased risk observed following Shigella, Salmonella or Yersinia

Importantly, deployment not only in- between military exercises and func- creases the potential risk of exposure, tional symptoms are the results of a re- but also likely increases the risk of psy- cent study of Singapore military per- chological stressors, independent risk sonnel in which increased intestinal factors of functional bowel disorders. It permeability during combat training has been hypothesized that stressful corresponded to an increase in post- events, such as deployment and/or training GI complaints (Li et al., 2013). combat situations may modify the ef- Recognition of this potential effect fect of infection of FGD risk (Dross- modification and development of meth- man, 2011). This was corroborated by ods to accurately capture levels of de- the 2011 study by Porter et al (although ployment-related stress are important to not statistically significant) (Porter et ensure appropriate interpretation of real., 2011b). Further supporting the link sults in this population.

Figure 5: Number of medical encounters for adverse health outcomes associated with enteric infection among US military personnel 2003-2012.

OUTCOMES OF INTEREST

Of equal importance to the numbers of et al., 2008), coeliac disease (3.6/100K available exposures are the outcomes p-y) (Riddle, 2012) and reactive arthri- of interest for studying the long-term tis (4.1/100K p-y) (Curry et al., 2010). impact of acute gastroenteritis. A re- Importantly, these outcomes are associ- cent study on the incidence of FGD ated with numerous medical encounters among this population from 1997-2007 annually (Figure 5). Over the last 10 estimated rates of any FGD of approxi- years, there have been 40,000-50,000 mately 231 per 100,000 person-years medical encounters each for dyspepsia, (/100K p-y) with the highest rate ob- CD, UC and IBS and 3,000-4,000 each served for constipation (127/100K p-y) for coeliac disease and reactive arthri- followed by IBS (66/100K p-y) and tis. This represents as substantial dyspepsia (49/100K p-y) and (Porter et population from which to identify well- al., 2011a). Similar estimates over defined cases and retrospectively fol- comparable time periods have been low the disease progression from pre- made for IBD (29.2/100K p-y) (Porter onset to present.

LIMITATIONS

The limitations of medical encounter view. Nonetheless, Table 3 outlines data systems have been described pre- some of the potential areas of bias viously and an exhaustive delineation inherent in generic systems as well as of the limitations specific to the DoD the DMSS. Importantly, studies as- system is beyond the scope of this re- sessing the quality of medical encoun-

Table 3: Potential sources of bias inherent in generic medical encounter repositories as well as the DMSS (modified from Schneeweiss and Avorn, 2005) ——————————————————————————————————————————————————————————— Generation of typical DMSS ‘Other’ systems medical encounter data ——————————————————————————————————————————————————————————— Population: - Well-defined demographic data for 1.4 million currently - Demographic data of total population may be active duty service members unavailable - Approximately 10 million total service members - System dependent (Kaiser, NHANES, country-wide) - Dissimilar to general population - Unique occupational exposures

‘Ill’ patient seeks care: - Potential variability in care-seeking behaviour across - Lack of insurance in specific populations may demographic characteristics decrease care-seeking behaviour - Potential use of serum repository to document outcomes - No ability to document outcomes among those not among those not seeking care seeking care - Frequent well-visits and vaccinations - Potential variability in care-seeking behaviour across -D ata linkable across global MTFs demographic characteristics - Lacks medical encounters during deployment - Care ‘out of network’ • pre-deployment health assessments • ER visits • Post-deployment health assessments • During travel

Medical examination/ - Medical history electronically available - Medical history may/may not be electronically history: - Medical examination data not available available - Medical examination data may/may not be available

Procedures & testing: - CPT codes readily available (potential for - CPT codes may/may not be available (potential for miscoding) miscoding)

Diagnoses: - ICD9-CM codes readily available (potential for - ICD9-CM codes readily available (potential for miscoding) miscoding) - Can be validated by serologic testing for specific outcomes ———————————————————————————————————————————————————————————

Table 3 (continued): Potential sources of bias inherent in generic medical encounter repositories as well as the DMSS (modified from Schneeweiss and Avorn, 2005) ——————————————————————————————————————————————————————————— Generation of typical DMSS ‘Other’ systems medical encounter data ——————————————————————————————————————————————————————————— Interventions (pharmacy): - Pharmaceutical data not readily linked with medical - Over the counter medications not routinely available encounter data - Free samples not documented - Over the counter medications not routinely available - Free samples not documented

Outcomes: - Out of military (theoretically linkable with Veteran’s - Lost to follow-up Administration data) • Provider/network change - All other administrative changes documented in • Death system • Other

Research purposes: - Supported through AFHSC (with DoD collaborator) - Limited to those with network-specific access - Human subjects’ protection - Methods exist to remove patient identifiers ———————————————————————————————————————————————————————————

ter data to identify exposures and out- (IBS, Crohn’s disease, ulcerative coli- comes of interest have been conducted tis, reactive arthritis, non-specific to varying degrees of success. Specific mono-arthropathy, coeliac disease, to the DMSS, Payne et al validated an- infectious gastroenteritis) of clinical thrax vaccination history and found outcomes is documented (250 pa- relatively highly levels of positive and tients/outcome). The medical chart of negative predictive values (Payne et each identified patient is then obtained al., 2007). While no studies assessing and data elements extracted to allow the PPV of the exposures and outcomes for adjudication by a third party. Com- of interests related to post-infectious pletion of the adjudication process will sequelae of acute enteric infectious enable the estimation of a positive pre- have been published to date, efforts are dictive value associated with each + underway by the authors to explore the ICD9-CM code with /- 5%. The results utility of these ICD-9 codes to identify of this study will support on-going incident outcomes. The on-going study efforts within and external to the DoD utilizes a total of 1750 subjects with a regarding outcomes of interest and the medical encounter in which an ICD9- utilization of electronic medical records CM code specific for one of a variety to identify novel associations.

UTILIZATION OF SERUM REPOSITORY

Perhaps one of the greatest strengths of utilization of the DoD’s serum the DoD system and an element that repository, has the potential to sets it apart from other medical encoun- transform our understanding of disease ter systems is the availability of se- processes and, related to post- quential serum samples that can be infectious sequelae of enteric infection, linked to the demographic, deployment, sample testing may elucidate novel vaccination and medical encounter mechanisms by which acute infection data. A quick PubMed® search indi- may lead to prolonged adverse health cates that investigators have utilized outcomes. the serum repository to identify Specifically related to potential se- serologic risk factors for disease (Levin quelae of enteric infection, genomic et al., 2012; Munger et al., 2013), analyses have identified that genes genotypic factors associated with associated with intestinal barrier and disease (Scher et al., 2011) and responses to enteric pathogens were temporal changes in antibody profiles associated with an increased risk of PI- preceding disease onset (Arbuckle et IBS among subjects with affected by a al., 2003). Despite these significant waterborne outbreak of entero-haemor- advancements in disease understanding rhagic E. coli and Campylobacter (Vil- brought about through utilization of the lani et al., 2010). Metabolomics have serum repository, these, and similar been utilized to identify baseline differ- studies have only begun to scratch the ences in IBS patients compared to con- surface of the potential utility of these trols and to subsequently measure the serum samples. Specifically, revolu- impact of probiotics on the metabo- tionary advances in genomics, lome and subsequent clinical improve- proteomics and metabolomics have ment of IBS patients (Hong et al., ushered in the systems’ biology era and 2011). Genetic biomarkers have shown

similar importance for Crohn’s disease storage temperature (-30°C), available with CARD15/NOD2 combined with aliquot volume (50 cc), number of bacterial infection shown to increase freeze/thaw cycles and potential gaps disease risk (Vaiopoulou et al., 2012). in the cold chain from specimen col- While not specific to post-infection lection to storage (Moore et al., 2010). IBD, novel disease biomarkers such as Furthermore, from a research perspec- IL-6, IL-23, ASCA and pANCA have tive, the temporality of specimen col- expanded our understanding of disease lection around an event(s) of interest patho-etiology (Yau et al., 2013). While may be sub-optimal. For example, we omics-based studies of samples ob- recently conducted a sero-epidemio- tained from the DoD serum repository logic study which required the last se- are not readily available in the peer-re- rum sample prior to initiating basic viewed literature, the accessibility of training during which an exposure of these serum samples combined with interest (norovirus outbreak) had oc- new technologies and platforms with curred. The mean time from sample which to conduct novel assays has the collection until the exposure of interest opportunity to enhance our understand- was approximately 6 months. Despite ing of the inter-relational association this (and other) limitation, one of the between the human genome, proteome most unique attributes of the DODSR and metabolome and further our under- is its longitudinal nature which enables standing of the pathophysiology of the measurements of seroconversion and/or sequelae of acute enteric infection. the development of novel biomarkers Importantly, these samples are not not present in previous (or subsequent) without limitations which have been samples. highlighted previously and include

OTHER LARGE EPIDEMIOLOGIC DATABASES

As alluded to previously, the US mili- This registry is inherently encrypted tary is not the sole source of medical with no link to personal identification encounter data. Country-wide systems information. However, encrypted data exists globally as do those that are spe- can be linked to other sources of data cific to given managed care organiza- utilizing identifiable information tions or those established to enable (Bakken et al., 2012). While an in- long-term cohort studies. For example, valuable resource to researchers, these Norway has a universal healthcare sys- data are not without limitations. Spe- tem in which all citizens have unre- cifically, linking data for single indi- stricted access to healthcare and in viduals across multiple years is diffi- which medical encounters at hospital- cult as identifiable information is often based clinics are documented (Lofthus replaced with a unique identifier spe- et al., 2005). In 2012, a total of 1.7 mil- cific to the year in which care was lion patients received care on at least sought (Lofthus et al., 2005). Addition- one occasion at one of the public hospi- ally, for specific years, available diag- tals, approximately 500K of which noses may be limited; furthermore, it is were associated with an in-patient stay unclear if procedure codes are incorpo- (https://www.ssb.no/en/pasient). A sub- rated into the available data (Lofthus et set of these data are querable online al., 2005). Similar systems exists in (https://www.ssb.no/en/helse). other Nordic countries including Den-

mark (Andersen et al., 1999), Finland laboratory tests. Data obtained from (Sund, 2012) and Sweden (Ludvigsson NHANES have been utilized in a mul- et al., 2011) with recent efforts to com- titude of research studies involving in- bine registries across countries to fur- fectious and non-infectious diseases ther expand the available study popula- (Tuteja et al., 2008). In addition to epi- tion (Furu et al., 2010; Olsen et al., demiologic data, NHANES also main- 2010). tains a biorepository with plasma, se- In addition to country-wide regis- rum and purified DNA that can be tries, other population-based cohort linked to health-related data. However, studies have been prospectively de- these samples are not collected on a signed to allow research among a longitudinal basis as are the DODSR cross-section of populations. One spe- samples. Furthermore, these data are cific study to note is the National inherently different than those de- Health and Nutrition Examination scribed in the US military system and Study (NHANES). This study, initiated the Nordic countries in that they are in the early 1960s, is designed as a sur- obtained for the sole purpose of re- vey-based cohort study on specific search while the medical encounter populations and health-related topics. data are collected as part of routine Since 1999, approximately 5,000 per- medical care. Complete understanding sons are surveyed annually for demo- of these differences, some of which are graphic, socioeconomic, dietary and outlined in Table 3, is key to ensuring other health-related information as well appropriate data interpretation and ex- as medical examinations to include trapolation to other populations. dental, psychological assessments and

CONCLUSION

The US DOD medical encounter data- studies are have enhanced our appre- bases and serum repository are invalua- ciation of the acute/chronic disease ble assets with the ability to modify our link, additional research is needed. The understanding of many disease pro- OMICs revolution has paved the way cesses. Historically, well-designed epi- for the utilization of the DOD SR demiologic studies have been at the linked with relevant medical encounter fore of linking exposures with out- data to further enhance our understand- comes and have enabled estimates of ing of disease patho-etiology. Im- outcome risk, identifying host- and portantly, these studies must be accom- pathogen-specific risk factors and un- panied by improved mechanistic pre- derstanding the timing from exposure clinical studies, which to date suggest to outcome. Studies of the post-infec- multifactorial biological processes tious sequelae of enteric diseases are no leading to PI-FGD. Together these different as highlighted by several sys- studies have the potential to begin un- tematic reviews on the epidemiologic ravelling the complexity of post-IGE evidence (Halvorson et al., 2006; Tha- sequelae and revolutionize our under- bane et al., 2007; Poropatich et al., standing of the long-term impacts of 2010; Deising et al., 2013; Pike et al., acute infection. 2013; Porter et al., 2013b). While these

LITERATURE

Andersen, T.F., Madsen, M., Jørgensen, J., tikainen, J.E., Almarsdottir, A.B., and Mellemkjoer, L., and Olsen, J.H.: The Sørensen, H.T.: The Nordic countries as a Danish National Hospital Register. A valu- cohort for pharmacoepidemiological re- able source of data for modern health sci- search. Basic Clin. Pharmacol. Toxicol. ences. Dan. Med. Bull. 46, 263-268 106, 86-94 (2010). (1999). Green, P.H. and Cellier, C.: Celiac disease. N. Arbuckle, M.R., McClain, M.T., Rubertone, Engl. J. Med. 357, 1731-1743 (2007). M.V., Scofield, R.H., Dennis, G.J., James, Halvorson, H.A., Schlett, C.D., and Riddle, J.A., and Harley, J.B.: Development of M.S.: Postinfectious irritable bowel syn- autoantibodies before the clinical onset of drome--a meta-analysis. Am. J. Gastroen- systemic lupus erythematosus. N. Engl. J. terol. 101, 1894-1899; quiz 1942 (2006). Med. 349, 1526-1533 (2003). Hong, Y.S., Hong, K.S., Park, M.H., Ahn, Bakken, I.J., Gystad, S.O., Christensen, Ø.O., Y.T., Lee, J.H., Huh, C.S., Lee, J., Kim, Huse, U.E., Larønningen, S., Nygård, J., I.K., Hwang, G.S., and Kim, J.S.: Metabo- Holmstrøm, L., Johannesen, T.B., Møller, nomic understanding of probiotic effects in B., and Larsen, K.: Comparison of data humans with irritable bowel syndrome. J. from the Norwegian Patient Register and Clin. Gastroenterol. 45, 415-425 (2011). the Cancer Registry of Norway. Tidsskr. Karlinger, K., Györke, T., Makö, E., Mester, Nor. Laegeforen, 132, 1336-40 (2012). A., and Tarján, Z.: The epidemiology and Curry, J.A., Riddle, M.S., Gormley, R.P., Trib- the pathogenesis of inflammatory bowel ble, D.R., and Porter, C.K.: The epidemiol- disease. Eur. J. Radiol. 35, 154-167 ogy of infectious gastroenteritis related (2000). reactive arthritis in U.S. military person- Koebnick, C., Smith, N., Huang, K., Martinez, nel: a case-control study. BMC Infect. Dis. M.P., Clancy, H.A., Williams, A.E., and 10, 266 (2010). Kushi, L.H.: OBAYA (obesity and adverse Dawber, T.R., Meadors, G.F., and Moore, F.E. health outcomes in young adults): feasibil- Jr.: Epidemiological approaches to heart ity of a population-based multiethnic co- disease: the Framingham Study. Am. J. hort study using electronic medical rec- Public Health Nations Health 41, 279-281 ords. Popul. Health Metr. 10, 15 (2012). (1951). Levin, L.I., Chang, E.T., Ambinder, R.F., Len- Deising, A., Gutierrez, R.L., Porter, C.K., and nette, E.T., Rubertone, M.V., Mann, R.B., Riddle, M.S.: Postinfectious functional Borowitz, M., Weir, E.G., Abbondanzo, gastrointestinal disorders: a focus on epi- S.L., and Mueller, N.E.: Atypical pre- demiology and research agendas. Gastro- diagnosis Epstein-Barr virus serology re- enterol. Hepatol. (NY) 9, 145-157 (2013). stricted to EBV-positive Hodgkin lym- Doll, R. and Hill, A.B.: The mortality of doc- phoma. Blood, 120, 3750-3755 (2012). tors in relation to their smoking habits; a Li, X., Kan, E.M., Lu, J., Cao, Y., Wong, R.K., preliminary report. Br. Med. J. 1 1451- Keshavarzian, A., and Wilder-Smith, C.H.: 1455 (1954). Combat-training increases intestinal Doll, R. and Pike, M.C.: Trends in mortality permeability, immune activation and gas- among British doctors in relation to their trointestinal symptoms in soldiers. Ali- smoking habits. J. R. Coll. Physicians ment. Pharmacol. Ther. 37, 799-809 Lond. 6, 216-222 (1972). (2013). Drossman, D.A.: Abuse, trauma, and GI ill- Lofthus, C.M., Cappelen, I., Osnes, E.K., ness: is there a link? Am. J. Gastroenterol. Falch, J.A., Kristiansen, I.S., Medhus, 106, 14-25 (2011). A.W., Nordsletten, L., and Meyer, H.E.: Furu, K., Wettermark, B., Andersen, M., Mar- Local and national electronic databases in

Norway demonstrate a varying degree of Medical Surveillance System, 1998-2004. validity. J. Clin. Epidemiol. 58, 280-285 Pharmacoepidemiol. Drug Saf. 16, 605- (2005). 611 (2007). Ludvigsson, J.F., Andersson, E., Ekbom, A., Pike, B.L., Porter, C.KJ., Sorell, T.J., and Rid- Feychting, M., Kim, J.L., Reuterwall, C., dle, M.S.: Acute gastroenteritis and the Heurgren, M., and Olausson, P.O.: Exter- risk of functional dyspepsia: a systematic nal review and validation of the Swedish review and meta-analysis. Am. J. Gastro- national inpatient register. BMC Public enterol. 108, 1558-1563 (2013). Health 11, 450 (2011). Poropatich, K.O., Walker, C.L., and Black, Marshall, J.K., Thabane, M., Garg, A.X., R.E.: Quantifying the association between Clark, W.F., Salvadori, M., Collins, S.M.; Campylobacter infection and Guillain- Walkerton Health Study Investigators: Barre syndrome: a systematic review. J. Incidence and epidemiology of irritable Health Popul. Nutr. 28, 545-552 (2010). bowel syndrome after a large waterborne Porter, C.K., Tribble, D.R., Aliaga, P.A., Hal- outbreak of bacterial dysentery. Gastro- vorson, H.A., and Riddle, M.S.: Infectious enterology 131, 445-450; quiz 660 (2006). gastroenteritis and risk of developing Mearin, F.: Postinfectious functional gastro- inflammatory bowel disease. Gastroen- intestinal disorders. J. Clin. Gastroenterol. terology 135, 781-786 (2008). 45 Suppl, S102-S105 (2011). Porter, C.K., Gormley, R., Tribble, D.R., Cash, Moore, M., Eiseman, E., Fisher, G., Olmsted, B.D., and Riddle, M.S.: The incidence and S.S., Sama, P.R., and Zambrano, J.A.: Har- gastrointestinal infectious risk of func- nessing full value from the DoD Serum tional gastrointestinal disorders in a Repository and the Defense Medical healthy US adult population. Am. J. Gas- Surveillance System. The RAND Corpora- troenterol. 106, 130-138 (2011a). tion, Santa Monica, CA, USA (2010). Porter, C.K., Gloor, K., Cash, B.D., and Riddle, Munger, K.L., Levin, L.I., Massa, J., Horst, R., M.S.: Risk of functional gastrointestinal Orban, T., and Ascherio, A.: Preclinical se- disorders in U.S. military following self- rum 25-hydroxyvitamin D levels and risk reported diarrhea and vomiting during de- of type 1 diabetes in a cohort of US mili- ployment. Dig. Dis. Sci. 56, 3262-3269 tary personnel. Am. J. Epidemiol. 177, (2011b). 411-419 (2013). Porter, C.K., Cash, B.D., Pimentel, M., Neal, K.R., Hebden, J., and Spiller, R.: Preva- Akinseye, A., and Riddle, M.S.: Risk of lence of gastrointestinal symptoms six inflammatory bowel disease following a months after bacterial gastroenteritis and diagnosis of irritable bowel syndrome. risk factors for development of the irritable BMC Gastroenterol. 12, 55 (2012a). bowel syndrome: postal survey of patients. Porter, C.K., Faix, D.J., Shiau, D., Espiritu, J., BMJ 314, 779-782 (1997). Espinosa, B.J., and Riddle, M.S.: Olsen, J., Brønnum-Hansen, H., Gissler, M., Postinfectious gastrointestinal disorders Hakama, M., Hjern, A., Kampeer-Jørgen- following norovirus outbreaks. Clin. In- sen, F., Rafnsson, V., Tell, G.S., Thaulow, fect. Dis. 55, 915-922 (2012b). I., and Thygesen, L.C.: High-throughput Porter, C.K., Choi, D., and Riddle, M.S.: epidemiology: combining existing data Pathogen-specific risk of reactive arthritis from the Nordic countries in health-related from bacterial causes of foodborne illness. collaborative research. Scand. J. Public J. Rheumatol. 40, 712-714 (2013a). Health 38, 777-779 (2010). Porter, C.K., Thura, N., and Riddle, M.S.: Payne, D.C., Rose, C.E. Jr., Aranas, A., Zhang, Quantifying the incidence and burden of Y., Tolentino, H., Weston, E., McNeil< postinfectious enteric sequelae. Mil. Med. M.M., and Ruscio, B.: Assessment of an- 178, 452-469 (2013b). thrax vaccination data in the Defense Porter, C.K., Choi, D., Cash, B., Pimentel, M.,

Murray, J., May, L., and Riddle, M.S.: Stephenson, J.J., Barghout, V., Kahler, K.H., Pathogen-specific risk of chronic gastro- Fernandes, J., Beaulieu, J.F., Joo, S., and intestinal disorders following bacterial Boccuzzi, S.J.: Effectiveness of tegaserod causes of foodborne illness. BMC Gastro- therapy on GI-related resource utilization enterol. 13, 46 (2013c). in a managed care population. Am. J. Riddle, M.S., Murray, J.A., and Porter, C.K.: Manag. Care 11 (1 Suppl), S35-S42 The incidence and risk of celiac disease in (2005). a healthy US adult population. Am. J. Sund, R.: Quality of the Finnish Hospital Dis- Gastroenterol. 107, 1248-1255 (2012). charge Register: a systematic review. Riddle, M.S., Murray, J.A., Cash, B.D., Pimen- Scand. J. Public Health 40, 505-515 tel, M., and Porter, C.K.: Pathogen-specific (2012). risk of celiac disease following bacterial Thabane, M., Kottachchi, D.T., and Marshall, causes of foodborne illness: a retrospective J.K.: Systematic review and meta-analysis: cohort study. Dig. Dis Sci. 58, 3242-3245 The incidence and prognosis of post-infec- (2013). tious irritable bowel syndrome. Aliment. Rubertone, M.V. and Brundage, J.F.: The De- Pharmacol. Ther. 26, 535-544 (2007). fense Medical Surveillance System and the Tuteja, A.K., Talley, N.J., Gelman, S.S., Alder, Department of Defense serum repository: S.C., Thompson, C., Tolman, K., and Hale, glimpses of the future of public health sur- D.C.: Development of functional diarrhea, veillance. Am. J. Public Health 92, 1900- constipation, irritable bowel syndrome, 1904 (2002). and dyspepsia during and after traveling Scallan, E., Griffin, P.M., Angulo, F.J., Tauxe, outside the USA. Dig. Dis. Sci. 53, 271- R.V., and Hoekstra, R.M.: Foodborne ill- 276 (2008). ness acquired in the United States --un- Vaiopoulou, A., Gazouli, M., Theodoropoulos, specified agents. Emerg. Infect. Dis. 17, G., and Zografos, G.: Current advantages 16-22 (2011a). in the application of proteomics in inflam- Scallan, E., Hoekstra, R.M., Angulo, F.J., matory bowel disease. Dig. Dis. Sci. 57, Tauxe, R.V., Widdowson, M.A., Roy, 2755-2764 (2012). S.L., Jones, J.L., and Griffin, P.M.: Food- Villani, A.C., Lemire, M., Thabane, M., borne illness acquired in the United States- Belisle, A., Geneau, G., Garg, A.X., Clark, -major pathogens. Emerg. Infect. Dis. 17, W.F., Moayyedi, P., Collins, S.M., 7-15 (2011b). Franchimont, D., and Marshall, J.L.: Ge- Scher, A.I., Wu, H., Tsao, J.W., Blom, H.J., netic risk factors for post-infectious irrita- Feit, P., Nevin, R.L., and Schwab, K.A.: ble bowel syndrome following a water- MTHFR C677T genotype as a risk factor borne outbreak of gastroenteritis. Gastro- for epilepsy including post-traumatic epi- enterology 138, 1502-1513 (2010). lepsy in a representative military cohort. J. Yau, Y., Leong, R.W., Zeng, M., and Neurotrauma 28, 1739-1745 (2011). Wasinger, V.C.: Proteomics and metabo- Schneeweiss, S. and Avorn, J.: A review of lomics in inflammatory bowel disease. J. uses of health care utilization databases for Gastroenterol. Hepatol. 28, 1076-1086 epidemiologic research on therapeutics. J. (2013). Clin. Epidemiol. 58, 323-337 (2005).

AN ENTERIC TRIANGLE: PROTOZOAN INFECTIONS, LINKS TO ENVIRONMENTAL ENTEROPATHY AND THE POTENTIAL INFLUENCE OF THE INTESTINAL BACTERIAL MICROBIOME ON THIS INTERACTION AND OVERALL HEALTH IN THE DEVELOPING WORLD

STACEY L. BURGESS and WILLIAM A. PETRI

University of Virginia School of Medicine, Carter Harrison Research Building, Charlottesville, VA, USA

SUMMARY

Lasting changes to the host intestinal mucosal surface may be caused by faecal enteropathogens such as Entamoeba histolytica, one of the causes of host diarrhoea, which when it repeatedly and frequently occurs instigates a subclinical condition, environmental enteropathy (EE), characterized by blunting of intestinal villi and intestinal inflammation. Amoebiasis and EE occur in the context of the host’s intestinal bacterial microbiome and the persistent changes driven by enteropathogens could be modulated by the composition of these intestinal bacterial communities. In this work, we will explore the long term changes to the host by protozoan infections such as E. histolytica in man and model organisms such as mice, how the intestinal bacterial microbiota and probiotic organisms might influence overall health and infections with these protozoa, and finally the role these interactions might have in health and wellness in the developing world where enteric diarrheal disease is endemic. We hypothesize that interactions between the host’s immune system, protozoan infections and the intestinal microbiome might influence EE and in turn vaccine failure and perhaps exacerbate nutritional deficits, increasing the risk of malnutrition in a food insecure household.

ENTERIC INFECTIONS AND ENVIRONMENTAL ENTEROPATHY

For many years our laboratory has (den Hartog et al., 2013). Furthermore, tracked the medical histories of a large diarrheal disease is a primary cause of cohort of children living in the Mirpur mortality in children less than five slums of Dhaka, Bangladesh. Here, years of age in these nations and ac- colonization with parasites is a major counts for nearly two million deaths cause of diarrheal illness, which is a annually (WHO). However, in tracking significant source of morbidity and these children, it has also become ap- mortality in the developing world. parent that these infections in early life Colonization with Giardia lamblia, may have lasting influences on health, Cryptosporidium parvum, and Enta- particularly growth, later nutritional moeba histolytica likely underlie 58 status, susceptibility to further infection million cases of childhood diarrhoea and perhaps even influence vaccine

efficacy (Korpe and Petri, 2012; ulceration, and submucosal inflamma- Mondal et al., 2012; Kotloff et al., tory infiltration that were reminiscent 2013). This cohort, and many others in not only of human amoebiasis but also developing tropical nations, has a high human EE. Thus, we have hypothe- incidence of environmental enteropathy sized that repeated infections with E. (EE), a subclinical condition caused by histolytica and other protozoans in chil- constant faecal–oral contamination that dren might contribute to EE by eliciting is characterized by blunting of intesti- long lasting changes to their intestinal nal villi and other histopathological ab- mucosa and to the infiltrating popula- normalities of the intestine such as in- tions of immune cells present. Our creased crypt lengthening and intestinal laboratory is currently exploring how inflammation, including lymphocyte this organism and other enteropatho- and monocyte infiltration. EE is be- gens might influence the long term se- lieved to significantly contribute to quelae of EE. However, while enter- malnutrition and stunting in these taining this notion, it is useful to under- populations by preventing normal de- stand the context of intestinal E. histo- velopment of the intestine. Much nutri- lytica infection. ent absorption occurs at the tips of Initial infection occurs after inges- intestinal villi and this is not possible in tion of faecally contaminated water or severely damaged and inflamed intes- food containing E. histolytica cysts tines (Fagundes-Neto et al., 1984, which then undergo excystation in the 1994; Korpe and Petri, 2012). Our la- lumen of the small intestine. The boratory has shown a correlation be- amoeba trophozoite then feeds on resi- tween E. histolytica driven childhood dent bacteria and possibly the intestinal diarrhoea and later stunting (Mondal et epithelium and in rare cases, it may al., 2006). In turn, malnourished chil- cause systemic amoebiasis by invading dren have higher rates of infection by the intestinal mucosa and traveling to E. histolytica and Cryptosporidium the blood stream, liver or brain (Petri (Korpe and Petri, 2012; den Hartog et and Singh, 1999; Haque et al., 2003; al., 2013a). Thus repeated colonization Verkerke et al., 2012). However the with enteropathogens, including these intestinal lumen is densely populated protozoa, likely instigates a feedback by a community of bacteria that may loop of poor nutrition and stunting. have a significant influence on the While E. histolytica and Cryptosporid- host’s immune response at baseline, ium do influence diarrheal diseases, and during amoeba infection, as well as they have not specifically been shown the virulence of the amoeba itself to cause EE in man. However, E. histo- (Mirelman et al., 1983; Noverr and lytica has been shown to instigate many Huffnagle, 2004; Frederick and Petri, of the characteristics of EE in murine 2005; Maslowski and Mackay, 2010; models of amoebiasis (Houpt et al., Cho and Blaser, 2012). These interac- 2002a; Mondal et al. 2012; Verkerke et tions may in turn also influence the al. 2012; den Hartog et al., 2013a). hosts nutritional status and ability to Houpt et al. (2002b) have shown, for mount a successful immune response to instance, that injection of E. histolytica later infections (Mondal et al., 2012). trophozoites into C3H/HeJ mice leads Thus when considering the persistent to chronic caecal infection in the ma- effects of protozoan infection and EE, jority of mice. They demonstrated that it is also pertinent to examine the con- infected mice had histological changes tribution of the intestinal bacterial mi- including crypt hyperplasia, epithelial crobiota to human health.

INFLUENCE OF COMMENSAL BACTERIA ON MAN AND MICE

The normal flora of the human gastro- coccus (enterotype 3) but have contri- intestinal tract is a large, complex com- butions from other genera (Arumugam munity of bacteria that is composed of et al., 2011). These enterotypes can fur- at least several hundred species and ther be represented by analysing clus- consists of 1012 bacteria per gram of ters of bacterial families and this analy- large bowel content. There are far more sis method shows that these first two bacterial cells than there are eukaryotic enterotypes are primarily characterized cells in the human body and these or- by the presence of Bacteroides and ganisms form a symbiosis that influ- Prevotella, whereas the third cluster is ences many aspects of human physiol- mostly characterized by related groups ogy including the composition of the of the family Clostridiaceae, and un- metabolome, which is the complete set classified Lachnospiraceae. Several of body wide small-molecule metabo- studies have suggested that specific en- lites including hormones, chemokines, terotypes might be associated with in- cytokines and other signalling mole- flammatory responses in the intestine cules (Arumugam et al., 2011; Siezen and deregulation of normal metabolic and Kleerebezem, 2011; Cho and controls. Specifically, intestinal auto- Blaser, 2012). However, understanding immunity, colitis, and obesity, have the mechanisms by which the human been associated with enterotypes 2 and microbiota influences the metabolome, 3 above, however, which specific immune system and nutrition is an organisms might influence these dis- emerging science. In studying the com- eases is still a point of contention position of the microbiota it has be- (Arumugam et al., 2011; Giongo et al., come clear that, while there is some 2011; Siezen and Kleerebezem, 2011). stability on a phyla and genera level Studies of type 1 diabetes have further within human populations, there is demonstrated that there are distinct dif- much variability in the milieu of spe- ferences in the microbiome of infants cific bacterial species between individ- that developed the disease compared to ual humans. Enterotypes are classifica- those that did not, with decreased mi- tion units based on the bacterial com- crobial diversity and a higher propor- position of the gut microbiome and tion of the Bacteroidetes phyla com- have been utilized to describe these pared to Firmicutes in the group with shared groups of bacterial phyla and the disease (Giongo et al., 2011). It has genera in humans and to study correla- also recently been shown that members tions with diseases that might be influ- of the Bacteroidetes and Firmicutes are enced by the microbiome. These enter- heavily involved in metabolism of otypes are independent of ethnic back- complex carbohydrates in the intestine ground but appear to be influenced to (Flint et al., 2012). Thus, the composi- some degree by the composition of diet tion of the intestinal bacterial microbi- (Arumugam et al., 2011; Wu et al., ota might significantly influence intes- 2011). How many enterotypes are pre- tinal inflammation, metabolism and sent is yet to be determined, however nutrition in man. However, the bulk of there are at least three (Arumugam et these studies have taken place in the al., 2011). They are defined largely by developed world. Changes in the mi- the variation in the levels of three gen- crobiome that might underlie the devel- era, Bacteroides (enterotype 1), opment of EE are not well described. Prevotella (enterotype 2), and Rumino- Mouse models have demonstrated

more directly that the microbiome can tine of mice. SFB are commensal, un- have a significant influence on the cultivable, obligate gut tropic, members function, structure and composition of of the Clostridiaceae that have been the immune system and intestine as shown to drive potent IgA induction well as malnutrition. Gnotobiotic, or and Th17 helper cell induction in the germ free, mice have structural defects intestine (Davis and Savage, 1974; in peyers patch formation in the small Ivanov II et al., 2008; Gaboriau- intestine, decreased or absent IgA pro- Routhiau et al., 2009; Kuwahara et al., duction, few intra epithelial lympho- 2011). Colonization with SFB has been cytes and a systemic defect in T regula- shown to exacerbate or influence a tory cell induction, all of which is re- number of intestinal and extra intestinal versed once the animals are recolo- models of human disease in mice, in- nized with a normal murine faecal mi- cluding colitis, autoimmune myelitis, crobiota (Umesaki and Setoyama, arthritis and type 1 diabetes via Th17 2000; Jiang et al., 2004). Components cell induction (Nutsch and Hsieh, of the intestinal microbiota such as 2012). The bacteria, while uncultiva- Lactobacillus casei and bacteria pre- ble, are well described both genetically sent on vegetables and in the soil such and morphologically in mice and a as Lactobacillus plantarum and Bifido- number of studies have suggested that bacterium bifidum can also drive Th1 they may be present in humans as well and T regulatory helper cell induction (Davis and Savage, 1974; Child et al., which may be antagonistic to inflam- 2006; Kuwahara et al., 2011; Yin et al., matory Th2 helper cell driven patholo- 2012; Caselli et al., 2013; Jonsson, gies such as asthma and eczema and 2013). SFB colonization also strongly have been shown to be protective in influences the composition of the intes- colitis models (Feleszko et al., 2007; tinal microbiota through its unique ki- Schwarzer et al., 2011). The microbi- netics of colonization and influence on ome can also have a significant influ- the immune system. For a short time ence on nutritional status in murine after weaning, the bacteria becomes the models. Smith et al. have recently dominate species within the murine gut shown that kwashiorkor, a type of se- and has been shown to outcompete en- vere acute malnutrition, may be influ- teropathic bacteria species (Chase and enced by the gut microbiome. They ob- Erlandsen, 1976; Heczko et al., 2000; served that when the microbiome from Stepankova et al., 2007; Kuwahara et malnourished versus healthy Malawian al., 2011). The ability of SFB, a com- twins was transplanted into gnotobiotic ponent of the mouse, and possibly hu- mice the kwashiorkor microbiome in- man, normal flora, to strongly influ- stigated marked weight loss in recipient ence the basic makeup of the immune mice, as well as altered the metabolism system as well as colonization by other of carbohydrates and amino acids, bacterial species illustrates the inter- when compared to the microbiome connectedness of individual bacterial from the healthy twins (Smith et al., species, the broader microbiome and 2013). health. However the influence of the Another striking example of a spe- intestinal bacterial microbiome is by no cific component of the bacterial micro- means limited to other bacterial species biota influencing physiology and the and may have profound influences on immune responses is the role seg- protozoa virulence and colonization mented filamentous bacteria (SFB) and in turn amoebiasis, diarrheal dis- play in immune maturation in the intes- ease and EE.

INTERACTIONS BETWEEN BACTERIA AND PROTOZOA

As mentioned previously, many proto- change in Ruminococcus compared to zoa inhabit the intestine for a signifi- healthy patients. These works suggest cant portion of their lives, and in doing that some of the pathology that results so interact intimately with other organ- during amoebiasis might be driven by a isms present there (Mirelman et al., disregulated microbiome or cross talk 1983). Berrilli et al. (2012) have re- between enteropathic bacteria and pro- cently highlighted some of the interac- tozoa and the intestinal immune sys- tions between intestinal microbial com- tem, particularly the intestinal epithe- munities, probiotics and bacterial path- lium. In fact, there are several works ogens and many types of parasite infec- suggesting that probiotics, particularly tions. Protozoa, including E. histolytica Lactobacillus species, might be protec- , are both influenced by the presence of tive in the context of protozoan infec- other enteropathic and probiotic bacte- tions (Preidis et al., 2011; Travers et ria and influence the composition of the al., 2011). Thus a decrease in protec- broader intestinal microbiota. Galván- tive, commensal, lactobacillus species Moroyoqui et al. (2008) have explored during E. histolytica infection might the effect of co-culture of trophozoites influence the severity of disease. One from E. histolytica and E. dispar with murine study has shown that daily ad- the enteropathogenic bacteria strains ministration of Lactobacillus acidophi- Escherichia coli (ETEC), Shigella dys- lus, a bacteria common in yogurt, for enteriae and a commensal Escherichia one week in Giardia lamblia infected coli on epithelial cell monolayers. In BALB/c mice significantly reduced G. doing this they determined that phago- lamblia infection burden in those mice. cytosis of pathogenic bacteria aug- Disease severity was also significantly mented the cytopathic effect of E. his- decreased. Histological analysis of the tolytica on the cell monolayer as well intestine showed that probiotic admin- as increased expression of the adher- istration protected mice against parasite ence lectin, Gal/GalNAc, on the induced mucosal damage and de- amoeba’s surface. Thus, interactions creased intestinal villous atrophy with enteropathic bacteria in humans (Shukla et al., 2010). Thus probiotic might serve to increase the virulence of interventions might provide an attrac- E. histolytica during amoebiasis. E. his- tive avenue to decrease intestinal dam- tolytica colonization in turn also influ- age in populations in which repeated ences the composition of the microbi- intestinal protozoal infections occur ome. Verma et al. (2012) have shown (Shukla et al., 2010). Further under- that during amoebiasis there is a sig- standing of how protozoa influence and nificant decrease in absolute quantifica- are influenced by the intestinal micro- tion of Bacteroides, Clostridium coc- biome, enteropathogens and probiotics coides, Clostridium leptum, Lactobacil- would thus be informative in designing lus and Campylobacter and an increase microbiome based interventions for in Bifdobacterium, while there was no diseases such as EE and malnutrition.

ENTERIC INFECTIONS, HEALTH AND EXPLORATION OF THE MICROBIOME IN THE DEVELOPING WORLD

The Petri laboratory has long sought to tions, diarrheal disease and diseases find connections between enteric infec- with persistent effects such as EE and

malnutrition in order to develop tar- tested this hypothesis in the Mirpur co- geted interventions that might decrease hort of children by measuring re- the burden of disease in the developing sponses to oral poliovirus vaccine in world. In our field site in Mipur, children who received a minimum of Dhaka, we followed children for the three doses of OPV by age 6 months. first year of life with every-other-day We observed that diminished antibody home visits and surveyed enteropatho- responses to OPV were associated with gens in diarrheal and monthly surveil- malnutrition, increased serum endocab lance stool samples. We also measured levels, and shorter breastfeeding dura- intestinal barrier function by endocab tion. We also examined potential im- antibodies, which correlate with trans- mune mechanisms that might underlie location of bacterial LPS into blood, vaccine failure in a smaller subset of and measured nutritional status and these children and found that children stunting by anthropometry. In this with OPV failure exhibited globally study we found that diarrhoea co- reduced cellular responsiveness to a occurred with infections caused by sev- range of cytokine stimulations, as well eral organisms including enteric proto- as elevated pro-inflammatory cytokine zoa (amoebiasis, cryptosporidiosis, and expression. These data indicated that giardiasis), rotavirus, astrovirus, and oral vaccine failure in these children is enterotoxigenic Escherichia coli influenced by a combination of (ETEC). We also observed that malnu- malnutrition, gut barrier dysfunction trition was present in 16.3% of children and shorter duration of breast-feeding at birth and 42.4% at 12 months of age in early childhood, and is associated and that children that were malnour- with a chronic inflammatory immune ished at birth had increased Entamoeba phenotype (unpublished data/abstract). histolytica, Cryptosporidium, and These studies further highlighted the ETEC infections as well as more severe complex interrelationship of malnutri- diarrhoea. The children who became tion, protozoan and enteric infections, malnourished at 12 months of age were diarrhoea and vaccine failure in infants also much more likely to have pro- in low-income settings and the persis- longed diarrhoea and intestinal barrier tent effects of these problems. Thus, dysfunction, a mother without educa- identification of probiotics, or particu- tion, and low family expenditure lar enterotypes, that are protective dur- (Korpe and Petri, 2012; Mondal et al., ing protozoan infection might help 2012; den Hartog et al., 2013b). mitigate destruction of the intestinal Our laboratory, along with many barrier and significantly improve health other collaborators, has also recently outcomes in these populations. begun exploring links between malnu- Unfortunately, the contribution, if trition, EE and oral polio vaccine fail- any, of specific components of the in- ure. This is particularly important as testinal bacterial microbiota to EE and vaccines for polio (OPV) and rotavirus vaccine failure in the developing world are far less effective in poor children in is not currently well described. How- the developing world and the underly- ever given the profound influence that ing cause for this failure is largely un- intestinal bacteria such as Lactobacil- known. Based on our previous experi- lus, Bifidobacterium, SFB, and entero- ence we thus hypothesized that failure pathic E. coli (ETEC) can have on host of oral vaccines such as OPV might be immunity, parasite burden and viru- due to EE and be driven by inflamma- lence in mice and man, and the relative tion from endotoxin exposure. We inexpensiveness of probiotic interven-

tions, it is certainly an area of study playing robust responses and those that should be pursued. Interestingly, with late, less robust responses. How- one such recent study has shown a link ever the vast majority of these OTUs between the composition of the bacte- were classified within the order Clos- rial microbiota and the effectiveness of tridiales (Eloe-Fadrosh et al., 2013). an oral typhoid vaccine in a small Another recent paper has shown that group of individuals. In this study the the presence of some commensal Clos- composition of the microbiome did not tridium related species is decreased influence the ability of responders to during autoimmune colitis in children mount an immune response to oral ty- (Michail et al., 2012). Thus the compo- phoid vaccination, but those with a sition of the microbiota, and Clostridia more complex microbiota mounted a related organisms, or commensals that more robust, multiphasic IFN-y re- induce immune responses similar to sponse to oral vaccination. Many dif- these bacteria, may very well be an im- ferent organisms represented by opera- portant factor influencing homeostasis tional taxonomic units (OTUs) were of the intestine and the success of vac- found to differ between individuals dis- cination in the developing world.

FINAL THOUGHTS

Infection with enteropathogens such as lence, and infectivity, of those protozoa the protozoa Entamoeba histolytica as well as the host’s ability to mount likely contributes to the development protective responses against future in- of environmental enteropathy and fections. A better understanding of the induces lasting effects on the intestinal interactions between these organisms mucosa that may negatively impact nu- and the intestinal microbiota, which tritional outcomes and vaccine success might include probiotics commonly in millions of children each year. These found in many foods, may lead to cost pathogenic organisms live in the con- effective treatments that could signifi- text of the intestinal bacteria microbi- cantly decrease the burden of entero- ota and interactions with these species pathogens in the developed world. may significantly influence the viru-

LITERATURE

Arumugam, M., Raes, J., Pelletier, E., Le Doré, J.; MetaHIT Consortium, Antolín, Paslier, D., Yamada, T., Mende, D.R., M., Artiguenave, F., Blottiere, H.M., Al- Fernandes, G.R., Tap, J., Bruls, T., Batto, meida, M., Brechot, C., Cara, C., Cher- J.M., Bertalan, M., Borruel, N., Casellas, vaux, C., Cultrone, A., Delorme, C., F., Fernandez, L., Gautier, L., Hansen, T., Denariaz, G., Dervyn, R., Foerstner, K.U., Hattori, M., Hayashi, T., Kleerebezem, M., Friss, C., van de Guchte, M., Guedon, E., Kurokawa, K., Leclerc, M., Levenez, F., Haimet, F., Huber, W., van Hylckama- Manichanh, C., Nielsen, H.B., Nielsen, T., Vlieg, J., Jamet, A., Juste, C., Kaci, G., Pons, N., Poulain, J., Qin, J., Sicheritz- Knol, J., Lakhdari, O., Layec, S., Le Roux, Ponten, T., Tims, S., Torrents, D., Ugarte, K., Maguin, E., Mérieux, A., Melo Mi- E., Zoetendal, E.G., Wang, J., Guarner, F., nardi, R., M'rini, C., Muller, J., Oozeer, R., Pedersen, O., de Vos, W.M., Brunak, S., Parkhill, J., Renault, P., Rescigno, M.,

Sanchez, N., Sunagawa, S., Torrejon, A., microbiota and correlations with s. Typhi- Turner, K., Vandemeulebrouck, G., specific immunological responses. PloS Varela, E., Winogradsky, Y., Zeller, G., One 8, e62026 (2013). Weissenbach, J., Ehrlich, S.D., and Bork, Fagundes-Neto, U., Viaro, T., Wehba, J., P.: Enterotypes of the human gut Patrício, F.R., and Machado, N.L.: Tropi- microbiome. Nature 473,174-180 (2011). cal enteropathy (environmental enteropa- Berrilli, F., Di Cave, D., Cavallero, S., and thy) in early childhood: a syndrome caused D’Amelio, S.: Interactions between by contaminated environment. J. Trop. parasites and microbial communities in the Pediatr., 30, 204-209 (1984). human gut. Front. Cell. Infect. Microbiol. Fagundes-Neto, U., Martins, M.C., Lima, F.L 2,141 (2012). Patricio, F.R, Toledo, M.R.: Asympto- Caselli, M., Tosini, D., Gafà, R., Gasbarrini, matic environmental enteropathy among A., and Lanza, G.: Segmented filamentous slum-dwelling infants. J. Am. Coll. Nutr. bacteria-like organisms in histological 13, 51-60 (1994). slides of ileo-cecal valves in patients with Feleszko, W., Jaworska, J., Rha, R.D., ulcerative colitis. Am. J. Gastroenterol. Steinhausen, S., Avagyan, A., Jaudszus, 108, 860-861 (2013). A., Ahrens, B., Groneberg, D.A., Wahn, Chase, D.G. and Erlandsen, S.L.: Evidence for U., and Hamelmann, E.: Probiotic-induced a complex life cycle and endospore suppression of allergic sensitization and formation in the attached, filamentous, airway inflammation is associated with an segmented bacterium from murine ileum. increase of T regulatory-dependent J. Bacteriol. 127, 572-583 (1976). mechanisms in a murine model of asthma. Child, M.W., Kennedy, A., Walker, A.W., Clin. Exp. Allergy 37, 498-505 (2007). Bahrami, B., Macfarlane, S., and Flint, H.J., Scott, K.P., Duncan, S.H., Louis, P., Macfarlane, G.T.: Studies on the effect of and Forano, E.: Microbial degradation of system retention time on bacterial complex carbohydrates in the gut. Gut populations colonizing a three-stage Microbes 3, 289-306 (2012). continuous culture model of the human Frederick, J.R. and Petri, W.A.: Roles for the large gut using FISH techniques. FEMS galactose-/N-acetylgalactosamine-binding Microbiol. Ecol. 55, 299-310 (2006). lectin of Entamoeba in parasite virulence Cho, I. and Blaser, M.J.: The human micro- and differentiation. Glycobiology 15, 53R– biome: at the interface of health and 59R (2005). disease. Nat. Rev. Genet. 13, 260-270 Gaboriau-Routhiau, V., Rakotobe, S., Lecuyer, (2012). E., Mulder, I., Lan, A., Bridonneau, C., Davis, C.P. and Savage, D.C.: Habitat, Rochet, V., Pisi, A., De Paepe, M., Brandi, succession, attachment, and morphology of G., Eberl, G., Snel, J., Kelly, D., Cerf- segmented, filamentous microbes Bensussan, N.: The key role of segmented indigenous to the murine gastrointestinal filamentous bacteria in the coordinated tract. Infect. Immun. 10, 948-956 (1974). maturation of gut helper T cell responses. den Hartog, J., Rosenbaum, L., Wood, Z., Burt, Immunity 31, 677-689 (2009). D., and Petri, W.A.: Diagnosis of multiple Galván-Moroyoqui, J.M., Del Carmen enteric protozoan infections by enzyme- Domínguez-Robles, M., Franco, E., and linked immunosorbent assay in the Meza, I.: The interplay between Guatemalan highlands. Am. J. Trop. Med. Entamoeba and enteropathogenic bacteria Hyg. 88, 167-71 (2013). modulates epithelial cell damage. PLoS Eloe-Fadrosh, E.A., McArthur, M.A., Seekatz, Negl. Trop. Dis. 2, e266 (2008). A.M., Drabek, E.F., Rasko, D.A., Sztein, Giongo, A., Gano, K.A., Crabb, D.B., M.B., and Fraser, C.M.: Impact of oral Mukherjee, N., Novelo, L.L., Casella, G., typhoid vaccination on the human gut Drew, J.C., Ilonen, J., Knip, M., Hyöty, H.,

Veijola, R., Simell, T., Simell, O., Neu, J., Kotloff, K.L., Nataro, J.P., Blackwelder, W.C., Wasserfall, C.H., Schatz, D., Atkinson, Nasrin, D., Farag, T.H., Panchalingam, S., M.A., and Triplett, E.W.: Toward defining Wu, Y., Sow, S.O., Su,r D., Breiman, R.F., the autoimmune microbiome for type 1 Faruque, A.S., Zaidi, A.K., Saha, D., diabetes. ISME J. 5, 82-91 (2011). Alonso, P.L., Tamboura, B., Sanogo, D., Haque, R., Huston, C.D., Hughes, M., Houpt, Onwuchekwa, U., Manna, B., Rama- E., and Petri, W.A.: Amebiasis. N. Engl. J. murthy, T., Kanungo, S., Ochieng, J.B., Med. 348, 1565-1573 (2003). Omore, R., Oundo, J.O., Hossain, A., Das, Heczko, U., Abe, A., and Finlay, B.B.: S.K., Ahmed, S., Qureshi, S., Quadri, F., Segmented filamentous bacteria prevent Adegbola, R.A., Antonio, M., Hossain, colonization of enteropathogenic M.J., Akinsola, A., Mandomando, I., Escherichia coli O103 in rabbits. J. Infect. Nhampossa, T., Acácio, S., Biswas, K., Dis. 181, 1027–1033 (2000). O'Reilly, C.E., Mintz, E.D., Berkeley, Houpt, E.R., Glembocki, D.J., Obrig, T.G., L.Y., Muhsen, K., Sommerfelt, H., Rob- Moskaluk, C.A., Lockhart, L.A., Wright, ins-Browne, R.M., and Levine, M.M.: R.L., Seaner, R.M., Keepers, T.R., Burden and aetiology of diarrhoeal disease Wilkins, T.D., and Petri, W.A. Jr.: The in infants and young children in mouse model of amebic colitis reveals developing countries (the Global Enteric mouse strain susceptibility to infection and Multicenter Study, GEMS): a prospective, exacerbation of disease by CD4+ T cells. case-control study. Lancet 382, 209-222 J. Immunol. 169, 4496-4503 (2002a). (2013). Houpt, E.R., Glembocki, D.J., Obrig, T.G., Kuwahara, T., Ogura, Y., Oshima, K., Moskaluk, C.A., Lockhart, L.A., Wright, Kurokawa, K., Ooka, T., Hirakawa, H., R.L., Seaner, R.M., Keepers, T.R., Itoh, T., Ishifune, C., Maekawa, Y., Ya- Wilkins, T.D., and Petri, W.A. Jr.: The sutomo, K., Hattori, M., and Hayashi, T.: mouse model of amebic colitis reveals The lifestyle of the segmented filamentous mouse strain susceptibility to infection and bacterium: a non-culturable gut-associated exacerbation of disease by CD4+ T cells. immunostimulating microbe inferred by J. Immunol. 169, 4496-4503 (2002b). whole-genome sequencing. DNA Res. 18, Ivanov, I.I., Frutos Rde, L., Manel, N., 291-303 (2011). Yoshinaga, K., Rifkin, D.B., Sartor, R.B., Maslowski, K.M. and Mackay, C.R.: Diet, gut Finlay, B.B., and Littman, D.R.: Specific microbiota and immune responses. Nat. microbiota direct the differentiation of IL- Immunol. 12, 5-9 (2010). 17-producing T-helper cells in the mucosa Michail, S., M. Durbin, D. Turner, A.M. of the small intestine. Cell. Host Microbe Griffiths, D.R. Mack, J. Hyams, N. 4, 337-349 (2008). Leleiko, Kenche, H., Stolfi, A., and Wine, Jiang, H.Q., Thurnheer, M.C., Zuercher, A.W., E.: Alterations in the gut microbiome of Boiko, N.V., Bos, N.A., and Cebra, J.J.: children with severe ulcerative colitis. Interactions of commensal gut microbes Inflamm. Bowel Dis. 18, 1799–1808 with subsets of B- and T-cells in the (2012). murine host. Vaccine 22,805-811 (2004). Mirelman, D., Feingold, C., Wexler, A., and Jonsson, H.: Segmented filamentous bacteria in Bracha, R.: Interactions between human ileostomy samples after high-fiber Entamoeba histolytica, bacteria and intake. FEMS Microbiol. Lett. 342, 24-29 intestinal cells. Ciba Found. Symp. 99, 2- (2013). 30 (1983). Korpe, P.S. and Petri, W.A.: Environmental Mondal, D., Petri, W.A., Sack, R.B., enteropathy: critical implications of a Kirkpatrick, B.D., and Haque, R.: poorly understood condition. Trends Mol. Entamoeba histolytica-associated diarrheal Med. 18, 328-336 (2012). illness is negatively associated with the

growth of preschool children: evidence Gut microbiomes of Malawian twin pairs from a prospective study. Trans. R. Soc. discordant for kwashiorkor. Science 339, Trop. Med. Hyg. 100, 1032-1038 (2006). 548-554 (2013). Mondal, D., Minak, J. Alam, M. Liu, Y. Dai, J. Stepankova, R., Powrie, F., Kofronova, O., Korpe, P. Liu, L., Haque, R., and Petri, Kozakova, H., Hudcovic, T., Hrncir, T., W.A. Jr.: Contribution of enteric infection, Uhlig, H., Read S, Rehakova Z, Benada O, altered intestinal barrier function, and Heczko P, Strus M, Bland P, Tlaskalova- maternal malnutrition to infant Hogenova H. Segmented filamentous malnutrition in Bangladesh. Clin. Infect. bacteria in a defined bacterial cocktail Dis. 54, 185-192 (2012). induce intestinal inflammation in SCID Noverr, M.C. and Huffnagle, G.B.: Does the mice reconstituted with CD45RBhigh microbiota regulate immune responses CD4+ T cells. Inflamm. Bowel Dis. 13, outside the gut? Trends. Microbiol. 12, 1202-1211 (2007). 562-568 (2004). Travers, M.A., Florent, I., Kohl, L., and Nutsch, K.M. and Hsieh, C.S.: T cell tolerance Grellier, P.: Probiotics for the control of and immunity to commensal bacteria. parasites: an overview. J. Parasitol. Res. Curr. Opin. Immunol. 24, 385-391 (2012). 2011, 610769 (2011). Petri, W.A. and Singh, U.: Diagnosis and Umesaki, Y. and Setoyama, H.: Structure of the management of amebiasis. Clin. Infect. intestinal flora responsible for Dis. 29, 1117-1125 (1999). development of the gut immune system in Preidis, G.A., Hill, C., Guerrant, R.L., a rodent model. Microbes Infect. 2, 1343- Ramakrishna, B.S., Tannock, G.W., and 1351 (2000). Versalovic, J.: Probiotics, enteric and Verkerke, H.P., Petri, W.A., and Marie, C.S.: diarrheal diseases, and global health. The dynamic interdependence of Gastroenterology 140, 8-14 (2011). amebiasis, innate immunity, and Schwarzer, M., Repa, A., Daniel, C., undernutrition. Semin. Immunopathol. 34, Schabussova, I., Hrncir, T., Pot, B., 771-785 (2012). Stepankova, R., Hudcovic T, Pollak A, Verma, A.K., Verma, R., Ahuja, V., and Paul, Tlaskalova-Hogenova H, Wiedermann U, J.: Real-time analysis of gut flora in Kozakova H. Neonatal colonization of Entamoeba histolytica infected patients of mice with Lactobacillus plantarum Northern India. BMC Microbiol. 12, 183 producing the aeroallergen Bet v 1 biases (2012). towards Th1 and T-regulatory responses Wu, G.D., Chen, J., Hoffmann, C., Bittinger, upon systemic sensitization. Allergy 66, K., Chen, Y.Y., Keilbaugh, S.A., Bewtra, 368-375 (2011). M., Knights, D., Walters, W.A., Knight, Shukla, G., Kaur, T., Sehgal, R., Rishi, P., and R., Sinha, R., Gilroy, E., Gupta, K., Prabha, V.: Protective potential of L. Baldassano, R., Nessel, L., Li, H., Bush- acidophilus in murine giardiasis. CEJ Med. man, F.D., and Lewis, J.D.: Linking long- 5, 456-463 (2010). term dietary patterns with gut microbial Siezen, R.J. and Kleerebezem, M.: The human enterotypes. Science 334, 105-108 (2011). gut microbiome: are we our enterotypes? Yin, Y., Wang, Y., Zhu, L., Liu, W., Liao, N., Microb. Biotechnol. 4, 550-553 (2011). Jiang, M., Zhu, B., Yu, H.D., Xiang, C., Smith, M.I., Yatsunenko, T., Manary, M.J., and Wang, X: Comparative analysis of the Trehan, I., Mkakosya, R., Cheng, J., Kau, distribution of segmented filamentous A.L., Rich, S.S., Concannon, P., My- bacteria in humans, mice and chickens. chaleckyj, J.C., Liu, J., Houpt, E., Li, J.V., ISME J. 7, 615-621 (2012). Holmes, E., Nicholson, J., Knights, D., Ursell, L.K., Knight, R., and Gordon, J.I.:

COELIAC DISEASE

JOSEPH A. MURRAY and ERIC V. MARIETTA

Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN, USA

SUMMARY

Coeliac disease, causes chronic inflammation of the proximal intestinal, is an increasingly common disorder impacting health and nutrition. It is also a model disease straddling gut epithelial responses and systemic autoimmunity. It is a disease about which we know much, but also presents mysteries. Why does it occur at any age and why is it increasing. Why do most individuals who carry the genetic predisposition and eat gluten but don’t get the disease? It differs substantially from the more classic inflammatory bowel diseases of Crohn’s disease and ulcerative colitis. Those latter diseases primarily affect the distal small intestine or colon and rarely the upper intestine. Furthermore, coeliac disease is a response to dietary glutens whereas the microbiota play the dominant role for triggering IBD. The colon and distal small intestine are filled with large numbers of bacteria and other microbiota, the proximal small intestine by contrast is almost sterile. It does contain some microbiota and is a target for pathogenic invasive organisms. The other difference is that the proximal small intestine is deal- ing primarily with ingesta, not resident microbiota. Indeed, there are further differences between the colon and the small intestine. The colon, which contains the largest resident biomass of our ecology, has built up effective barriers to the microbiota which when breached can result in devastating inflammation. By contrast, the small intestine is a far more permeable organ with its primary role being digestion and absorption. There is more cross talk between the luminal contents and the mucosal immune system. Hence, regulation of the inflammatory response within the small intestine is even more crucial to permit normal digestive functioning to occur than in the colon. The microbiota and coeliac disease may intersect in several ways. The microbiota may be inherently different in children/adults at risk for coeliac disease. Furthermore, pathogenic infections may perturb the intestinal milieu to an extent that individuals genetically prone to coeliac disease will lose tolerance to gluten. The gluten-free diet may change the microbiota that may affect the host. This review summarizes what is known about the aetiology, epidemiology, diagnosis, treatment, as well as the microbiology of coeliac disease.

INTRODUCTION

Coeliac disease is a chronic inflamma- the proximal small intestine (Murray et tory condition predominately affecting al., 2008). It occurs in people who have

Figure 1: Coeliac disease has two required co-factors: 1) genetic predisposition, and 2) environmental factors, primarily the ingestion of gluten.

a specific genetic type, in particular the the digestive and secretory function of MHC class 2 gene pairs encoding the the intestine. This leads to a wide vari- HLA molecules DQ2 or DQ8. Indi- ety of symptoms that can impact diges- viduals must be exposed to dietary glu- tive function and also impact extra- ten for the disease to occur and, in most intestinal sites (Reilly et al., 2012). patients, the disease regresses and While the rate of diagnosis of coeliac eventually heals when gluten is re- disease has increased dramatically moved from the diet (Murray et al., (Ludvigsson et al., 2013), this rate of 2004). The disease mostly affects Cau- diagnosis probably still greatly under- casians; however, this broadly affects estimates the proportion of patients many ethnic and racial groups across affected, with most patients remaining Europe, the Middle East, North Africa, undiagnosed (Rubio-Tapia et al., India, and the Near East. It also affects 2012). The treatment of coeliac disease Caucasians living in other areas, such is based on avoidance of dietary gluten as North America, South America, and and, when diagnosed and treated in Australia (Rubio-Tapia and Murray, childhood, usually healing is prompt 2007). The damage to the intestine pro- and complete. However, patients diag- duces inflammation as well as impacts nosed as adults heal much less com-

monly and often take much longer to disease, the impact of the microbiome heal (Lebwohl et al., 2013). This re- on coeliac disease, as well as the poten- view will focus on what is known tial impact of gluten on the microbiome about the triggering factors for coeliac and gut function.

AETIOLOGY

Coeliac disease affects individuals who to genes that regulate immune response have acquired genetic predisposition as and inflammation predominately (Du- well as ingesting gluten, the storage bois et al., 2010). Some of these genes protein from wheat, barley and rye may regulate immune responsiveness (Figure 1). If all that was required for to microbial stimulation of the innate coeliac disease to occur was the car- system. However, the attributable risk riage of the appropriate risk factors as of coeliac disease to these other loci is well as the ingestion of sufficient glu- relatively low, probably contributing ten on a daily basis, then fully 30% of no more than 10-15% of additional ge- the Caucasian population would de- netic risks. It is likely that coeliac dis- velop the disease. However, between 1- ease is the result of the major HLA sus- 2% of Caucasian population developed ceptibility genes combined with several the disease, indicating that there must other common genetic polymorphisms be other factors responsible for trigger- that increase immune responsiveness. ing the disease (Walker et al., 2010). Many of these genes associated with these loci are common to other inflam- Genetic Basis of the Disease matory conditions, such as rheumatoid Coeliac disease is associated primarily arthritis, Crohn’s disease and type 1 with the HLA genes initially thought to diabetes, though there are some inter- be Class 1 genes A1B8, but sub- esting genes that are negatively associ- sequently identified to be the HLA ated with particularly type 1 diabetes. molecules DQ2 or DQ8. These are en- The precise risk within family mem- coded by DQA1’05:DQB1’02XX and bers varies depending on how close DQA1’03XXDQB1’0302, respective- genetically the proband is to the pa- ly. The carriage of one or other gene tient. Monozygous twins have the high- pairs is essential for the development of est concordance rate of 80%. Note, this disease; however, this is not sufficient is not 100% as there must be some en- for the disease to occur. The known vironmental triggering differences be- HLA genes contribute probably no tween even identical twins. Siblings more than 50% of the genetic familial who share HLA risk factors have a risk for coeliac disease (Bevan et al., greater risk than parents or children of 1999). Siblings who carry the HLA patients with coeliac disease (Book et type have an increased risk of coeliac al., 2003). Second-degree relatives disease compared to those who do not likely have a lower risk of coeliac dis- (Murray et al., 2007). Several genome- ease (Fasano et al., 2003). wide association studies (GWAS studies), now incorporating many thou- Dietary Gluten sands of patients and controls, have The primary and required environmen- identified many other gene loci that are tal factor for coeliac disease is dietary associated with risk of disease (Garner gluten. Gluten in this context represents et al., 2009). These other loci are close the stored proteins from wheat, barley,

and rye, often given the term prola- flammation. There are also microbial mines. These storage proteins provide transglutaminases present within the the nitrogen store needed for seed ger- gut lumen and some have been used in mination. The original work identifying food processing. The transglutaminases the protein fraction of these grains as act to remove the amine side chain spe- being deleterious was performed in a cific glutamines in the gliadin peptides. series of challenge and withdrawal When these amine groups are removed, studies done in children in the Nether- so-called deamidation, it renders the lands during and subsequent to the 2nd peptide far higher binding affinity to World War (Dicke, 1951). It was these the class 2 HLA molecule binding site seminal observations that laid the (Shan et al., 2002). By so binding, it groundwork for the modern treatment then results in dramatically increased of coeliac disease. Since that time, fur- affinity to the T cells, which then pro- ther work was done to identify the most duce both proliferate and trigger an in- immunogenic fragments of gluten, and flammatory cascade within the intes- these fragments are characterized by a tine. Further complicating matters is large percentage of amino acids that the very rich genetic material of these glutamines and prolines. Particular mo- grains. Wheat used for bread is hexa- tifs characterized by sequences of glu- ploid and has multiple repeat regions tamines interspersed with prolines ap- where the genes for these storage pro- pear to be particularly immunogenic teins. Hence, there are many (probably (Jabri and Sollid, 2006). In addition to 50-100) epitopes within each wheat their immunogenicity, is also their re- protein that can produce coeliac-resistance of endopeptidase activity sponses from T cells derived from the within the intestine. Within the human small intestine. Children seem to re- intestine, endopeptidases failed to spond more to native gliadin peptides break down some of the most immuno- than do adults and, in general, the de- genic fragments of the gluten-derived amidated peptides produce a much proteins, particularly a gliadin. A spe- more potent effect than do the native, cific 33-mer peptide or a gliadin is es- non-deamidated peptides. The peptides pecially resistant to digestion. This operant in DQ2 positive coeliacs differ same 33-mer peptide contains within it from those in DQ8 positive coeliacs. 3 very immunogenic peptides that, when each is complexed with the T Triggers cell, are presented to the T cells within It has also become clear from epidemi- the small intestine by the HLA mole- ologic studies that coeliac disease can cule on the antigen-presenting cells, occur at any age (Lohi et al., 2007). Pa- produce a very potent inflammatory tients, in particular children, at genetic response. The binding characteristics of risk don’t necessarily develop immune these peptides can be greatly enhanced responses to gluten immediately. This by deamidation of specific glutamine may occur at discontinuous times over peptides, particularly glutamine amino the 1st years of life. In addition, patients acids (Molbert et al., 1998). Deami- with type 1 diabetes who are often dation occurs in response to transglu- screened regularly for coeliac disease taminase enzyme effects. Transglu- may develop coeliac disease later at the taminases are enzymes present within approximate age of 10 or 11, despite the intestine and elsewhere. They are having been negative for some years expressed constitutively but are espe- previously despite being on a gluten- cially expressed in the context of in- containing diet. Most intriguingly, el-

derly patients who were negative for Work has been done showing that coeliac disease may subsequently turn there are changes in the microbiota in positively (Norris et al., 2005; Vilppula coeliac disease. For example, there are et al., 2009). The fact that patients, who changes in the population of microbiota apparently have tolerated gluten, then on the duodenal mucosa. One study has can lose tolerance to gluten and de- suggested increased bacterial adher- velop coeliac disease suggests that ence to the duodenal mucosa of rod- there may be environmental triggers. like bacteria in coeliac disease (Ou et These factors may start at birth. Two al., 2009). It is not known if this studies have suggested that children phenomenon is primary or secondary to born by Caesarean section are at in- the damaged epithelium and loss of creased risk of developing coeliac dis- mucus that occurs in the context of ease in childhood (Marild et al., 2012; coeliac disease. Reductions in the Decker et al., 2010). Children born in proportions of Lactobacillus and the summer months who will likely Bifidobacterium species have also been wean in the winter are more prone to reported in the faecal samples from the disease (Ivarsson et al., 2003). treated coeliac patients (Nistal et al., While not studied in this largely epide- 2012). Another group has shown that miologic study, it is well known that there are differences in Bacteroides route of delivery affects the gut micro- species and pathogenic features in coe- biota and even humoral immunity liac patients as compared to healthy (Huurre et al., 2008). controls (Sanchez et al., 2012) based on In addition, coeliac disease may faecal samples. also be associated with intrauterine These results seem to be contradic- growth retardation and neonatal infec- tory, but there are many explanations tions (Sandberg-Bennich et al., 2002). as to why they are not. The first expla- After birth, further exposures may be nation is that the majority of earlier important for the occurrence of coeliac identification analyses were based on disease; in particular, many studies on culture screening methods; whereas breast-feeding have suggested that the today, the majority of studies are based failure of overlap between breast-feed- on the sequencing of specific regions of ing and the introduction of gluten the 16srRNA gene. The second expla- might increase the risk of coeliac dis- nation is that some studies evaluate du- ease in childhood. Indeed, the now odenal biopsies, while others evaluate well-described Swedish epidemic of faecal samples, both of which would be infant coeliac disease was ascribed to a sampling different niches that different combination of an abrupt termination bacteria can survive. Thus, all of these of breast-feeding with a dramatic in- studies have provided valuable infor- creased concentration of gluten in fol- mation as to how the composition of low-on formula (Ivarsson et al., 2002). the intestinal microbiome of coeliac pa- Recurrent rotavirus infection alone was tients are different from healthy con- initially thought to be an independent trols; the next step is to determine if risk factor for coeliac disease occurring any of these changes are causative for in infancy. However, subsequent work or consequential to intestinal damage. has demonstrated that perhaps it is Previous studies have suggested recurrent rotavirus infection coinciding that two different types of situations with cereal introduction that truly in- could trigger intestinal inflammation in creases the risk for developing coeliac coeliac disease. The one type would be disease (Stene et al., 2006). decreased levels of anti-inflammatory

bacteria such as Lactobacillus and occurrence of an inflammatory insult, Bifidobacterium, the other, an increase such as rotavirus in childhood, along in inflammation promoting bacteria. In with the introduction of a novel pep- one study, Bacteroides fragilis was in- tide, such as gluten, especially one to creased in both untreated and treated which the child is genetically suscepti- coeliac patients, and the Bacteroides ble to react to, may result in activation fragilis associated virulence factors, bft of dendritic cells and presenting cells and MPII , were also increased in the towards a pro-inflammatory response, coeliac patients as compared to healthy thereby skewing naïve T cells to be- controls (Nistal et al., 2012). All of come effector T cells producing in- these studies demonstrate then, that flammation as opposed to T-reg cells. dysbiosis of the intestinal microbiota is It is also possible that in patients who present in coeliac patients and that this have already established tolerance to may indeed be causative as opposed to gluten that this tolerance could be bro- consequential. Potentially then, a com- ken in the context of severe on-going bination of probiotics and targeted re- inflammation. Furthermore, wheat and moval of specific intestinal bacterial like cereals contain other substances species would greatly improve the that can incite innate responses (Junker health of coeliac patients. et al. 2012) and indeed do so in ways There also appears to be differences that simulate an LPS-like effect of bac- in immunoglobulin binding of bacteria teria (Yamazaki et al., 2008). within the gut. A recent paper suggested that in patients with coeliac dis- Uptake of Gliadin Molecules ease, both treated and untreated, there There are two primary pathways by is reduction of immunoglobulin-bound which the peptides are thought to be bacteria within the intestine as com- taken up. One is an active pathway pared to normal. It is interesting that transcellularly stimulated by IFN-γ patients with selective IgA deficiency, through the aberrant expression of IgA who do not secrete any IgA, are at receptors on the surface (Bethune et al., greatly increased risk of coeliac dis- 2009). In this circumstance, IgA-bound ease, with about 10% of patients devel- gliadin is taken up and transported oping coeliac disease. Recent cases and through the enterocyte into the basolat- work by Naval Medical Research in the eral surface where the gliadin peptide is Department of Defense has suggested released and then processed by antigen- that coeliac disease may be more likely presenting cells and presented to the T to be diagnosed after an infectious cells responsive to gliadin. The other gastroenteritis event (Verdu et al., pathway is the paracellular pathway. It 2007; Welander et al., 2010; Riddle et is has been shown the gluten acutely al., 2012). It can be speculated that increases gut permeability (Lammers et these infectious events may result in al., 2008). It causes disruption of inter- temporary inflammation activation of cellular tight junctions and this is likely stress responses within the small modulated through the release of a pep- intestinal mucosa, stimulate the innate tide called zonulin, also recently shown immune system, which will then drive to be prehaptoglobin 2 (Tripathi et al., the adaptive response. Excess numbers 2009). The zonulin appears to cause of bacteria can also be seen in the uncoupling of tight junctions that may duodenum of patients with unrespon- allow the transit of some gluten pep- sive coeliac disease or even before tides in a paracellular pathway, but also treatment (Tursi et al., 2003). Also, co- allow access of other antigens, particu-

Figure 2: The maelstrom of inflammation that leads to established coeliac disease requires many conspirators affecting both the innate and adaptive parts of the immune system. Reprinted with permission from: Nehra, V., Marietta, E., and Murray, J.A.: Coeliac disease. In: Encyclopedia of human nutrition (Caballero, B., Allen, L., Prentice, A., Eds.). Elsevier, Oxford, 1, 407-418 (2006).

larly bacterial antigens that may excite fects on permeability (Kelly et al., an innate and amplify an adaptive re- 2012). This permeability can be exacer- sponse to gluten (Jabri et al., 2005). bated by the use of an agent that in- Recent clinical trials have shown po- creases inflammation and blocked by tential benefit of larazotide acetate, a larazotide acetate (Natividad et al., modulatory inhibitory of gluten’s ef- 2009).

IMMUNOPATHOGENESIS

The immunopathogenesis of coeliac cells, while generally thought to have a disease involves several cell types regulatory effect, in the context of coe- within the intestine (Jabri et al., 2009). liac disease can become cytotoxic and The enterocyte or epithelial cells lining can express NK receptors on their sur- the intestine become distressed and ex- face (Meresse et al., 2004). These NK press aberrant class 1 molecules on receptors interacting with the class 1 their surface. These also can express molecules on the enterocyte can cause IL15. Intraepithelial lymphocytes are enterocyte injury. These intraepithelial predominately CD8+ T cells, and these enterocytes also respond to IL15, either

produced from the lamina propria or by glutaminase are also elaborated within enterocytes on the surface in response the intestinal mucosa they are secreted, to stress. Key essential ingredients to and also increased in circulation. The coeliac disease pathogenesis are gluten- IgA isotype antibodies are more responsive CD4 cells in the lamina pro- commonly expressed, particularly pria. These CD4 cells respond to spe- directed against the autoantigen tissue cific gluten peptides presented in the transglutaminase. However, for gliadin context of class 2 molecules and pro- or deamidated gliadin antibodies, the duce a cascade of cytokines character- IgG is equally expressed, suggesting ized by INFγ and to a lesser extent IL2 that the primary target for the humoral and TNF-α. These cytokines drive a response is the exogenous antigen and cascade of responses that produce both not the autoantigen. Making coeliac cellular as well as humoral responses disease further different from many (Figure 2). The cellular responses fur- autoimmune diseases, with which it ther cascade, drawing in macrophages shares genetic predisposition, is the ob- and activating other cells, including a servation that the antibodies diminish complement system, neutrophils, eosin- in quantity once the exogenous antigen ophils and mast cells. These combina- is removed. Indeed typically patients tion cells like lead to the destruction of who have been on a gluten-free diet for the architecture of the small intestine, >1 year will become IgA negative. resulting in increased thickening or Other antibody responses are often crypt hyperplasia, villous atrophy, dis- seen. These are antibodies directed at ruption of enterocyte function and ab- microbial antigens present within the sorption, increased secretion, and con- intestine. For example, false positive sequent inflammation and malabsorp- ASCA antibodies are seen in the con- tion. Metalloproteases are elaborated, text of coeliac disease and usually these which further alter the architecture. The diminish or disappear with treatment of inflammatory response also includes a the patient with a gluten-free diet, sug- potent humoral response and antibodies gesting that these are seen because of directed against the deamidated gliadin injury. peptides as well as against tissue trans-

DIAGNOSIS

Coeliac disease is typically detected transglutaminase was the endomysial first by serologic tests. The primary Ig antibody test done by indirect immu- serologic test and most accurate sero- nofluorescence. This test has a very logic test is the tissue transglutaminase high specificity, virtually 100% in most IgA antibody with sensitivities of ap- laboratories, but a sensitivity that is proximately 95% and specificities of quite variable, likely due to variability 95% (Rostom et al., 2005; Hadithi et in laboratory methods and interpreta- al., 2007). The performance of this test tions (Li et al., 2009). Accompaniment has been further refined in that patients of the tissue transglutaminase IgA test with extremely positive tests >10 times is a measurement of total IgA. This is the upper limit of normal are essen- performed in order to detect those pa- tially almost guaranteed to have coeliac tients with selective IgA deficiency that disease (Klapp et al., 2013). The test not only are at greatly increased risk of which immediately preceded tissue coeliac disease but for whom the stand-

ard IgA-based tests will be negative antibodies have reduced sensitivity even in the setting of coeliac disease when the patient reduces gluten intake (Rubio-Tapia et al., 2013). For patients (Rashtak et al., 2008b). who have known or suspected selective Intestinal biopsies showing the IgA deficiency, deamidated gliadin IgG classic features of villous atrophy, or tissue transglutaminase IgG tests can crypt hyperplasia, and increased intra- be performed, though their sensitivity epithelial lymphocytes are still re- is not perfect for coeliac disease garded as the mainstay of diagnosis. (Rashtak et al., 2008a). Tissue transglu- Adequate numbers of samples should taminase IgG is not useful in the con- be taken (Lebwohl et al., 2012). Recent text of normal IgA levels (Rashtak et guidelines promulgated by the Euro- al., 2008a). Antibodies directed against pean Society for Pediatric Gastroenter- native gliadin antibodies are not of any ology and Hepatology and Nutrition additional benefit and indeed are have suggested that in patients meeting fraught with greatly reduced specificity certain very strict criteria, a biopsy may and can negatively impact the accuracy be unnecessary to confirm the diagno- of diagnostic testing when included in sis (Husby et al., 2012, 2013; Rubio- diagnostic panels for coeliac disease Tapia et al., 2013). (Rubio-Tapia et al., 2013). All of the

TREATMENT

Once the disease is detected and con- often necessitated in patients who have firmed by biopsy, then treatment is B12 deficiency due to ileal resection or usually instituted. Treatment includes a pernicious anaemia (Murray and Ross, strict, gluten-free diet. While theoreti- 2004). cally simple, there are many challenges In adult patients diagnosed with to be able to adhere to a gluten-free diet coeliac disease, bone density is typi- long-term. Patients are usually evalu- cally performed as diminished bone ated for any vitamin deficiencies that density, either due to osteomalacia or may require repletion. Most common is osteoporosis, is very common. Fracture iron deficiency, calcium, vitamin D and risk is increased in patients with coeliac other fat-soluble vitamins. B12 defi- disease prior to diagnosis, and the rate ciency can affect over 20% of patients does not apparently drop in follow-up with coeliac disease. It is unclear if this (Jafri et al., 2008). mandates parenteral B12 for life as is

PROGNOSIS

Most patients diagnosed as children risk of complications and increased will promptly respond to a gluten-free mortality, especially within the 1st year diet and will resume normal growth of diagnosis (Rubio-Tapia et al., 2010; and development. As long as they re- Lanzini et al., 2009). This seems to be a main on a gluten-free diet, they will relatively modest increase in overall remain healed and well. Adult patients mortality in diagnosed coeliac disease. diagnosed with coeliac disease heal Patients with undiagnosed coeliac dis- much more slowly and are at increased ease, however, may be at greatly in-

creased risk of mortality with an almost tion of coeliac disease, in particular 4-fold increase in mortality over 45 geography, may impact the likely long- years (Rubio-Tapia et al., 2009). The term mortality. Early detection couple increase in mortality associated with with vigorous treatment with a gluten- coeliac disease is not universally found free diet is most likely to result in a and variances such as the rate of detec- good outcome.

FUTURE TREATMENTS

Several new treatments are in develop- den Broeck et al., 2009). Other ap- ment for coeliac disease given that we proaches, such as an immunotherapy know much about the processes in- approach to try to reduce tolerance to volved. Treatments directed at altering gluten, are also under study (Senger et the nature of wheat itself, detoxifying al., 2003; Vickery et al., 2009; Keech et the wheat within the gastrointestinal al., 2009). Other targets, such as block- tract or even before ingestion, binding ing the deamidation by tissue transglu- the peptides within the intestine to pre- taminase and blocking the binding of vent presentation to the immune system the peptides to DQ2 or DQ8, are also (Liang et al., 2010). The modulation of areas that have sparked development of the permeability or prevention of in- agents in preclinical study (Xia et al., creased permeability induced by glu- 2007; Klock et al., 2011). Also, use of ten, as well as the degradation of the worms to skew the immune response immune peptides within the gut by po- away from a coeliac-like response has tent endopeptidases all are in pre-clini- been used in humans (Daveson et al., cal or even clinical trials (Gass et al., 2011). 2007; Cerf-Bensussan et al., 2007; van

PREVENTION

Coeliac disease has dramatically in- coeliac disease are under study. The creased in prevalence over the last 50 Prevent CD trial primarily in Europe years (Rubio-Tapia et al., 2009). This and North America addresses the issue is almost certainly due to environmen- of timing of introduction and quantity tal factors. Identifying and abrogating of gluten in the infant diet. Other stud- these environmental factors is crucial if ies looking at using systems biology we are to stem the tide of coeliac dis- approach to address the role of the mi- ease. Whilst it affects still somewhat crobial community on tolerance to glu- just less than 1% of the population of ten is also under study. It is interesting the U.S., it now affects over 2% of to note that the phenomena of oral other Caucasian populations such as tolerance require the presence of Sweden and Finland, rates that appear microbiota within the gut. Studies look- to be increasing (Dube et al., 2005). It ing at the co-administration of benefi- is vitally important to identify the fac- cial bacterial with gliadin molecules tors that have led to this increase so have also been studied and may prevent that this increase can be mitigated. Sev- or reverse gluten sensitization (Hui- eral approaches to the prevention of bregtse et al., 2009).

SUMMARY

In summary, coeliac disease is an in- interaction between the environment creasingly common chronic disease and immune system, and may present affecting primarily the upper small in- an example of a disorder that can be the testine associated with significant mor- result of triggering of loss of homeosta- bidity and mortality. It is often over- sis due to food or microbial influences looked, but presents both opportunities within the intestine. and challenges for understanding of the

LITERATURE

Book, L., Zone, J.J., and Neuhausen, S.L.: disease--a randomised double-blind Prevalence of celiac disease among placebo controlled trial. PLoS One 6, relatives of sib pairs with celiac disease in e17366 (2011). U.S. families. Am. J. Gastroenterol. 98, Decker, E., Engelmann, G., Findeisen, A., 377-381 (2003). Gerner, P., Laass, M., Ney, D., Posovszky, Bethune, M.T., Siegel, M., Howles-Banerji,S., C., Hoy, L., and Hornef, M.W.: Cesarean and Khosla, C.: Interferon-gamma released delivery is associated with celiac disease by gluten-stimulate celiac disease-specific but not inflammatory bowel disease in intestinal T cells enhances the children. Pediatrics 125, e1433-e1440 transepithelial flux of gluten peptides. J. (2010). Pharmacol. Exp. Ther. 329, 657-668 Dicke, W.K.: Treatment of celiac disease. Ned. (2009). Tijdschr. Geneeskd. 95, 124-130 (1951). Bevan, S., Popat, S., Braegger, C.P., Busch, A., Dube, C., Rostom, A., Sy, R., Cranney, A., O’Donoghue, D., Faith-Magnusson, K., Saloojee, N., Garritty, C., Sampson, M., Ferguson, A., Godkin, A., Hogberg, L., Zhang, L., Yazdi, F., Mamaladze, F., Pan, Holmes, G., Hosie, K.B., Howdle, P.D., I., Macneil, J., Mack, D., Patel, D., and Jenkins, H., Jewell, D., Johnston, S., Moher, D.: The prevalence of celiac Kennedy, N.P., Kerr, G., Kumar, P., disease in average-risk and at-risk Western Logan, R.F.A., Love, A.H.G., Marsh, M., European populations: a systematic Mulder, C.J.J., Sjoberg, K., Stenhammer, review. Gastroenterology 128, S56-S67 L., Walker-Smith, J., Marossy, A.M., and (2005). Houlston, R.S.: Contribution of the MHC Dubois, P.C.A., Trynka, G., Franke, L., Hunt, region to the familial risk of coeliac K.A., Romanose, J., Curtotti, A., disease. J. Med. Genet. 36, 687-690 Zhernakova, A., Heap, G.A.R., Adany, R., (1999). Aromaa, A., Bardella, M.T., van den Berg, Cerf-Bensussan, N., Matysiak-Budnik, T., L.H., Bockett, N.A., de la Concha, E.G., Cellier, C., and Heyman, M.: Oral Dema, B., Fehrmann, R.S.N., Fernandez- proteases: a new approach to managing Arquero, M., Fiatal, S., Grandone, E., coeliac disease. Gut 56, 157-160 (2007). Green, P.M., Groen, H.J.M., Gwiliam, R., Daveson, A.J., Jones, D.M., Gaze, S., Houwen, R.H.J., Hunt, S.E., Kaudinen, K., McSorley, H., Clouston, A., Pascoe, A., Kelleher, D., Korponay-Szabo, I., Kurppa, Cooke, S., Speare, R., MacDonald, G.A., K., MacMathuna, P., Maki, M., Mazzilli, Anderson, R., McCarthy, J.S., Loukas, A., M.C., McCan, O.T., Mearin, M.L., Mein, and Croese, J.: Effect of hookworm C.A., Mirza, M.M., Mistry, V., Mora, B., infection on wheat challenge in celiac Morley, K.I., Mulder, C.J., Murray, J.A.,

Nunez, C., Oosterom, E., Ophoff, R.A., Koning, F., Rottiers, P., David, C.S., van Polanco, I., Peltonen, L., Plattel, M., Deventer, S.J., and Murray, J.A.: Induction Rybak, A., Salomaa, V., Schweizer, J.J., of antigen-specific tolerance by oral Sperandeo, M.P., Tack, G.J., Turner, G., administration of Lactococcus lactis Veldink, J.H., Verbeek, W.H., Weersma, delivered immunodominant DQ8-restricted R.K., Rolters, V.M., Urcelay, E., gliadin peptide in sensitized nonobese Cukrowska, B., Greco, L., Neuhausen, diabetic Abo Dq8 transgenic mice. J. S.L., McManus, R., Barisani, D., Immunol. 183, 2390-2396 (2009). Deloukas, P., Barrett, J.C., Saavalainen, P., Husby, S., Koletzko, S., Korponay-Szabo, I.R., Wijmegna, C., and van Heel, D.A.: Mearin, M.L., Phillips, A., Shamir, R., Multiple common variants for celiac Troncone, R., Giersiepen, K., Branski, D., disease influencing immune gene Catassi, C., Leigeman, M., Maki, M., expression. Nat. Genet. 42, 295-302 Ribes-Koninckx, C., Ventura, A., Zimmer, (2010). K.P., ESPGHAN Working Group on Falth-Magnusson, K., Franzen, L., Jansson, G., Coeliac Disease Diagnosis, ESPGHAN Laurin, P., and Stenhammar, L.: Infant Gastroenterology Committee, and feeding history shows distinct differences European Society for Pediatric Gastro- between Swedish celiac and reference enterology, Hepatology and Nutrition.: children. Pediatr. Allergy Immunol. 7, 1-5 European Society for Pediatric Gastro- (1996). enterology, Hepatology, and Nutrition Fasano, A., Berti, I., Gerarduzzi, T., Not, T., Guidelines for the diagnosis of coeliac Colletti, R.B., Drago, S., Elitsur, Y., disease. J. Pediatr. Gastroenterol. Nutr. 54, Green, P.H., Guandalini, S., Hill, I.D., 136-60 (2012). Pietzak, M., Ventura, A., Thorpe, M., Husby, S. and Murray, J.A.: New aspects of the Kryszak, D., Fornaroli, F., Wasserman, diagnosis of celiac disease in children, S.S., Murray, J.A., and Horvath, K.: adolescents, and adults. Mayo Clin. Proc. Prevalence of celiac disease in at-risk and 88, 540-543 (2013). not-at-risk groups in the United States: a Huurre, A., Kalliomaki, M., Rautava, S., large multicenter study. Arch. Intern. Med. Rinne, M., Salminen, S., and Isolauri, E.: 163, 286-292 (2003). Mode of delivery - effects on gut Garner, C.P., Murray, .JA., Ding, Y.C., Tien, microbiota and humoral immunity. Z., van Heel, D.A., and Neuhausen, S.L.: Neonatology 93, 236-240 (2008). Replication of celiac disease UK genome- Ivarsson, A., Hernell, O., Stenlund, H., and wide association study results in a US Persson, L.A.: Breast-feeding protects population. Hum. Mol. Genet. 18, 4219- against celiac disease. Am. J. Clin. Nutr. 4225 (2009). 75, 914-921 (2002). Gass, J., Bethune, M.T., Siegel, M., Spencer, Ivarsson, A., Hernell, O., Nystrom, L., and A., and Khosla, C.: Combination enzyme Persson, L.A.: Children born in the therapy for gastric digestion of dietary summer have increased risk for coeliac gluten in patients with celiac sprue. disease. J. Epidemiol. Community Health Gastroenterology 133, 472-480 (2007). 57, 36-39 (2003). Hadithi, M., von Blomberg, B.M., Crusius, Jabri, B., Kasarda, D.D., and Green, P.H.: J.B., Bloemena, E., Kostense, P.J., Meijer, Innate and adaptive immunity: the yin and J.W., Mulder, C.J., Stehouwer, C.D., and yang of celiac disease. Immunol. Rev. 206, Pena, A.S.: Accuracy of serologic tests and 219-231 (2005). HLA-DQ typing for diagnosing celiac Jabri, B. and Sollid, L.M.: Mechanisms of disease. Ann. Intern. Med. 147, 294-302 disease: immunopathogeneis of celiac (2007). disease. Nat. Clin. Pract. Gastroenterol. Huibregtse, I.L., Marietta, E.V., Rashtak, S., Hepatol. 3, 516-525 (2006).

Jabri, B. and Sollid, L.M.. Tissue-mediated Lett. 21, 2692-2696 (2011). control of immunpathology in coeliac Lammers, K.M., Lu, R., Brownley, J., Lu, B., disease. Nat. Rev. Immunol. 9, :858-870 Gerard, C., Thomas, K., Rallabhandi, P., (2009). Shea-Donohue, T., Tamiz, A., Alkan, S., Jafri, M.R., Nordstrom, C.W., Murray, J.A., Netzel-Arnett, S., Antalis, T., Vogel, S.N., Van Dyke, C.T., Dierkhising, R.A., and Fasano, A.: Gliadin induces an Zinsmeister, A.R., and Melton, L.J. 3rd: increase in intestinal permeability and Long-term fracture risk in patients with zonulin release by binding to the celiac disease: a population-based study in chemokine receptor CXCR3. Olmsted County, Minnesota. Dig. Dis. Sci. Gastroenterology 135, 194-204 (2008). 53, 964-971 (2008). Lanzini,A., Lanzarotto, F., Villanacci, V., Junker, Y., Zeissig, S., Kim, S.J., Barisani, D., Mora, A., Bertolazzi, S., Turini, D., Wieser, H., Leffler, D.A., Zevallos, V., Carella, G., Malagoli, A., Ferrante, G., Libermann, T.A., Dillon, S., Freitag, T.L., Cesana, B.M., and Ricci, C.: Complete Kelly, C.P., and Schuppan, D.: Wheat recovery of intestinal mucosa occurs very amylase trypsin inhibitors drive intestinal rarely in adult coeliac patients despite inflammation via activation of toll-like adherence to gluten-free diet. Aliment. receptor 4. J. Exp. Med. 209, 2395-2408 Pharmacol. Ther. 15, 1299-1308 (2009). (2012). Lawler, M., Humphries, P., O’Farrelly, C., Keech, C.L., Dromey, J., Chen, Z.J., Anderson, Hoey, H., Sheils, O., Jeffers, M., O’Briain, R.P., and McCluskey, J.: 355 Immune D.S., and Kelleher, D.: Adenovirus 12 tolerance induced by peptide E1A gene detection by polymerase chain immunotherapy in an HLA DQ2- reaction in both the normal and coeliac dependent mouse model of gluten duodenum. Gut 35, 1226-1232 (1994). immunity. Gastroenterology 136, Suppl. 1, Lebwohl, B., Rubio-Tapia, A., Assiri, A., A-57 (2009). Newland, C., and Guandalini, S.: Kelly, C.P., Green, P.H., Murray, J.A., Diagnosis of celiac disease. Gastrointest. Dimarino, A., Colatrella, A., Leffler, D.A., Endosc. Clin. N. Am. 22, 661-677 (2012). Alexander, T., Arsenescu, R., Leon, F., Lebwohl, B., Granath, F., Ekbom, A., Jiag, J.G., Arterburn, L.A., Paterson, B.M., Montgomery, S.M., Murray, J.A., Rubio- Fedorak,R.N., and Larazotide Acetate Tapia, A., Green, P.H., and Ludvigsson, Celiac Disease Study Group: Larazotide J.F.: Mucosal healing and mortality in acetate in patients with coeliac disease coeliac disease. Aliment. Pharmacol. Ther. undergoing a gluten challenge: a 37, 332-339 (2013). randomized placebo-controlled study. Li, M., Yu, L., Tiberti, C., Bonamico, M., Taki, Aliment. Pharmacol. Ther. 37, 252-262 I., Miao, D., Murray, J.A., Rewers, M.J., (2013). Hoffenberg, E.J., Agardh, D., Mueller, P., Klapp, G., Masip, E., Bolonio, M., Donat, E., Stern, M., Bonifacio, E., and Liu, E.: A Polo, B., Ramos, D., and Ribes-Koninckx, report on the Internatonal Translutaminase C.: Celiac disease: the new proposed Autoantibody Workshop for Celiac ESPGHAN diagnostic criteria do work Disease. Am. J. Gastroenterol. 104, 154- well in a selected population. J. Pediatr. 163 (2009). Gastroenterol. Nutr. 56, 251-256 (2013). Liang, L., Piier, M., Leroux, J.C., and Klock, C., Jin, X., Choi, K., Khosla, C., Subirade, M.: Interaction of alpha-gliadin Madrid, P.B., Spencer, A., Raimundo, with polyanions: design considerations for B.C., Boardman, P., Lanza, G., and sequestrants used in supportive treatment Griffin, J.H.: Acylideneoxoindoles: a new of celiac disease. Biopolymers 93, 418-428 class of reversible inhibitors of human (2010). transglutaminase 2. Bioorg. Med. Chem. Lohi, S., Mustalahti, K., Kaukinen, K., Laurila,

K., Collin, P., Rissanen, H., Lohi, O., gene dosage and risk and severity of celiac Bravi, E., Gasparin, M., Reunanen, A., and disease. Clin. Gastroenterol. Hepatol. 5, Maki, M.: Increasing prevalence of coeliac 1406-1412 (2007). disease over time. Aliment. Pharmacol. Murray, J.A., Rubio-Tapia, A., Van Dyke, Ther. 26, 1217-1225 (2007). C.T., Brogan, D.L., Knipschield, M.A., Ludvigsson, J.F., Rubio-Tapia, A., van Dyke, Lahr, B., Rumalla, A., Zinsmeister, A.R., C.T., Melton, L.J., 3rd, Zinsmeister, A.R., and Gostout, C.J.: Mucosal atrophy in Lahr, B.D., and Murray, J.A.: Increasing celiac disease: extent of involvement, incidence of celiac disease in a North correlation with clinical presentation, and American population. Am. J. response to treatment. Clin. Gastroenterol. Gastroenterol. 108, 818-824 (2013). Hepatol. 6, 186-193 (2008). Marild, K., Stephansson, O., Montgomery, S., Natividad, J.M., Huang, X., Slack, E., Jury, J., Murray, J.A., and Ludvigsson, J.F.: Sanz, Y., David, C., Denou, E., Yang, P., Pregnancy outcome and risk of celiac Murray, J., McCoy, K.D., and Verdu, E.F.: disease in offspring: a nationwide case- Host responses to intestinal microbial control study. Gastroenterology 142, 39-45 antigens in gluten-sensitive mice. PLoS ( 2012). One 4, e6472 (2009). Meresse, B., Chen, Z., Ciszewski, C., Nistal, E., Caminero, A., Vivas, S., Ruiz de Tretiakova, M., Bhagat, G., Krausz, T.N., Morales, J.M., Sáenz de Miera, L.E., Raulet, D.H., Lanier, L.L., Groh, V., Spies, Rodriguez-Aparicio, L.B., and Casqueiro, T., Ebert, E.C., Green, P.H., and Jabri, B.: J. Differences in faecal bacteria Coordinated induction by IL15 of a TCR- populations and faecal bacteria metabolism independent NKG2D signaling pathway in healthy adults and celiac disease converts CTL into lymphokine-activated patients. Biochemie 94, 1724-1729 (2012). killer cells in celiac disease. Immunity 21, Norris, J.M., Barriga, K., Hoffenberg, E.J., 357-366 (2004). Taki, I., Miao, D., Haas, J.E., Emery, Molberg, O., Mcadam, S.N., Körner, R., L.M., Sokol, R.J., Erlich, H.A., Eisenbarth, Quarsten, H., Kristiansen, C., Madsen, L., G.S., and Rewers, M.: Risk of celiac Fugger, L., Scott, H., Norén, O., disease autoimmunity and timing of gluten Roepstorff, P., Lundin, K.E., Sjöström, H., introduction in the diet of infants at Sollid, L.M.: Tissue transglutaminase increased risk of disease. JAMA 293, selectively modifies gliadin peptides that 2343-2351 (2005). are recognized by gut-derived T cells in Ou, G., Hedberg, M., Hörstedt, P., Baranov, V., celiac disease. Nat. Med. 4, 713-717 Forsberg, G., Drobni, M., Sandsröm, O., (1998). Wai, S.N., Johansson, I., Hammarström, Murray, J.A., Watson, T., Clearman, B., and M.-L., Hernell, O., and Hammarström, S.: Mitros, F.: Effect of a gluten-free diet on Proximal small intestinal microbiota and gastrointestinal symptoms in celiac identification of rod-shaped bacteria disease. Am. J. Clin. Nutr. 79, 669-673 associated with childhood celiac disease. (2004). Am. J. Gastroenterol. 104, 3058-3067 Murray, J.A. and Ross, M.A.: Malabsorption (2009). and neurological disease. In: Neurogastro- Pinier, M., Verdu,E.F., Nasser-Eddine, M., enterology (Quigley, E.M.M., Pfeiffer, David, C.S., Vezina, A., Rivard, N., and R.F., Eds.). Butterworth-Heinemann, Phil- Leroux, J.C.: Polymeric binders suppress adelpha, 203-53 (2004). gliadin-induced toxicity in the intestinal Murray, J.A., Moore, S.B., Van Dyke, C.T., epithelium. Gastroenterology 136, 288-298 Lahr, B.D., Dierkhising, R.A., Zinsmeister, (2009). A.R., Melton, L.J., 3rd, Kroning, C.M., El- Rashtak, S., Ettore, M.W., Homburger, H.A., Yousseff, M., and Czaja, A.J.: HLA DQ and Murray, J.A.: Comparative usefulness

of deamidated gliadin antibodies in the 1544; quiz 1537, 1545 (2012). diagnosis of celiac disease. Clin. Rubio-Tapia, A., Hill, I.D., Kelly, C.P., Gastroenterol. Hepatol. 6, 426-432, quiz Calderwood, A.H., and Murray, J.A.: ACG 370 (2008a). clinical guidelines: diagnosis and Rashtak, S., Ettore, M.W., Homberger, H.A., management of celiac disease. Am. J. and Murray, J.A.: Combination testing for Gastroenterol. 108, 656-676; quiz 77 antibodies in the diagnosis of coeliac (2013). disease: coparison of multiplex Sanchez, E., Laparra, J.M., and Sanz, Y.: immunoassay and ELISA methods. Discerning the role of Bacteriodes fragilis Aliment. Pharmacol. Ther. 28, 805-813 in celiac disease pathogenesis. Appl. (2008b). Environ. Micobiol. 78, 6507-6515 (2012). Reilly, N.R., Fasano, A., and Green, P.H.: Sandberg-Bennich, S., Dahlquist, G., and Presentation of celiac disease. Gastrointest. Kallen, B.: Coeliac disease is associated Endosc. Clin. N. Am. 22, 613-221 (2012). with intrauterine growth and neonatal Riddle, M.S., Murray, J.A., and Porter, C.K.: infections. Acta Paediatr. 91, 30-33 (2002). The incidence and risk of celiac disease in Senger, S., Luongo, D., Maurano,F., Mazzeo, a healthy U.S. adult population. Am. J. M.F., Sicilliano, R.A., Gianfrani, C., Gastroenterol. 107, 1248-1255 (2012). David, C., Troncone, R., Auricchio, S., and Rostom, A., Dube, C., Cranney, A., Saloojee, Rossi, M.: Intranasal administration of a N., Sy, R., Garritty, C., Sampson, M., recombinant alpha-gliadin down-regulates Zhang, L., Yazdi, F., Mamaladze, V., Pan, the immune response to wheat gliadin in I., MacNeil, J., Mack, D., Patel, D., and DQ8 transgenic mice. Immunol. Lett. 88, Moher, D.: The diagnostic accuracy of 127-134 (2003). serologic tests for celiac disease: a Shan, L., Molberg, O., Parrot, I., Hausch, F., systematic review. Gastroenterology 128, Filiz, F., Gray, G.M., Sollid, L.M., and S38-S46 (2005). Khosla, C.: Structural basis for gluten Rubio-Tapia, A. and Murray, J.A.: Celiac intolerance in celiac sprue. Science 297, disease. In: GI Epidemiology (Talley, N.J., 227-2279 (2002). Locke, G.R. III, and Saito, Y.A., Eds.). Stene, L.C., Honeyman, M.C., Hoffenberg, Blackwell Publishing, Malden, 157-163 E.J., Haas, J.E., Sokol, R.J., Emery, L., (2007). Taki, I., Norris, J.M., Erlich, H.A., Rubio-Tapia, A., Kyle, R.A., Kaplan, E.L., Eisenbarth, G.S., and Rewers, M.: Johnson, D.R., Page, W., Erdtmann, F., Rotavirus infection frequency and risk of Brantner, T.L., Kim, W.R., Phelps, T.K., celiac disease autoimmunity in early Lahr, B.D., Zinsmeister, A.R., Melton, childhood: a longitudinal study. Am. J. L.J., 3rd, and Murray, J.A.: Increased Gastroneterol. 101, 2333-2340 (2006). prevalence and mortality in undiagnosed Tripathi, A., Lammers, K.M., Goldblum, S., celiac disease. Gastroenterology 137, 88- Shea-Donohue, T., Netzel-Arnett, S., 93 (2009). Buzza, M.S., Antalis, T.M., Vogel, S.N., Rubio-Tapia, A., Rahim, M.W., See, J.A., Zhao, A., Yang, S., Arrietta, M.C., Lahr, B.D., Wu,T.-T., and Murray, J.A.: Meddings, J.B., and Fasano, A.: Mucosal recovery and mortality in adults Identification of human zonulin, a with celiac disease after treatment with a physiological modulator of tight junctions, gluten-free diet. Am. J. Gastroenterol. 105, as prehaptoglobin-2. Proc. Natl. Acad. Sci. 1412-1420 (2010). USA 106, 16799-16804 (2009). Rubio-Tapia, A., Ludvigsson, J.F., Brantner, Tursi, A., Brandimarte, G., and Giogetti, G.: T.L., Murray, J.A., and Everhart, J.E.: The High prevalence of small intestinal prevalence of celiac disease in the United bacterial overgrowth in celiac patients with States. Am. J. Gastroenterol. 107, 1538- persistence of gastrointestinal symptoms

after gluten withdrawal. Am. J. Gastroen- Krekela, I., Patrikainen, H., Valve, R., terol. 98, 839-843 (2003). Luostarinen, M., Laurila, K., Maki, M., van den Broeck, H.C., van Herpen, T.W., and Collin, P.: Clinical benefit of gluten- Schuit, C., Salentijn, E.M., Dekking, L., free diet in screen-detected older celiac Bosch, D., Hamer, R.J., Smulders, M.J., disese patietns. BMC Gastoenterol 11, 135 Gilissen, L.J., and van der Meer, I.M.: (2011). Removing celiac disease-related gluten Walker, M.M., Murray, J.A., Ronkainen, J., proteins from bread wheat while retaining Aro, P., Storskrubb, T., D’Amato, M., technological properties: a study from Lahr, B., Talley, N.J., and Agreus, L.: Chinese Spring deletion lines. BMC Plant Detection of celiac disease and lympho- Biol. 9, 41 (2009). cytic enteropathy by parallel serology and Verdu, E.F., Mauro, M., Bourgeois, J., and histopathology in a population-based Armstrong, D.: Clinical onset of celiac study. Gastroenterology 139, 112-119 disease after an episode of Campylobacter (2010). jejuni enteritis. Can. J. Gastroenterol. 21, Welander A., Tjernberg, A.R., Montgomery, 453-455 (2007). S.M., Ludvigsson, J., and Ludvigsson, J.F.: Vickery, B.P. and Burks, A.W.: Infectious disease and risk of later celiac Immunotherapy in the tretment of food disease in childhood. Pediatrics 125, e530- allergy: focus on oral tolerance. Curr. e536 (2010). Opin. Allergy Clin. Immunol. 9, 364-370 Xia, J., Bergseng, E., Fleckenstein, B., Siegel, (2009). M., Kim, C.Y., Khosla, C., and Sollid, Vilppula, A., Kaukinen, K., Luostarinen, L., L.M.: Cyclic and dimeric gluten peptide Krekela, I., Patrikainen, H., Valve, R., analogues inhibiting DQ2-mediated anti- Maki, M., and Collin, P.: Increasing gen presentation in celiac disease. Bioorg. prevalance and high incidence of celiac Med. Chem. 15, 6565-6573 (2007). disease in elderly people: a population- Yamazaki, K., Murray, J.A., and Kita, H.: In- based study. BMC Gastoenterol 9, 49 nate immunomodulatory effects of cereal (2009). grains through induction of IL-10. J. Al- Vilppula, A., Kaukinen, K., Luostarinen, L., lergy Clin. Immunol. 121, 172-178 (2008).

ACUTE ENTERIC INFECTIONS ALTER COMMENSAL MICROBIOTA: NEW MECHANISMS IN POST-INFECTIOUS INTESTINAL INFLAMMATORY DISORDERS

ANDRE G. BURET and KRISTEN RETI

Department of Biological Sciences, Inflammation Research Network, Host-Parasite Interactions Program, Snyder Institute for Chronic Diseases, University of Calgary, AB, Canada

SUMMARY

This chapter discusses how acute enteric infections may lead to post- infectious complications. Particular emphasis is given to infections with Campylobacter jejuni and Giardia intestinalis, two of the most common causes of enteric infections worldwide. The review provides a critical discussion of the biology of the human intestinal microbiota. Chronic post-infectious sequelae of these infections include malnutrition, stunting, failure to thrive, and impaired cognitive functions. They may also cause post-infectious disease at extra-intestinal sites, including the joints, the skin, the eyes, the lungs, the heart, the muscles, the kidneys, and the central nervous system. Findings from recent and on- going research suggest that enteropathogen-induced disruptions of the commensal microbiota may at least in part play a role in triggering the sequence of events that result in these presentations. These disruptions include a promotion of the transcellular and paracellular translocation of non-invasive commensal bacteria, as well as disruptions of the composition and structure of microbiota biofilms.

INTRODUCTION

Diarrhoeal disease resulting from en- nutritional and immune status, as well teric infections remains one of the ma- as environmental factors, like co-infec- jor causes of morbidity and mortality tions, are important in determining worldwide. Each year, an estimated 2 symptom severity. Indeed, some en- to 2.5 million children under the age of teropathogens worsen the outcome of 5 years die from the 1.4 billion yearly concurrent infections while others, like diarrhoeal episodes in the paediatric Giardia intestinalis, may partly protect population of developing countries children against diarrhoeal disease in (O'Ryan et al., 2005). Approximately developing countries (Moore, 2001; 20 parasitic, viral, and bacterial patho- Jensen et al., 2009; Veenemans et al., gens are known to be the most common 2012). Intestinal parasitic helminths are causative aetiologies (Table 1). Infec- known to posses potent immune-regu- tion most commonly occurs through lating properties that may help attenu- ingestion of contaminated food or wa- ate tissue damage (Maizels et al., ter, or through direct faecal oral infec- 2003), but overall, the mechanisms tion. Host factors, such as the host’s directing the clinical outcome of co-

Table 1: Enteropathogens most commonly reported as causes of acute diarrhoea worldwide. Most of these have been reported to have long-term sequelae via mechanisms that remain incompletely understood. ——————————————————————————————————————— Parasites: Giardia intestinalis (syn. lamblia, or duodenalis) Cryptosporidium hominis and C. parvum Entamoeba histolytica

Bacteria: Campylobacter jejuni and C. coli Diarrhoeagenic Escherichia coli (ETEC- enterotoxigenic; EPEC - enteropathogenic; STEC - shiga-toxin producing; EAEC; enteroadherent; EIEC - enteroinvesive) Salmonella sp. Shigella sp. Vibrio cholera Aeromonas sp.

Viruses: Rotavirus Norovirus Sapovirus Astrovirus Enteric adenovirus ———————————————————————————————————————

infections and polyparasitism remain thrive, stunted growth, and impaired obscure. In developed countries of the cognitive functions. Recent studies fol- World, public media warn that preva- lowing outbreaks of intestinal infec- lence of enteric infections seems to in- tions, and large retroactive cohort stud- crease. Indeed, the Wall Street Journal ies, also found that these infections (3/15/2012, Martin) reported that ac- may be responsible for chronic fatigue cording to CDC estimates, every year syndrome, arthritis, irritable bowel syn- approximately 48 million Americans drome (IBS), and flare-ups in patients become ill through contaminated food, with Inflammatory Bowel diseases and 3,000 die. In its "National Brief- (IBD) (Rodriguez and Ruigomez, 1999; ing," the New York Times (3/15/2012, Riddle et al., 2001; Berkman et al., A20, Grady) reported that the CDC ob- 2002; Gradel et al., 2009; Kalischuk serves that "Deaths from gastrointesti- and Buret, 2010; Moore et al., 2011; nal infections more than doubled in the Wensaas et al., 2012). As a result, from United States from 1997 to 2007, to being a leading cause of global child more than 17,000 a year from 7,000 a death, infectious diarrhoea now appears year. to have become a key source of life- Adding to the raising concerns long morbidity (Table 2). The causes of caused by acute enteric infections, re- the post-infectious clinical manifesta- cent observations indicate that post- tions due to enteric infections, even af- infectious complications may arise fol- ter complete elimination of the entero- lowing exposure to a variety of enteropathogen, remain obscure. However, pathogens, including Campylobacter the commonality of these post-infec- jejuni, diarrhoeagenic Escherichia coli, tious disorders raises the intriguing Salmonella sp., Shigella sp., Crypto- possibility that they may share at least sporidium parvum, and Giardia intesti- some of their basic biological mecha- nalis. In developing countries, acute nisms, hence offering great potential diarrhoeal disease caused by these en- for the identification of novel therapeu- teropathogens can lead to failure to tic targets. This chapter elaborates on

Table 2: Examples of post-infectious complications and disorders reported following enteric infections, with C. jejuni, E. coli, Salmonella sp., Shigella sp., Cryptosporidium parvum, G. intestinalis, or viral enteropathogens. ——————————————————————————————————————————————————————————— Affected site: Disorder/condition: References: ——————————————————————————————————————————————————————————— Intestine: Food allergies Farthing et al., 1986; Hardin et al., 1997; Di Prisco et al., 1998 Post-infectious irritable bowel syndrome Rodriguez and Ruigomez, 1999; Spiller et al., 2000; Riddle et al., 2001;; Dizdar et al., 2007; Stark et al., 2007; Thabane et al., 2007; Gradel et al., 2009; Hanevik et al., ,2009; Marshall et al., 2010; Robertson et al., 2010; Wensaas et al., 2012; Buret et al., 2013; Simren et al., 2013 Flare-ups in inflammatory bowel diseases Kalischuk and Buret, 2010; Riddle et al., 2001; Swidsinsky et al., 2009; Reiff and Kelly, 2010; Buret et al., 2013 Coeliac disease Riddle et al., 2001

Joints: Arthritis Keat, 1991; Borman et al., 2001; Carlson and Finger, 2004; Schiellerup et al., 2008; Scher and Abramson, 2011

Skin: Urticaria/Pruritus/Dermatitis Di Prisco et al., 1998; Giacometti et al., 2003; Pietrzak et al., 2005

Eyes: Iridocyclitis/Choroiditis/Retinal haemorrhages Pettoelo-Mantovani et al., 1990; Corsi et al., 1998

Lungs: Asthma, Di Prisco et al., 1998; Han et al., 2012 Obstructive lung disease Hity et al., 2010

Heart: Endocarditis Miki et al., 2005

Muscles: Hypokalaemic myopathy Cervelló et al., 1993; Addiss and Lengerich, 1994; Genovese et al., 1996

CNS: Impaired cognitive function Guerrant et al., 1999; Berkman et al., 2002; Niehaus et al., 2002; Simsek et al., 2004; Koruk et al., 2010; Forsythe and Kunze, 2013; Bergman and Graham, 2005 Autism Mulle et al., 2013 Guillain Barré Syndrome (paralysis) Willison and O'Hanlon, 2000; Willison, 2005

Kidneys/ Haemolytic uremic syndrome Delans et al., 1984 Urethra:

Entire body: Stunting Farthing et al., 1986; Guerrant et al., 1999; Berkman et al., 2002; Simsek et al., 2004; Ignatius et al., 2012 Chronic Fatigue Syndrome Mørch et al., 2009; Naess et al., 2012; Wensaas et al., 2012 ———————————————————————————————————————————————————————————

Figure 1: Hypothetical mechanisms, inferred from published research, by which acute enteric infections may lead to post-infectious, chronic complications. Enteropathogens, including Campylobacter jejuni or Giardia intestinalis cause microbiota biofilm dysbiosis, and disruptions of gut homeostasis and barrier function (Kalischuk and Buret, 2010; Cotton et al., 2011; Buret et al., 2013). These effects may be further compounded by a direct breakdown of the mucus barrier via mucin degradation by the enteropathogen (Macfarlane et al., 1999, 2005; Moncada et al., 2005; Derrien et al., 2010). In turn this causes proliferation of autoreactive cells like Th17 and Th1 lymphocytes, and release of pro- inflammatory cytokines (including IL-17, IFN-γ, TNF, etc.). Local dendritic cells, macrophages, and T Regulatory cells play a central role in regulating these processes. As the effects reach the circulation, this leads to immune complex deposition, tissue mast cell degranulation, NF-κB activation, and/or other yet unknown processes that cause organ pathology and disease (Hardin et al., 1997; Hity et al., 2010; Philpott et al., 2011; Scher and Abramson, 2011; Han et al., 2012; Forsythe and Kunze, 2013; Mulle et al., 2013).

recent findings suggesting that acute intestinalis are some of the most com- enteritis may disrupt the intestinal mon causes of bacterial and parasitic homeostatic balance between the intes- enteritis worldwide, the chapter will tine of the host and its own microbiota, emphasize on findings using these two and that these alterations in turn may disease models primarily (Savioli et al., be responsible for initiating the se- 2006; WHO: Campylobacter. Fact quence of events culminating in post- sheet N°255, October 2011 infectious symptomatology. Because http://www.who.int/mediacentre/factsh Campylobacter jejuni and Giardia eets/fs255/en/).

BIOLOGY OF THE INTESTINAL MICROBIOTA

Intestinal microbiota and host have physiology are shaped by the microbi- evolved to live in tolerant commensal- ota, which itself is modulated by host ism. Indeed, host immunity and gut immune, genetic, dietary and other

environmental factors. Recent evidence versity of the microbiota, which may also indicates that disruptions of these occur via mechanisms that remain un- microbiota influences homeostasis to clear, has been associated with disease cause disease at extra-intestinal sites, (Frank et al., 2007). Taken together, including asthma, obstructive lung dis- these new insights from DNA se- ease, arthritis, and even disorders of the quence-based analyses of gut microbial central nervous system such as autism communities suggest that the microbi- (Hity et al., 2010; Scher and Abramson, ome represents a key environmental 2011; Han et al., 2012; Forsythe and factor that can influence disease mani- Kunze, 2013; Mulle et al., 2013). Much festation (Figure 1). How in turn acute of the findings remain association- enteric infections may be at the source based, and cause-to-effect relationship of pathogenic microbiota disruptions studies are now sorely needed. Over has become a very intriguing part of 70% of the gut microbiota have not yet this puzzle. been cultured or classified, but new The population of industrialized culture-independent techniques have countries, with their characteristic high established that these microbial com- fat, high protein diets, harbour different munities are host- and GI tract loca- microbiota than those living in rural tion-specific, and that they play a key areas of developing countries, with a role in health (Zoetendal et al., 2008; polysaccharide-rich diet (De Filippo et de Weerth et al., 2013; Lepage et al., al., 2010)). The differences mainly re- 2013). Indeed, recent advances in se- flect an increased representation of quencing, metagenomics, and bioinfor- Bacteroidetes in the latter group, a matics technology have found that the group of bacteria known for its high estimated 1014 human gut microorgan- genetic ability to hydrolyse xyloses. isms, weighing a rough total of 3 The relative sensitivity of these distinct pounds, contain an overall genome size microbiota to enteropathogens, and 150 times larger than that of the human how in turn disruptions in their respec- genome (Zoetendal et al., 2008; de tive flora may differentially regulate Weerth et al., 2013; Lepage et al., post-infectious disorders, is unknown. 2013). Moreover, these studies found Microbial communities colonizing that gene exchanges between represent- the gut do so in a gradient, from few atives of the gut microbiota were in organisms in the oesophagus and stom- fact much more common than previ- ach, to the much more heavily colo- ously anticipated (Zoetendal et al., nized colon (Hokins et al., 2002). Ever 2008). This raises the question as to since the late 19th century, when Robert whether or not such exchanges may Koch’s studies in Germany developed also be common between enteropatho- the germ theory of disease, bacteria gens and the normal microbiota. In the were envisioned as single cells that early stages of life, the gut microbiota float or swim through some kind of wa- undergoes some degree of shifting, but tery habitat, within the human body, or overall, in the later parts of life, it re- in the environment in which they lived. mains stable in the absence of major With the giant progresses made in mi- disturbances of the host’s health condi- crobiology since then, and in the large tions and diet (Dethlefsen and Relman, part only since the 1960’s,we now un- 2011). To date, the phylogenetic core derstand that the swimming bacteria in of the human microbiota is thought to typical laboratory cultures act nothing be composed of 60-70 highly prevalent like the ones encountered in nature. In- species (Tap et al., 2009). Loss of di- deed, many of these organisms do not,

in the real world, spend much time charolytic biofilm bacteria, particularly swimming freely as isolated cells. In- at site where fermentable carbohydrates stead, they stick to wetted surfaces in may be in small supply, such as in the organized colonies that form a complex colon (Macfarlane et al., 1992). This multi-species slime-enclosed film, al- breakdown of mucin appears to be a lowing them to withstand environmen- cooperative undertaking by the micro- tal flushing forces, as well as resist biota, meant to benefit the entire bacte- against antimicrobials and competing rial biofilm community (Macfarlane microbes. These slime-producing com- and Macfarlane, 1992). Intriguingly, munities are known as bacterial “bio- intestinal biofilm bacteria living on films” (Hall-Stoodley et al., 2004). The mucin differ metabolically and phylo- slimy coating in which they live con- genetically from those living in a tains more than 90% water and com- planktonic state (Macfarlane et al., plex microbially secreted exo-polysac- 2005). Together, these data demon- charides that confer viscosity. The GI strate the physiological significance of tract, with its constant luminal flow and the biofilm mode of life of the intesti- high nutrient availability make it an nal microflora. A number of reports ideal site for bacteria to live as bio- have now established that patients with films. While we now understand that IBD and IBS have a reduced microflora gut microbiota do indeed live in a bio- diversity, and an apparent reduction in film phenotype, much remains to be the commensal phyla Firmicutes and learnt on how the integrity and physiol- Bacteroidetes, as well as increased ogy of these communities may bear on numbers of Proteobacteria (Reiff and gut homeostasis and disease (Von Kelly, 2010; Simren et al., 2013). Fur- Rosenvinge et al., 2013). Importantly thermore, recent evidence from fluores- however, as these biofilms live in very cent in situ hybridization labelling close proximity to host epithelial cells, (FISH) indicates that mucosal bacterial they are the microbial communities colonies live in closer proximity to the most likely to interact with host physi- epithelial lining in IBD than in normal ology, immunity, and metabolism. As individuals (Swidsinsky et al., 2009). such, more so than studies on faecal Whether this results from a more microbiota, research now needs to in- “adherent-invasive” phenotype in the vestigate the bacterial biofilms living bacteria found in patients with IBD re- on the mucosal surface of the intestine, quires further clarification. Indeed, in- as it is likely to shed key elements of volvement of the commensal micro- our understanding of host-gut microbi- flora in the initiation, persistence, and ota interactions. Indeed, biofilm micro- flare-ups of IBD has been suggested as colonies exist on the intestinal mucosal early as 1971 (Hill et al., 1971). How- surface in healthy people, and these ever, much remains to be learnt about bacterial populations are different from the role of microbiota biofilm disrup- those living in the intestinal lumen tions in the aetiology of post-infectious (Macfarlane and Macfarlane, 1992). complications following acute enteric The layer of mucus coating the epithe- infections. Moreover, research is lial surfaces of the gut prevents most needed to assess the effects of entero- microorganisms from reaching and/or pathogens on gut biofilm integrity, in persisting on the mucosal surface. an attempt to identify new mechanisms However, mucus glycoproteins, includ- that lead to the severe intestinal and ing mucins, represent an important extra-intestinal post-infectious compli- source of carbohydrates for the sac- cations of acute enteric infections.

CAMPYLOBACTER JEJUNI

Campylobacter jejuni is one of the tion of affected intestinal tissues re- most prevalent causes of human bacte- veals a broad spectrum of tissue altera- rial enteritis in the World (Moore, tions, and, most commonly, infiltration 2001; O'Ryan et al., 2005; Public of neutrophils into the lamina propria Health Agency of Canada, URL: secondary to NF-κB activation (Wasse- http://www.phac-aspc.gc.ca/id-mi/az- naar et al., 1999). The disease seems to index-eng.php). About 90% of human be less severe in developing countries campylobacteriosis is caused by C. je- (Coker et al., 2002). juni, and the remaining 10% is induced predominantly by C. coli. In the pae- Pathogenesis diatric population of developing coun- The host-pathogen interactions respon- tries however, diarrhoeagenic E. coli sible for inciting inflammation remain (EPEC, ETEC, STEC, EAEC, and incompletely understood (Janssen et EIEC) remain the most common bacte- al., 2008). Tissue damage appears to be ria detected during diarrhoea, and these largely due to the effects of cytotoxins, bacteria represent 30-40% of acute di- and/or host-cell invasion (Wassenaar et arrhoeal episodes in children (O'Ryan al., 1999; Russell et al., 1993). Recent et al., 2005). Though polymicrobial findings suggest that a C. jejuni lipo- infections involving Campylobacter protein (JlpA) promotes epithelial ad- appear to be less common in developed hesion, and the subsequent induction of countries, the bacterium is frequently pro-inflammatory signalling (Jin et al., isolated with other enteropathogens in 2003). Moreover, C. jejuni produces the developing World; co-infecting or- cytolethal distending toxin (Cdt), a ganisms include Escherichia coli, Sal- DNase-like toxin produced by several monella sp., Shigella sp., Giardia intes- species of bacteria. This holotoxin is tinalis, and rotavirus (Coker et al., composed of three subunits: CdtB be- 2002; Janssen et al., 2008). longs to the family of DNase I-like nu- Inadequate hygiene represents a cleases and is the active subunit, while significant contributor to campylobac- the CdtA and C subunits help deliver teriosis, as is the case for infections CdtB into target cells (Bang et al., with Shigella sp., Salmonella sp., and 2001). Upon epithelial invasion by C. other enteropathogens. C. jejuni lives jejuni, Cdt translocate into the nucleus, as a non-pathogenic enteric bacterium where it activates the G1 and/or G2/M in poultry and cattle. Upon ingestion of checkpoint responses, resulting in cell contaminated water or food (e.g. raw cycle arrest, and ultimately cell death milk), infected humans exhibit a range via mechanisms that are poorly under- of symptoms varying from mild to se- stood (Whitehouse et al., 1998; Bang et vere diarrhoea. Clinically therefore, al., 2001). The effects of Cdt have pri- acute campylobacteriosis is not readily marily been described for lymphocytes distinguishable from other enteric in- and monocytes, in which it appears to fections. These clinical features typi- induce apoptosis (Hickey et al., 2005). cally arise 2-4 days post-infection and However, Cdt has also been shown re- are indicative of the inflammatory cently to induce non-apoptotic death of response that is occurring (Public endothelial cells (Bielaszewska et al., Health Agency of Canada, URL: 2005). In addition, C. jejuni is thought http://www.phac-aspc.gc.ca/id-mi/az- to produce several other cytotoxins be- index-eng.php). Histological examina- sides Cdt, including shiga-like toxin,

haemolysin, and hepatotoxin (Wasse- 2013). Further research is warranted to naar et al., 1997). These cytotoxic fac- clarify the sequence of events responsi- tors require further characterization, ble for the post-infectious complica- and their possible implication in C. je- tions of campylobacteriosis, which will juni-induced epithelial injury is poorly help unravel common pathways understood. C. jejuni flagella represent through which acute enteropathogens another important virulence factor, as cause post-infectious inflammatory dis- they promote motility through the mu- orders. cus layer, and are important in the adherence to and invasion of epithelial Post infectious complications cells (Wassenaar et al., 1991; Konkel et Long after the C. jejuni infection has al., 2004; Christensen et al., 2009). In- been cleared, some individuals experi- terestingly, strain-dependent induction ence abnormal bowel patterns (more of epithelial cell oncosis by C. jejuni frequent, watery stools; or fewer correlates with invasion ability, but hard/lumpy stools) that may persist for may occur independently of Cdt (Kalis- years (Riddle et al., 2001; Marshall et chuk et al., 2007). Regardless, the acute al., 2010). These abnormalities may be stage of the infection is able to break associated with intestinal enteroendo- the intestinal barrier, which in turn al- crine cell hyperplasia, and low-grade lows luminal antigens to stimulate sub- inflammation, including an increase in epithelial immunity. In the acute stage CD3 lymphocytes and proliferation of of infection, C. jejuni also disrupts pro- intraepithelial lymphocytes (Spiller et tective TLR-9 signalling in epithelial al., 2000). cells (O'Hara et al., 2012). These alter- The post-infectious complications ations in turn prime the intestine for caused by acute campylobacteriosis heightened inflammatory injury upon include Guillain-Barré syndrome, reac- mild pro-inflammatory stimulation, via tive arthritis, Reiter syndrome (an in- mechanisms that remain incompletely flammatory disease with either con- understood (O'Hara et al., 2012). junctival or urethral inflammation), Irritable Bowel Syndrome, flare-ups in Microbiota disruptions by Campylo- patients with Inflammatory Bowel Dis- bacter sp. eases, and possibly celiac disease (Keat C. jejuni facilitates the translocation of and Rowe, 1991; Riddle et al., 2001; non-invasive, commensal bacteria, both Coker et al., 2002; Janssen et al., 2008; paracellularly as well as transcellularly. Marshall et al., 2010). On rare occa- The latter occurs by hijacking the host sions, C. jejuni infection may also lipid raft pathway as well as via epithe- cause haemolytic-uremic syndrome, a lial M cells (Lamb-Rosteski et al., well-known consequence of infection 2008; Kalischuk et al., 2009, 2010). with enterotoxigenic Escherichia coli On-going research has also uncovered (Delans et al., 1984). Another rare ex- that C. jejuni is able to directly disrupt tra-intestinal complication of campylo- the composition and integrity of human bacteriosis is endocarditis (Miki et al., microbiota biofilms, which may lead to 2005). Intriguingly, the symptoms of a loss of tolerance by the host to its post-infectious arthritis appear to be own commensal microbiota (Buret et similar regardless of the infecting bac- al., 2013). Interestingly, the effects of terial species, indicating a role for fac- C. jejuni on microbiota biofilms can be tors common to a range of pathogens duplicated with G. intestinalis, but not (Schiellerup et al., 2008). Guillain- with a commensal E. coli (Buret et al., Barré Syndrome may develop in ap-

proximately 1 in 1,000 infected indi- ciated with rheumatic manifestations, viduals infected with C. jejuni, and is a further linking gut disturbance to osteo- serious autoimmune neurological disor- articular disorders (Gradel et al., 2009; der; symptoms may range from weak- Rodriguez-Reyna et al., 2009). The ness of extremities to complete paraly- mechanisms remain unclear. Several sis and respiratory insufficiency (Willi- inflammatory factors implicated in IBD son, 2005). The majority of patients implicate the NF-κB pathway. In keep- may recover completely within 6 to 12 ing with a prominent role for microbes months (Willison, 2005). Guillain- in the pathogenesis of IBD, a variety of Barré Syndrome is thought to occur bacterial products, including bacterial because of molecular mimicry between lipopolysaccharide (LPS), are also po- the lipo-oligosaccharide of the C. jejuni tent activators of this pathway. Poly- cell envelope, and sugar moieties on morphic mutations of NOD2 (also nerve gangliosides (Willison and called CARD15), which acts as an O'Hanlon, 2000). In turn, antibodies intracellular sensor of bacteria-derived raised during infection with C. jejuni muramyl dipeptide (a component of may cross-react with nerve gangli- Gram-positive and Gram-negative bac- osides in some individuals, leading to terial cells walls) are the product of the the demyelinization of nerves, and sub- IBD1 gene mutation present in some sequent degeneration of axons (Willi- patients with IBD, and significantly son, 2005). increase disease susceptibility by alter- Irritable bowel syndrome (IBS), is ing the NF-κB pathway (Ahmad et al., the most commonly diagnosed func- 2002; Podolsky, 2002). The cytosolic tional gastrointestinal disorder by gas- NOD2 receptor may also activate the troenterologists, and is characterized by NF-κB pathway upon exposure to LPS abdominal hypersensitivity and abnor- which may have entered the cytoplasm mal bowel movement (diarrhoea and/or via mechanisms that have yet to be elu- constipation). It is a common long-term cidated (Ahmad et al., 2002; Podolsky, consequence of acute gastroenteritis 2002). The increased numbers of E. caused by a variety of enteropathogens, coli and Proteobacteria detected in the including C. jejuni, Samonella sp., intestinal mucosa of IBD patients may diarroeagenic E. coli, and Giardia heighten exposure to pathogenic prod- intestinalis (Riddle et al., 2001; Tha- ucts such as lipoproteins, proteogly- bane et al., 2007; Ohman et al., 2010). cans, and LPS. Other components of Altered motility patterns as well as ab- the resident microflora, within the dominal pain in post-infectious IBS stressed and inflamed environment of have been associated with mast cell the IBD intestine, may also activate secretions such as mast cell tryptase host inflammation via mechanisms that and serotonin (5-hydrocytryptamin, 5- are incompletely understood (Mac- HT), also released from enterochromaf- Pherson et al., 2004). In addition, up- fin cells (Cenac et al., 2007; Cremon et regulated expression and/or polymor- al., 2011). phic mutations of receptors for LPS, Finally, it was recently demon- e.g. TLR-4, have been found in epithe- strated that acute gastroenteritis with C. lial cells of IBD patients (Cario et al., jejuni, diarrhoeagenic E. coli, or Sal- 2000). LPS has the ability to break the monella sp. may lead to the initiation intestinal barrier (Qi et al., 2005; Yu et and or exacerbation of Inflammatory al., 2005; Chin et al., 2006). It has also Bowel Diseases (IBD; Crohn’s Disease been recently reported that peripheral and ulcerative colitis), themselves asso- blood monocytes from IBD patients

exhibit increased TLR2 expression, and ceptible host. Findings from on-going this is correlated with a marked in- research indicate that C. jejuni- or crease of TLR-2 -mediated TNF-α pro- Giardia-induced disruptions of the mi- duction (Cantó et al., 2006). The notion crobiota biofilm composition and integ- that circulating LPS and anti-endotoxin rity may help trigger a sequence of antibodies can be found in the plasma events that may lead to post-infectious of IBD patients is consistent with a complications (Buret et al., 2013). A breach in the epithelial barrier (Gardi- better understanding of the disruptions ner et al., 1995). Activation of TLR-4 to the resident microflora and deregu- appears to be implicated in the devel- lated bacterial recognition secondary to opment of pathology during infectious acute C. jejuni infection may shed new colitis (McKay, 1999; Cario et al., light on the mechanisms responsible 2000). Interestingly, a recent study also for the initiation and/or exacerbation of demonstrated that C. jejuni infection inflammation in IBD patients. disrupts TLR-9 signalling, which Taken together, findings have makes the intestinal mucosa more started to establish processes through prone to inflammatory injury (O'Hara which campylobacteriosis may lead to et al., 2012). Furthermore, in situ post-infectious sequelea. These include examination of biopsies from patients inflammatory disorders in the gut, but with IBD revealed the increased uptake may also affect extra-intestinal sites, of non-invasive, commensal E. coli via including the central nervous system, the follicle-associated epithelial M the lungs, the kidneys, the eyes, the cells, a phenomenon known to be facil- joints, and even the heart. This charac- itated by C. jejuni (Keita et al., 2008). teristic is shared with the post-infec- These invading commensal E. coli tious complications caused by a variety were shown to co-localize with den- of other enteropathogens, further sup- dritic cells, which correlated with in- porting the hypothesis of common creased levels of the pro-inflammatory pathogenic pathways. At least part of cytokine TNF-α. Disruptions of the in- these processes appear to be triggered testinal barrier by enteropathogens may by enteropathogen-induced disruptions permit luminal material, including of the host microbiota biofilms. More commensal bacteria and/or their prod- research needs to identify the mecha- ucts, to activate baso-lateral pro-in- nisms through which Campylobacter, flammatory sensors like TLR’s which and other enteropathogens, may trigger otherwise may have been inaccessible. events in the microbiota and the intesti- Therefore, luminal factors capable of nal mucosa that ultimately set the stage breaching epithelial integrity, and/or for chronic inflammatory disorders in altering the polar distribution of TLR’s, the gut, as well as at extra-intestinal may predispose the intestine to height- sites. ened intestinal inflammation in a sus-

GIARDIA INTESTINALIS

Giardiasis, caused by G. intestinalis tion's Neglected Disease Initiative (synonymous G. lamblia or G. duode- (Savioli et al., 2006; WHO: Guidelines nalis), is the most common waterborne for drinking water quality, 3rd Edition parasitic infection of the human intes- http://www.who.int/water_sanitation_h tine worldwide, and was recently in- ealth/dwq/gdwq3/en/). The prevalence cluded in the World Health Organisa- of human giardiasis is highest in devel-

oping countries, where it ranges from 2009; Cotton et al., 2011). Whether 20% to 100% of the population, versus these effects may be further com- its prevalence of 3-7% in developed pounded by degradation of local mu- countries (Jensen et al., 2009; Ankarlev cins by Giardia, as it was found for et al., 2010). People infected with other enteric microbes like E. histoloyt- Giardia may develop a broad range of ica, requires further investigation clinical manifestations, ranging from (Macfarlane and Macfarlane, 1999; asymptomatic infection, to acute or Macfarlane et al., 2005; Moncada et chronic diarrhoeal disease associated al., 2005; Derrien et al., 2010). As is with abdominal pain and nausea (De the case with other enteropathogens, Filippo et al., 2010; Cotton et al., induction of apoptosis in enterocytes 2011). Most infections are self-limit- by Giardia represents a key component ing, although re-infection and chronic in the pathogenesis of the infection infection can occur. Recent evidence (Chin et al., 2002; Panaro et al., 2007; indicates that G. intestinalis, like C. Troeger et al., 2007; Buret et al., 2013). jejuni and other enteropathogens, is The mechanisms responsible are un- responsible for chronic post-infectious known, and the identification of a Gi- complications, via mechanisms that ardia “enterotoxin” has remained elu- remain obscure. sive. Giardia-mediated increases in intestinal permeability result from altera- Pathogenesis tions to the apical junctional com- Pathophysiology in giardiasis occurs plexes, including disruptions of F-actin, without invasion of the small intestinal zonula-occludens (ZO)-1, claudin-1, tissues by the trophozoites, and in the and α-actinin, a component of the acto- absence of any overt inflammatory cell myosin ring that regulates paracellular infiltration, with the exception of a flow, under the control of epithelial modest increase in intraepithelial lym- myosin light chain kinase (Teoh et al., phocytes; some of the acute pathology, 2000; Scott et al., 2002; Cotton et al., which involves a diffuse shortening of 2011; Maia-Brigagão et al., 2012). The epithelial microvilli, is caused by acti- mechanisms leading to loss of intesti- vated CD8+ lymphocytes (Buret et al., nal barrier function caused by Giardia 1992; Scott et al., 2004). As is the case sp. are shared among a broad range of for enteric infections caused by Cam- enteropathogens (O'Hara and Buret, pylobacter sp., diarrhoeagenic E. coli, 2008). Salmonella sp., and others, the pathophysiology of acute diarrhoea in giar- Microbiota disruptions by Giardia diasis implicates a disruption of the in- Bacterial components of the microbiota testinal barrier function. In giardiasis, from patients with symptomatic giar- heightened rates of enterocytes apopto- diasis appear to heighten G. intestinalis sis, intestinal barrier dysfunction, acti- virulence in gnotobiotic mice, via un- vation of host lymphocytes, shortening clear mechanisms (Torres et al., 2000). of brush border microvilli with or with- Little is known of the effects of acute out coinciding villous atrophy, disac- giardiasis on the human commensal charidase deficiencies, small intestinal microbiota. But findings from on-going malabsorption, anion hypersecretion studies indicate that indeed, G. intesti- and increased intestinal transit rates all nalis is able to disrupt the composition seem to contribute to disease (Buret et and integrity of human intestinal al., 1992; Chin et al., 2002; Scott et al., microbiota biofilms, in a fashion simi- 2004; Troeger et al., 2007; Koot et al., lar to what C. jejuni does (Buret et al.,

2013). More research in this area will 2002; Bergman and Graham, 2005; help identify common pathways Ettehad et al., 2010). The persistence through which acute enteritis may lead of infection and its association with to long term inflammatory disorders by diarrhoea are key factors associated disrupting commensal bacterial bio- with growth disturbance and failure to films, as well as how these may lead to thrive, and diarrhoea caused by enteric extra-intestinal complications. infections in early childhood has become a predictor of stunting (Berkman Post-infectious complications of et al., 2002; Botero-Garcés et al., 2009; giardiasis Ettehad et al., 2010). In giardiasis and Infection with Giardia sp. may lead to cryptosporidiosis, as well as other en- food allergies, negatively affect nutri- teric infections, diarrhoea may lead to tional and growth status, and impair poor cognitive function by causing zinc cognitive function in humans (Farthing and iron micronutrient deficiencies, as et al., 1986; Berkman et al., 2002; well as defects in the anti-oxidant sys- Niehaus et al., 2002; Ettehad et al., tem, which may all affect neuroplas- 2010; Ignatius et al., 2012). Recent evi- ticity (Ajjampur et al., 2011). Moreo- dence also indicates that 5-10% of pa- ver, diarrhoea during early childhood tients diagnosed with giardiasis will was also found to impair visual-motor develop post-infectious irritable syn- coordination, auditory short-term drome and functional dyspepsia, long memory, information processing, and after clearance of the parasite (Dizdar cortical cognitive function (Guerrant et et al., 2007; Stark et al., 2007; Hanevik al., 1999; Ajjampur et al., 2011). Com- et al., 2009; Robertson et al., 2010). bined with these complications, some When the infection persists for months, individuals may develop post-giardiasis microscopic duodenal inflammation fatigue and musculoskeletal pain may develop (Hanevik et al., 2009; (Naess et al., 2012). Viral, bacterial, as Mørch et al., 2009), further underscor- well as parasitic pathogens have the ing the need for rapid parasitic elimina- ability to cause chronic fatigue syn- tion to reduce the risk of chronic com- drome. Recent studies have reported a plications in giardiasis. Infection with high prevalence of post-infectious fa- Giardia can cause iron deficiency tigue following a giardiasis outbreak in anaemia, micronutrient deficiencies, Bergen, Norway, in 2004 (Dizdar et al., protein-energy malnutrition, which all 2007; Hanevik et al., 2009; Mørch et have been linked to growth and cogni- al., 2009; Robertson et al., 2010; Naess tive retardation (Simsek et al., 2004; et al., 2012; Wensaas et al., 2012). Koruk et al., 2010). Studies conducted Beyond its long-term consequences in Brazil and Peru found that diarrhoeal on intestinal and overall metabolic pa- disease occurring in the first 2 years of rameters, giardiasis, like campylobac- life negatively correlates with cognitive teriosis, and other enteric infections, function, verbal fluency, and physical also has the ability to cause post-infec- fitness, and may lead to long-term tious complications at extra-intestinal growth faltering (Guerrant et al., 1999; sites (Cantey et al., 2011). Sites af- Berkman et al., 2002). Long-term se- fected include the eyes (Pettoelo-Man- quelae of wasting and/or stunting often tovani et al., 1990; Corsi et al., 1998), include general behavioural and devel- the joints (Borman et al., 2001; Carlson opmental consequences that present as and Finger, 2004), the skin (Hardin et failure to thrive, which has also been al., 1997; Di Prisco et al., 1998; Giaco- linked to giardiasis (Berkman et al., metti et al., 2003; Pietrzak et al., 2005),

and on rare occasions, the muscles tious consequences again are similar to (Cervelló et al., 1993; Addiss and Leng- what has been reported for other enteric erich, 1994; Genovese et al., 1996). infections (Table 2). These patterns of chronic post-infec-

CONCLUSIONS

Recent findings clearly demonstrate Beyond the well-established need that the health consequences of enteric to better understand host-pathogen infections go far beyond their acute cross-talks, as well as interactions be- diarrhoeal symptoms, as they can lead tween the host and its microbiota, these to severe chronic post-infectious intes- observations lay the foundations for tinal inflammatory disorders, failure to future research into enteropathogen- thrive, and serious growth and cogni- microbiota interactions. This research tive impairment. Moreover, the chronic will help better understand gut homeo- sequelae may also cause post-infectious stasis, and will help unravel new patho- disease at extra-intestinal sites, includ- physiological pathways. ing the joints, the skin, the eyes, the The direct benefits of microbiota lungs, the heart, the muscles, the kid- are not well understood mechanisti- neys, and the central nervous system. cally. Most information is derived from The mechanisms responsible for these the use of probiotics, and more recently long-term effects remain obscure. the use of faecal microbiota transplant However, findings from recent and on- (FMT). With the recent decision of the going research suggest that entero- Federal Drug Administration (U.S.A.) pathogen-induced disruptions of the to call FMT a “drug” and hence to only commensal microbiota may at least in allow its use under an Investigational part play a role in triggering the se- New Drug application, (with a recently quence of events that result in these implemented exception for the treat- presentations. These disruptions in- ment of Clostridium difficile infection), clude a promotion of the transcellular mechanistic insights into the now well- and paracellular translocation of non- established beneficial effects of FMT invasive commensal bacteria, as well as may take a long time to become clear. disruptions of the composition and This further underscores the need to structure of microbiota biofilms. These uncover model systems that will allow in turn contribute, at least in part, to the to develop well-characterized, safe activation of the host autoimmune reac- “synthetic microbiota” of “laboratory tions that are implicated in the produc- prepared” FMT-like cocktails. tion of post-infectious complications.

ACKNOWLEDGMENTS

This work discusses studies that were performed with the financial support of the Natural Sciences and Engineering Council of Canada, The Canadian Institutes for Health Research, the Crohn’s and Colitis Foundation of Canada, the Alberta Live- stock and Meat Industry, and the Alberta Health and Innovation IBD Consortium.

LITERATURE

Addiss, D.G. and Lengerich, E.J.: Hypokalemic Cravens, J., Reller, L.B., and Wang, W.L.: myopathy induced by Giardia lamblia. N. Campylobacter enteritis: clinical and Engl. J. Med. 330, 66-67 (1994). epidemiologic features. Ann. Intern. Med. Ahmad, T., Armuzzi, A., Bunce, M., Mulcahy- 91, 179-185 (1979). Hawes, K., Marshall, S.E., Orchard, T.R., Borman, P., Seçkin, Ü., and Özoran, K.: Bea- Crawshaw, J., Large, O., de Silva, A., ver fever -- a rare cause of reactive arthri- Cook, J.T., Barnardo, M., Cullen, S., tis. J. Rheumatol. 28, 683-688 (2001). Welsh, K.I., and Jewell, D.P.: The molecu- Botero-Garcés, J.H., García-Montoya, G.M., lar classification of the clinical manifesta- and Grisales-Patiño, D.: Giardia intesti- tions of Crohn’s disease. Gastroenterology nalis and nutritional status in children par- 122, 854-866 (2002). ticipating in the complementary nutrition Ajjampur, S.S.R., Koshy, B., Venkataramani, program, Antioquia, Columbia, May to M., Sarkar, R., Joseph, A.A., Jacob, K.S., October 2006. Rev. Inst. Med. Trop. São Ward, H., and Kang, G.: Effect of cryptos- Paulo 51, 155-162 (2009). poridial and giardial diarrhea on social ma- Buret, A.G., Hardin, J.A., Olson, M.E., and turity, intelligence and physical growth in Gall, D.G.: Pathophysiology of small children in a semi-urban slum in south In- intestinal malabsorption in gerbils infected dia. Ann. Trop. Paediatr. 31, 205-212 with Giardia lamblia. Gastroenterology (2011). 103, 506-513 (1992). Ankarlev, J., Jerlstrom-Hultqvist, J., Ringqvist, Buret, A.G., Akierman, S.V., Feener, T., E., Troell, K., and Svard, S.G.: Behind the McKnight, W., Rioux, K.P., Beck, P.L., smile: cell biology and disease mecha- Wallace, J.L., and Beatty, J.: Campylobac- nisms of Giardia species. Nat. Rev. Micro- ter jejuni- or Giardia duodenalis-mediated biol. 8, 413-422 (2010). disruptions of human intestinal microbiota Bang, D.D.,, Scheutz F., Ahrens, P., Pedersen, biofilms: novel mechanisms producing K., Blom, J., and Madsen, M.: Prevalence post-infectious intestinal inflammatory dis- of cytolethal distending toxin (cdt) genes orders? Gastroenterology 144, Suppl. 1, S- and CDT production in Campylobacter 309 (2013). spp. isolated from Danish broilers. J. Med. Buret, A.G. and Bhargava, A.: Modulatory Microbiol. 50, 1087-1094 (2001). mechanisms of enterocyte apoptosis by Bergman, P. and Graham, J.: An approach to viral, bacterial and parasitic pathogens, "failure to thrive". Aust. Fam. Physician Critical Rev. Microbiol. Epub. 34, 725-729 (2005). DOI:10.3109/1040841X.2012.746952 Berkman, D.S., Lescano, A.G., Gilman, R.H., (2013). Lopez, S.L., and Black, M.M.: Effects of Cantey, P.T., Roy, S., Lee, B., Cronquist, A., stunting, diarrhoeal disease, and parasitic Smith, K., Liang, J., and Beach, M.J.: infection during infancy on cognition in Study of nonoutbreak giardiasis: novel late childhood: a follow-up study. Lancet findings and implication for research. Am. 359, 564-571 (2002). J. Med. 124, 1175.e1-1175.e8 (2011). Bielaszewska, M., Sinha, B., Kuczius, T., and Cantó, E., Ricart, E., Monfort, D., González- Karch, H.: Cytolethal distending toxin Juan, D., Balanzó, J., Rodríguez-Sánchez, from Shiga toxin-producing Escherichia J.L., and Vidal S.: TNF alpha production coli O157 causes irreversible G2/M arrest, to TLR2 ligands in active IBD patients. inhibition of proliferation, and death of hu- Clin. Immunol. 119, 156-165 (2006). man endothelial cells. Infect. Immun. 73, Cario, E. and Podolsky, D.K.: Differential al- 552-562 (2005). teration in intestinal epithelial cell expres- Blaser, M.J., Berkowitz, I.D., LaForce, F.M., sion of TLR3 and TLR4 in inflammatory

100

bowel disease. Infect. Immun. 68, 7010- neuron activation, and abdominal pain in 7017 (2000). irritable bowel syndrome. Am. J. Carlson, D.W. and Finger, D.R.: Beaver fever Gastroenterol. 106, 1290-1298 (2011). arthritis. J. Clin. Rheumatol. 10, 86-88 De Filippo, C., Cavalieri, D., Di Paola, M., (2004). Ramazzotti, M., Poullet, J.B., Massart, S., Cenac, N., Andrews, C.N., and Vergnolle, N.: Collini, S., Pieraccini, G., and Lionetti, P.: Role of protease activity in visceral pain in Impact of diet in shaping gut microbiota irritable bowel syndrome. J. Clin. Invest. revealed by a comparative study in chil- 117 ,636-647 (2007). dren from Europe and rural Africa. Proc. Cervelló, A., Alfaro, A., and Chumillas, M.J.: Natl. Acad. Sci. USA 107,14691-14696 Hypokalemic myopathy induced by Giar- (2010). dia lamblia. N. Engl. J. Med. 329, 210-211 de Weerth, C., Funetes, S., Puylaert, P., and de (1993). Vos, W.M.: Intestinal microbiota of infants Chin, A.C., Teoh, D.A., Scott, K.G., Meddings, with colic: development and specific J.B., Macnaughton, W.K., and Buret, A.G.: signatures. Paediatrics 131, e550-e558 Strain-dependent induction of enterocytes (2013). apoptosis by Giardia lamblia disrupts epi- Delans, R.J., Biuso, J.D., Saba, S.R., and thelial barrier function in a caspase-3-de- Ramirez, G.: Hemolytic uremic syndrome pendent manner. Infect. Immun. 70, 3673- after Campylobacter-indcued diarrhea in 3680 (2002). an adult. Arch. Intern. Med. 144, 1074- Chin, A.C., Flynn, A.N., Fedwick, J.P., and 1076 (1984). Buret, A.G.: The role of caspase-3 in lipo- Derrien, M., van Passel, M.W.J., and Dekker, polysaccharide-mediated disruption of epi- J.: Mucin-bacterial interactions in the hu- thelial tight junctions. Can. J. Physiol. man oral cavity and digestive tract. Gut Pharmacol. 84, 1043-1050 (2006). Microbes 1, 254-268 (2010). Christensen, J.E., Pacheco, S.A., and Konkel, Dethlefsen, L. and Relman, D.A.: Incomplete M.E.: Identification of a Campylobacter recovery and individualized responses of jejuni-secreted protein required for the human distal gut microbiota to re- maximal invasion of host cells. Mol. peated antibiotic perturbation. Proc. Natl. Microbiol. 73, 650-662 (2009). Acad. Sci. USA 108, Suppl. 1, 4554-4561 Coker, A., Isokphei, R.D., and Obi, C.L.: Hu- (2011). man campylobacteriosis in developing Di Prisco, M.C., Hagel, I., Lynch, N.R., Jimé- countries. Emerg. Infect. Dis. 8, 237-243 nez, J.C., Rojas, R., Gil, M., and Mata, E.: (2002). Association between giardiasis and allergy. Corsi, A., Nucci, C., Knafelz, D., Bulgarini, D., Ann. Allergy Asthma Immunol. 81, 261- Di Iorio, L., Polito, A., De Risi, F., Ardenti 265 (1998). Morini, F., and Paone, F.M.: Ocular Dizdar, V., Gilja, O.H., and Hausken, T.: In- changes associated with Giardia lamblia creased visceral sensitivity in Giardia-in- infection in children. Br. J. Ophthalmol. duced postinfectious irritable bowel syn- 82, 59-62 (1998). drome and functional dyspepsia. Effect of Cotton, J.A., Beatty, J.K., and Buret, A.G.: the 5HT3-antagonist ondansetron. Neuro- Host parasite interactions and patho- gastroenterol. Motil. 19, 977-982 (2007). physiology of Giardia infections. Int. J. Ettehad, G.H., Daryani, A., and Nemati, A.: Parasitol. 41, 925-933 (2011). Effect of Giardia infection on nutritional Cremon, C., Carini, G., Wang, B., Vasina, V., status in primary schoolchildren, in north- Cogliandro, R.F., De Giorgio, R., west Iran. Pak. J. Biol. Sci. 13, 229-234 Stanghellini, V., Grundy, D., Tonini, M., (2010). De Ponti, F., Corinaldesi, R., and Barbara, Farthing, M.J.G., Mata, L., Urutia, J.J., and G.: Intestinal serotonin release, sensory Kronmal, R.A.: Natural history of Giardia

101

infections of infants and children in rural Han, M.K., Hunag, Y.J., LiPuma, J.J., Guatemala and its impact on physical Boushey, H.A., Boucher, R.C., Cookson, growth. Am. J. Clin. Nutr. 43 395-405 W.O., Curtis, J.L., Erb-Downward, J., (1986). Lynch, S.V., Sethi, S., Toews, G.B., Forsythe, P. and Kunze, W.A.: Voices from Young, V.B., Wolfgang, M.C., Huffnagle, within: gut microbes and the CNS. Cell. G.B., and Martinez, F.J.: Significance of Mol. Life Sci. 70, 55-69 (2013). the microbiome in obstructive lung dis- Frank, D.N., St. Amand, A.L., and Feldman, ease. Thorax 67, 456-463 (2012). R.A.: Molecular phylogenetic characteriza- Hanevik, K., Dizdar, V., Langeland, N., and tion of microbial community imbalances in Hausken, T.: Development of functional human inflammatory bowel diseases. Proc. gastrointestinal disorders after Giardia Natl. Acad. Sci. USA 104, 13780-13785 lamblia infection. BMC Gastroenterol. 9, (2007). 27-31 (2009). Gardiner, K.R., Halliday, M.I., Barclay, G.R., Hardin, J.A., Buret, A.G., Olson, M.E., Kimm, Milne, L., Brown, D., Stephens, S., Max- M., and Gall, D.G.: Mast cell hyperplasia well, R.J., and Rowlands, B.J.: Signifi- and increased macromolecular uptake in an cance of systemic endotoxaemia in inflam- animal model of giardiasis. J. Parasitol. 83, matory bowel disease. Gut 36, 897-901 908-912 (1997). (1995). Hickey, T.E., Majam, G., and Guerry, P.: Intra- Genovese, A., Spadaro, G., Santoro, L., Gas- cellular survival of Campylobacter jejuni paro Rippa, P., Onorati, A.M., and in human monocytic cells and induction of Marone, G.: Giardiasis as a cause of hypo- apoptotic death by cytholethal distending kalemic myopathy in congenital immuno- toxin. Infect. Immun. 73, 5194-5197 deficiency. Int. J. Clin. Lab. Res. 26, 132- (2005). 135 (1996). Hill, M.J., Drasar, B.S., Hawksworth, G., Ar- Giacometti, A., Cirioni, O., Antonicelli, L., ies, V., Crowther, J.S., and Williams, R.E.: D’Amato, G., Silvestri, C., Del Prete, Bacteria and the aetiology of cancer of M.S., and Scalise, G.: Prevalence of intes- large bowel. Lancet 1, 95-100 (1971). tinal parasites amongt individuals with al- Hity, M., Burke, C., Pedro, H., Cardenas, P., lergic skin diseases. J. Parasitol. 89, 490- Bush, A., Bossley, C., Davies, J., Ervine, 493 (2003). A., Poulter, L., Pachter, L., Moffatt, M.F., Gradel, K.O., Nielsen, H.L., Schønheyder, and Cookson, W.O.: Disordered microbial H.C., Ejlertsen, T., Kristensen, B., and communities in asthmatic airways. PLoS Nielsen, H.: Increased short and long term One 5, e8578, (2010). risk of inflammatory bowel disease after Hokins, M.J., Sharp, R., and Macfarlane, G.T.: Salmonella or Campylobacter gastroenteri- Variation in human intestinal microbiota tis. Gastroenterology 137, 495-501 (2009). with age. Dig. Liver Dis. 34, Suppl. 2, Guerrant, D.I., Moore, S.R., Lima, A.A.M., S12-S18 (2002). Patrick, P.D., Schorling, J.B., and Guer- Ignatius, R., Gahutu, J.B., Klotz, C., Steininger, rant, R.L.: Association of early childhood C., Shyirambere, C., Lyng, M., Muse- diarrhea and cryptosporidiosis with im- makweri, A., Aebischer, T., Martus, P., paired fitness and cognitive function four- Harms, G., and Mockenhaupt, F.P.: High seven years later in a poor urban commu- prevalence of Giardia duodenalis assem- nity in northeast Brazil. Am. J. Trop. Med. blage B infection and association with un- Hyg. 61, 707-713 (1999). derweight in Rwandan children. PLoS Hall-Stoodley, L., Costerton, J.W., and Negl. Trop. Dis. 6, e1677 (2012). Stoodley, P.: Bacterial biofilms: from the Janssen, R., Krogfelt, K.A., Cawthraw, S.A., natural environment to infectious diseases. van Pelt, W., Wagenaar, J.A., and Owen, Nat. Rev. Microbiol. 2, 95-108 (2004). R.J.: Host-pathogen interactions in Cam-

102

pylobacter infections: the host perspective. proteins from Campylobacter jejuni is Clin. Microbiol. Rev. 21, 505-518 (2008). dependent on a functional flagellar export Jensen, L.A., Marlin, J.W., Dyck, D.D., and apparatus. J. Bacteriol. 186, 3296-3303 Laubach, H.E.: Prevalence of multi-gastro- (2004). intestinal infections with helminth, proto- Koot, B.G.P., ten Kate, F.J.W., Juffrie, M., zoan and Campylobacter spp. in Guatema- Rosalina, I., Taminiau, J.J.A.M., and Ben- lan children. J. Infect. Dev. Ctries. 3, 229- ninga, M.A.: Does Giardia lamblia cause 234 (2009). villous atrophy in children?: a restrospec- Jin, S., Song, Y.C., Emili, A., Sherman, P.M., tive cohort study of the histological abnor- and Chan, V.L.: JlpA of Campylobacter je- malities in giardiasis. J. Pediatr. Gastroen- juni interacts with surface-exposed heat terol. Nutr. 49, 304-308 (2009). shock protein 90alpha and triggers signal- Koruk, I., Simsek, Z., Tekin Koruk, S., Doni, ling pathways leading to the activation of N., and Gürses, G.: Intestinal parasites, nu- NF-kappaB and p38 MAP kinase in epi- tritional status and psychomotor develop- thelial cells. Cell. Microbiol. 5, 165-174 ment delay in migratory farm worker's (2003). children. Child Care Health Dev. 36, 888- Kalischuk, L.D., Inglis, G.D., and Buret, A.G.: 894 (2010). Strain-dependent induction of epithelial Lamb-Rosteski, J.M., Kalischuk, L.D., Inglis, cell oncosis by Campylobacter jejuni cor- G.D., and Buret, A.G.: Epidermal growth relates with invasion ability and is inde- factor inhibits Campylobacter jejuni- pendent of cytolethal distending toxin. Mi- induced claudin-4 disruption, loss of crobiology 153, 2952-2963 (2007). epithelial barrier function, and Escherichia Kalischuk, L.D., Inglis, G.D., and Buret, A.G.: coli translocation. Infect. Immun. 76, Campylobacter jejuni induces transcellular 3390-3398 (2008). translocation of commensal bacteria via Lepage, P., LeClerc, M.C., Jossens, M., Mon- lipid rafts. Gut Pathog. 1, 2 (2009). dot, S., Blottiere, H.M., Raes, J., Erlich, Kalischuk, L.D. and Buret, A.G.: A role for D., and Dore, J.: A metagenomic insight Campylobacter jejuni-induced enteritis in into our gut’s microbiome. Gut 62, 146- inflammatory bowel disease? Am. J. Phys- 158 (2013). iol. Gastrointest. Liver Physiol. 298, G1- Macfarlane, G.T., Gibson, G.R., and Cumming, G9 (2010). J.H.: Comparison of fermentation reactions Kalischuk, L.D., Leggett, F., and Inglis, G.D.: in different regions of the human colon. J. Campylobacter jejuni induces transcytosis Appl. Bacteriol. 72, 57-64 (1992). of commensal bacteria across the intestinal Macfarlane, S. and Macfarlane, G.: Bacterial epithelium through M-like cells. Gut colonization of surfaces in the large intes- Pathog. 2, 14 (2010). tine. In: Colonic microflora, nutrition and Keat, A. and Rowe, I.: Reiter's syndrome and health. (Gibson, G. and Roberfroid, M., associated arthritides. Rheum. Dis. Clin. Eds.). Chapman and Hall, London, 71-87 N. Am. 17, 25-42 (1991). (1999). Keita, Å., Salim, S.Y., Jiang, T., Yang, P.C., Macfarlane, S., Woodmansey, E.J., and Franzén, L., Söderkvist, P., Magnusson, Macfarlane, G.T.: Colonization of mucin K.E., and Söderholm, J.D.: Increased by human intestinal bacteria and establish- uptake of non-pathogenic E. coli via the ment of biofilm communities in a two- follicle-associated epithelium in stage continuous culture system. Appl. longstanding ileal Crohn's disease. J. Environ. Microbiol. 71, 7483-7492 (2005). Pathol. 215, 135-144 (2008). MacPherson, A.J. and Harris, N.L.: Interactions Konkel, M.E., Klena, J.D., Rivera-Amill, V., between commensal intestinal bacteria and Monteville, M.R., Biswas, D., Raphael, B., the immune system. Nat. Rev. Immunol. 4, and Mickelson, J.: Secretion of virulence 478-485 (2004).

103

Maia-Brigagão, C., Morgado-Díaz, J.A., and tism research. Curr. Psychiatry Rep. 15, De Souza, W.: Giardia disrupts the ar- 337-347 (2013). rangement of tight, adherens and desmoso- Naess, H., Nyland, M., Hausken, T., Follestad, mal junction protein of intestinal cells. I., and Nyland, H.I.: Chronic fatigue syn- Parasitol. Int. 61, 280-287 (2012). drome after Giardia enteritis: clinical char- Maizels, R.M. and Yazdanbakhsh, M.: Immune acteristics, disability and long-term sick- regulation by helminth parasites: cellular ness absence. BMC Gastroenterol. 12, 13- and molecular mechanisms. Nat. Rev. Im- 19 (2012). munol. 3, 733-744 (2003). Niehaus, M.D., Moore, S.R., Patrick, P.D., Marshall, J.K., Thabane, M., Garg, A.X., Derr, L.L., Lorntz, B., Lima, A.A., and Clark, W.F., Moayyedi, P., and Collins, Guerrant, R.L.: Early childhood diarrhea is S.M.: Eight year prognosis of associated with diminished cognitive func- postinfectious irritable bowel syndrome tion 4 to 7 years later un children in a following waterborne bacterial dysentery. northeast Brazilian shantytown. Am. J. Gut 59, 605-611 (2010). Trop. Med. Hyg. 66, 590-593 (2002). McKay, D.M.: Intestinal inflammation and the O'Hara, J.R. and Buret, A.G.: Mechanisms of gut microflora. Can. J. Gastroenterol. 13, intestinal tight junctional disruption during 509-516 (1999). infection. Front. Biosci. 13, 7008-7021 Miki, K., Maekura, R., Hiraga, T., Hirotani, A., (2008). Hashimoto, H., Kitada, S., Miki, M., O’Hara, J.R., Feener, T.D, Fischer, C.D., and Yoshimura, K., Naka, N., Motone, M., Fu- Buret , A.G.: Campylobacter jejuni dis- jikawa, T., Takashima, S., Kitazume, R., rupts protective TLR9 signaling in colonic Kanzaki, H., Nakatani, S., Watanuki, H., epithelial cells and increases the severity Tagusari, O., Kobayashi, J., and Ito M.: In- of DSS-induced colitis in mice. Infect. Im- fective tricuspid valve endocarditis with mun. 80, 1563-1571 (2012). pulmonary emboli casued by Campylobac- O’Ryan, M., Prado, V., and Pickering, L.K.: A ter fetus after tooth extraction. Intern. Med. millennium update on pediatric diarrheal 44, 1055-1059 (2005). illness in the developing World. Sem. Moncada, D., Keller, K., and Chadee, K.: Enta- Pediatr. Infect. Dis. 16, 125-136 (2005). moeba histolytica-secreted products de- Ohman, L. and Simren, M.: Pathogenesis of grade colonic mucin oligosaccharides. In- IBS: role of inflammation, immunity and fect. Immun. 73, 3790-3793 (2005). neuroimmune interactions. Nat. Rev. Moore, S.R.: Early childhood diarrhea and hel- Gastroenterol. Hepatol. 7, 163-173 (2010). minthiases associated with lond-term lin- Panaro, M.A., Cianciulli, A., Mitolo, V., Ac- ear growth faltering. Int. J. Epidemiol. 30, quafreda, A., Brandonisio, O., adn 1457-1464 (2001). Cavallo, P.: Caspase-dependent apoptosis Moore, S.R., Lima, A.M., and Guerrant, R.L.: of the HCT-8 epithelial cell line induced Infection: Preventing 5 million child by the parasite Giardia intestinalis. FEMS deaths from diarreha in the next 5 years. Immunol. Med. Microbiol. 51, 302-309 Nature Rev. Gastroenterol. Hepatol. 8, (2007). 363-364 (2011). Pettoelo-Mantovani, M., Giardino, I., Magli, Mørch, K., Hanevik, K., Rortveit, G., Wensaas, A., di Martino, L., and Guandalini, S.: K.A., Eide, G.E., Hausken, T., and Lange- Intestinal giardiasis associated with oph- land, N.: Severity of Giardia infection talmic changes. J. Pediatr. Gastroenterol. associated with post-infectious fatigue and Nutr. 11, 196-200 (1990). abdominal symptoms two years after. Philpott, H., Gibson, P., and Thien, F.: Irritable BMC Infect. Dis. 9, 206-214 (2009). Bowel Syndrome - an inflammatory dis- Mulle, J.G., Sharp, W.G., and Cubells, J.F.: ease involving mast cells. Asia Pac. Al- The gut microbiome: a new frontier in au- lergy 1, 36-42 (2011).

104

Pietrzak, A., Chodorowska, G., Urban, J., Bo- Rheumatol. 35, 480-487 (2008). gucka, V., and Dybiec, E.: Cutaneous man- Scott, K.G., Meddings, J.B., Kirk, D.R., Lees- ifestation of giardiasis - case report. Ann. Miller, S.P., and Buret, A.G.: Intestinal Agric. Environ. Med. 12, 299-303 (2005). infection with Giardia spp. reduces epithe- Podolsky, D.K.: Inflammatory bowel disease. lial barrier function in a myosin light N. Engl. J. Med. 347, 417-429 (2002). chain-kinase dependent fashion. Gastro- Qi, W., Ebbert, K.V., Craig, A.W., Greer, P.A., enterology 123, 1179-1190 (2002). and McCafferty, D.M.: Absence of Fer Scott, K.G., Yu, L.C., and Buret, A.G.: Role of protein tyrosine kinase exacerbates endo- CD8+ and CD4+ T lymphocytes in jejunal toxin induced intestinal epithelial barrier mucosal injury during murine giardiasis. dysfunction in vivo. Gut 54, 1091-1097 Infect. Immun. 72, 3536-3542 (2004). (2005). Simren, M., Barbara, G., Flint, H.J., Spiegel, Reiff, C. and Kelly, D.: Inflammatory bowel B.M.R., Spiller, R.C., Vanner, S., Verdu, disease, gut bacteria and probiotic therapy. E.F., Worwell, P.J., and Zoetendal, E.G.: Int. J. Med. Microbiol. 300, 25-33 (2010). Intestinal microbiota in functional bowel Riddle, M.S., Gutierrez, R.L., and Verdu, E.F.: disorders; a Rome foundation report. Gut The Chronic gastrointestinal consequences 62, 159-176 (2013). associated with Campylobacter. Curr. Gas- Simsek, Z., Zeyreck, F.Y., and Kurcer, M.A.: troenterol. Rep. 14, 395-405 (2001). Effect of Giardia infection on growth and Robertson, L.J., Hanevik, K., Escobedo, A.A., psychomotor development of children Mørch, K., and Langeland, N.: Giardiasis - aged 0-5 years. J. Trop. Pediatr. 50, 90-93 - why do the symptoms sometimes never (2004). stop? Trends Parasitol. 26, 75-82 (2010). Spiller, R.C., Jenkins, D., Thornley, J.P., Rodriguez, L.A.G. and Ruigomez, A.: In- Hebden, J.M., Wright, T., Skinner, M., and creased risk of irritable bowel syndrome Neal, K.R.: Increased rectal mucosal after bacterial gastroenteritis: cohort study. enteroendocrine cells, T lymphocytes, and BMJ 318, 565-566 (1999). increased gut permeability following acute Rodriguez-Reyna, T.S., Martinez-Reyes, C., Campylobacter enteritis and in post- and Yamamoto-Furusho, J.K.: Rheumatic dysenteric irritable bowel syndrome. Gut manifestations of inflammatory bowel dis- 47, 804-811 (2000). ease. World J. Gastroenterol. 15, 5517- Stark, D., van Hal, S., Marriott, D., Ellis, J., 5524 (2009). and Harkness J.: Irritable bowel syndrome: Russell, R.G., O'Donnoghue, M., Blake, D.C. a review on the role of intestinal protozoa Jr., Zulty, J., and DeTolla, L.J.: Early co- and the importance of their detection and lonic damage and invasion of Campylo- diagnosis. Int. J. Parasitol. 37, 11-20 bacter jejuni in experimentally challenged (2007). infant Macaca mulatta. J. Infect. Dis. 168, Swidsinsky, A., Loering-Bauke, V., and Her- 210-215 (1993). ber, A.: Mucosal flora in Crohn’s disease Savioli, L., Smith, H., and Thompson, A.: Giar- and ulcerative colitis - an overview. J. dia and Cryptosporidium join the 'ne- Physiol. Pharmacol. 60, Suppl. 6, 61-71 glected diseases initiative'. Trends Parasi- (2009). tol. 22, 203-208 (2006). Tap, J., Mondot, S., and Levenez, F.: Towards Scher, J.U. and Abramson, S.B.: The microbi- the human intestinal phylogenetic core. ome and rheumatoid arthritis. Nat. Rev. Environ. Microbiol. 11, 2574-2584 (2009). Rheumatol. 7, 569-578 (2011). Teoh, D.A., Kamieniecki, D., Pang, G., and Schiellerup, P., Krogfelt, K.A., and Locht, H.: Buret, A.G.: Giardia lamblia rearranges F- A comparison of self-reported joint symp- actin and alpha-actinin in human colonic toms following infection with different en- and duodenal monolayers and reduces teric pathogens: effect of HLA-B27. J. transepithelial electrical resistance. J.

105

Parasitol. 86, 800-806 (2000). invasion. The EMBO J. 10, 2055-2061 Thabane, M., Kottachchi, D.T., and Marshall, (1991). J.K.: Systematic review and meta-analysis: Wassenaar, T.M.: Toxin production by Cam- the incidence and prognosis of post- pylobacter spp. Clin. Microbiol. Rev. 10, infectious irritable bowel syndrome. 466-476 (1997). Aliment. Pharmacol. Ther. 26, 535-544 Wassenaar, T.M. and Blaser, M.J.: Pathophysi- (2007). ology of Campylobacter jejuni infections Torres, M.F., Uetanabaro, A.P.T., Costa, A.F., of humans. Microbes Infect. 1, 1023-1033 Alves, C.A., Farias, L.M., Bambirra, E.A., (1999). Penna, F.J., Vieira, E.C., and Nicoli, J.R.: Wensaas, K.A., Langeland, N., Hanevik, K., Influence of bacteria from the duodenal Mørch, K., Eide, G.E., and Rortveit G. microbiota of patients with symptomatic Irritable bowel syndrome and chronic fa- giardiasis on the pathogenicity of Giardia tigue 3 years after acute giardiasis: historic duodenalis in gnotoxenic mice. J. Med. cohort study. Gut, 61, 214-219 (2012). Microbiol. 49, 209-215 (2000). Whitehouse, C.A., Balbo, P.B., Pesci, E.C., Troeger, H., Eppel, H.J., Schneider, T., Wahn- Cottle, D.L., Mirabito, P.M., and Pickett, schaffe, U., Ulrich, R., Burchard, G.D., C.L.: Campylobacter jejuni cytolethal dis- Jelinek, T., Zeitz, M., Fromm, M., and tending toxin causes a G2-phase cell cycle Schulzke, J.D.: Effect of chronic Giardia block. Infect. Immun. 66, 1934-1940 lamblia infection on epithelial transport (1998). and barrier function in human duodenum. Willison, H.J. and O'Hanlon, G.M.: Anti-gly- Gut 56, 328-335 (2007). cosphingolipid antibodies and Guillain- Veenemans, J., Schouten, L.R.A., Ottenhof, Barré syndrome. In: Campylobacter, 2nd M.J., Mank, T.G., Uges, D.R.A., Mbugi, ed. (Nachamkin, I. and Blaser, M.J., Eds.). E.V., Demir, A.Y., Kraaijenhagen, R.J., ASM Press, Washington, DC, USA, 241- Savelkoul, H.F.J., and Verhoef, H.: Effect 258 (2000). of preventive supplementation with zinc Willison, H.J.: The immunobiology of Guil- and other micronutrients on non-malarial lain-Barré syndromes. J. Peripher. Nerv. morbidity in Tanzanian pre-school chil- Syst. 10, 94-112 (2005). dren: a randomized trial. Plos One 7, Yu, L.C., Flynn, A.N., Turner, J.R., and Buret, e41630 (2012). A.G.: SGLT-1-mediated glucose uptake Von Rosenvinge, E.C., O’May, G.A., Macfar- protects intestinal epithelial cells against lane, S., Macfarlane, G.T., and Shirtliff, LPS-induced apoptosis and barrier defects: M.E.: Microbial biofilms and gastrointesti- a novel cellular rescue mechanism? nal diseases. Pathog. Dis. 67, 25-38 FASEB J. 19, 1822-1835 (2005). (2013). Zoetendal, E.G., Rajilic-Stojanovic, M., and de Wassenaar, T.M., Bleumink-Pluym, N.M., and Vos, W.M.: High-throughput diversity and van der Zeijst, B.A.: Inactivation of functionality analysis of the gastrointesti- Campylobacter jejuni flagellin genes by nal tract microbiota. Gut 57, 1605-1615 homologous recombination demonstrates (2008). that flaA but not flaB is required for

106

THE ROLES OF PATHOGENIC AND COMMENSAL BACTERIA IN THE INTESTINAL AND BEHAVIOURAL MANIFESTATIONS OF FUNCTIONAL GI DISORDERS

STEPHEN M. COLLINS

Farncombe Family Digestive Health Research Institute, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada

SUMMARY

This paper reviews clinical and animal-based studies linking pathogenic microbes to the onset of functional GI disorders, and commensal microbes to the maintenance of these chronic disorders. The review will focus on post-infectious irritable bowel syndrome (PI-IBS). The link between acute gastroenteritis and PI-IBS is now well established on the basis of clinical studies. Risk factors include the severity of acute infection, pre-existing psychiatric morbidity and age <60. Host genetic determinants point to susceptibility loci in bacterial recognition, cytokine secretion and intestinal barrier function. Low-grade colonic inflammation is evident PI-IBS as well as a subset of IBS patients without prior gastroenteritis. Risk factors for IBS include infection, stress and antibiotic usage and are known to disrupt the intestinal microbiota. Recent work has shown that destabilization of the microbiota alters gut physiology, brain chemistry and behaviour in animals. Thus a vicious cycle is established in which the initial perturbation of the microbiota results in altered colonic function and the resulting change in the colonic environment selects for different bacteria. This microbial centred model of IBS accommodates key features of IBS including chronicity, variability in symptom expression, and psychiatric co-morbidity. Growing evidence of the therapeutic benefit of microbiome-directed therapies, including pre- and probiotics and as well as selected antibiotics, for this common condition, also support this model.

FUNCTIONAL GI DISORDERS

Functional gastrointestinal disorders and Di Lorenzo, 2012). Up to 80% of are chronic abdominal symptom com- patients exhibit behavioural changes plexes for which there is no discernible that include anxiety, depression or underlying structural abnormality. The somatization. There are no biomarkers most common functional disorder is the and despite the availability of diagnos- Irritable Bowel Syndrome (IBS) that is tic criteria, the diagnosis is often made characterized by abdominal pain or dis- by exclusion of other pathologies. In comfort and altered bowel habit and is the absence of an underlying cause, generally considered to reflect dysfunc- there is no cure and treatment is invari- tion of the gut-brain axis (Camilleri ably symptom-based and of limited

107

efficacy. Together, these factors ac- Each of these factors has been shown count for the very high socio-economic to induce changes in the microbial burden of IBS (Maxion-Bergemann et composition of the gut. Antibiotics in- al., 2006). duce a transient disruption of the intes- It is now well established that acute tinal microbiota, but in some cases the enteric infection precedes the onset of effect may be long lasting (Jernberg et chronic symptoms in a subset of IBS al., 2010). Stress has also been shown patients. It is estimated that between 5 to disrupt the microbial community of and 32% of patients with acute gastro- the gut in animal models (Bailley et al., enteritis develop post-infectious IBS 2010, 2011). Acute gastroenteritis pro- (PI-IBS). Bacterial infection with Cam- duces changes in the intestinal microbi- pylobacter, Shigella, Salmonella or ota in humans (Nelson et al., 2010). Escherichia coli, parasitic or viral in- Disruption of the intestinal microbi- fections have been associated with the ota results in changes in gut function development of PI-IBS. Risk factors reminiscent of that found in IBS. For include the existence of psychological example, antibiotic-induced perturba- disorders at the time of infection, the tion of the intestinal microbiota in mice severity of infection and host genetic resulted in a small increase in inflam- factors (for review, see: Thabane and matory activity in the gut mucosa and Marshall, 2009). The latter includes an increased response to visceral dis- polymorphisms in genes that encode tension, interpreted to reflect hyper- bacterial recognition, cytokine secre- algesia (Verdu et al., 2006). Similarly, tion and the integrity of the intestinal disruption of the intestinal microbiota epithelium (Villani et al., 2006). There by antibiotics results in changes in gut is evidence of low-grade inflammation transit mediated by Toll-like receptor-4 (without tissue destruction) in the co- TLR-4 (Anitha et al., 2012). Thus, dis- lonic mucosa of PI-IBS patients and ruption of the intestinal microbiota this may arise from inefficient down- changes the physiology and the physic- regulation of the acute inflammatory chemical properties of the colon; these response to the acute gastroenteritis changes in turn generate different se- (Gwee et al., 2003). Studies in an ani- lection pressures on the microbial com- mal model of acute parasitic infection munity, resulting in further destabiliza- provide proof of concept that transient tion of the microbiota. As illustrated in infection can induce persistent gut dys- Figure 1, the resulting vicious cycle function that can be reversed by anti- produces inter-related changes in the inflammatory therapy (Barbara et al., microbial composition of the gut and in 2001). A recent study showed that PI- gut physiology, resulting in chronic IBS may last for at least 8 years post- dysfunction and symptom generation infection but the factors that maintain (Collins and Bercik, 2009). This model gut dysfunction in the long term are is supported by clinical observations poorly understood (Marshall et al., showing (a) variation in symptom ex- 2010). Attention now focuses on pression over time in IBS patients changes in the intestinal microbiota as (Mearin et al., 2004) and (b) temporal a driver of immune activation, low- instability of the intestinal microbiome grade inflammation and symptom gen- in IBS patients (Maukonen et al., 2005; eration in IBS. Matto et al., 2006). Interestingly, a re- Factors known to trigger the onset cent study showed that duodenal instil- of IBS include infection, stress and an- lation of faecal bacteria from healthy tibiotic usage (Villareal et al., 2012). subjects delayed the development of

108

Figure 1: Schematic representation of a microbial centred model of IBS. Pathogenic microbes and other triggers cause an initial destabilization of the microbiota that creates a vicious and self sustaining cycle of host-microbial imbalance, resulting in chronic gut dysfunction and altered behaviour – IBS.

IBS symptoms in patients recovering testinal microbiota resulted in changes from Giardia infection (Morken et al., in brain chemistry and behaviour 2009), supporting the notion that (Bercik et al., 2011). Taken together infection-induced destabilization of the these observations support the integra- microbiota plays a critical role in the tion of the intestinal microbiome into expression of PI-IBS. the gut-brain axis, thereby providing a Up to 80% of IBS patients exhibit basis for considering the role of com- psychiatric co-morbidities that include mensal bacteria in both the gastroin- anxiety and depression (North et al., testinal and behavioural manifestations 2007). Recent studies have shown that of IBS (Collins et al., 2012). The ap- the intestinal microbiota influences plicability of these animal-based stud- brain chemistry and behaviour. Germ- ies to man is supported by the recent free mice show less anxiety than colo- demonstration that probiotic bacteria nized mice (Neufeld et al., 2011) and alter brain activity in healthy human antibiotic-induced disruption of the in- subjects (Tillisch et al., 2013).

ACKNOWLEDGEMENTS

Dr. S.M. Collins is the recipient of the GSK Chair in Gastroenterological Research and is funded by grants from the Canadian Institutes of Health Research.

109

LITERATURE

Anitha, M., Vijay-Kumar, M., Sitaraman, S.V., kova, A., Deng, Y., Graham, J.C., Gewirtz, A.T., and Srinivasan, S.: Gut mi- McKendrick, M.W., and Moochhala, S.M.: crobial products regulate murine gastroin- Increased rectal mucosal expression of intestinal motility via Toll-like receptor 4 terleukin 1beta in recently acquired post- signaling. Gastroenterology 143, 1006- infectious irritable bowel syndrome. Gut 1016 (2012). 52, 523-526 (2003). Bailey, M.T., Dowd, S.E., Parry, N.M., Galley, Jernberg, C., Löfmark, S., Edlund, C., and J.D., Schauer, D.B., and Lyte, M.: Stressor Jansson, J.K.: Long-term impacts of antibi- exposure disrupts commensal microbial otic exposure on the human intestinal mi- populations in the intestines and leads to crobiota. Microbiology 156, 3216-3223 increased colonization by Citrobacter ro- (2010). dentium. Infect. Immun. 78, 1509-1519 Marshall, J.K., Thabane, M., Garg, A.X., (2010). Clark, W.F., Moayyedi, P., Collins, S.M.; Bailey, M.T., Dowd, S.E., Galley, J.D., Walkerton Health Study Investigators: Hufnagle, A.R., Allen, R.G., and Lyte, M.: Eight year prognosis of postinfectious irri- Exposure to a social stressor alters the table bowel syndrome following water- structure of the intestinal microbiota: im- borne bacterial dysentery. Gut 59, 605-611 plications for stressor-induced immuno- (2010). modulation. Brain Behav. Immun. 25, 397- Mättö, J., Maunuksela, L., Kajander, K., Palva, 407 (2011). A., Korpela, R., Kassinen, A., and Saarela, Barbara, G., De Giorgio, R., Deng, Y., Val- M.: Composition and temporal stability of lance, B., Blennerhassett, P., and Collins, gastrointestinal microbiota in irritable S.M.: Role of immunologic factors and cy- bowel syndrome--a longitudinal study in clooxygenase 2 in persistent postinfective IBS and control subjects. FEMS Immunol. enteric muscle dysfunction in mice. Gas- Med. Microbiol. 43, 213-222 (2005). troenterology 120, 1729-1736 (2001). Maukonen, J., Satokari, R., Mättö, J., Söder- Bercik, P., Denou, E., Collins, J., Jackson, W., lund, H., Mattila-Sandholm, T., and Lu, J., Jury, J., Deng, Y., Blennerhassett, Saarela, M.: Prevalence and temporal sta- P., Macri, J., McCoy, K.D., Verdu, E.F., bility of selected clostridial groups in irri- and Collins, S.M.: The intestinal microbi- table bowel syndrome in relation to pre- ota affect central levels of brain-derived dominant faecal bacteria. J. Med. Micro- neurotropic factor and behavior in mice. biol. 55, 625-633 (2006). Gastroenterology 141, 599-609 (2011). Maxion-Bergemann, S., Thielecke, F., Abel, F., Camilleri, M. and Di Lorenzo, C.: Brain-gut and Bergemann, R.: Costs of irritable axis: from basic understanding to treatment bowel syndrome in the UK and US. Phar- of IBS and related disorders. J. Pediatr. macoeconomics 24, 21-37, (2006). Gastroenterol. Nutr. 54, 446-453 (2012). Mearin, F., Baró, E., Roset, M., Badía, X., Zá- Collins, S.M. and Bercik, P.: The relationship rate, N., and Pérez, I.: Clinical patterns between intestinal microbiota and the cen- over time in irritable bowel syndrome: tral nervous system in normal gastrointesti- symptom instability and severity variabil- nal function and disease. Gastroenterology ity. Am. J. Gastroenterol. 99, 113-121 136, 2003-2014 (2009). (2004). Collins, S.M., Surette, M., and Bercik, P.: The Morken, M.H., Valeur, J., Norin, E., Midtvedt, interplay between the intestinal microbiota T., Nysaeter, G., and Berstad, A.: Antibi- and the brain. Nat. Rev. Microbiol. 10, otic or bacterial therapy in post-giardiasis 735-742 (2012). irritable bowel syndrome. Scand. J. Gas- Gwee, K.A., Collins, S.M., Read, N.W., Rajna- troenterol. 44, 1296-1303 (2009).

110

Nelson, A.M., Walk, S.T., Taube, S., Taniuchi, mented milk product with probiotic modu- M., Houpt, E.R., Wobus, C.E., and Young, lates brain activity. Gastroenterology 144, V.B.: Disruption of the human gut micro- 1394-1401(2013). biota following Norovirus infection. PLoS Verdú, E.F., Bercik, P., Verma-Gandhu, M., One. 7, e4822 (2012). Huang, X.X., Blennerhassett, P., Jackson, Neufeld, K.M., Kang, N., Bienenstock, J., and W., Mao, Y., Wang, L., Rochat, F., and Foster, J.A.: Reduced anxiety-like behav- Collins, S.M.: Specific probiotic therapy ior and central neurochemical change in attenuates antibiotic induced visceral hy- germ-free mice. Neurogastroenterol. Motil. persensitivity in mice. Gut 55, 182-190 23, 255-264 (2011). (2006). North, C.S., Hong, B.A., and Alpers, D.H.: Villani, A.C., Lemire, M., Thabane, M., Relationship of functional gastrointestinal Belisle, A., Geneau, G., Garg, A.X., Clark, disorders and psychiatric disorders: impli- W.F., Moayyedi, P., Collins, S.M., cations for treatment. World J. Gastroen- Franchimont, D., and Marshall, J.K.: Ge- terol. 13, 2020-2027 (2007). netic risk factors for post-infectious irrita- Thabane, M. and Marshall, J.K.: Post-infec- ble bowel syndrome following a water- tious irritable bowel syndrome. World J. borne outbreak of gastroenteritis. Gastro- Gastroenterol. 15, 3591-3596 (2009). enterology 138, 1502-1513 (2010). Tillisch, K., Labus, J., Kilpatrick, L., Jiang, Z., Villarreal, A.A., Aberger, F.J., Benrud, R., and Stains, J., Ebrat, B., Guyonnet, D., Gundrum, J.D.: Use of broad-spectrum Legrain-Raspaud, S., Trotin, B., Naliboff, antibiotics and the development of irritable B., and Mayer, E.A.: Consumption of fer- bowel syndrome. WMJ. 111, 17-20 (2012).

111

112

WHY BACTERIA MATTER: INSIGHTS FROM THE HYDRA HOLOBIONT

THOMAS C.G. BOSCH and SEBASTIAN FRAUNE

Zoological Institute, Christian-Albrechts University, Kiel, Germany

SUMMARY

Animals, ranging from basal metazoans to primates, are engaged in symbiotic relationships with complex microbial ecosystems. These resident microbes influence fitness and thus ecologically important traits of their hosts, ultimately forming a meta-organism consisting of a multicellular host and a community of associated microorganisms. The evolutionary dynamics within such a meta-organism and the involved molecular interactions are rather complex and often difficult to investigate experimentally. Untangling the complex interactions requires simple animal models with only a few specific symbiotic partners. Here we show that organisms on at the base of the evolutionary scale such as the freshwater polyp Hydra may be key to dissecting the fundamental principles that underlie all host-microbe interactions.

FROM KARL AUGUST MÖBIUS TO THE HOLOGENOME THEORY OF EVOLUTION

In 1877, Karl Möbius, Professor of later it became obvious that not only Zoology at Kiel University, coined the ecological systems but also complex term “biocoenosis” for a community of “environmental” diseases can only be living beings belonging to different understood if the relationships between species and associated by way of inter- the interacting infectious agents present species interdependence. In one of the at a given time in a given territory are first studies, later to become a classic, recognized. By analogy with “bio- to be conducted in the emerging sci- coenosis”, the understanding of a dis- ence of ecology, Möbius was seeking ease as a complex dynamic phenome- to determine why some oyster beds in non was conceptualized with the word the Atlantic were becoming exhausted, “pathocoenosis” (Grmek, 1969). while the oyster beds in the British Today we realize that all epithelia river estuaries and the Schleswig-Hol- in animals are colonized by microbial stein oyster beds were very rich communities and that, therefore, any (Möbius, 1877). He related this phe- multicellular organism must be consid- nomenon to the other species present, ered a meta-organism comprised of the rather than to the oysters in the beds macroscopic host and synergistic inter- themselves. Möbius thus was the first dependence with bacteria, archaea, to recognise that an ecological system fungi, and numerous other microbial must be taken as a whole and coined and eukaryotic species. The «meta- the term "biocoenosis" for a living organism» concept (Bosch and McFall- community. About a hundred years Ngai, 2011) considers the dynamic

113

Figure 1: Cnidaria are a sister group of all Bilateria. A: Phylogeny of basal metazoan animals. B: Hydra oligactis (taken from Fraune and Bosch, 2007). C: Hydractinia milleri (printed with permission from Gary McDonald). D: Blue coral (taken from http://www.jcu.edu.au/cgc/CoralGenomicsHP.html). E: Hydra viridis with Chlorella symbionts. F: Phase-contrast micrograph of Hydra viridis. G: Fluorescence microscopy of the same area shown in F. Chlorella algae appears red, Hydra tissue green. H: Phase contrast micrograph of a single macerated endodermal epithelial cell containing symbiotic algae in the basal part below the nucleus (stained blue) (F-H taken from Habetha et al 2003).

communities of bacteria on epithelial are not solitary, homogenous entities surfaces as an integral part of the func- but consist of complex communities of tionality of the respective organism it- many species that likely evolved during self. Today there is also an increasing a billion years of coexistence led to the appreciation that microbes are an es- hologenome theory of evolution sential part of the animal phenotype (Rosenberg et al., 2007, 2009; Zilber- influencing fitness and thus ecologi- Rosenberg and Rosenberg, 2008) cally-important traits of their hosts which considers the holobiont with its (O'Hara and Shanahan, 2006; McFall- hologenome as the unit of selection in Ngai, 2007; Fraune and Bosch, 2010). evolution. Thus, modern symbiosis re- Disease onset is seen as a complex set search has become an emerging cross- of interactions among a variety of asso- disciplinary field focused on under- ciated partners that affect the fitness of standing the general principles by the collective holobiont (Rosenstiel et which these complex host-microbe al., 2009). Discovering that individuals communities function and evolve.

114

What is the complexity in species num- sition? Finally, what are the conse- ber and structural organization of these quences of the associations on molecu- associations? What is the physiological lar pathways and the reactive genomes? role of temporal differences of associ- Here we show that for addressing these ated microbiota during life cycles? questions and untangling the complex Which selective forces drive the evolu- interactions that influence the host’s tion of these interactions, i.e. how do health and development, members of the associated organisms influence the ancient animal phylum Cnidaria each other's fitness? Which forces may serve as simple but highly in- shape the colonizing microbial compo- formative models.

THE HYDRA HOLOBIONT

Hydra is member of the animal phylum two-party interactions of polyp and Cnidaria which are not only among the symbiotic algae, but also in more com- earliest known phyletic lineages known plex systems comprising three or more to contain stem cells (Figure 1A) but associates including bacteria and vi- also possess most of the gene families ruses (Bosch, 2012a, 2012b). Thus, be- found in bilaterians and have retained side photosynthetic algae (Figure 1G- many ancestral genes that have been 1H), bacteria are another important lost in Drosophila and C. elegans component of the cnidarian holobiont. (Kortschak et al., 2003; Miller et al., In Hydra, the 36 identified bacterial 2005; Technau et al., 2005; Putnam et phylotypes represent three different al., 2007; Hemmrich et al., 2012). bacterial divisions and are dominated Similar to other animals, Cnidaria are by Proteobacteria and Bacteroidetes complex holobionts consisting of the (Fraune and Bosch, 2007, 2010). Dis- animal and its associated endogenous turbance or shifts in any of these part- microbiota. Inter-species interactions in ners can compromise the health of the several Cnidaria species (Figure 1A- whole animal (Fraune et al., 2009). 1D) between symbiotic algae and host Loss of symbiotic algae from coral tis- cells have been the subject of research sues, for example, can lead to coral since decades since they not only pro- bleaching and death. Since healthy in- vide insights into the basic "tool kit" dividuals of the same coral species necessary to establish symbiotic inter- from different location are colonized actions, but are also of relevance in un- by similar bacterial communities derstanding the resulting evolutionary (Rohwer et al., 2002) but diseased or selection processes (e.g. Muscatine and bleached corals contain changed bacte- Lenhoff, 1963; Pool, 1979; Thorington rial communities that differ greatly and Margulis, 1981; O'Brien, 1982; for from healthy ones (Ritchie, 2006; review see: Bosch, 2012a). In the Rosenberg et al., 2007), it seems that meantime it is becoming evident that in similar to complex “environmental” Cnidaria such as green Hydra viridis diseases in human, understanding dis- (Figure 1E-1H) or many coral species, eases within corals requires an in-depth a long term persistence of mutualistic knowledge of the basic biology of each associations is prevalent not only in holobiont member.

115

Figure 2: Hydra polyps are colonized by species-specific microbiota. A: Bacterial communities identified from four different Hydra species. B: Comparison of the phylogenetic tree from Hydra and the environmental cluster tree of the corresponding microbiota.

THE HOST ACTIVELY SHAPES THE COLONIZING MICROBIOTA

For decades a number of Hydra species cific microbial communities over long have been cultivated under standard periods of time. Bacteria in Hydra are conditions at constant temperature and specific for any given species (Figure identical food. It came as a complete 2A) (Fraune and Bosch, 2007; Fraune surprise, therefore, that examining the et al., 2010). Closely related Hydra microbiota in different Hydra species species as Hydra vulgaris and Hydra kept in the laboratory for more than 20 magnipapillata are associated with a years under controlled conditions re- very similar microbial community. In vealed an epithelium colonized by a contrast, Hydra oligactis, the most ba- complex community of microbes, and sal Hydra species analysed so far that individuals from different species (Hemmrich et al., 2007), is associated differed greatly in their microbiota. with the most distinct microbial com- Even more astonishing was the finding munity compared to the other Hydra that individuals living in the wild were species. In line with this, comparing the colonized by a group of microbes that phylogenetic tree of the Hydra species is similar to that in polyps grown in the with the according cluster tree of asso- lab, pointing to the maintenance of spe- ciated bacterial communities reveals a

116

very similar microbial community. In ciated bacterial communities reveals a contrast, Hydra oligactis, the most ba- high degree of congruency (Figure 2B). sal Hydra species analysed so far This strongly indicates that distinct (Hemmrich et al., 2007), is associated selective pressures are imposed on and with the most distinct microbial com- within the Hydra epithelium. The munity compared to the other Hydra forces that shape the colonizing micro- species. In line with this, comparing the bial composition are the focus of much phylogenetic tree of the Hydra species current investigation (Bevins and Salz- with the according cluster tree of asso- man, 2011).

HOW DOES THE HOST CONTROL THE MICROBIOTA IN THE CONTEXT OF SPECIFIC DEVELOPMENTAL OR ENVIRONMENTAL CONDITIONS?

In the same way that microbial com- family represents a shift from maternal munities are expected to change in dif- to zygotic protection of the embryo ferent parts of a body, they are also dy- (Fraune et al., 2011). In adult Hydra namic in time. For a first understanding polyps, additional AMPs including hy- of the temporal dynamics in Hydra-mi- dramacin (Bosch et al., 2009) and ar- crobe interactions we investigated the minin (Augustin et al., 2009) contribute establishment of the microbiota during to the host-derived control of bacterial oogenesis and embryogenesis. Early colonization. embryonic stages in Hydra are colo- After hatching from the “cutical” nized by a limited number of microbes stage the Hydra polyps get colonized (Fraune et al., 2010). During embryo- by its specific bacterial community. genesis the number of bacterial colo- The processes controlling community nizers changes in number and composi- membership and influencing the estab- tion. For example, Curvibacter-related lishment of the microbial ecosystem Betaproteobacteria are present only in during development are poorly under- late developmental stages while they stood. Therefore, the microbial com- appear to be absent in the early em- munities in polyps at various time bryo. Thus, early developmental stages points after hatching was profiled have a microbiota that is clearly dis- (Franzenburg et al., 2013). Distinct tinct from later developmental stages. features included high diversity of Interestingly, the differential coloniza- community profiles in the first week, tion is reflected in differences in anti- followed by progressive emergence of microbial activity. Hydra embryos are a stable adult-like pattern characterized protected by a maternally produced by low species diversity and the pre- antimicrobial peptide (AMP) of the ponderance of the Betaproteobacterium periculin peptide family, which con- Curvibacter. trols the establishment of the microbi- In adult Hydra polyps, additional ota during embryogenesis. Beginning AMPs including hydramacin (Bosch et with the gastrula stage, Hydra embryos al., 2009) and arminin (Augustin et al., express a set of periculin peptides 2009) and the bacterial signalling via (periculin 2a and 2b), which replaces MyD88 (Franzenburg et al., 2012) the maternal produced periculin pep- contribute to the host-derived control tides 1a and 1b. This shift in the ex- of bacterial colonization. pression within the periculin peptide

117

ANTIMICROBIAL PEPTIDES - KEY FACTORS FOR HOST-BACTERIA CO-EVOLUTION

Antimicrobial peptides (AMPs) are pressing periculin we apparently have known as prominent effector molecules created a new holobiont that is different which get often secreted after external from all investigated Hydra species. stimuli. Do they have, in addition to From these results we assume that spe- their killing activity against pathogens, cific associations between hosts and key regulatory functions in host-mi- bacteria are a result of bacterial adapta- crobe homeostasis as the driving force tion to different repertoires on AMPs in that leads to changes in microbiota different host species. Evolutionary composition? To investigate whether changes in the AMP repertoire of host the ectotopic expression of an AMP species, therefore, are expected to lead may affect the number and composition to changes in the composition of the of the colonizing microbiota at the ec- associated bacterial community. Future todermal epithelial surface, we gener- efforts will be directed towards analys- ated transgenic Hydra expressing peri- ing the performance of this new pheno- culin1a in ectoderm epithelial cells type under different environmental (Fraune et al., 2010). Comparing the conditions. Interestingly, patients with bacterial load of these transgenic Crohn’s disease often have strongly polyps with that of wild-type control reduced α-defensin expression and polyps revealed not only a significantly drastically altered endogeneous micro- lower bacterial load in transgenic biota (Wehkamp et al., 2005). Moreo- polyps overexpressing periculin1a but ver, mice expressing human alpha- also, unexpectedly, drastic changes in defensin-5 (DEFA5) and mice lacking the bacterial community structure. an enzyme required for the processing Analysing the identity of the colonizing of mouse alpha-defensins show signifi- bacteria showed that the dominant β- cant changes in intestinal microbiota Proteobacteria decreased in number, composition (Salzman et al., 2010). whereas α-Proteobacteria were more These findings support the view that prevalent. Thus, overexpression of epithelial-derived AMP may represent periculin causes not only a decrease in an important regulatory mechanism the number of associated bacteria but shaping the composition of epithelial also a changed bacterial composition. microbiota. With the transgenic polyps overex-

WHAT ARE THE MICROBES FOR?

The intimacy of the interaction be- polyps that are devoid of any bacteria. tween host and microbiota, as well as While morphologically no differences the high evolutionary pressure to main- could be observed to control polyps, tain a specific microbiota, points to the we are currently finding evidence that significance of the interkingdom asso- Hydra lacking bacteria suffer from fun- ciation and implies that hosts deprived gal infections unknown in normally of their microbiota should be at a dis- cultured polyps (Franzenburg, Fraune advantage. To investigate the effect of and Bosch, unpublished). Thus, do absence of microbiota in Hydra we beneficial microbes associated with have produced gnotobiotic Hydra Hydra produce anti-fungal compounds?

118

Future efforts are directed towards iso- continue its development, it needs an lating the active substances from these external trigger that appears to be pro- bacteria that eventually may lead to the vided in the natural habitat by certain development of novel antimycotics. sedentary bacteria of the genus Alter- Microbes also provide signals for omonas. A lipophilic substance pro- multiple developmental steps. One of duced by these bacteria is thought to the most pervasive examples of micro- act as this trigger (Leitz and Wagner, bial impact in animal development is in 1993). The mechanisms by which Hy- the induction of settlement and meta- dractinia sense bacteria-derived envi- morphosis of many marine invertebrate ronmental cues to form colonies and to larvae (Hadfield, 2011). This transition reproduce may provide crucial insights is an absolute requirement for comple- into the genetic and developmental tion of the animal’s life cycle, and is foundations of life cycles, but little is dependent upon induction by exoge- known about their natural history or nous morphogenetic cues, many of biochemistry. Observations in a num- which are produced by bacteria associ- ber of other invertebrates and verte- ated with a particular environmental brates strongly support the view that surface. Hydractinia, for example, microbes should be considered partners (Figure 1C) a marine colonial Cnidaria in animal development. Bacterial con- frequently found in the North Sea, tributions are indispensable, for exam- commonly covers shells inhabited by ple, in shaping the immune system and hermit crabs. Fertile colonies, male and development of organs such as the female, produce eggs and sperm, vertebrate intestine or the squid light respectively, and within less than three organ (reviewed in Fraune and Bosch, days the fertilized egg develops into a 2010). Animal development has tradi- mature planula larva. "Mature" larva tionally been viewed as an autonomous means a larva that is able to metamor- process directed by the genome. It phose into a polyp, but under sterile seems that we have to rethink develop- laboratory conditions it will never do. ment at least in part, as an orchestration It will rather die, as it is unable to take of both animal-encoded ontogeny and up food (Leitz and Wagner, 1993; Wal- inter-kingdom communication. ther et al., 1996; Frank et al., 2001). To

THE HOLOBIONT IN A CONSTANTLY CHANGING ENVIRONMENT

The association between host and mi- and food source. These changes have crobes is strongly affected by the envi- significant effects on the composition ronment. To determine the impact of of the bacterial community. For exam- different environmental conditions on ple, while one bacterial phylotype be- the bacterial community in Hydra, we longing to the α-Proteobacteria could cultured polyps, which were taken from be identified as the most dominant spe- the wild, for two months under stand- cies in long term culture, in polyps ard laboratory conditions. Thereafter, from the wild and two month after the we analysed the associated bacteria in shift to the laboratory this bacterium comparison to the bacteria from polyps was present only in relative low abun- taken directly from the wild. Culturing dance (Fraune and Bosch, 2007). Other of polyps from the wild under labora- bacterial species completely disap- tory conditions involves a change in peared from the tissue due to the culture temperature, culture medium change in culturing conditions. Thus,

119

Figure 3: The meta-organism under changing environmental conditions

Hydra is not only associated with spe- given host species. Since this, for cies specific bacteria but also responds example, may provide corals with re- to changes in the environment with sistance against certain pathogens ena- changes in the bacterial community. In bling them to adapt much faster to sum, the holobiont appears to be a dy- novel environmental conditions than by namic system being characterized by mutation and selection, host-microbe functional redundancy and fast adapta- interactions may be considered as sig- tions to altered environmental condi- nificant drivers of animal evolution and tions. diversification. This hypothesis is sup- Based on the holobiont concept, ported by at least three observations: (i) Rosenberg and colleagues in 2007 pro- corals are associated with diverse mi- posed that corals are able to adapt rap- crobiota (Rohwer et al., 2002; Bourne idly to changing environmental condi- et al., 2008); (ii) the associated micro- tions by altering their associated microbiota change in response to environ- biota (Figure 3) (Rosenberg et al., mental stress (Ritchie and Smith, 1995; 2007). Depending on the variety of dif- Pantos et al., 2003) or seasons (Koren ferent niches provided by the host, and Rosenberg, 2006); and (iii) corals which can change with developmental are able to develop resistance against stage, diet or other environmental fac- pathogens although they lack adaptive tors, a more or less diverse microbial immune response (Reshef et al., 2006). community can be established within a

CONCLUDING REMARKS

The beneficial microbiota is a complex sential to the development, protection, and multifunction ecosystem that is es- and overall health of its host. Thus, the

120

microbiota appears to function as an biosis research. Cnidaria not only offer extra organ, to which the host has out- valuable models for exploring the basis sourced numerous crucial metabolic, of interkingdom-communication and nutritional, and protective functions. the role of bacterial signalling in ani- Studies from Cnidaria to primates indi- mal development. Findings derived cate that the host’s role far outweighs from the in vivo context of the Cnidaria other environmental factors in molding models may also provide one of the the composition of the microbiota. simplest possible systems to address Antimicrobial peptides appear to be questions of how a stable host-microbe key factors for host-bacteria co-evolu- community is established and remains tion and the driving force that leads to in balance over time. The uncovered changes in microbiota composition. basic molecular machinery can be Finally, and maybe most important, the transliterated to more complex organ- dynamic relationship between symbi- isms, providing conceptual insights otic microorganisms and environmental into the complexity of host-microbe conditions results in the selection of the interactions. Symbiosis research in most advantageous holobiont. In corals, Cnidaria, therefore, is an emerging changing their microbial partners may field in which scientists from many dis- allow them to adapt to changing envi- ciplines can make fundamental discov- ronmental conditions much more rap- eries and rapidly advance scientific un- idly than via mutation and selection derstanding of a strictly microbe-de- (Figure 3). pendent life style and its evolutionary Taken together, studying host-mi- consequences while combining labora- crobe interactions in basal metazoans is tory and field studies. a challenging and exciting field of sym-

ACKNOWLEDGMENTS

The authors apologize that, because of the selective focus, many interesting investigations and reviews were not included. The authors’ work related to this topic was supported in part by grants from the Deutsche Forschungsgemeinschaft (DFG) and grants from the DFG Cluster of Excellence programs “The Future Ocean” and “Inflammation at Interfaces”.

LITERATURE

Augustin, R., Anton-Erxleben, F., Jungnickel, Bosch, T.C.G.: What Hydra has to say about S., Hemmrich, G., Spudy, B., Podschun, the role and origin of symbiotic interac- R., and Bosch, T.C: Activity of the novel tions. Biol. Bull. 223, 78-84 (2012a) peptide arminin against multiresistant hu- Bosch, T.C.G.: Understanding complex host- man pathogens shows the considerable po- microbe interactions in Hydra. Gut Mi- tential of phylogenetically ancient organ- crobes 3, 345-351 (2012b). isms as drug sources. Antimicrob. Agents Bosch, T.C., Augustin, R., Anton-Erxleben, F., Chemother. 53, 5245-5250 (2009). Fraune, S., Hemmrich, G., Zill, H., Rosen- Bevins, C.L. and Salzman, N.H.: The potter's stiel, P., Jacobs, G., Schreiber, S., Leippe, wheel: the host's role in sculpting its M., Stanisak, M., Grötzinger, J., Jung, S., microbiota. Cell. Mol. Life Sci. 68, 3675- Podschun, R., Bartels, J., Harder, J., and 3685 (2011). Schröder, J.M.: Uncovering the evolution-

121

ary history of innate immunity: the simple larvae use to choose settlement sites. Ann. metazoan Hydra uses epithelial cells for Rev. Mar. Sci. 3, 453-470 (2011). host defence. Dev. Comp. Immunol. 33, Hemmrich, G., Anokhin, B., Zacharias, H., and 559-569 (2009). Bosch, T.C.: Molecular phylogenetics in Bosch, T.C. and McFall-Ngai, M.J.: Meta- Hydra, a classical model in evolutionary organisms as the new frontier. Zoology developmental biology. Mol. Phylogenet. (Jena) 114, 185-190 (2011). Evol. 44, 281-290 (2007). Bourne, D., Iida, Y., Uthicke, S., and Smith- Hemmrich, G., Khalturin, K., Boehm, A.M., Keune, C.: Changes in coral-associated mi- Puchert, M., Anton-Erxleben, F., Wittlieb, crobial communities during a bleaching J., Klostermeier, U.C., Rosenstiel, P., event. ISME J. 2, 350-363 (2008). Oberg, H.H., Domazet-Lošo, T., Frank, U., Leitz, T., and Muller, W.A.: The Sugimoto, T., Niwa, H., and Bosch, hydroid Hydractinia: a versatile, informa- T.C.G.: Molecular signatures of the three tive cnidarian representative. Bioessays 23, stem cell lineages in Hydra and the emer- 963-971 (2001). gence of stem cell function at the base of Franzenburg, S., Fraune, S., Altrock, P.M., multicellularity. Mol. Biol. Evol. 29, 3267- Kunzel, S., Baines, J.F., Traulsen, A., and 3280 (2012) Bosch, T.C.: Bacterial colonization of Hy- Koren, O. and Rosenberg, E.: Bacteria associ- dra hatchlings follows a robust temporal ated with mucus and tissues of the coral pattern. ISME J. 7, 781-790 (2013). Oculina patagonica in summer and winter. Fraune, S., Abe, Y., and Bosch, T.C.: Disturb- Appl. Environ. Microbiol. 72, 5254-5259 ing epithelial homeostasis in the metazoan (2006). Hydra leads to drastic changes in associ- Kortschak, R.D., Samuel, G., Saint, R., and ated microbiota. Environ Microbiol 11, Miller, D.J.: EST analysis of the cnidarian 2361-2369 (2009). Acropora millepora reveals extensive gene Fraune, S., Augustin, R., Anton-Erxleben, F., loss and rapid sequence divergence in the Wittlieb, J., Gelhaus, C., Klimovich, V.B., model invertebrates. Curr Biol. 13, 2190- Samoilovich, M.P., and Bosch, T.C.: In an 2195 (2003). early branching metazoan, bacterial coloni- Leitz, T. and Wagner, T.: The marine bacte- zation of the embryo is controlled by ma- rium Alteromonas espejiana induces meta- ternal antimicrobial peptides. Proc. Natl. morphosis of the hydroid Hydractinia echi- Acad. Sci. USA 107, 18067-18072 (2010). nata. Mar. Biol. 115, 173-178 (1993). Fraune, S., Augustin, R., and Bosch, T.C.: Em- McFall-Ngai, M.: Adaptive immunity: care for bryo protection in contemporary immunol- the community. Nature 445, 153 (2007). ogy: Why bacteria matter. Commun. In- Miller, D.J., Ball, E.E., and Technau, U.: Cni- tegr. Biol. 4, 369-372 (2011). darians and ancestral genetic complexity in Fraune, S. and Bosch, T.C.: Long-term mainte- the animal kingdom. Trends Genet. 21, nance of species-specific bacterial micro- 536-539 (2005). biota in the basal metazoan Hydra. Proc. Möbius, K.A.: Die Auster und die Austern- Natl. Acad. Sci. USA 104, 13146-13151 wirthschaft. Wiegandt, Hempel and Parey, (2007). Berlin, Germany (1877). Fraune, S. and Bosch, T.C.: Why bacteria mat- Muscatine, L. and Lenhoff, H.M.: Symbiosis: ter in animal development and evolution. On the role of algae symbiotic with Hydra. Bioessays 32, 571-580 (2010). Science 142, 956-958 (1963). Grmek, M.D.: Préliminaire d’une étude his- O'Brien, T.L.: Inhibition of vacuolar membrane torique des maladies. Annales ESC 24, fusion by intracellular symbiotic algae in 1437-1483 (1969). Hydra viridis (Florida strain). J, Exp. Zool, Hadfield, M.G.: Biofilms and marine inverte- 223, 211-218 (1982). brate larvae: what bacteria produce that O'Hara, A.M. and Shanahan, F.: The gut flora

122

as a forgotten organ. EMBO Rep. 7, 688- Darwinian framework. Environ. Microbiol. 693 (2006). 11, 2959-2962 (2009). Pantos, O., Cooney, R.P., Le Tissier, M.D., Rosenstiel, P., Philipp, E.E., Schreiber, S., and Barer, M.R., O'Donnell, A.G., and Bythell, Bosch, T.C.: Evolution and function of in- J.C.: The bacterial ecology of a plague-like nate immune receptors--insights from ma- disease affecting the Caribbean coral Mon- rine invertebrates. J. Innate Immun. 1, 291- tastrea annularis. Environ. Microbiol. 5, 300 (2009). 370-382 (2003). Salzman, N.H., Hung, K., Haribhai, D., Chu, Pool, R.R.: The role of algal antigenic determi- H., Karlsson-Sjoberg, J., Amir, E., Teg- nants in the recognition of potential algal gatz, P., Barman, M., Hayward, M., East- symbionts by cells of chlorohydra. J, Cell. wood, D., Stoel, M., Zhou, Y., Sodergren, Sci. 35, 367-379 (1979). E., Weinstock, G.M., Bevins, C.L., Wil- Putnam, N.H., Srivastava, M., Hellsten, U., liams, C.B., and Bos, N.A.: Enteric defen- Dirks, B., Chapman, J., Salamov, A., sins are essential regulators of intestinal Terry, A., Shapiro, H., Lindquist, E., microbial ecology. Nat. Immunol. 11, 76- Kapitonov, V.V., Jurka, J., Genikhovich, 83 (2010). G., Grigoriev, I.V., Lucas, S.M., Steele, Technau, U., Rudd, S., Maxwell, P., Gordon, R.E., Finnerty, J.R., Technau, U., Martin- P.M., Saina, M., Grasso, L.C., Hayward, dale, M.Q., AND Rokhsar, D.S.: Sea D.C., Sensen, C.W., Saint, R., Holstein, anemone genome reveals ancestral T.W., Ball, .EE., and Miller, D.J.: Mainte- eumetazoan gene repertoire and genomic nance of ancestral complexity and non- organization. Science 317, 86-94 (2007). metazoan genes in two basal cnidarians. Reshef, L., Koren, O., Loya, Y., Zilber-Rosen- Trends Genet. 21, 633-639 (2005). berg, I., and Rosenberg, E.: The coral pro- Thorington, G. and Margulis, L.: Hydra viridis: biotic hypothesis. Environ. Microbiol. 8, transfer of metabolites between Hydra and 2068-2073 (2006). symbiotic algae. Biol. Bull. 160, 175-188 Ritchie, K.B.: Regulation of microbial popula- (1981). tions by coral surface mucus and mucus- Walther, M., Ulrich, R., Kroiher, M., and Berk- associated bacteria. Mar. Ecol. Prog. Ser. ing, S.: Metamorphosis and pattern for- 322, 1-14 (2006). mation in Hydractinia echinata, a colonial Ritchie, K.B. and Smith, W.G.: Carbon-source hydroid. Int. J. Dev. Biol. 40, 313-322 utilization of coral-associated marine (1996). heterotrophs. J. Mar. Biotechnol. 3, 107- Wehkamp, J., Salzman, N.H., Porter, E., Nud- 109 (1995). ing, S., Weichenthal, M., Petras, R.E., Rohwer, F., Seguritan, V., Azam, F., and Shen, B., Schaeffeler, E., Schwab, M., Knowlton, N.: Diversity and distribution Linzmeier, R., Feathers, R.W., Chu, H., of coral-associated bacteria. Mar. Ecol. Lima, H. Jr., Fellermann, K., Ganz, T., Prog. Ser. 243, 1-10 (2002). Stange, E.F., and Bevins, C.L.: Reduced Rosenberg, E., Koren, O., Reshef, L., Efrony, Paneth cell alpha-defensins in ileal Crohn's R., and Zilber-Rosenberg, I.: The role of disease. Proc. Natl. Acad. Sci. USA 102, microorganisms in coral health, disease 18129-18134 (2005). and evolution. Nat. Rev. Microbiol. 5, 355- Zilber-Rosenberg, I. and Rosenberg, E.: Role 362 (2007). of microorganisms in the evolution of ani- Rosenberg, E., Sharon, G., and Zilber-Rosen- mals and plants: the hologenome theory of berg, I.: The hologenome theory of evolu- evolution. FEMS Microbiol. Rev. 32, 723- tion contains Lamarckian aspects within a 735 (2008).

123

124

PERSISTING CONSEQUENCE OF INTESTINAL INFECTION: SUMMARY OF THE SEMINAR

MARK RIDDLE1 and RICHARD WALKER2

1Naval Medical Research Center, Silver Spring, MD, USA; 2Enteric Vaccine Initiative, PATH, Washington, DC, USA

INTRODUCTION

Infectious diarrhoea is a global public While the acute consequences health problem with high mortality and would appear to be of global signifi- morbidity, particularly among children cance and drive science and public of the developing world. Each year, health efforts to mitigate the problem, approximately 750,000 children under there is growing evidence linking such five years old die from severe, dehy- infections with a myriad of chronic drating diarrhoea and dysentery world- health consequences including neuro- wide, and millions more are hospital- logical, haematological and rheumato- ized, mostly in low-resource countries logical systems (Lindsay, 1997). The (Liu et al., 2012). In addition, many accounting of these chronic health con- more children suffer from diarrheal dis- sequences beyond that of acute disease ease-associated malnutrition and its needs to be understood and considered adverse consequences on physical and in the global burden of disease assess- cognitive development, which perpetu- ment to inform policy and decision ates the cycle of poverty. making around food safety and sanita- While the brunt of the morbidity tion policy globally, and emphasize the and mortality burden due to these in- reduction of enteric infection among fections are in the developing world, those at high risk (e.g. travellers’ and acute infectious diarrhoea is also a fre- deployed military and children living in quent cause for outpatient visits and resource-poor environments) through hospitalization throughout the devel- primary and possibly secondary pre- oped world and is a significant health vention strategies. problem. For example, Scallan and col- The difficulty of understanding leagues published recently updated es- how such infections may cause chronic timates for foodborne illness in the health problems cannot be overstated, United States (Scallan et al., 2011a, given the range of pathogens (viruses, 2011b). Based on empirical modelling bacteria, parasites), the genetic and ac- of active, passive and outbreak surveil- quired host factors, the often repeated lance data it was estimated that each infections in a person’s life, and the year 31 major pathogens acquired in interactions between complex neuro- the United States caused 9.4 million endocrine, immunological, and micro- episodes of diarrhoeal illness, 55,961 biological systems, many which are not hospitalizations, and 1,351 deaths. In well understood. But what is clear is addition, it is estimated that unspecified that acute infections can colonize, in- agents, resulting in 71,878 hospitaliza- vade and exert their effects locally and tions and 1,686 deaths, caused approxi- systemically at the individual level and mately 38.4 million episodes of domes- chronic pathological changes to these tically acquired foodborne illnesses. organ systems have been noted. Thus,

125

Figure 1: Persisting consequences of intestinal infection in developing (oval on right) and/or developed world (oval on left). GI – gastrointestinal; HUS – haemolytic uremic syndrome; GBS – Guillain Barré Syndrome.

the aim of this conference was to assess gaps in knowledge, and describe future our position in terms of epidemiologi- directions related to the challenge of cal and pathological-aetiological under- persisting consequences of intestinal standing of the phenomenon, identify infection.

EPIDEMIOLOGY: DEFINING THE PROBLEM

Epidemiological research is fundamen- monly used to describe complex rela- tal and complementary to our under- tionships and their epidemiology (Bird, standing of disease and development of 2011). Hill’s criteria include strength of primary, secondary and tertiary inter- association, consistency of effect, spec- ventions. To put the current evidence in ificity of effect, temporality, biological context, epidemiological research can gradient or dose response, and biologi- identify knowledge gaps and define re- cal plausibility to form the basis of an search priorities for increasing under- argument for causation and have been standing in the areas of disease attribu- used successfully to establish the tion, burden of disease, clinical charac- pathogenic role of H. pylori, HIV and terization and management. A number toxins (Szklo and Nieto, 2007). of frameworks designed to elucidate Emerging from the literature is an the epidemiologic determination of understanding that there are persistent causation have been advanced over the consequences which are both common years (Parascandola, 2011). Koch’s and unique to populations when strati- original postulates proved effective at fied by geo-economic strata. In Figure establishing disease-pathogen relation- 1 the overlapping ovals shows those ships but fall short with more complex issues that are primarily problems of associations (Evans, 1976; Marshall et the developed world (oval on left), al., 1985). In recent years, Bradford while individuals in resource-poor set- Hill’s criteria have been more com- tings suffer more from the problems

126

shown on the right. In the overlapping and disease processes which may have area is shown those problems currently different manifestations in a different understood to be more common to both human-environmental setting. Be that populations. While such a divide is as it may, there is an emergence of evi- convenient and may be due to differ- dence which describes disparate per- ences such as genetics, diet, and other sisting consequences of acute enteric unique environmental influences, it infections among those who have ac- may also artificially de-emphasize pos- cess to clean food water and sanitation, sible similar pathogenic mechanisms and those who do not.

POST-INFECTIOUS CONSEQUENCES: A DEVELOPED WORLD PERSPECTIVE

Despite food safety regulations, public controlling for unmeasured factors and education and advanced agricultural generally are only able to account for a systems, occurrence of foodborne ill- fraction of explainable risk (though ness and other enteric infections in the there are notable exceptions where the developed world are not infrequent. In attributable fraction is stronger such as this seminar, Porter and colleagues in the case of some strains of Campylo- (page 41) describe the utilization, chal- bacter associated with GBS), and mis- lenges and opportunities of the US De- classification of both exposures and partment of Defense (DoD) medical outcomes generally will bias associa- encounter databases and serum reposi- tive estimates towards the null. Rela- tory in understanding of the problem of tively rare events are also harder to chronic consequences of enteric dis- study in rigorous cohort study designs eases. Historically, well-designed epi- without large and expensive studies. demiologic studies have been at the Despite these limitations, continued fore of linking exposures with out- epidemiological studies from a diver- comes and have enabled estimates of sity of populations and designs are outcome risk, identification of host- needed to validate initial findings as and pathogen-specific risk factors, and well as explore new associations be- understanding the timing from expo- tween acute enteric infections and the sure to outcome. Studies of the post- growing spectrum of illnesses. infectious sequelae of enteric diseases An example of one such chronic are no different as highlighted by sev- health illness, coeliac disease, is an eral systematic reviews on the epide- illustrative example of the complexity miologic evidence where functional in which the immune system, the envi- gastrointestinal disorders appear to be ronment, and infection may interact to most substantiated in addition to reac- cause disease. In this seminar Murray tive arthritis (Halvorson et al., 2006; (page 71) describes some unique obser- Thabane et al., 2007; Deising et al., vations on the pathogenesis and trig- 2013; Pike et al., 2013; Poropatich et gers of coeliac disease (CD), an in- al., 2010; Porter et al., 2013). creasingly common chronic disease While these studies have enhanced affecting primarily the upper small in- our appreciation of the acute/chronic testine associated with significant mor- disease link, they have certain limita- bidity and mortality in much of the de- tions. Epidemiological studies do veloped world. While there is certainly poorly in informing disease mecha- a genetic predisposition to coeliac dis- nisms, are fraught with challenges in ease which is required, not everyone

127

with this predisposition will develop irritable bowel syndrome (IBS)-like disease, and the onset of coeliac dis- symptoms after a confirmed episode of ease can occur at any time in some- Campylobacter jejuni enteritis was sub- one’s life suggesting an environmental sequently diagnosed with new onset triggering event. It appears that infec- CD (Verdu et al., 2007). Studies among tion-colonization events in early life the previously described DoD medical may have an impact on risk given the encounter system database have also findings of higher risk of CD in indi- recently been reported which support viduals born via caesarean section association between bacterial acute en- (Decker et al., 2010; Marild et al., teric infection (in particular Campylo- 2012), and weaning in the winter bacter) and the onset of CD. (Riddle et months (Ivarsson et al., 2003). These al., 2012). One explanatory hypothesis events can alter the nature of the micro- which needs to be confirmed is that biota. It has been clearly shown that Campylobacter jejuni, which has been microbiota are important for immuno- shown experimentally to permit the logical maturation of the host (Lanning translocation of normal non-invasive and Knight, 2005; Sjogren et al., 2009; microflora (Chen et al., 2006; Chung et al., 2012), but how the early Kalischuk et al., 2009; Kalischuk and constitution of an individual’s microbi- Buret, 2010), could trigger aberrant im- ota can have affects such as susceptibil- mune responses/loss of tolerance to co- ity to the loss of tolerance a grain pro- transported luminal antigens, including tein remains a mystery and opportunity gluten peptides, across the intestinal to study. barrier. In certain susceptible individu- Emerging evidence is also suggest- als primed towards a mucosal immune ing that CD may be triggered after an response towards such antigens this acute enteric infection in some indi- could result in loss of tolerance to these viduals. Anecdotal reports and case se- antigens due to an inappropriate in- ries suggesting an association have flammatory response (Jabri et al., been described (Landzberg and 2005; Jabri and Sollid, 2009). As Connor, 2005; Ginsburg and Bayless, demonstrated in animal models of glu- 2008; Chae et al., 2010). It has also ten sensitivity (Verdu et al., 2008; been described that exposure to three or Natividad et al., 2009), gastrointestinal more infectious gastroenteritis events infection may trigger or facilitate the in young children at or around the time onset of clinical CD, either by increas- of introduction of follow-on formula ing intestinal permeability or enhancing was associated with a substantial in- uptake and dysfunctional anti-gliadin creased risk of childhood diagnosis of immune response in the genetically coeliac disease (Falth-Magnusson et susceptible host (DeMeo et al., 2002; al., 1996). More recently, a case of a Fasano and Shea-Donohue, 2005). healthy subject who developed sudden

POST-INFECTIOUS CONSEQUENCES: A DEVELOPING WORLD PERSPECTIVE

Human enteric infection in the devel- (multiple repeated exposures), variety oping world is different from that of of pathogens (more diverse), nutritional the developed world on a number of status of the host, as well as a number features including the age of onset of of other factors including co-infection, infection (earlier), force of infection diet and genetics. As such, approaching

128

Figure 2. Framework to understand the patho-aetiology of persisting consequences of acute enteric infections.

an understanding of the link between new insights on the often described “vi- infections and chronic illnesses in this cious cycle” of enteric infections lead- population is challenged as the “one- ing to enteropathy leading to malnutri- infection” leads to “one outcome” ap- tion and effects on growth, cognition proach is difficult to tease out. A and immunity, which themselves con- unique study entitled the “Malnutrition tribute to increased susceptibility to and Enteric Infections network” (“Mal- enteric infections (Guerrant et al., ED”) exploring the aetiology, risk fac- 2008; Moore et al., 2010). The results tors and interactions of enteric infec- presented by McCormick and col- tions on child health was presented leagues (page 23) expand on this con- with a unique design constructed for ceptual model but also challenge our this environment (Malnutrition & understanding of the significance of the Enteric Infections Network MAL-ED: repeated asymptomatic episodes of in- The Interactions of Malnutrition & fections with disease-causing potential Enteric Infections: Consequences for on growth and nutrition outcomes. It Child Health and Development. Project would appear that infection with patho- website (accessed Dec 10, 2013) gens which may or may not be associ- http://mal-ed.fnih.org/). While results ated with symptoms is the driving fac- presented were preliminary and did not tor for the chronic consequences in include cognitive and nutritional met- these populations. rics, the results are shedding light on

129

Expanding on the findings of ability to respond to additional infec- MAL-ED, Petri and colleagues (page tions, and evidence also suggests that 61) described how recurrent infections, such a condition may negate the valua- and in particular some combinations of ble contributions of some orally-admin- protozoal and bacterial infections, can istered vaccines through evidence of result in a subclinical condition of envi- hyporesponsiveness (see: Old Herborn ronmental enteropathy (EE) character- University Seminar Monograph 24; ized by disruption of intestinal archi- Development of strategies to overcome tecture and chronic inflammation. Such barriers to effective mucosal immun- changes would appear to have direct ization of infants in developing coun- effects on nutrition, the individual’s tries).

PATHOGENIC MECHANISMS: A GROWING FRAMEWORK OF COMPLEXITY

While epidemiological studies are in- harmful bacteria at bay. Underlying creasing our confidence that there are this extrinsic barrier is what is often real associations between enteric infec- referred to as the intrinsic barrier which tions and chronic health effects, these is made up of a layer of host cells. data are limiting and only tell us that While the figure is a simplification, this there is a problem and, given the mag- layer is not homogenous and includes a nitude, ought to do something about it. number of cell types which are derived It is critical is to understand what is oc- from stem cells and include absorptive curring in order to explain these ob- enterocytes, mucus-secreting goblet served associations. Based on presenta- cells, entero-endocrine cells and Paneth tions and discussions at this Old Her- cells. This cell layer is highly dynamic born University Seminar, a framework and includes complete turnover every has emerged by which to understand few days and relies on a delicate bal- the problem (Figure 2). At the gut sur- ance between cell proliferation and cell face there is an increasing density and death. This cell layer has a diverse set complexity gradient of microbial flora of functions which include production (both bacteria and viruses) that are of antibacterial peptides, sampling of found in the lumen of the GI tract in luminal contents, and absorbing im- the oral to aboral direction. Within the portant nutrients and fluid into the host lumen are also nutrients (taken in by circulatory system. This layer is at an the individual and produced by the interface between the luminal extrinsic flora), chemicals, antibiotics and non- barrier and the host innate and adaptive digestible matter which shape and con- immune system which largely takes in tribute in part to the diversity of this the sub-epithelial level where neuro- microbial milieu. There is also a micro- endocrine, circulatory and cellular and biome gradient which exists from the humoral immune cells systems inter- lumen to the surface of the enterocyte face. While the basic systems involved where there is a decreasing density as are delineated in such simple terms, you get closer to the cell surface. It is each of these systems are complex, and within this interface, often referred to it is likely that it is the interaction of as the extrinsic barrier which acts as a each of these systems which needs to two-way filter, where many important be understood - like a puzzle which is processes take place including allowing made up of hundreds of other puzzles. nutrients to be absorbed and keeping Several presenters revealed insight

130

into pathogenic mechanisms which have a disrupting effect on the balance have advanced our understanding of a within the gut epithelial barrier and mi- piece of the puzzle. Buret and col- crobiome. This disturbance in suscepti- leagues (page 87) reported on recent ble individuals can result in abnormal findings from an animal model sug- immune activation which may further gesting that enteropathogen-induced alter the balance of the microbiome in disruptions of the commensal microbi- the gut (dysbiosis). This dysbiosis may ota have a part play in triggering the have direct effects on the neuroendo- sequence of events that result in vari- crine system with central effects on the ous intestinal and extra-intestinal brain (perception/pain), as well as local chronic health outcomes. Data from effects on the enteric system (e.g. mo- animal models of campylobacteriosis tility). While studies have shown that and giardiasis suggest that infection by risk for functional disorders are asso- enteric pathogens may promote the ciated with psychiatric co-morbidities, transcellular and paracellular transloca- the opposite has been described in ani- tion of non-invasive commensal bacte- mal models where microbiome changes ria, as well as disruptions of the com- can change behaviour and brain neuro- position and structure of microbiota chemistry (Bercik et al., 2011). Finally, biofilms. Such actions then lead to the it must not be forgotten that stress itself activation of the host autoimmune reac- can have an impact on the gut epithelial tions that are implicated in the produc- barrier and increase risk of infections tion of post-infectious complications. including enteric infections, which in Beyond the well-established need to combination with stress appear to in- better understand host-pathogen cross- crease the risk of functional disorders talks, as well as interactions between (Lyte et al., 2011; O'Malley et al., the host and its microbiota, these obser- 2011). Thus, in our framework of un- vations lay the foundations for future derstanding, it is important to consider research into enteropathogen-microbi- the impact of the important two way ota interactions and highlight that the cross-talk between the brain and the gut mechanism of intestinal dysbiosis (a (see: Old Herborn University Seminar process whereby an enteric pathogen Monograph 26; The gut microbiome disrupts the normal commensal flora and the nervous system), and how these such that an imbalance among micro- interact with the microbiome to result bial populations occur on a body sur- in dysbiosis and disease associated face, often with deleterious effects on with nervous system disturbances. health) is a central piece of the puzzle Finally, Chang and colleagues to solve. The availability of animal (page 13) described evidence from a models like these is quite useful to help Campylobacter model in rats whereby better understand gut homeostasis. infection, potentially through molecular Building on this theme (and adding mimicry, can have an effect on gastric an additional layer of complexity), Col- motility, which in itself may lead to a lins (page 107) describes the phenom- type of dysbiosis where the GI sys- enon of acute enteric infections and tem’s ability to maintain healthy levels functional gastrointestinal disorders as of bacteria in the small intestine can be an important model to understand the disrupted. The findings of this model relationship between the gut-brain axis, suggest that there may be multiple the microbiome and acute infection. In mechanisms by where a single patho- simple terms, infection, stress and antigen my cause a chronic health conse- biotics (alone or in combination) can quence.

131

THE FUTURE: UNDERSTANDING COMPLEXITY AND EXPLORING NEW FRONTIERS

As described, each of the systems that genic organisms affect the holobiome may influence the determination of per- is an important gap. In addition, with sistent infectious consequences is inde- the increasing number of chronic health pendently complex, and together the problems in which epidemiological as- complexity is compounded. Clearly, sociations are being found, we are chal- better tools and model systems are lenged by understanding how big the needed to piece together the puzzles. problem is or may well be. Microbiome One such tool which may have poten- changes are now being associated with tial is in the area of intestinal stem cell obesity and liver disease in developed research described by van der Flier and world populations (Henao-Mejia et al., colleagues (page 1). The stem cell 2013; Karlsson et al., 2013; Zhao, intestinal epithelium model represents a 2013). Natural questions which follow unique opportunity to study adult stem include what are the factors behind cell biology and lineage specification. these changes and could infection be The combination of a rapid self-renew- one of them? Epidemiological studies ing tissue, evident compartmentaliza- and perhaps animal models could be tion of proliferating and differentiated directed at looking for such associa- cell types and a relatively simple, re- tions. A study of certain value would petitive tissue architecture, is ideal for be to conduct a developed world post- the visualization and identification of infectious microbiome cohort studies stem cell types, cell fate specification among populations at high risk for and cellular behaviour. In the past, ge- acute enteric infection (e.g. travellers). netic studies in mice have created a Evaluating baseline microbiome as wealth of new insights on the biology well as changes that may occur with of the intestinal epithelium. The re- travel, travel-related infections and an- cently established long term culture tibiotics, and the persistence of such conditions of intestinal epithelium, es- changes which may be linked to well pecially mini-organs from human defined post-infectious functional gas- origin, may boost the research field for trointestinal disorders may offer new the next generation by providing a vari- insights. ety of possibilities for research and Finally, we need a framework therapeutic applications of intestinal which approaches the problem not only biology. The system in its current con- by considering the contributions of struct lacks the enteric neuroendocrine pathogens, or the host or the commen- and microbiome integration, but addi- sal flora, but rather as an integrated tions of such components may be feasi- system. The concept and importance of ble and add to the utility. the holobiont has emerged over the last New data challenge the way we un- two decades. Bacterial cells outnumber derstand the relationship between in- human cells by a factor of 10 to 1, and fection and illness. It is apparent that the collective genes of these bacteria even though one can recover a patho- outnumber our genes by 150 to 1 (Qin gen from an individual it is often noted et al., 2010). We know that these bacte- that that pathogen is not causing ill- ria are critical to our survival and in- ness, which begs the question of is it a clude important functions such as food pathogen? Clearly understanding how break down, biotransformation of nutri- asymptomatic infections with patho- ents, development and function of a

132

Box 1: Remaining gaps in our understanding and questions for future directions.

• How does the viriome interact and what of the effects of viral enteric infection on the viriome and bacteriome and vice versa? • How do antibiotic challenges to the system interact? • What is the relationship of enteropathy with nutrition? Can complex nutrient effects be isolated and rebalanced through nutrition? • What is the impact of environmental chemicals on the microbiome and host factors associated with disease? • How do neuro-immune interactions work? • What is the nature of inflammation in various disease models? Why is inflammation sometimes healthful and sometimes deleterious? • How do we facilitate inflammation resolution (not stopping, but helping resolution)? • What could generational/adoptive studies teach us about the association between acute enteric infection and chronic health consequences? • What are the short-term opportunities to treat these diseases? For example what role might faecal microbiota transplant play? Could well-characterized synthetic faecal microbiome products be useful? What might the value of microbiome feeding and topical anti-inflammatories be? • Can continued epidemiological studies from a diversity of populations and designs validate initial findings as well as explore new associations between acute enteric infections and the growing spectrum of illnesses? • How does Campylobacter jejuni trigger aberrant immune responses/loss of tolerance to co-transported luminal antigens across the intestinal barrier? • How does the early constitution of an individual’s microbiota have affects such as susceptibility to the loss of tolerance a grain protein? • How do enteropathogen-microbiota interactions trigger intestinal dysbiosis (a process whereby an enteric pathogen disrupts the normal commensal flora such that an imbalance among microbial populations occur on a body surface, often with deleterious effects on health)? • Could microbial toxins from enteric pathogens like ETEC and Campylobacter contribute to microbiome disruption and its consequent effects?

normal immune system, angiogenesis, system in Hydra, where the microbiota and regulation of fat accumulation is a complex and multifunctional eco- (Singh et al., 2013). In addition to these system that is essential to the develop- functions, the microbiota compete for ment, protection, and overall health of space with pathogens and can exert its host. Furthermore the dynamic rela- other local effects to protect the host tionship between symbiotic microor- from infections. The microbiota is also ganisms and environmental conditions adaptable which provides an advantage results in the selection of the most ad- to the host which may need to rely on vantageous holobiont. Systems like the flexibility of the microbiota struc- these may provide tools to further study ture to adapt to changes of diet or cli- and explore and potentially lead to in- mate or infection. Bosch and col- sights to prevent the unintended conse- leagues (page 113) describe an elegant quences of holobiome disturbances.

133

CONCEPTS AND QUESTIONS FOR FURTHER CONSIDERATION

Continued epidemiological studies standing how asymptomatic infection from a diversity of populations and de- with pathogenic organisms affects the signs are needed to validate initial find- holobiome is an important gap. How ings as well as explore new associa- does the viriome interact and what of tions between acute enteric infections the effects of viral enteric infection on and the growing spectrum of illnesses. the viriome and bacteriome and vice Studies are also needed to better under- versa? Non-infectious events can also stand the mechanism by which an contribute to the problem of dysbiosis, enteropathogen such as Campylobacter as stress and antibiotics (alone or in jejuni which induces disruptions of the combination) can have a disrupting ef- commensal microbiota can affect the fect on the balance within the gut epi- sequence of events that result in vari- thelial barrier and microbiome ous intestinal and extra-intestinal Clearly better tools and model sys- chronic health outcomes. Future re- tems are needed to help piece together search into enteropathogen-microbiota the puzzles with which we are con- interactions should highlight dysbiosis fronted (Box 1). Although these ques- as a central piece of the puzzle to solve. tions are challenging, the benefits to There may be multiple mechanisms by human life which can be achieved which a single pathogen my cause a make the undertaking well worth the chronic health consequence. Under- effort.

LITERATURE

Bercik, P., Denou, E., Collins, J., Jackson, W., Villablanca, E.J., Wang, S., Mora, J.R., Lu, J., Jury, J., Deng, Y., Blennerhassett, Umesaki, Y., Mathis, D., Benoist, C., Rel- P., Macri, J., McCoy, K.D., Verdu, E.F., man, D.A., and Kasper, D.L.: Gut immune and Collins, S.M.: The intestinal microbi- maturation depends on colonization with a ota affect central levels of brain-derived host-specific microbiota. Cell 149, 1578- neurotropic factor and behavior in mice. 1593 (2012). Gastroenterology 141, 599-609 (2011). Decker, E., Engelmann, G., Findeisen, A., Bird, A.: The epistemological function of Hill's Gerner, P., Laass, M., Ney, D., Posovszky, criteria. Prev. Med. 53, 242-245 (2011). C., Hoy, L., and Hornef, M.W.: Cesarean Chae, S.C., Yu, J.L., Oh, G.J., Choi, C.S., delivery is associated with celiac disease Choi, S.C., Yang, Y.S., and Yun, K.J.: but not inflammatory bowel disease in Identification of single nucleotide children. Pediatrics 125, e1433-e1440 polymorphisms in the TNFRSF17 gene (2010). and their association with gastrointestinal Deising, A., Gutierrez, R.L., Porter, C.K., and disorders. Mol. Cells 29, 21-28 (2010). Riddle, M.S.: Postinfectious functional Chen, M.L., Ge, Z., Fox, J.G., and Schauer, gastrointestinal disorders: a focus on D.B.: Disruption of tight junctions and epidemiology and research agendas. induction of proinflammatory cytokine Gastroenterol. Hepatol. (NY) 9, 145-157 responses in colonic epithelial cells by (2013). Campylobacter jejuni. Infect. Immun. 74, DeMeo, M.T., Mutlu, E.A., Keshavarzian, A., 6581-6589 (2006). and Tobin, M.C.: Intestinal permeation and Chung, H., Pamp, S.J., Hill, J.A., Surana, N.K., gastrointestinal disease. J. Clin. Gastro- Edelman, S.M., Troy, E.B., Reading, N.C., enterol. 34, 385-396 (2002).

134

Evans, A.S.: Causation and disease: the Henle- Kalischuk, L.D. and Buret, A.G.: A role for Koch postulates revisited. Yale J. Biol. Campylobacter jejuni-induced enteritis in Med. 49, 175-195 (1976). inflammatory bowel disease? Am. J. Falth-Magnusson, K., Franzen, L., Jansson, G., Physiol. Gastrointest. Liver Physiol. 298, Laurin, P., and Stenhammar, L.: Infant G1-G9 (2010). feeding history shows distinct differences Karlsson, F., Tremaroli, V., Nielsen, J., and between Swedish celiac and reference Backhed, F.: Assessing the human gut children. Pediatr. Allergy Immunol. 7, 1-5 microbiota in metabolic diseases. Diabetes (1996). 62, 3341-3349 (2013). Fasano, A. and Shea-Donohue, T.: Mechanisms Landzberg, B.R. and Connor, B.A.: Persistent of disease: the role of intestinal barrier diarrhea in the returning traveler: think function in the pathogenesis of beyond persistent infection. Scand. J. gastrointestinal autoimmune diseases. Nat. Gastroenterol. 40, 112-114 (2005). Clin. Pract. Gastroenterol. Hepatol. 2, 416- Lanning, D.K. and Knight, K.L.: Intestinal 422 (2005). bacteria and development of the antibody Ginsburg, P.M. and Bayless, T.M.: How can repertoire. Discov. Med. 5, 393-398 IBD be distinguished from IBS? Inflamm. (2005). Bowel Dis. 14, Suppl. 2, S152-S154 Lindsay, J.A.: Chronic sequelae of foodborne (2008). disease. Emerg. Infect. Dis. 3, 443-452 Guerrant, R.L., Oria, R.B., Moore, S.R., Oria, (1997). M.O., and Lima, A.A.: Malnutrition as an Liu, L., Johnson, H.L., Cousens, S., Perin, J., enteric infectious disease with long-term Scott, S., Lawn, J.E., Rudan, I., Campbell, effects on child development. Nutr. Rev. H., Cibulskis, R., Li, M., Mathers, C., 66, 487-505 (2008). Black, R.E.; Child Health Epidemiology Halvorson, H.A., Schlett, C.D., and Riddle, Reference Group of WHO and UNICEF: M.S.: Postinfectious irritable bowel Global, regional, and national causes of syndrome--a meta-analysis. Am. J. child mortality: an updated systematic Gastroenterol. 101, 1894-1899 (2006). analysis for 2010 with time trends since Henao-Mejia, J., Elinav, E., Thaiss, C.A., 2000. Lancet 379, 2151-2161 (2012). Licona-Limon, P., and Flavell, R.A.: Role Lyte, M., Vulchanova, L., and Brown, D.R.: of the intestinal microbiome in liver Stress at the intestinal surface: catechol- disease. J. Autoimmun. 46, 66-73 (2013). amines and mucosa-bacteria interactions. Ivarsson, A., Hernell, O., Nystrom, L., and Cell Tissue Res. 343, 23-32 (2011). Persson, L.A.: Children born in the Marild, K., Stephansson, O., Montgomery, S., summer have increased risk for coeliac Murray, J.A., and Ludvigsson, J.F.: disease. J. Epidemiol. Community Health Pregnancy outcome and risk of celiac 57, 36-39 (2003). disease in offspring: a nationwide case- Jabri, B., Kasarda, D.D., and Green, P.H.: control study. Gastroenterology 142, 39-45 Innate and adaptive immunity: the yin and e33 (2012). yang of celiac disease. Immunol. Rev. 206, Marshall, B.J., Armstrong, J.A., McGechie, 219-231 (2005). D.B., and Glancy, R.J.: Attempt to fulfil Jabri, B. and Sollid, L.M.: Tissue-mediated Koch's postulates for pyloric Campylo- control of immunopathology in coeliac bacter. Med. J. Aust. 142, 436-439 (1985). disease. Nat. Rev. Immunol. 9, 858-870 Moore, S.R., Lima, N.L., Soares, A.M., Oria, (2009). R.B., Pinkerton, R.C., Barrett, L.J., Kalischuk, L.D., Inglis, G.D., and Buret, A.G.: Guerrant, R.L., and Lima, A.A.: Prolonged Campylobacter jejuni induces transcellular episodes of acute diarrhea reduce growth translocation of commensal bacteria via and increase risk of persistent diarrhea in lipid rafts. Gut Pathog. 1, 2 (2009). children. Gastroenterology 139, 1156-1164

135

(2010). sequencing. Nature 464, 59-65 (2010). Natividad, J.M., Huang, X., Slack, E., Jury, J., Riddle, M.S., Murray, J.A., and Porter, C.K.: Sanz, Y., David, C., Denou, E., Yang, P., The incidence and risk of celiac disease in Murray, J., McCoy, K.D., and Verdú, E.F.: a healthy US adult population. Am. J. Host responses to intestinal microbial Gastroenterol. 107, 1248-1255 ( 2012). antigens in gluten-sensitive mice. PLoS Scallan, E., Griffin, P.M., Angulo, F.J., Tauxe, One 4, e6472 (2009). R.V., and Hoekstra, R.M.: Foodborne O'Malley, D., Quigley, E.M., Dinan, T.G., and illness acquired in the United States-- Cryan, J.F.: Do interactions between stress unspecified agents. Emerg. Infect. Dis. 17, and immune responses lead to symptom 16-22 (2011a). exacerbations in irritable bowel syndrome? Scallan, E., Hoekstra, R.M., Angulo, F.J., Brain Behav. Immun. 25, 1333-1341 Tauxe, R.V., Widdowson, M.A., Roy, (2011). S.L., Jones, J.L., and Griffin, P.M.: Parascandola, M.: Causes, risks, and Foodborne illness acquired in the United probabilities: probabilistic concepts of States--major pathogens. Emerg. Infect. causation in chronic disease epidemiology. Dis. 17, 7-15 (2011b). Prev. Med. 53, 232-234 (2011). Singh, Y., Ahmad, J., Musarrat, J., Ehtesham, Pike, B.L., Porter, C.K., Sorrell, T.J., and N.Z., and Hasnain, S.E.: Emerging Riddle, M.S.: Acute Gastroenteritis and the importance of holobionts in evolution and Risk of Functional Dyspepsia: A in probiotics. Gut Pathog. 5, 12 (2013). Systematic Review and Meta-Analysis. Sjögren, Y.M., Tomicic, S., Lundberg, A., Am. J. Gastroenterol. 108, 1558-1563 Fagerås-Bottcher, M., Björkstén, B., (2013). Sverremark-Ekström, E., and Jenmalm, Poropatich, K.O., Walker, C.L., and Black, M.C.: Influence of early gut microbiota on R.E.: Quantifying the association between the maturation of childhood mucosal and Campylobacter infection and Guillain- systemic immune responses. Clin. Exp. Barré syndrome: a systematic review. J. Allergy 39, 1842-1851 (2009). Health Popul. Nutr. 28, 545-552 (2010). Szklo, M. and Nieto, F.J.: Epidemiology: Porter, C.K., Thura, N., and Riddle, M.S.: beyond the basics, 2nd edition. Jones and Quantifying the incidence and burden of Bartlett Publishers, Sudbury, MA, USA postinfectious enteric sequelae. Mil. Med. (2007).. 178, 452-469 (2013). Thabane, M., Kottachchi, D.T., and Marshall, Qin, J., Li, R., Raes, J., Arumugam, M., J.K.: Systematic review and meta-analysis: Burgdorf, K.S., Manichanh, C., Nielsen, the incidence and prognosis of post- T., Pons, N., Levenez, F., Yamada, T., infectious irritable bowel syndrome. Mende, D.R., Li, J., Xu, J., Li, S., Li, D., Aliment. Pharmacol. Ther. 26, 535-544 Cao, J., Wang, B., Liang, H., Zheng, H., (2007). Xie, Y., Tap, J., Lepage, P., Bertalan, M., Verdu, E.F., Huang, X., Natividad, J., Lu, J., Batto, J.-M., Hansen, T., Le Paslier, D., Blennerhassett, P.A., David, C.S., McKay, Linneberg, A., Nielsen, H.B., Pelletier, E., D.M., and Murray, J.A.: Gliadin- Renault, P., Sicheritz-Ponten, T., Turner, dependent neuromuscular and epithelial K., Zhu, H., Yu, C., Li, S., Jian, M., Zhou, secretory responses in gluten-sensitive Y., Li, Y., Zhang, X., Li, S., Qin, N., HLA-DQ8 transgenic mice. Am. J. Yang, H., Wang, J., Brunak, S., Doré, J., Physiol. Gastrointest. Liver Physiol. 294, Guarner, F., Kristiansen, K., Pedersen, O., G217-225 (2008). Parkhill, J., Weissenbach, J., MetaHIT Verdu, E.F., Mauro, M., Bourgeois, J., and Consortium, Bork, P., Ehrlich, S.D., and Armstrong, D.: Clinical onset of celiac Wang, J.: A human gut microbial gene disease after an episode of Campylobacter catalog established by metagenomic jejuni enteritis. Can. J. Gastroenterol. 21,

136

453-455 (2007). correlation to causality. Nat. Rev. Zhao, L.: The gut microbiota and obesity: from Microbiol. 11, 639-647 (2013).

137